1 /* Definitions for data structures and routines for the regular 2 expression library. 3 Copyright (C) 1985,1989-93,1995-98,2000,2001,2002,2003,2005,2006,2008 4 Free Software Foundation, Inc. 5 This file is part of the GNU C Library. 6 7 The GNU C Library is free software; you can redistribute it and/or 8 modify it under the terms of the GNU Lesser General Public 9 License as published by the Free Software Foundation; either 10 version 2.1 of the License, or (at your option) any later version. 11 12 The GNU C Library is distributed in the hope that it will be useful, 13 but WITHOUT ANY WARRANTY; without even the implied warranty of 14 MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU 15 Lesser General Public License for more details. 16 17 You should have received a copy of the GNU Lesser General Public 18 License along with the GNU C Library; if not, write to the Free 19 Software Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 20 02111-1307 USA. */ 21 22 #ifndef _REGEX_H 23 #define _REGEX_H 1 24 25 #include <sys/types.h> 26 #include <gnu/option-groups.h> 27 28 /* Allow the use in C++ code. */ 29 #ifdef __cplusplus 30 extern "C" { 31 #endif 32 33 /* The following two types have to be signed and unsigned integer type 34 wide enough to hold a value of a pointer. For most ANSI compilers 35 ptrdiff_t and size_t should be likely OK. Still size of these two 36 types is 2 for Microsoft C. Ugh... */ 37 typedef long int s_reg_t; 38 typedef unsigned long int active_reg_t; 39 40 /* The following bits are used to determine the regexp syntax we 41 recognize. The set/not-set meanings are chosen so that Emacs syntax 42 remains the value 0. The bits are given in alphabetical order, and 43 the definitions shifted by one from the previous bit; thus, when we 44 add or remove a bit, only one other definition need change. */ 45 typedef unsigned long int reg_syntax_t; 46 47 #ifdef __USE_GNU 48 /* If this bit is not set, then \ inside a bracket expression is literal. 49 If set, then such a \ quotes the following character. */ 50 # define RE_BACKSLASH_ESCAPE_IN_LISTS ((unsigned long int) 1) 51 52 /* If this bit is not set, then + and ? are operators, and \+ and \? are 53 literals. 54 If set, then \+ and \? are operators and + and ? are literals. */ 55 # define RE_BK_PLUS_QM (RE_BACKSLASH_ESCAPE_IN_LISTS << 1) 56 57 /* If this bit is set, then character classes are supported. They are: 58 [:alpha:], [:upper:], [:lower:], [:digit:], [:alnum:], [:xdigit:], 59 [:space:], [:print:], [:punct:], [:graph:], and [:cntrl:]. 60 If not set, then character classes are not supported. */ 61 # define RE_CHAR_CLASSES (RE_BK_PLUS_QM << 1) 62 63 /* If this bit is set, then ^ and $ are always anchors (outside bracket 64 expressions, of course). 65 If this bit is not set, then it depends: 66 ^ is an anchor if it is at the beginning of a regular 67 expression or after an open-group or an alternation operator; 68 $ is an anchor if it is at the end of a regular expression, or 69 before a close-group or an alternation operator. 70 71 This bit could be (re)combined with RE_CONTEXT_INDEP_OPS, because 72 POSIX draft 11.2 says that * etc. in leading positions is undefined. 