summaryrefslogtreecommitdiff
path: root/numeric.c
AgeCommit message (Collapse)Author
2006-09-24Minor rdoc fixryan
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/branches/ruby_1_8@11013 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2006-09-07* numeric.c (fix_plus): addition in Fixnum will never overflowmatz
long. a patch from Ondrej Bilka <neleai at seznam.cz>. [ruby-core:08794] * numeric.c (fix_minus): ditto. * bignum.c (rb_big_pow): eagerly truncate resulting bignum. [ruby-core:08794] git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/branches/ruby_1_8@10880 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2006-09-04* numeric.c (flo_hash): improve collision. fixed: [ruby-dev:29352]nobu
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/branches/ruby_1_8@10864 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2006-09-01* common.mk, configure.in, defines.h, eval.c, gc.c, main.c,akr
numeric.c, ruby.h, ia64.s: backport IA64 HP-UX support. git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/branches/ruby_1_8@10827 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2006-05-01* numeric.c (num_div): use floor rather than rb_Integer().matz
[ruby-dev:28589] * numeric.c (flo_divmod): the first element of Float#divmod should be an integer. [ruby-dev:28589] * test/ruby/test_float.rb: add tests for divmod, div, modulo and remainder. * util.c (ruby_strtod): fixed wrong conversion. git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/branches/ruby_1_8@10123 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2006-02-05* numeric.c (fix_to_s): removed workaround for radix 2. Historically,ocean
rb_fix2str can only handle radix 8, 10, 16. (Rev1.37) But for now, it can handle radix 2..36. [ruby-Bugs#3438] git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/branches/ruby_1_8@9893 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2006-02-04* numeric.c (fix_to_s): (2**32).to_s(2) fails with exception whereocean
sizeof(int) == 4 < sizeof(long). [ruby-core:7300] I think the function name of rb_int2big is quite misleading. This should be "rb_long2big". git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/branches/ruby_1_8@9882 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2005-10-30* ext/readline/readline.c (readline_readline): type check.matz
[ruby-core:6089] * numeric.c (fix_rshift): RDoc fix. [ruby-core:6351] * util.h (strtod): add #undef for platforms defines strtod() macro. [ruby-dev:27563] git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/branches/ruby_1_8@9483 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2005-09-28* class.c (rb_mod_init_copy, rb_class_init_copy), file.c (rb_stat_init_copy),nobu
numeric.c (num_init_copy), object.c (rb_obj_init_copy, Init_Object), re.c (match_init_copy, rb_reg_init_copy), time.c (time_init_copy): undocumented. git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/branches/ruby_1_8@9332 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2005-07-22* bignum.c (rb_big_eq): reduce isnan(). [ruby-dev:26600]matz
* numeric.c (flo_eq, flo_gt, flo_ge, flo_lt, flo_le): ditto. git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/branches/ruby_1_8@8823 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2005-05-24* numeric.c (fix_pow): support Fixnum ** Float case directlymatz
without coercing. [ruby-talk:142697] [ruby-talk:143054] * ruby.c (require_libraries): caused SEGV when continuation jumped in to the required library code. git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/branches/ruby_1_8@8515 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2005-05-10* numeric.c (flo_plus): small typo fix. [ruby-core:4636]ocean
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/branches/ruby_1_8@8417 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2005-02-28* exception error messages updated. [ruby-core:04497]matz
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/branches/ruby_1_8@8035 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2005-01-11* numeric.c (Init_Numeric): turn off floating point exceptionsocean
on bcc32. "1e300".to_f had crashed by overflow. git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/branches/ruby_1_8@7765 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2004-11-18* process.c (proc_getpgrp): prohibit for $SAFE=2.matz
