diff options
Diffstat (limited to 'ext/stringio/stringio.c')
| -rw-r--r-- | ext/stringio/stringio.c | 665 |
1 files changed, 401 insertions, 264 deletions
diff --git a/ext/stringio/stringio.c b/ext/stringio/stringio.c index bc2d7a8ccc..cc2294a795 100644 --- a/ext/stringio/stringio.c +++ b/ext/stringio/stringio.c @@ -12,11 +12,15 @@ **********************************************************************/ -#define STRINGIO_VERSION "3.0.5" +static const char *const +STRINGIO_VERSION = "3.2.1.dev"; + +#include <stdbool.h> #include "ruby.h" #include "ruby/io.h" #include "ruby/encoding.h" +#include "ruby/version.h" #if defined(HAVE_FCNTL_H) || defined(_WIN32) #include <fcntl.h> #elif defined(HAVE_SYS_FCNTL_H) @@ -32,84 +36,21 @@ # define rb_class_new_instance_kw(argc, argv, klass, kw_splat) rb_class_new_instance(argc, argv, klass) #endif -#ifndef HAVE_RB_IO_EXTRACT_MODEENC -#define rb_io_extract_modeenc strio_extract_modeenc -static void -strio_extract_modeenc(VALUE *vmode_p, VALUE *vperm_p, VALUE opthash, - int *oflags_p, int *fmode_p, struct rb_io_enc_t *convconfig_p) +static inline bool +str_chilled_p(VALUE str) { - VALUE mode = *vmode_p; - VALUE intmode; - int fmode; - int has_enc = 0, has_vmode = 0; - - convconfig_p->enc = convconfig_p->enc2 = 0; - - vmode_handle: - if (NIL_P(mode)) { - fmode = FMODE_READABLE; - } - else if (!NIL_P(intmode = rb_check_to_integer(mode, "to_int"))) { - int flags = NUM2INT(intmode); - fmode = rb_io_oflags_fmode(flags); - } - else { - const char *m = StringValueCStr(mode), *n, *e; - fmode = rb_io_modestr_fmode(m); - n = strchr(m, ':'); - if (n) { - long len; - char encname[ENCODING_MAXNAMELEN+1]; - has_enc = 1; - if (fmode & FMODE_SETENC_BY_BOM) { - n = strchr(n, '|'); - } - e = strchr(++n, ':'); - len = e ? e - n : (long)strlen(n); - if (len > 0 && len <= ENCODING_MAXNAMELEN) { - rb_encoding *enc; - if (e) { - memcpy(encname, n, len); - encname[len] = '\0'; - n = encname; - } - enc = rb_enc_find(n); - if (e) - convconfig_p->enc2 = enc; - else - convconfig_p->enc = enc; - } - if (e && (len = strlen(++e)) > 0 && len <= ENCODING_MAXNAMELEN) { - convconfig_p->enc = rb_enc_find(e); - } - } - } - - if (!NIL_P(opthash)) { - rb_encoding *extenc = 0, *intenc = 0; - VALUE v; - if (!has_vmode) { - ID id_mode; - CONST_ID(id_mode, "mode"); - v = rb_hash_aref(opthash, ID2SYM(id_mode)); - if (!NIL_P(v)) { - if (!NIL_P(mode)) { - rb_raise(rb_eArgError, "mode specified twice"); - } - has_vmode = 1; - mode = v; - goto vmode_handle; - } - } - - if (rb_io_extract_encoding_option(opthash, &extenc, &intenc, &fmode)) { - if (has_enc) { - rb_raise(rb_eArgError, "encoding specified twice"); - } - } - } - *fmode_p = fmode; +#if (RUBY_API_VERSION_MAJOR == 3 && RUBY_API_VERSION_MINOR >= 4) || RUBY_API_VERSION_MAJOR >= 4 + // Do not attempt to modify chilled strings on Ruby 3.4+ + // RUBY_FL_USER2 == STR_CHILLED_LITERAL + // RUBY_FL_USER3 == STR_CHILLED_SYMBOL_TO_S + return FL_TEST_RAW(str, RUBY_FL_USER2 | RUBY_FL_USER3); +#else + return false; +#endif } + +#ifndef HAVE_TYPE_RB_IO_MODE_T +typedef int rb_io_mode_t; #endif struct StringIO { @@ -117,7 +58,7 @@ struct StringIO { rb_encoding *enc; long pos; long lineno; - int flags; + rb_io_mode_t flags; int count; }; @@ -127,7 +68,13 @@ static long strio_write(VALUE self, VALUE str); #define IS_STRIO(obj) (rb_typeddata_is_kind_of((obj), &strio_data_type)) #define error_inval(msg) (rb_syserr_fail(EINVAL, msg)) -#define get_enc(ptr) ((ptr)->enc ? (ptr)->enc : rb_enc_get((ptr)->string)) +#define get_enc(ptr) ((ptr)->enc ? (ptr)->enc : !NIL_P((ptr)->string) ? rb_enc_get((ptr)->string) : NULL) + +static bool +readonly_string_p(VALUE string) +{ + return OBJ_FROZEN_RAW(string); +} static struct StringIO * strio_alloc(void) @@ -171,7 +118,7 @@ static const rb_data_type_t strio_data_type = { strio_free, strio_memsize, }, - 0, 0, RUBY_TYPED_FREE_IMMEDIATELY + 0, 0, RUBY_TYPED_FREE_IMMEDIATELY | RUBY_TYPED_WB_PROTECTED }; #define check_strio(self) ((struct StringIO*)rb_check_typeddata((self), &strio_data_type)) @@ -245,9 +192,27 @@ writable(VALUE strio) static void check_modifiable(struct StringIO *ptr) { - if (OBJ_FROZEN(ptr->string)) { + if (NIL_P(ptr->string)) { + /* Null device StringIO */ + } + else if (OBJ_FROZEN_RAW(ptr->string)) { rb_raise(rb_eIOError, "not modifiable string"); } + else { + rb_str_modify(ptr->string); + } +} + +static inline bool +outside_p(struct StringIO *ptr, long pos) +{ + return NIL_P(ptr->string) || pos >= RSTRING_LEN(ptr->string); +} + +static inline bool +eos_p(struct StringIO *ptr) +{ + return outside_p(ptr, ptr->pos); } static VALUE @@ -260,17 +225,32 @@ strio_s_allocate(VALUE klass) * call-seq: * StringIO.new(string = '', mode = 'r+') -> new_stringio * - * Note that +mode+ defaults to <tt>'r'</tt> if +string+ is frozen. - * * Returns a new \StringIO instance formed from +string+ and +mode+; - * see {Access Modes}[rdoc-ref:File@Access+Modes]: + * the instance should be closed when no longer needed: * - * strio = StringIO.new # => #<StringIO> + * strio = StringIO.new + * strio.string # => "" + * strio.closed_read? # => false + * strio.closed_write? # => false * strio.close * - * The instance should be closed when no longer needed. + * If +string+ is frozen, the default +mode+ is <tt>'r'</tt>: + * + * strio = StringIO.new('foo'.freeze) + * strio.string # => "foo" + * strio.closed_read? # => false + * strio.closed_write? # => true + * strio.close + * + * Argument +mode+ must be a valid + * {Access Mode}[rdoc-ref:File@Access+Modes], + * which may be a string or an integer constant: + * + * StringIO.new('foo', 'w+') + * StringIO.new('foo', File::RDONLY) * - * Related: StringIO.open (accepts block; closes automatically). + * Related: StringIO.open + * (passes the \StringIO object to the block; closes the object automatically on block exit). */ static VALUE strio_initialize(int argc, VALUE *argv, VALUE self) @@ -356,17 +336,18 @@ strio_init(int argc, VALUE *argv, struct StringIO *ptr, VALUE self) { VALUE string, vmode, opt; int oflags; - struct rb_io_enc_t convconfig; + rb_io_enc_t convconfig; argc = rb_scan_args(argc, argv, "02:", &string, &vmode, &opt); rb_io_extract_modeenc(&vmode, 0, opt, &oflags, &ptr->flags, &convconfig); - if (argc) { + if (!NIL_P(string)) { StringValue(string); } - else { + else if (!argc) { string = rb_enc_str_new("", 0, rb_default_external_encoding()); } - if (OBJ_FROZEN_RAW(string)) { + + if (!NIL_P(string) && readonly_string_p(string)) { if (ptr->flags & FMODE_WRITABLE) { rb_syserr_fail(EACCES, 0); } @@ -376,11 +357,11 @@ strio_init(int argc, VALUE *argv, struct StringIO *ptr, VALUE self) ptr->flags |= FMODE_WRITABLE; } } - if (ptr->flags & FMODE_TRUNC) { + if (!NIL_P(string) && (ptr->flags & FMODE_TRUNC)) { rb_str_resize(string, 0); } - ptr->string = string; - if (argc == 1) { + RB_OBJ_WRITE(self, &ptr->string, string); + if (argc == 1 && !NIL_P(string)) { ptr->enc = rb_enc_get(string); } else { @@ -397,30 +378,27 @@ static VALUE strio_finalize(VALUE self) { struct StringIO *ptr = StringIO(self); - ptr->string = Qnil; + RB_OBJ_WRITE(self, &ptr->string, Qnil); ptr->flags &= ~FMODE_READWRITE; return self; } /* * call-seq: - * StringIO.open(string = '', mode = 'r+') {|strio| ... } - * - * Note that +mode+ defaults to <tt>'r'</tt> if +string+ is frozen. + * StringIO.open(string = '', mode = 'r+') -> new_stringio + * StringIO.open(string = '', mode = 'r+') {|strio| ... } -> object * - * Creates a new \StringIO instance formed from +string+ and +mode+; - * see {Access Modes}[rdoc-ref:File@Access+Modes]. + * Creates new \StringIO instance by calling <tt>StringIO.new(string, mode)</tt>. * - * With no block, returns the new instance: + * With no block given, returns the new instance: * * strio = StringIO.open # => #<StringIO> * - * With a block, calls the block with the new instance + * With a block given, calls the block with the new instance * and returns the block's value; - * closes the instance on block exit. + * closes the instance on block exit: * - * StringIO.open {|strio| p strio } - * # => #<StringIO> + * StringIO.open('foo') {|strio| strio.string.upcase } # => "FOO" * * Related: StringIO.new. */ @@ -446,7 +424,7 @@ strio_s_new(int argc, VALUE *argv, VALUE klass) } /* - * Returns +false+. Just for compatibility to IO. + * Returns +false+; for compatibility with IO. */ static VALUE strio_false(VALUE self) @@ -456,7 +434,7 @@ strio_false(VALUE self) } /* - * Returns +nil+. Just for compatibility to IO. + * Returns +nil+; for compatibility with IO. */ static VALUE strio_nil(VALUE self) @@ -466,7 +444,7 @@ strio_nil(VALUE self) } /* - * Returns an object itself. Just for compatibility to IO. + * Returns +self+; for compatibility with IO. */ static VALUE strio_self(VALUE self) @@ -476,7 +454,7 @@ strio_self(VALUE self) } /* - * Returns 0. Just for compatibility to IO. + * Returns 0; for compatibility with IO. */ static VALUE strio_0(VALUE self) @@ -536,7 +514,7 @@ strio_get_string(VALUE self) * call-seq: * string = other_string -> other_string * - * Assigns the underlying string as +other_string+, and sets position to zero; + * Replaces the stored string with +other_string+, and sets the position to zero; * returns +other_string+: * * StringIO.open('foo') do |strio| @@ -550,7 +528,7 @@ strio_get_string(VALUE self) * "foo" * "bar" * - * Related: StringIO#string (returns the underlying string). + * Related: StringIO#string (returns the stored string). */ static VALUE strio_set_string(VALUE self, VALUE string) @@ -560,21 +538,27 @@ strio_set_string(VALUE self, VALUE string) rb_io_taint_check(self); ptr->flags &= ~FMODE_READWRITE; StringValue(string); - ptr->flags = OBJ_FROZEN(string) ? FMODE_READABLE : FMODE_READWRITE; + ptr->flags = readonly_string_p(string) ? FMODE_READABLE : FMODE_READWRITE; ptr->pos = 0; ptr->lineno = 0; - return ptr->string = string; + RB_OBJ_WRITE(self, &ptr->string, string); + return string; } /* * call-seq: * close -> nil * - * Closes +self+ for both reading and writing. + * Closes +self+ for both reading and writing; returns +nil+: * - * Raises IOError if reading or writing is attempted. + * strio = StringIO.new + * strio.closed? # => false + * strio.close # => nil + * strio.closed? # => true + * strio.read # Raises IOError: not opened for reading + * strio.write # Raises IOError: not opened for writing * - * Related: StringIO#close_read, StringIO#close_write. + * Related: StringIO#close_read, StringIO#close_write, StringIO.closed?. */ static VALUE strio_close(VALUE self) @@ -588,9 +572,16 @@ strio_close(VALUE self) * call-seq: * close_read -> nil * - * Closes +self+ for reading; closed-write setting remains unchanged. + * Closes +self+ for reading; + * closed-write setting remains unchanged; + * returns +nil+: * - * Raises IOError if reading is attempted. + * strio = StringIO.new + * strio.closed_read? # => false + * strio.close_read # => nil + * strio.closed_read? # => true + * strio.closed_write? # => false + * strio.read # Raises IOError: not opened for reading * * Related: StringIO#close, StringIO#close_write. */ @@ -609,11 +600,16 @@ strio_close_read(VALUE self) * call-seq: * close_write -> nil * - * Closes +self+ for writing; closed-read setting remains unchanged. + * Closes +self+ for writing; closed-read setting remains unchanged; returns +nil+: * - * Raises IOError if writing is attempted. + * strio = StringIO.new + * strio.closed_write? # => false + * strio.close_write # => nil + * strio.closed_write? # => true + * strio.closed_read? # => false + * strio.write('foo') # Raises IOError: not opened for writing * - * Related: StringIO#close, StringIO#close_read. + * Related: StringIO#close, StringIO#close_read, StringIO#closed_write?. */ static VALUE strio_close_write(VALUE self) @@ -630,8 +626,16 @@ strio_close_write(VALUE self) * call-seq: * closed? -> true or false * - * Returns +true+ if +self+ is closed for both reading and writing, - * +false+ otherwise. + * Returns whether +self+ is closed for both reading and writing: + * + * strio = StringIO.new + * strio.closed? # => false # Open for reading and writing. + * strio.close_read + * strio.closed? # => false # Still open for writing. + * strio.close_write + * strio.closed? # => true # Now closed for both. + * + * Related: StringIO.closed_read?, StringIO.closed_write?. */ static VALUE strio_closed(VALUE self) @@ -645,7 +649,14 @@ strio_closed(VALUE self) * call-seq: * closed_read? -> true or false * - * Returns +true+ if +self+ is closed for reading, +false+ otherwise. + * Returns whether +self+ is closed for reading: + * + * strio = StringIO.new + * strio.closed_read? # => false + * strio.close_read + * strio.closed_read? # => true + * + * Related: StringIO#closed?, StringIO#closed_write?, StringIO#close_read. */ static VALUE strio_closed_read(VALUE self) @@ -659,7 +670,14 @@ strio_closed_read(VALUE self) * call-seq: * closed_write? -> true or false * - * Returns +true+ if +self+ is closed for writing, +false+ otherwise. + * Returns whether +self+ is closed for writing: + * + * strio = StringIO.new + * strio.closed_write? # => false + * strio.close_write + * strio.closed_write? # => true + * + * Related: StringIO#close_write, StringIO#closed?, StringIO#closed_read?. */ static VALUE strio_closed_write(VALUE self) @@ -673,20 +691,26 @@ static struct StringIO * strio_to_read(VALUE self) { struct StringIO *ptr = readable(self); - if (ptr->pos < RSTRING_LEN(ptr->string)) return ptr; - return NULL; + if (eos_p(ptr)) return NULL; + return ptr; } /* * call-seq: * eof? -> true or false * - * Returns +true+ if positioned at end-of-stream, +false+ otherwise; - * see {Position}[rdoc-ref:File@Position]. + * Returns whether +self+ is positioned at end-of-stream: * - * Raises IOError if the stream is not opened for reading. + * strio = StringIO.new('foo') + * strio.pos # => 0 + * strio.eof? # => false + * strio.read # => "foo" + * strio.pos # => 3 + * strio.eof? # => true + * strio.close_read + * strio.eof? # Raises IOError: not opened for reading * - * StreamIO#eof is an alias for StreamIO#eof?. + * Related: StringIO#pos. */ static VALUE strio_eof(VALUE self) @@ -699,15 +723,19 @@ strio_eof(VALUE self) static VALUE strio_copy(VALUE copy, VALUE orig) { - struct StringIO *ptr; + struct StringIO *ptr, *old_ptr; + VALUE old_string = Qundef; orig = rb_convert_type(orig, T_DATA, "StringIO", "to_strio"); if (copy == orig) return copy; ptr = StringIO(orig); - if (check_strio(copy)) { - strio_free(DATA_PTR(copy)); + old_ptr = check_strio(copy); + if (old_ptr) { + old_string = old_ptr->string; + strio_free(old_ptr); } DATA_PTR(copy) = ptr; + RB_OBJ_WRITTEN(copy, old_string, ptr->string); RBASIC(copy)->flags &= ~STRIO_READWRITE; RBASIC(copy)->flags |= RBASIC(orig)->flags & STRIO_READWRITE; ++ptr->count; @@ -719,7 +747,7 @@ strio_copy(VALUE copy, VALUE orig) * lineno -> current_line_number * * Returns the current line number in +self+; - * see {Line Number}[rdoc-ref:IO@Line+Number]. + * see {Line Number}[rdoc-ref:StringIO@Line+Number]. */ static VALUE strio_get_lineno(VALUE self) @@ -732,7 +760,7 @@ strio_get_lineno(VALUE self) * lineno = new_line_number -> new_line_number * * Sets the current line number in +self+ to the given +new_line_number+; - * see {Line Number}[rdoc-ref:IO@Line+Number]. + * see {Line Number}[rdoc-ref:StringIO@Line+Number]. */ static VALUE strio_set_lineno(VALUE self, VALUE lineno) @@ -746,7 +774,7 @@ strio_set_lineno(VALUE self, VALUE lineno) * binmode -> self * * Sets the data mode in +self+ to binary mode; - * see {Data Mode}[rdoc-ref:File@Data+Mode]. + * see {Data Mode}[rdoc-ref:StringIO@Data+Mode]. * */ static VALUE @@ -807,9 +835,7 @@ strio_reopen(int argc, VALUE *argv, VALUE self) * pos -> stream_position * * Returns the current position (in bytes); - * see {Position}[rdoc-ref:IO@Position]. - * - * StringIO#tell is an alias for StringIO#pos. + * see {Position}[rdoc-ref:StringIO@Position]. */ static VALUE strio_get_pos(VALUE self) @@ -822,7 +848,7 @@ strio_get_pos(VALUE self) * pos = new_position -> new_position * * Sets the current position (in bytes); - * see {Position}[rdoc-ref:IO@Position]. + * see {Position}[rdoc-ref:StringIO@Position]. */ static VALUE strio_set_pos(VALUE self, VALUE pos) @@ -857,9 +883,9 @@ strio_rewind(VALUE self) * call-seq: * seek(offset, whence = SEEK_SET) -> 0 * - * Sets the current position to the given integer +offset+ (in bytes), + * Sets the position to the given integer +offset+ (in bytes), * with respect to a given constant +whence+; - * see {Position}[rdoc-ref:IO@Position]. + * see {IO#seek}[rdoc-ref:IO#seek]. */ static VALUE strio_seek(int argc, VALUE *argv, VALUE self) @@ -881,7 +907,11 @@ strio_seek(int argc, VALUE *argv, VALUE self) offset = ptr->pos; break; case 2: - offset = RSTRING_LEN(ptr->string); + if (NIL_P(ptr->string)) { + offset = 0; + } else { + offset = RSTRING_LEN(ptr->string); + } break; default: error_inval("invalid whence"); @@ -914,10 +944,9 @@ strio_get_sync(VALUE self) * call-seq: * each_byte {|byte| ... } -> self * - * With a block given, calls the block with each remaining byte in the stream; - * see {Byte IO}[rdoc-ref:IO@Byte+IO]. + * :include: stringio/each_byte.rdoc * - * With no block given, returns an enumerator. + * Related: StringIO#each_char, StringIO#each_codepoint, StringIO#each_line. */ static VALUE strio_each_byte(VALUE self) @@ -935,10 +964,10 @@ strio_each_byte(VALUE self) /* * call-seq: - * getc -> character or nil + * getc -> character, byte, or nil + * + * :include: stringio/getc.rdoc * - * Reads and returns the next character from the stream; - * see {Character IO}[rdoc-ref:IO@Character+IO]. */ static VALUE strio_getc(VALUE self) @@ -950,7 +979,7 @@ strio_getc(VALUE self) int len; char *p; - if (pos >= RSTRING_LEN(str)) { + if (eos_p(ptr)) { return Qnil; } p = RSTRING_PTR(str)+pos; @@ -961,17 +990,17 @@ strio_getc(VALUE self) /* * call-seq: - * getbyte -> byte or nil + * getbyte -> integer or nil + * + * :include: stringio/getbyte.rdoc * - * Reads and returns the next 8-bit byte from the stream; - * see {Byte IO}[rdoc-ref:IO@Byte+IO]. */ static VALUE strio_getbyte(VALUE self) { struct StringIO *ptr = readable(self); int c; - if (ptr->pos >= RSTRING_LEN(ptr->string)) { + if (eos_p(ptr)) { return Qnil; } c = RSTRING_PTR(ptr->string)[ptr->pos++]; @@ -993,8 +1022,17 @@ strio_extend(struct StringIO *ptr, long pos, long len) if (pos > olen) MEMZERO(RSTRING_PTR(ptr->string) + olen, char, pos - olen); } - else { - rb_str_modify(ptr->string); +} + +static void +strio_unget_string(struct StringIO *ptr, VALUE c) +{ + const char *cp = NULL; + long cl = RSTRING_LEN(c); + if (cl > 0) { + if (c != ptr->string) cp = RSTRING_PTR(c); + strio_unget_bytes(ptr, cp, cl); + RB_GC_GUARD(c); } } @@ -1012,6 +1050,7 @@ strio_ungetc(VALUE self, VALUE c) rb_encoding *enc, *enc2; check_modifiable(ptr); + if (NIL_P(ptr->string)) return Qnil; if (NIL_P(c)) return Qnil; if (RB_INTEGER_TYPE_P(c)) { int len, cc = NUM2INT(c); @@ -1019,19 +1058,22 @@ strio_ungetc(VALUE self, VALUE c) enc = rb_enc_get(ptr->string); len = rb_enc_codelen(cc, enc); - if (len <= 0) rb_enc_uint_chr(cc, enc); + if (len <= 0) { + rb_enc_uint_chr(cc, enc); /* to raise an exception */ + UNREACHABLE; + } rb_enc_mbcput(cc, buf, enc); return strio_unget_bytes(ptr, buf, len); } else { - SafeStringValue(c); + StringValue(c); + if (RSTRING_LEN(c) == 0) return Qnil; enc = rb_enc_get(ptr->string); enc2 = rb_enc_get(c); if (enc != enc2 && enc != rb_ascii8bit_encoding()) { c = rb_str_conv_enc(c, enc2, enc); } - strio_unget_bytes(ptr, RSTRING_PTR(c), RSTRING_LEN(c)); - RB_GC_GUARD(c); + strio_unget_string(ptr, c); return Qnil; } } @@ -1049,21 +1091,17 @@ strio_ungetbyte(VALUE self, VALUE c) struct StringIO *ptr = readable(self); check_modifiable(ptr); + if (NIL_P(ptr->string)) return Qnil; if (NIL_P(c)) return Qnil; if (RB_INTEGER_TYPE_P(c)) { - /* rb_int_and() not visible from exts */ - VALUE v = rb_funcall(c, '&', 1, INT2FIX(0xff)); - const char cc = NUM2INT(v) & 0xFF; - strio_unget_bytes(ptr, &cc, 1); + /* rb_int_and() not visible from exts */ + VALUE v = rb_funcall(c, '&', 1, INT2FIX(0xff)); + const char cc = NUM2INT(v) & 0xFF; + strio_unget_bytes(ptr, &cc, 1); } else { - long cl; - SafeStringValue(c); - cl = RSTRING_LEN(c); - if (cl > 0) { - strio_unget_bytes(ptr, RSTRING_PTR(c), cl); - RB_GC_GUARD(c); - } + StringValue(c); + strio_unget_string(ptr, c); } return Qnil; } @@ -1094,7 +1132,7 @@ strio_unget_bytes(struct StringIO *ptr, const char *cp, long cl) if (rest > cl) memset(s + len, 0, rest - cl); pos -= cl; } - memcpy(s + pos, cp, cl); + memcpy(s + pos, (cp ? cp : s), cl); ptr->pos = pos; return Qnil; } @@ -1131,12 +1169,11 @@ strio_readbyte(VALUE self) /* * call-seq: - * each_char {|c| ... } -> self + * each_char {|char| ... } -> self * - * With a block given, calls the block with each remaining character in the stream; - * see {Character IO}[rdoc-ref:IO@Character+IO]. + * :include: stringio/each_char.rdoc * - * With no block given, returns an enumerator. + * Related: StringIO#each_byte, StringIO#each_codepoint, StringIO#each_line. */ static VALUE strio_each_char(VALUE self) @@ -1155,10 +1192,9 @@ strio_each_char(VALUE self) * call-seq: * each_codepoint {|codepoint| ... } -> self * - * With a block given, calls the block with each remaining codepoint in the stream; - * see {Codepoint IO}[rdoc-ref:IO@Codepoint+IO]. + * :include: stringio/each_codepoint.rdoc * - * With no block given, returns an enumerator. + * Related: StringIO#each_byte, StringIO#each_char, StringIO#each_line. */ static VALUE strio_each_codepoint(VALUE self) @@ -1221,38 +1257,57 @@ struct getline_arg { }; static struct getline_arg * -prepare_getline_args(struct getline_arg *arg, int argc, VALUE *argv) +prepare_getline_args(struct StringIO *ptr, struct getline_arg *arg, int argc, VALUE *argv) { - VALUE str, lim, opts; + VALUE rs, lim, opts; long limit = -1; int respect_chomp; - argc = rb_scan_args(argc, argv, "02:", &str, &lim, &opts); - respect_chomp = argc == 0 || !NIL_P(str); + argc = rb_scan_args(argc, argv, "02:", &rs, &lim, &opts); + respect_chomp = argc == 0 || !NIL_P(rs); switch (argc) { case 0: - str = rb_rs; + rs = rb_rs; break; case 1: - if (!NIL_P(str) && !RB_TYPE_P(str, T_STRING)) { - VALUE tmp = rb_check_string_type(str); + if (!NIL_P(rs) && !RB_TYPE_P(rs, T_STRING)) { + VALUE tmp = rb_check_string_type(rs); if (NIL_P(tmp)) { - limit = NUM2LONG(str); - str = rb_rs; + limit = NUM2LONG(rs); + rs = rb_rs; } else { - str = tmp; + rs = tmp; } } break; case 2: - if (!NIL_P(str)) StringValue(str); + if (!NIL_P(rs)) StringValue(rs); if (!NIL_P(lim)) limit = NUM2LONG(lim); break; } - arg->rs = str; + if (!NIL_P(ptr->string) && !NIL_P(rs)) { + rb_encoding *enc_rs, *enc_io; + enc_rs = rb_enc_get(rs); + enc_io = get_enc(ptr); + if (enc_rs != enc_io && + (rb_enc_str_coderange(rs) != ENC_CODERANGE_7BIT || + (RSTRING_LEN(rs) > 0 && !rb_enc_asciicompat(enc_io)))) { + if (rs == rb_rs) { + rs = rb_enc_str_new(0, 0, enc_io); + rb_str_buf_cat_ascii(rs, "\n"); + rs = rs; + } + else { + rb_raise(rb_eArgError, "encoding mismatch: %s IO with %s RS", + rb_enc_name(enc_io), + rb_enc_name(enc_rs)); + } + } + } + arg->rs = rs; arg->limit = limit; arg->chomp = 0; if (!NIL_P(opts)) { @@ -1262,9 +1317,9 @@ prepare_getline_args(struct getline_arg *arg, int argc, VALUE *argv) keywords[0] = rb_intern_const("chomp"); } rb_get_kwargs(opts, keywords, 0, 1, &vchomp); - if (respect_chomp) { + if (respect_chomp) { arg->chomp = (vchomp != Qundef) && RTEST(vchomp); - } + } } return arg; } @@ -1288,7 +1343,7 @@ strio_getline(struct getline_arg *arg, struct StringIO *ptr) long w = 0; rb_encoding *enc = get_enc(ptr); - if (ptr->pos >= (n = RSTRING_LEN(ptr->string))) { + if (NIL_P(ptr->string) || ptr->pos >= (n = RSTRING_LEN(ptr->string))) { return Qnil; } s = RSTRING_PTR(ptr->string); @@ -1304,7 +1359,7 @@ strio_getline(struct getline_arg *arg, struct StringIO *ptr) str = strio_substr(ptr, ptr->pos, e - s - w, enc); } else if ((n = RSTRING_LEN(str)) == 0) { - const char *paragraph_end = NULL; + const char *paragraph_end = NULL; p = s; while (p[(p + 1 < e) && (*p == '\r') && 0] == '\n') { p += *p == '\r'; @@ -1314,18 +1369,18 @@ strio_getline(struct getline_arg *arg, struct StringIO *ptr) } s = p; while ((p = memchr(p, '\n', e - p)) && (p != e)) { - p++; - if (!((p < e && *p == '\n') || - (p + 1 < e && *p == '\r' && *(p+1) == '\n'))) { - continue; - } - paragraph_end = p - ((*(p-2) == '\r') ? 2 : 1); - while ((p < e && *p == '\n') || - (p + 1 < e && *p == '\r' && *(p+1) == '\n')) { - p += (*p == '\r') ? 2 : 1; - } - e = p; - break; + p++; + if (!((p < e && *p == '\n') || + (p + 1 < e && *p == '\r' && *(p+1) == '\n'))) { + continue; + } + paragraph_end = p - ((*(p-2) == '\r') ? 2 : 1); + while ((p < e && *p == '\n') || + (p + 1 < e && *p == '\r' && *(p+1) == '\n')) { + p += (*p == '\r') ? 2 : 1; + } + e = p; + break; } if (arg->chomp && paragraph_end) { w = e - paragraph_end; @@ -1340,8 +1395,9 @@ strio_getline(struct getline_arg *arg, struct StringIO *ptr) str = strio_substr(ptr, ptr->pos, e - s - w, enc); } else { - if (n < e - s) { - if (e - s < 1024) { + if (n < e - s + arg->chomp) { + /* unless chomping, RS at the end does not matter */ + if (e - s < 1024 || n == e - s) { for (p = s; p + n <= e; ++p) { if (MEMCMP(p, RSTRING_PTR(str), char, n) == 0) { e = p + n; @@ -1372,22 +1428,22 @@ strio_getline(struct getline_arg *arg, struct StringIO *ptr) * gets(limit, chomp: false) -> string or nil * gets(sep, limit, chomp: false) -> string or nil * - * Reads and returns a line from the stream; - * assigns the return value to <tt>$_</tt>; - * see {Line IO}[rdoc-ref:IO@Line+IO]. + * :include: stringio/gets.rdoc + * */ static VALUE strio_gets(int argc, VALUE *argv, VALUE self) { + struct StringIO *ptr = readable(self); struct getline_arg arg; VALUE str; - if (prepare_getline_args(&arg, argc, argv)->limit == 0) { - struct StringIO *ptr = readable(self); + if (prepare_getline_args(ptr, &arg, argc, argv)->limit == 0) { + if (NIL_P(ptr->string)) return Qnil; return rb_enc_str_new(0, 0, get_enc(ptr)); } - str = strio_getline(&arg, readable(self)); + str = strio_getline(&arg, ptr); rb_lastline_set(str); return str; } @@ -1410,32 +1466,30 @@ strio_readline(int argc, VALUE *argv, VALUE self) } /* + * :markup: markdown + * * call-seq: * each_line(sep = $/, chomp: false) {|line| ... } -> self * each_line(limit, chomp: false) {|line| ... } -> self * each_line(sep, limit, chomp: false) {|line| ... } -> self * - * Calls the block with each remaining line read from the stream; - * does nothing if already at end-of-file; - * returns +self+. - * See {Line IO}[rdoc-ref:IO@Line+IO]. + * :include: stringio/each_line.md * - * StringIO#each is an alias for StringIO#each_line. */ static VALUE strio_each(int argc, VALUE *argv, VALUE self) { VALUE line; + struct StringIO *ptr = readable(self); struct getline_arg arg; - StringIO(self); RETURN_ENUMERATOR(self, argc, argv); - if (prepare_getline_args(&arg, argc, argv)->limit == 0) { + if (prepare_getline_args(ptr, &arg, argc, argv)->limit == 0) { rb_raise(rb_eArgError, "invalid limit: 0 for each_line"); } - while (!NIL_P(line = strio_getline(&arg, readable(self)))) { + while (!NIL_P(line = strio_getline(&arg, ptr))) { rb_yield(line); } return self; @@ -1453,15 +1507,15 @@ static VALUE strio_readlines(int argc, VALUE *argv, VALUE self) { VALUE ary, line; + struct StringIO *ptr = readable(self); struct getline_arg arg; - StringIO(self); - ary = rb_ary_new(); - if (prepare_getline_args(&arg, argc, argv)->limit == 0) { + if (prepare_getline_args(ptr, &arg, argc, argv)->limit == 