73 We already implemented a previous draft which made those constructs 74 invalid, though, so we haven't changed the code back. */ 75 # define RE_CONTEXT_INDEP_ANCHORS (RE_CHAR_CLASSES << 1) 76 77 /* If this bit is set, then special characters are always special 78 regardless of where they are in the pattern. 79 If this bit is not set, then special characters are special only in 80 some contexts; otherwise they are ordinary. Specifically, 81 * + ? and intervals are only special when not after the beginning, 82 open-group, or alternation operator. */ 83 # define RE_CONTEXT_INDEP_OPS (RE_CONTEXT_INDEP_ANCHORS << 1) 84 85 /* If this bit is set, then *, +, ?, and { cannot be first in an re or 86 immediately after an alternation or begin-group operator. */ 87 # define RE_CONTEXT_INVALID_OPS (RE_CONTEXT_INDEP_OPS << 1) 88 89 /* If this bit is set, then . matches newline. 90 If not set, then it doesn't. */ 91 # define RE_DOT_NEWLINE (RE_CONTEXT_INVALID_OPS << 1) 92 93 /* If this bit is set, then . doesn't match NUL. 94 If not set, then it does. */ 95 # define RE_DOT_NOT_NULL (RE_DOT_NEWLINE << 1) 96 97 /* If this bit is set, nonmatching lists [^...] do not match newline. 98 If not set, they do. */ 99 # define RE_HAT_LISTS_NOT_NEWLINE (RE_DOT_NOT_NULL << 1) 100 101 /* If this bit is set, either \{...\} or {...} defines an 102 interval, depending on RE_NO_BK_BRACES. 103 If not set, \{, \}, {, and } are literals. */ 104 # define RE_INTERVALS (RE_HAT_LISTS_NOT_NEWLINE << 1) 105 106 /* If this bit is set, +, ? and | aren't recognized as operators. 107 If not set, they are. */ 108 # define RE_LIMITED_OPS (RE_INTERVALS << 1) 109 110 /* If this bit is set, newline is an alternation operator. 111 If not set, newline is literal. */ 112 # define RE_NEWLINE_ALT (RE_LIMITED_OPS << 1) 113 114 /* If this bit is set, then `{...}' defines an interval, and \{ and \} 115 are literals. 116 If not set, then `\{...\}' defines an interval. */ 117 # define RE_NO_BK_BRACES (RE_NEWLINE_ALT << 1) 118 119 /* If this bit is set, (...) defines a group, and \( and \) are literals. 120 If not set, \(...\) defines a group, and ( and ) are literals. */ 121 # define RE_NO_BK_PARENS (RE_NO_BK_BRACES << 1) 122 123 /* If this bit is set, then \<digit> matches <digit>. 124 If not set, then \<digit> is a back-reference. */ 125 # define RE_NO_BK_REFS (RE_NO_BK_PARENS << 1) 126 127 /* If this bit is set, then | is an alternation operator, and \| is literal. 128 If not set, then \| is an alternation operator, and | is literal. */ 129 # define RE_NO_BK_VBAR (RE_NO_BK_REFS << 1) 130 131 /* If this bit is set, then an ending range point collating higher 132 than the starting range point, as in [z-a], is invalid. 133 If not set, then when ending range point collates higher than the 134 starting range point, the range is ignored. */ 135 # define RE_NO_EMPTY_RANGES (RE_NO_BK_VBAR << 1) 136 137 /* If this bit is set, then an unmatched ) is ordinary. 138 If not set, then an unmatched ) is invalid. */ 139 # define RE_UNMATCHED_RIGHT_PAREN_ORD (RE_NO_EMPTY_RANGES << 1) 140 141 /* If this bit is set, succeed as soon as we match the whole pattern, 142 without further backtracking. */ 143 # define RE_NO_POSIX_BACKTRACKING (RE_UNMATCHED_RIGHT_PAREN_ORD << 1) 144 145 /* If this bit is set, do not process the GNU regex operators. 