[ruby-dev:24899] * process.c (get_pid): ditto. [ruby-dev:24904] * process.c (get_ppid): ditto. * array.c (rb_ary_delete): defer rb_ary_modify() until actual modification. [ruby-dev:24901] * parse.y (newline_node): should not use FL_SET. [ruby-dev:24874] * parse.y (string_content): should not use FL_UNSET. * node.h (NODE_NEWLINE): remove unused bit to utilize flag field in nodes. * string.c (rb_str_splice): move rb_str_modify() after StringValue(), which may alter the receiver. [ruby-dev:24878] git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/branches/ruby_1_8@7307 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2004-11-16* numeric.c (flo_divmod): protect float values from GC bymatz
assignment to local variables. [ruby-dev:24873] * string.c (str_mod_check): frozen check should be separated. [ruby-core:3742] * array.c (rb_ary_update): pedantic check to detect rb_ary_to_ary() to modify the receiver. [ruby-dev:24861] * string.c (rb_str_justify): typo fixed. [ruby-dev:24851] git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/branches/ruby_1_8@7290 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2004-10-27* string.c (RESIZE_CAPA): check string attribute before modifyingmatz
capacity member of string structure. [ruby-dev:24594] * ext/zlib/zlib.c (gzreader_gets): use memchr() to to gain performance. [ruby-talk:117701] * sprintf.c (rb_f_sprintf): raise ArgumentError for extra arguments, unless (digit)$ style used. * ext/zlib/zlib.c (gzreader_gets): use memchr() to to gain performance. [ruby-talk:117701] * sprintf.c (rb_f_sprintf): raise ArgumentError for extra arguments, unless (digit)$ style used. * eval.c (frame_free): Guy Decoux solved the leak problem. Thanks. [ruby-core:03549] * ext/zlib/zlib.c (zstream_append_input): clear klass for z->input to avoid potential vulnerability. git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/branches/ruby_1_8@7119 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2004-07-09* array.c, enum.c, pack.c: rdoc patch from Johan Holmbergocean
<holmberg@iar.se> [ruby-core:3132] [ruby-core:3136] * numeric.c: rdoc patch. git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/branches/ruby_1_8@6607 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2004-06-16* eval.c (rb_mod_freeze): prepare string representation beforematz
freezing. [ruby-talk:103646] git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/branches/ruby_1_8@6469 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2004-06-04* io.c (rb_io_gets_m): set lastline ($_) even when read line ismatz
nil. [ruby-dev:23663] git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/branches/ruby_1_8@6427 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2004-05-20* numeric.c (flo_eq): alway check if operands are NaN.matz
[ruby-list:39685] * lib/cgi/session.rb: use LOCK_SH to read, and a few other improvements. [ruby-core:02328] git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/branches/ruby_1_8@6371 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2004-05-17* numeric.c (flo_to_s): p 0.0 should be '0.0' not '0.0e+00'.ocean
* numeric.c (flo_to_s): the number of significand is correctly handled, there is assumption that DBL_DIG == 15 though. (p 0.00000000000000000001 was '9.999999999999999e-21', now is '1.0e-20') git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/branches/ruby_1_8@6346 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2004-05-07* eval.c (rb_eval): too many line trace call. (ruby-bugs PR#1320)matz
* numeric.c (flo_to_s): tweak output string based to preserve decimal point and to remove trailing zeros. [ruby-talk:97891] * string.c (rb_str_index_m): use unsigned comparison for T_FIXNUM search. [ruby-talk:97342] * hash.c (rb_hash_equal): returns true if two hashes have same set of key-value set. [ruby-talk:97559] * hash.c (rb_hash_eql): returns true if two hashes are equal and have same default values. * string.c (rb_str_equal): always returns true or false, never returns nil. [ruby-dev:23404] * io.c (rb_io_reopen): should use rb_io_check_io(). git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/branches/ruby_1_8@6263 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2004-04-14* numeric.c (flo_eq): workaround for bcc32's bug.eban
(ruby-bugs-ja:PR#594) git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/branches/ruby_1_8@6159 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2004-04-14* array.c, enum.c, eval.c, file.c, io.c, numeric.c, object.c, prec.c,nobu
process.c, re.c, string.c: typos in RDoc comments. [ruby-core:02783] git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/branches/ruby_1_8@6158 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2004-04-05* eval.c (top_include): include in the wrapped load is done formatz