0) { rb_raise(rb_eArgError, "invalid limit: 0 for readlines"); } - while (!NIL_P(line = strio_getline(&arg, readable(self)))) { + ary = rb_ary_new(); + while (!NIL_P(line = strio_getline(&arg, ptr))) { rb_ary_push(ary, line); } return ary; @@ -1500,6 +1554,7 @@ strio_write(VALUE self, VALUE str) if (!RB_TYPE_P(str, T_STRING)) str = rb_obj_as_string(str); enc = get_enc(ptr); + if (!enc) return 0; enc2 = rb_enc_get(str); if (enc != enc2 && enc != ascii8bit && enc != (usascii = rb_usascii_encoding())) { VALUE converted = rb_str_conv_enc(str, enc2, enc); @@ -1525,6 +1580,7 @@ strio_write(VALUE self, VALUE str) } else { strio_extend(ptr, ptr->pos, len); + rb_str_modify(ptr->string); memmove(RSTRING_PTR(ptr->string)+ptr->pos, RSTRING_PTR(str), len); } RB_GC_GUARD(str); @@ -1559,9 +1615,10 @@ strio_write(VALUE self, VALUE str) /* * call-seq: - * strio.putc(obj) -> obj + * putc(object) -> object + * + * :include: stringio/putc.rdoc * - * See IO#putc. */ static VALUE strio_putc(VALUE self, VALUE ch) @@ -1571,10 +1628,12 @@ strio_putc(VALUE self, VALUE ch) check_modifiable(ptr); if (RB_TYPE_P(ch, T_STRING)) { + if (NIL_P(ptr->string)) return ch; str = rb_str_substr(ch, 0, 1); } else { char c = NUM2CHR(ch); + if (NIL_P(ptr->string)) return ch; str = rb_str_new(&c, 1); } strio_write(self, str); @@ -1591,9 +1650,10 @@ strio_putc(VALUE self, VALUE ch) /* * call-seq: - * strio.read([length [, outbuf]]) -> string, outbuf, or nil + * read(maxlen = nil, out_string = nil) → new_string, out_string, or nil + * + * :include: stringio/read.rdoc * - * See IO#read. */ static VALUE strio_read(int argc, VALUE *argv, VALUE self) @@ -1617,15 +1677,16 @@ strio_read(int argc, VALUE *argv, VALUE self) if (len < 0) { rb_raise(rb_eArgError, "negative length %ld given", len); } - if (len > 0 && ptr->pos >= RSTRING_LEN(ptr->string)) { + if (eos_p(ptr)) { if (!NIL_P(str)) rb_str_resize(str, 0); - return Qnil; + return len > 0 ? Qnil : rb_str_new(0, 0); } binary = 1; break; } /* fall through */ case 0: + if (NIL_P(ptr->string)) return Qnil; len = RSTRING_LEN(ptr->string); if (len <= ptr->pos) { rb_encoding *enc = get_enc(ptr); @@ -1643,7 +1704,7 @@ strio_read(int argc, VALUE *argv, VALUE self) } break; default: - rb_error_arity(argc, 0, 2); + rb_error_arity(argc, 0, 2); } if (NIL_P(str)) { rb_encoding *enc = binary ? rb_ascii8bit_encoding() : get_enc(ptr); @@ -1654,16 +1715,64 @@ strio_read(int argc, VALUE *argv, VALUE self) if (len > rest) len = rest; rb_str_resize(str, len); MEMCPY(RSTRING_PTR(str), RSTRING_PTR(ptr->string) + ptr->pos, char, len); - if (binary) - rb_enc_associate(str, rb_ascii8bit_encoding()); - else + if (!binary) { rb_enc_copy(str, ptr->string); + } } ptr->pos += RSTRING_LEN(str); return str; } /* + * call-seq: + * pread(maxlen, offset, out_string = nil) -> new_string or out_string + * + * :include: stringio/pread.rdoc + * + */ +static VALUE +strio_pread(int argc, VALUE *argv, VALUE self) +{ + VALUE rb_len, rb_offset, rb_buf; + rb_scan_args(argc, argv, "21", &rb_len, &rb_offset, &rb_buf); + long len = NUM2LONG(rb_len); + long offset = NUM2LONG(rb_offset); + + if (len < 0) { + rb_raise(rb_eArgError, "negative string size (or size too big): %" PRIsVALUE, rb_len); + } + + if (len == 0) { + if (NIL_P(rb_buf)) { + return rb_str_new("", 0); + } + return rb_buf; + } + + if (offset < 0) { + rb_syserr_fail_str(EINVAL, rb_sprintf("pread: Invalid offset argument: %" PRIsVALUE, rb_offset)); + } + + struct StringIO *ptr = readable(self); + + if (outside_p(ptr, offset)) { + rb_eof_error(); + } + + if (NIL_P(rb_buf)) { + return strio_substr(ptr, offset, len, rb_ascii8bit_encoding()); + } + + long rest = RSTRING_LEN(ptr->string) - offset; + if (len > rest) len = rest; + rb_str_resize(rb_buf, len); + rb_enc_associate(rb_buf, rb_ascii8bit_encoding()); + MEMCPY(RSTRING_PTR(rb_buf), RSTRING_PTR(ptr->string) + offset, char, len); + return rb_buf; +} + + +/* * call-seq: * strio.sysread(integer[, outbuf]) -> string * strio.readpartial(integer[, outbuf]) -> string @@ -1711,8 +1820,14 @@ strio_read_nonblock(int argc, VALUE *argv, VALUE self) return val; } +/* + * See IO#write + */ #define strio_syswrite rb_io_write +/* + * See IO#write_nonblock + */ static VALUE strio_syswrite_nonblock(int argc, VALUE *argv, VALUE self) { @@ -1730,17 +1845,17 @@ strio_syswrite_nonblock(int argc, VALUE *argv, VALUE self) /* * call-seq: - * strio.length -> integer - * strio.size -> integer + * size -> integer + * + * :include: stringio/size.rdoc * - * Returns the size of the buffer string. */ static VALUE strio_size(VALUE self) { VALUE string = StringIO(self)->string; if (NIL_P(string)) { - rb_raise(rb_eIOError, "not opened"); + return INT2FIX(0); } return ULONG2NUM(RSTRING_LEN(string)); } @@ -1757,10 +1872,12 @@ strio_truncate(VALUE self, VALUE len) { VALUE string = writable(self)->string; long l = NUM2LONG(len); - long plen = RSTRING_LEN(string); + long plen; if (l < 0) { error_inval("negative length"); } + if (NIL_P(string)) return 0; + plen = RSTRING_LEN(string); rb_str_resize(string, l); if (plen < l) { MEMZERO(RSTRING_PTR(string) + plen, char, l - plen); @@ -1769,12 +1886,20 @@ strio_truncate(VALUE self, VALUE len) } /* - * call-seq: - * strio.external_encoding => encoding + * call-seq: + * external_encoding -> encoding or nil + * + * Returns an Encoding object that represents the encoding of the string; + * see {Encodings}[rdoc-ref:StringIO@Encodings]: + * + * strio = StringIO.new('foo') + * strio.external_encoding # => #<Encoding:UTF-8> + * + * Returns +nil+ if +self+ has no string and is in write mode: + * + * strio = StringIO.new(nil, 'w+') + * strio.external_encoding # => nil * - * Returns the Encoding object that represents the encoding of the file. - * If the stream is write mode and no encoding is specified, returns - * +nil+. */ static VALUE @@ -1786,10 +1911,9 @@ strio_external_encoding(VALUE self) /* * call-seq: - * strio.internal_encoding => encoding + * internal_encoding -> nil * - * Returns the Encoding of the internal string if conversion is - * specified. Otherwise returns +nil+. + * Returns +nil+; for compatibility with IO. */ static VALUE @@ -1823,21 +1947,31 @@ strio_set_encoding(int argc, VALUE *argv, VALUE self) else { enc = rb_find_encoding(ext_enc); if (!enc) { - struct rb_io_enc_t convconfig; - int oflags, fmode; + rb_io_enc_t convconfig; + int oflags; + rb_io_mode_t fmode; VALUE vmode = rb_str_append(rb_str_new_cstr("r:"), ext_enc); rb_io_extract_modeenc(&vmode, 0, Qnil, &oflags, &fmode, &convconfig); enc = convconfig.enc2; } } ptr->enc = enc; - if (WRITABLE(self)) { + if (!NIL_P(ptr->string) && WRITABLE(self) && !str_chilled_p(ptr->string)) { rb_enc_associate(ptr->string, enc); } return self; } +/* + * call-seq: + * strio.set_encoding_by_bom => strio or nil + * + * Sets the encoding according to the BOM (Byte Order Mark) in the + * string. + * + * Returns +self+ if the BOM is found, otherwise +nil. + */ static VALUE strio_set_encoding_by_bom(VALUE self) { @@ -1848,16 +1982,9 @@ strio_set_encoding_by_bom(VALUE self) } /* - * \IO streams for strings, with access similar to - * {IO}[rdoc-ref:IO]; - * see {IO}[rdoc-ref:IO]. - * - * === About the Examples - * - * Examples on this page assume that \StringIO has been required: - * - * require 'stringio' + * :markup: markdown * + * :include: stringio/stringio.md */ void Init_stringio(void) @@ -1865,15 +1992,20 @@ Init_stringio(void) #undef rb_intern #ifdef HAVE_RB_EXT_RACTOR_SAFE - rb_ext_ractor_safe(true); + rb_ext_ractor_safe(true); #endif VALUE StringIO = rb_define_class("StringIO", rb_cObject); + /* The version string */ rb_define_const(StringIO, "VERSION", rb_str_new_cstr(STRINGIO_VERSION)); rb_include_module(StringIO, rb_mEnumerable); rb_define_alloc_func(StringIO, strio_s_allocate); + + /* Maximum length that a StringIO instance can hold */ + rb_define_const(StringIO, "MAX_LENGTH", LONG2NUM(LONG_MAX)); + rb_define_singleton_method(StringIO, "new", strio_s_new, -1); rb_define_singleton_method(StringIO, "open", strio_s_open, -1); rb_define_method(StringIO, "initialize", strio_initialize, -1); @@ -1923,6 +2055,7 @@ Init_stringio(void) rb_define_method(StringIO, "gets", strio_gets, -1); rb_define_method(StringIO, "readlines", strio_readlines, -1); rb_define_method(StringIO, "read", strio_read, -1); + rb_define_method(StringIO, "pread", strio_pread, -1); rb_define_method(StringIO, "write", strio_write_m, -1); rb_define_method(StringIO, "putc", strio_putc, 1); @@ -1951,7 +2084,9 @@ Init_stringio(void) rb_define_method(StringIO, "set_encoding_by_bom", strio_set_encoding_by_bom, 0); { + /* :stopdoc: */ VALUE mReadable = rb_define_module_under(rb_cIO, "generic_readable"); + /* :startdoc: */ rb_define_method(mReadable, "readchar", strio_readchar, 0); rb_define_method(mReadable, "readbyte", strio_readbyte, 0); rb_define_method(mReadable, "readline", strio_readline, -1); @@ -1961,7 +2096,9 @@ Init_stringio(void) rb_include_module(StringIO, mReadable); } { + /* :stopdoc: */ VALUE mWritable = rb_define_module_under(rb_cIO, "generic_writable"); + /* :startdoc: */ rb_define_method(mWritable, "<<", strio_addstr, 1); rb_define_method(mWritable, "print", strio_print, -1); rb_define_method(mWritable, "printf", strio_printf, -1); |