146 If not set, then the GNU regex operators are recognized. */ 147 # define RE_NO_GNU_OPS (RE_NO_POSIX_BACKTRACKING << 1) 148 149 /* If this bit is set, turn on internal regex debugging. 150 If not set, and debugging was on, turn it off. 151 This only works if regex.c is compiled -DDEBUG. 152 We define this bit always, so that all that's needed to turn on 153 debugging is to recompile regex.c; the calling code can always have 154 this bit set, and it won't affect anything in the normal case. */ 155 # define RE_DEBUG (RE_NO_GNU_OPS << 1) 156 157 /* If this bit is set, a syntactically invalid interval is treated as 158 a string of ordinary characters. For example, the ERE 'a{1' is 159 treated as 'a\{1'. */ 160 # define RE_INVALID_INTERVAL_ORD (RE_DEBUG << 1) 161 162 /* EGLIBC: Old regex implementation does not support these. */ 163 # ifdef __OPTION_POSIX_REGEXP_GLIBC 164 /* If this bit is set, then ignore case when matching. 165 If not set, then case is significant. */ 166 # define RE_ICASE (RE_INVALID_INTERVAL_ORD << 1) 167 168 /* This bit is used internally like RE_CONTEXT_INDEP_ANCHORS but only 169 for ^, because it is difficult to scan the regex backwards to find 170 whether ^ should be special. */ 171 # define RE_CARET_ANCHORS_HERE (RE_ICASE << 1) 172 173 /* If this bit is set, then \{ cannot be first in an bre or 174 immediately after an alternation or begin-group operator. */ 175 # define RE_CONTEXT_INVALID_DUP (RE_CARET_ANCHORS_HERE << 1) 176 177 /* If this bit is set, then no_sub will be set to 1 during 178 re_compile_pattern. */ 179 # define RE_NO_SUB (RE_CONTEXT_INVALID_DUP << 1) 180 # endif /* __OPTION_POSIX_REGEXP_GLIBC */ 181 #endif 182 183 /* This global variable defines the particular regexp syntax to use (for 184 some interfaces). When a regexp is compiled, the syntax used is 185 stored in the pattern buffer, so changing this does not affect 186 already-compiled regexps. */ 187 extern reg_syntax_t re_syntax_options; 188 189 #ifdef __USE_GNU 191 /* Define combinations of the above bits for the standard possibilities. 192 (The [[[ comments delimit what gets put into the Texinfo file, so 193 don't delete them!) */ 194 /* [[[begin syntaxes]]] */ 195 #define RE_SYNTAX_EMACS 0 196 197 #define RE_SYNTAX_AWK \ 198 (RE_BACKSLASH_ESCAPE_IN_LISTS | RE_DOT_NOT_NULL \ 199 | RE_NO_BK_PARENS | RE_NO_BK_REFS \ 200 | RE_NO_BK_VBAR | RE_NO_EMPTY_RANGES \ 201 | RE_DOT_NEWLINE | RE_CONTEXT_INDEP_ANCHORS \ 202 | RE_UNMATCHED_RIGHT_PAREN_ORD | RE_NO_GNU_OPS) 203 204 #define RE_SYNTAX_GNU_AWK \ 205 ((RE_SYNTAX_POSIX_EXTENDED | RE_BACKSLASH_ESCAPE_IN_LISTS | RE_DEBUG) \ 206 & ~(RE_DOT_NOT_NULL | RE_INTERVALS | RE_CONTEXT_INDEP_OPS \ 207 | RE_CONTEXT_INVALID_OPS )) 208 209 #define RE_SYNTAX_POSIX_AWK \ 210 (RE_SYNTAX_POSIX_EXTENDED | RE_BACKSLASH_ESCAPE_IN_LISTS \ 211 | RE_INTERVALS | RE_NO_GNU_OPS) 212 213 #define RE_SYNTAX_GREP \ 214 (RE_BK_PLUS_QM | RE_CHAR_CLASSES \ 215 | RE_HAT_LISTS_NOT_NEWLINE | RE_INTERVALS \ 216 | RE_NEWLINE_ALT) 217 218 #define RE_SYNTAX_EGREP \ 219 (RE_CHAR_CLASSES | RE_CONTEXT_INDEP_ANCHORS \ 220 | RE_CONTEXT_INDEP_OPS | RE_HAT_LISTS_NOT_NEWLINE \ 221 | RE_NEWLINE_ALT | RE_NO_BK_PARENS \ 222 | RE_NO_BK_VBAR) 223 224 #define RE_SYNTAX_POSIX_EGREP \ 225 (RE_SYNTAX_EGREP | RE_INTERVALS | RE_NO_BK_BRACES \ 226 | RE_INVALID_INTERVAL_ORD) 227 228 /* P1003.2/D11.2, section 4.20.7.1, lines 5078ff. */ 229 #define