the wrapper, not for a singleton class for wrapped main. [ruby-dev:23305] * bignum.c (rb_big_eq): use temporary double variable to save the result (internal float register may be bigger than 64 bits, for example, 80 bits on x86). [ruby-dev:23311] * eval.c (block_pass): should generate unique identifier of the pushing block. [ruby-talk:96363] * ext/socket/socket.c (make_hostent): fix memory leak, based on the patch from HORIKAWA Hisashi <vzw00011@nifty.ne.jp>. git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/branches/ruby_1_8@6092 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2003-12-27Add RDoc comments to numeric.cdave
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/branches/ruby_1_8@5316 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2003-12-23RDoc comments for Fixnum, Float, and Hash. Add heuristic to RDoc to handle ↵dave
yaml/stringio git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@5265 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2003-12-22* configure.in (ac_cv_func_setitimer): moved from defines.hnobu
* defines.h, rubysig.h, signal.c: removed macro handling which should be done in configure. * configure.in (intrinsics.h): check if present. * ruby.h: include intrinsics.h if available. * bignum.c, marshal.c: include ieeefp.h if available. * missing.h (isinf): define as a macro if finite() and isnan() are available. [ruby-core:02032] git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@5244 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2003-12-20* eval.c (rb_with_disable_interrupt): prohibit thread contextmatz
switch during proc execution. [ruby-dev:21899] git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@5232 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2003-12-11* configure.in (ieeefp.h), numeric.c: needed for finite() onnobu
Solaris. [ruby-core:01921] * file.c (rb_stat_inspect): adjust format specifier. * parse.c (arg_prepend): nodetype() is for debug use. * ruby.h (ISASCII, etc): cast to int to get rid of warning. * ruby.h (alloca.h): include even in GCC. [ruby-core:01925] * ext/bigdecimal/bigdecimal.c (GetVpValue): adjust format specifier. * ext/bigdecimal/bigdecimal.c (BigDecimal_prec, BigDecimal_coerce, BigDecimal_divmod): use rb_assoc_new() to suppress memory usage. * ext/bigdecimal/bigdecimal.c (BigDecimal_split): ditto. * ext/dl/sym.c (rb_dlsym_guardcall): guard itself should be volatile. * ext/iconv/iconv.c (iconv_convert): ensure actual parameter with format specifier. * ext/pty/pty.c (MasterDevice, SlaveDevice, deviceNo): do not define unless used. * ext/pty/pty.c (getDevice): get rid of warning. * ext/socket/socket.c (port_str, sock_s_getaddrinfo, sock_s_getnameinfo): FIX2INT() now returns long. * ext/socket/socket.c (init_inetsock_internal): uninitialized variable. * ext/syck/rubyext.c (syck_parser_assign_io): add prototype. * ext/syck/rubyext.c (rb_syck_mktime, yaml_org_handler): use ISDIGIT() instead of isdigit() to avoid warnings and for platforms which don't support non-ascii charater. git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@5163 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2003-12-01* numeric.c (num_sadded): prohibit singleton method definition formatz
Numerics. fill yet another gap between Fixnum and Bignum. git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@5076 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2003-12-01* pack.c (htov16): converts endian using swap16. htov32(), hton16,matz
hton32 as well. [ruby-talk:85377] * pack.c (swap16): swap 2 bytes no matter how big short is on the platform. swap32() is also prepared. * numeric.c (rb_num2int): returns long to preserve information. rb_fix2int(), rb_num2uint(), rb_fix2uint() as well. [ruby-talk:85377] * numeric.c (rb_num2uint): should not check for value range if the source value is negative. git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@5075 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2003-08-28* variable.c (rb_const_get_0): should check constants defined inmatz
included modules, if klass is Object. [ruby-talk:79302] * numeric.c (check_uint): check should be done using UINT_MAX, not INT_MAX. this fix is submitted by <lyle@knology.net> in [ruby-core:01486] git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@4454 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2003-08-04* eval.c (method_proc): should specify YIELD_FUNC_SVALUE.matz
[ruby-dev:21107] * marshal.c (w_object): should not call w_extended for USRMARSHAL dump. [ruby-dev:21106] git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@4314 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2003-08-02* numeric.c (flo_to_s): get rid of buffer overflow.matz
* io.c (appendline): clearerr(3) before raising exception, since exception may be captured by rescue. [ruby-talk:77794] git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@4290 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2003-07-31* numeric.c (rb_num_coerce_relop): export function.matz