RE_SYNTAX_ED RE_SYNTAX_POSIX_BASIC 230 231 #define RE_SYNTAX_SED RE_SYNTAX_POSIX_BASIC 232 233 /* Syntax bits common to both basic and extended POSIX regex syntax. */ 234 #define _RE_SYNTAX_POSIX_COMMON \ 235 (RE_CHAR_CLASSES | RE_DOT_NEWLINE | RE_DOT_NOT_NULL \ 236 | RE_INTERVALS | RE_NO_EMPTY_RANGES) 237 238 #ifdef __OPTION_POSIX_REGEXP_GLIBC 239 #define RE_SYNTAX_POSIX_BASIC \ 240 (_RE_SYNTAX_POSIX_COMMON | RE_BK_PLUS_QM | RE_CONTEXT_INVALID_DUP) 241 #else 242 #define RE_SYNTAX_POSIX_BASIC \ 243 (_RE_SYNTAX_POSIX_COMMON | RE_BK_PLUS_QM) 244 #endif 245 246 /* Differs from ..._POSIX_BASIC only in that RE_BK_PLUS_QM becomes 247 RE_LIMITED_OPS, i.e., \? \+ \| are not recognized. Actually, this 248 isn't minimal, since other operators, such as \`, aren't disabled. */ 249 #define RE_SYNTAX_POSIX_MINIMAL_BASIC \ 250 (_RE_SYNTAX_POSIX_COMMON | RE_LIMITED_OPS) 251 252 #define RE_SYNTAX_POSIX_EXTENDED \ 253 (_RE_SYNTAX_POSIX_COMMON | RE_CONTEXT_INDEP_ANCHORS \ 254 | RE_CONTEXT_INDEP_OPS | RE_NO_BK_BRACES \ 255 | RE_NO_BK_PARENS | RE_NO_BK_VBAR \ 256 | RE_CONTEXT_INVALID_OPS | RE_UNMATCHED_RIGHT_PAREN_ORD) 257 258 /* Differs from ..._POSIX_EXTENDED in that RE_CONTEXT_INDEP_OPS is 259 removed and RE_NO_BK_REFS is added. */ 260 #define RE_SYNTAX_POSIX_MINIMAL_EXTENDED \ 261 (_RE_SYNTAX_POSIX_COMMON | RE_CONTEXT_INDEP_ANCHORS \ 262 | RE_CONTEXT_INVALID_OPS | RE_NO_BK_BRACES \ 263 | RE_NO_BK_PARENS | RE_NO_BK_REFS \ 264 | RE_NO_BK_VBAR | RE_UNMATCHED_RIGHT_PAREN_ORD) 265 /* [[[end syntaxes]]] */ 266 267 /* Maximum number of duplicates an interval can allow. Some systems 269 (erroneously) define this in other header files, but we want our 270 value, so remove any previous define. */ 271 # ifdef RE_DUP_MAX 272 # undef RE_DUP_MAX 273 # endif 274 /* If sizeof(int) == 2, then ((1 << 15) - 1) overflows. */ 275 # define RE_DUP_MAX (0x7fff) 276 #endif 277 278 279 /* POSIX `cflags' bits (i.e., information for `regcomp'). */ 280 281 /* If this bit is set, then use extended regular expression syntax. 282 If not set, then use basic regular expression syntax. */ 283 #define REG_EXTENDED 1 284 285 /* If this bit is set, then ignore case when matching. 286 If not set, then case is significant. */ 287 #define REG_ICASE (REG_EXTENDED << 1) 288 289 /* If this bit is set, then anchors do not match at newline 290 characters in the string. 291 If not set, then anchors do match at newlines. */ 292 #define REG_NEWLINE (REG_ICASE << 1) 293 294 /* If this bit is set, then report only success or fail in regexec. 295 If not set, then returns differ between not matching and errors. */ 296 #define REG_NOSUB (REG_NEWLINE << 1) 297 298 299 /* POSIX `eflags' bits (i.e., information for regexec). */ 300 301 /* If this bit is set, then the beginning-of-line operator doesn't match 302 the beginning of the string (presumably because it's not the 303 beginning of a line). 304 If not set, then the beginning-of-line operator does match the 305 beginning of the string. */ 306 #define REG_NOTBOL 1 307 308 /* Like REG_NOTBOL, except for the end-of-line. */ 309 #define REG_NOTEOL (1 << 1) 310 311 #ifdef __OPTION_POSIX_REGEXP_GLIBC 312 /* Use PMATCH[0] to delimit the start and end of the search in the 313 buffer. */ 314 #define REG_STARTEND (1 << 2) 315 #endif 316 317 318 /* If any error codes are removed, changed, or added, update the 319 `re_error_msg' table in regex.c. */ 320 typedef enum 321 { 322 #if defined _XOPEN_SOURCE || defined __USE_XOPEN2K 323 REG_ENOSYS = -1, /* This will never happen for this implementation. */ 324 #endif 325 326 REG_NOERROR = 0, /* Success. */ 327 REG_NOMATCH, /* Didn't find a match (for regexec). */ 328 329 /* POSIX regcomp return error codes. (In the order listed in the 330 standard.) */ 331 REG_BADPAT, /* Invalid pattern. */ 332 REG_ECOLLATE, /* Inalid collating element. */ 333 REG_ECTYPE, /* Invalid character class name. */ 334 REG_EESCAPE, /* Trailing backslash. */ 335 REG_ESUBREG, /* Invalid back reference. */ 336 REG_EBRACK, /* Unmatched left bracket. */ 337 REG_EPAREN, /* Parenthesis imbalance. */ 338 REG_EBRACE, /* Unmatched \{. */ 339 REG_BADBR, /* Invalid contents of \{\}. */ 340 REG_ERANGE, /* Invalid range end. */ 341 REG_ESPACE, /* Ran out of memory. */ 342 REG_BADRPT, /* No preceding re for repetition op. */ 343 344 /* Error codes we've added. */ 345 REG_EEND, /* Premature end. */ 346 REG_ESIZE, /* Compiled pattern bigger than 2^16 bytes. */ 347 REG_ERPAREN /* Unmatched ) or \); not returned from regcomp. */ 348 } reg_errcode_t; 349 350 /* This data structure represents a compiled pattern. Before calling 352 the pattern compiler, the fields `buffer', `allocated', `fastmap', 353 `translate', and `no_sub' can be set. After the pattern has been 354 compiled, the `re_nsub' field is available. All other fields are 355 private to the regex routines. */ 356 357 #ifndef RE_TRANSLATE_TYPE 358 # define __RE_TRANSLATE_TYPE unsigned char * 359 # ifdef __USE_GNU 360 # define RE_TRANSLATE_TYPE __RE_TRANSLATE_TYPE 361 # endif 362 #endif 363 364 #ifdef __USE_GNU 365 # define __REPB_PREFIX(name) name 366 #else 367 # define __REPB_PREFIX(name) __##name 368 #endif 369 370 struct re_pattern_buffer 371 { 372 /* Space that holds the compiled pattern. It is declared as 373 `unsigned char *' because its elements are sometimes used as 374 array indexes. */ 375 unsigned char *__REPB_PREFIX(buffer); 376 377 /* Number of bytes to which `buffer' points. */ 378 unsigned long int __REPB_PREFIX(allocated); 379 380 /* Number of bytes actually used in `buffer'. */ 381 unsigned long int __REPB_PREFIX(used); 382 383 /* Syntax setting with which the pattern was compiled. */ 384 reg_syntax_t __REPB_PREFIX(syntax); 385 386 /* Pointer to a fastmap, if any, otherwise zero. re_search uses the 387 fastmap, if there is one, to skip over impossible starting points 388 for matches. */ 389 char *__REPB_PREFIX(fastmap); 390 391 /* Either a translate table to apply to all characters before 392 comparing them, or zero for no translation. The translation is 393 applied to a pattern when it is compiled and to a string when it 394 is matched. */ 395 __RE_TRANSLATE_TYPE __REPB_PREFIX(translate); 396 397 /* Number of subexpressions found by the compiler. */ 398 size_t re_nsub; 399 400 /* Zero if this pattern cannot match the empty string, one else. 401 Well, in truth it's used only in `re_search_2', to see whether or 402 not we should use the fastmap, so we don't set this absolutely 403 perfectly; see `re_compile_fastmap' (the `duplicate' case). */ 404 unsigned __REPB_PREFIX(can_be_null) : 1; 405 406 /* If REGS_UNALLOCATED, allocate space in the `regs' structure 407 for `max (RE_NREGS, re_nsub + 1)' groups. 