* marshal.c (w_object): check has been dropped. "_dump must return string." [ruby-dev:21024] git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@4243 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2003-07-25* ext/socket/socket.c (tcp_s_gethostbyname): was usingmatz
uninitialized size_t value. [ruby-talk:76946] * Minor cleanups. git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@4156 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2003-07-21* numeric.c (num_equal): should not use rb_equal().matz
* object.c (rb_equal): back out the last change. git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@4113 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2003-06-16* numeric.c (rb_fix2uint): renamed from rb_fix2int on IA64.eban
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@3951 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2003-05-30* numeric.c (rb_num2uint, rb_fix2int): new function to convertnobu
values over INT_MAX. [ruby-core:01099] * ruby.h (NUM2UINT, FIX2INT): ditto. git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@3885 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2003-05-19* object.c (init_copy): rename copy_object as initialize_copy,matz
since it works as copy constructor. * eval.c (rb_add_method): initialize_copy should always be private, like initialize. git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@3820 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2003-05-08* gc.c (rb_gc): check odd alignment stack on m68k machines.matz
* numeric.c (num_step): better error treatment of float values. git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@3769 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2003-05-08* compar.c (rb_cmperr): raise comparison failure.nobu
* intern.h: prototype; rb_cmperr * numeric.c (flo_gt, flo_ge, flo_lt, flo_le, fix_gt, fix_ge, fix_lt, fix_le): should fail unless the argument is comparable. (ruby-bugs-ja:PR#456) git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@3768 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2003-05-07* numeric.c (int_upto, int_downto): should fail unless thenobu
argument is comparable. (ruby-bugs-ja:PR#454) git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@3766 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2003-05-06* numeric.c (num_step): remove epsilon; add margin of 0.5, to makematz
"1.1.step(1.5,0.1)" to work (third try). git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@3764 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2003-05-06* object.c (rb_obj_methods): list singleton methods if recurmatz
argument is false; list all methods otherwise. * numeric.c (num_step): double epsilon to make "1.1.step(1.5,0.1)" to work. * ext/gdbm/gdbm.c (fgdbm_values_at): new method to replace select(index..). * ext/sdbm/init.c (fsdbm_values_at): ditto. * ext/dbm/dbm.c (fdbm_values_at): ditto. * ext/dbm/dbm.c (DBM::VERSION): defined. * ext/gdbm/testgdbm.rb: replace select with values_at. * ext/sdbm/testsdbm.rb: ditto. * ext/dbm/testdbm.rb: ditto. git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@3758 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2003-05-04* array.c (rb_ary_values_at): new method to replace select(index..).matz
* hash.c (rb_hash_values_at,env_values_at): ditto. * re.c (match_values_at): ditto. * struct.c (rb_struct_values_at): ditto. * re.c (match_select): add iterator behavior. * ext/curses/curses.c, ext/digest/sha2/sha2.c, ext/iconv/iconv.c, ext/racc/cparse/cparse.c: include "ruby.h" at the top to shut up "_FILE_OFFSET_BITS redefined" warning on Solaris. * class.c (rb_class_protected_instance_methods): now gives warnings to show migration path. The default will be reversed on Jan 2004. * numeric.c (num_step): "1.1.step(1.5,0.1)" to work. git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@3754 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2003-04-21* numeric.c (flo_is_finite_p): use finite() if available.nobu
* win32/win32.h (isinf, isnan): define as macro. [ruby-win32:00533] * bcc32/Makefile.sub, win32/Makefile.sub: no longer use missing/isinf.c, missing/isnan.c. git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@3708 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2003-04-15* numeric.c (rb_fix2str): buffer was insufficient.nobu
(ruby-bugs-ja:PR#431) git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@3681 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
generated by cgit v1.2.3 (git 2.25.1) at 2024-04-19 04:50:37 +0000