408 If REGS_REALLOCATE, reallocate space if necessary. 409 If REGS_FIXED, use what's there. */ 410 #ifdef __USE_GNU 411 # define REGS_UNALLOCATED 0 412 # define REGS_REALLOCATE 1 413 # define REGS_FIXED 2 414 #endif 415 unsigned __REPB_PREFIX(regs_allocated) : 2; 416 417 /* Set to zero when `regex_compile' compiles a pattern; set to one 418 by `re_compile_fastmap' if it updates the fastmap. */ 419 unsigned __REPB_PREFIX(fastmap_accurate) : 1; 420 421 /* If set, `re_match_2' does not return information about 422 subexpressions. */ 423 unsigned __REPB_PREFIX(no_sub) : 1; 424 425 /* If set, a beginning-of-line anchor doesn't match at the beginning 426 of the string. */ 427 unsigned __REPB_PREFIX(not_bol) : 1; 428 429 /* Similarly for an end-of-line anchor. */ 430 unsigned __REPB_PREFIX(not_eol) : 1; 431 432 /* If true, an anchor at a newline matches. */ 433 unsigned __REPB_PREFIX(newline_anchor) : 1; 434 }; 435 436 typedef struct re_pattern_buffer regex_t; 437 438 /* Type for byte offsets within the string. POSIX mandates this. */ 440 typedef int regoff_t; 441 442 443 #ifdef __USE_GNU 444 /* This is the structure we store register match data in. See 445 regex.texinfo for a full description of what registers match. */ 446 struct re_registers 447 { 448 unsigned num_regs; 449 regoff_t *start; 450 regoff_t *end; 451 }; 452 453 454 /* If `regs_allocated' is REGS_UNALLOCATED in the pattern buffer, 455 `re_match_2' returns information about at least this many registers 456 the first time a `regs' structure is passed. */ 457 # ifndef RE_NREGS 458 # define RE_NREGS 30 459 # endif 460 #endif 461 462 463 /* POSIX specification for registers. Aside from the different names than 464 `re_registers', POSIX uses an array of structures, instead of a 465 structure of arrays. */ 466 typedef struct 467 { 468 regoff_t rm_so; /* Byte offset from string's start to substring's start. */ 469 regoff_t rm_eo; /* Byte offset from string's start to substring's end. */ 470 } regmatch_t; 471 472 /* Declarations for routines. */ 474 475 #ifdef __USE_GNU 476 /* Sets the current default syntax to SYNTAX, and return the old syntax. 477 You can also simply assign to the `re_syntax_options' variable. */ 478 extern reg_syntax_t re_set_syntax (reg_syntax_t __syntax); 479 480 /* Compile the regular expression PATTERN, with length LENGTH 481 and syntax given by the global `re_syntax_options', into the buffer 482 BUFFER. Return NULL if successful, and an error string if not. */ 483 extern const char *re_compile_pattern (const char *__pattern, size_t __length, 484 struct re_pattern_buffer *__buffer); 485 486 487 /* Compile a fastmap for the compiled pattern in BUFFER; used to 488 accelerate searches. Return 0 if successful and -2 if was an 489 internal error. */ 490 extern int re_compile_fastmap (struct re_pattern_buffer *__buffer); 491 492 493 /* Search in the string STRING (with length LENGTH) for the pattern 494 compiled into BUFFER. Start searching at position START, for RANGE 495 characters. Return the starting position of the match, -1 for no 496 match, or -2 for an internal error. Also return register 497 information in REGS (if REGS and BUFFER->no_sub are nonzero). */ 498 extern int re_search (struct re_pattern_buffer *__buffer, const char *__string, 499 int __length, int __start, int __range, 500 struct re_registers *__regs); 501 502 503 /* Like `re_search', but search in the concatenation of STRING1 and 504 STRING2. Also, stop searching at index START + STOP. */ 505 extern int re_search_2 (struct re_pattern_buffer *__buffer, 506 const char *__string1, int __length1, 507 const char *__string2, int __length2, int __start, 508 int __range, struct re_registers *__regs, int __stop); 509 510 511 /* Like `re_search', but return how many characters in STRING the regexp 512 in BUFFER matched, starting at position START. */ 513 extern int re_match (struct re_pattern_buffer *__buffer, const char *__string, 514 int __length, int __start, struct re_registers *__regs); 515 516 517 /* Relates to `re_match' as `re_search_2' relates to `re_search'. */ 518 extern int re_match_2 (struct re_pattern_buffer *__buffer, 519 const char *__string1, int __length1, 520 const char *__string2, int __length2, int __start, 521 struct re_registers *__regs, int __stop); 522 523 524 /* Set REGS to hold NUM_REGS registers, storing them in STARTS and 525 ENDS. Subsequent matches using BUFFER and REGS will use this memory 526 for recording register information. STARTS and ENDS must be 527 allocated with malloc, and must each be at least `NUM_REGS * sizeof 528 (regoff_t)' bytes long. 529 530 If NUM_REGS == 0, then subsequent matches should allocate their own 531 register data. 532 533 Unless this function is called, the first search or match using 534 PATTERN_BUFFER will allocate its own register data, without 535 freeing the old data. */ 536 extern void re_set_registers (struct re_pattern_buffer *__buffer, 537 struct re_registers *__regs, 538 unsigned int __num_regs, 539 regoff_t *__starts, regoff_t *__ends); 540 #endif /* Use GNU */ 541 542 #if defined _REGEX_RE_COMP || (defined _LIBC && defined __USE_BSD) 543 # ifndef _CRAY 544 /* 4.2 bsd compatibility. */ 545 extern char *re_comp (const char *); 546 extern int re_exec (const char *); 547 # endif 548 #endif 549 550 /* GCC 2.95 and later have "__restrict"; C99 compilers have 551 "restrict", and "configure" may have defined "restrict". */ 552 #ifndef __restrict 553 # if ! (2 < __GNUC__ || (2 == __GNUC__ && 95 <= __GNUC_MINOR__)) 554 # if defined restrict || 199901L <= __STDC_VERSION__ 555 # define __restrict restrict 556 # else 557 # define __restrict 558 # endif 559 # endif 560 #endif 561 /* gcc 3.1 and up support the [restrict] syntax. */ 562 #ifndef __restrict_arr 563 # if (__GNUC__ > 3 || (__GNUC__ == 3 && __GNUC_MINOR__ >= 1)) \ 564 && !defined __GNUG__ 565 # define __restrict_arr __restrict 566 # else 567 # define __restrict_arr 568 # endif 569 #endif 570 571 /* POSIX compatibility. */ 572 extern int regcomp (regex_t *__restrict __preg, 573 const char *__restrict __pattern, 574 int __cflags); 575 576 extern int regexec (const regex_t *__restrict __preg, 577 const char *__restrict __string, size_t __nmatch, 578 regmatch_t __pmatch[__restrict_arr], 579 int __eflags); 580 581 extern size_t regerror (int __errcode, const regex_t *__restrict __preg, 582 char *__restrict __errbuf, size_t __errbuf_size); 583 584 extern void regfree (regex_t *__preg); 585 586 587 #ifdef __cplusplus 588 } 589 #endif /* C++ */ 590 591 #endif /* regex.h */ 592
