1 /* -*- buffer-read-only: t -*- vi: set ro: */ 2 /* DO NOT EDIT! GENERATED AUTOMATICALLY! */ 3 /* Definitions for data structures and routines for the regular 4 expression library. 5 Copyright (C) 1985,1989-93,1995-98,2000,2001,2002,2003,2005,2006 6 Free Software Foundation, Inc. 7 This file is part of the GNU C Library. 8 9 This program is free software; you can redistribute it and/or modify 10 it under the terms of the GNU General Public License as published by 11 the Free Software Foundation; either version 3, or (at your option) 12 any later version. 13 14 This program is distributed in the hope that it will be useful, 15 but WITHOUT ANY WARRANTY; without even the implied warranty of 16 MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the 17 GNU General Public License for more details. 18 19 You should have received a copy of the GNU General Public License along 20 with this program; if not, write to the Free Software Foundation, 21 Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. */ 22 23 #ifndef _REGEX_H 24 #define _REGEX_H 1 25 26 #include <sys/types.h> 27 28 /* Allow the use in C++ code. */ 29 #ifdef __cplusplus 30 extern "C" { 31 #endif 32 33 /* Define __USE_GNU_REGEX to declare GNU extensions that violate the 34 POSIX name space rules. */ 35 #undef __USE_GNU_REGEX 36 #if (defined _GNU_SOURCE \ 37 || (!defined _POSIX_C_SOURCE && !defined _POSIX_SOURCE \ 38 && !defined _XOPEN_SOURCE)) 39 # define __USE_GNU_REGEX 1 40 #endif 41 42 #ifdef _REGEX_LARGE_OFFSETS 43 44 /* Use types and values that are wide enough to represent signed and 45 unsigned byte offsets in memory. This currently works only when 46 the regex code is used outside of the GNU C library; it is not yet 47 supported within glibc itself, and glibc users should not define 48 _REGEX_LARGE_OFFSETS. */ 49 50 /* The type of the offset of a byte within a string. 51 For historical reasons POSIX 1003.1-2004 requires that regoff_t be 52 at least as wide as off_t. However, many common POSIX platforms set 53 regoff_t to the more-sensible ssize_t and the Open Group has 54 signalled its intention to change the requirement to be that 55 regoff_t be at least as wide as ptrdiff_t and ssize_t; see XBD ERN 56 60 (2005-08-25). We don't know of any hosts where ssize_t or 57 ptrdiff_t is wider than ssize_t, so ssize_t is safe. */ 58 typedef ssize_t regoff_t; 59 60 /* The type of nonnegative object indexes. Traditionally, GNU regex 61 uses 'int' for these. Code that uses __re_idx_t should work 62 regardless of whether the type is signed. */ 63 typedef size_t __re_idx_t; 64 65 /* The type of object sizes. */ 66 typedef size_t __re_size_t; 67 68 /* The type of object sizes, in places where the traditional code 69 uses unsigned long int. */ 70 typedef size_t __re_long_size_t; 71 72 #else 73 74 /* Use types that are binary-compatible with the traditional GNU regex 75 implementation, which mishandles strings longer than INT_MAX. */ 76 77 typedef int regoff_t; 78 typedef int __re_idx_t; 79 typedef unsigned int __re_size_t; 80 typedef unsigned long int __re_long_size_t; 81 82 #endif 83 84 /* The following two types have to be signed and unsigned integer type 85 wide enough to hold a value of a pointer. For most ANSI compilers 86 ptrdiff_t and size_t should be likely OK. Still size of these two 87 types is 2 for Microsoft C. Ugh... */ 88 typedef long int s_reg_t; 89 typedef unsigned long int active_reg_t; 90 91 /* The following bits are used to determine the regexp syntax we 92 recognize. The set/not-set meanings are chosen so that Emacs syntax 93 remains the value 0. The bits are given in alphabetical order, and 94 the definitions shifted by one from the previous bit; thus, when we 95 add or remove a bit, only one other definition need change. */ 96 typedef unsigned long int reg_syntax_t; 97 98 #ifdef __USE_GNU_REGEX 99 100 /* If this bit is not set, then \ inside a bracket expression is literal. 101 If set, then such a \ quotes the following character. */ 102 # define RE_BACKSLASH_ESCAPE_IN_LISTS ((unsigned long int) 1) 103 104 /* If this bit is not set, then + and ? are operators, and \+ and \? are 105 literals. 106 If set, then \+ and \? are operators and + and ? are literals. */ 107 # define RE_BK_PLUS_QM (RE_BACKSLASH_ESCAPE_IN_LISTS << 1) 108 109 /* If this bit is set, then character classes are supported. They are: 110 [:alpha:], [:upper:], [:lower:], [:digit:], [:alnum:], [:xdigit:], 111 [:space:], [:print:], [:punct:], [:graph:], and [:cntrl:]. 112 If not set, then character classes are not supported. */ 113 # define RE_CHAR_CLASSES (RE_BK_PLUS_QM << 1) 114 115 /* If this bit is set, then ^ and $ are always anchors (outside bracket 116 expressions, of course). 117 If this bit is not set, then it depends: 118 ^ is an anchor if it is at the beginning of a regular 119 expression or after an open-group or an alternation operator; 120 $ is an anchor if it is at the end of a regular expression, or 121 before a close-group or an alternation operator. 122 123 This bit could be (re)combined with RE_CONTEXT_INDEP_OPS, because 124 POSIX draft 11.2 says that * etc. in leading positions is undefined. 125 We already implemented a previous draft which made those constructs 126 invalid, though, so we haven't changed the code back. */ 127 # define RE_CONTEXT_INDEP_ANCHORS (RE_CHAR_CLASSES << 1) 128 129 /* If this bit is set, then special characters are always special 130 regardless of where they are in the pattern. 131 If this bit is not set, then special characters are special only in 132 some contexts; otherwise they are ordinary. Specifically, 133 * + ? and intervals are only special when not after the beginning, 134 open-group, or alternation operator. */ 135 # define RE_CONTEXT_INDEP_OPS (RE_CONTEXT_INDEP_ANCHORS << 1) 136 137 /* If this bit is set, then *, +, ?, and { cannot be first in an re or 138 immediately after an alternation or begin-group operator. */ 139 # define RE_CONTEXT_INVALID_OPS (RE_CONTEXT_INDEP_OPS << 1) 140 141 /* If this bit is set, then . matches newline. 142 If not set, then it doesn't. */ 143 # define RE_DOT_NEWLINE (RE_CONTEXT_INVALID_OPS << 1) 144 145 /* If this bit is set, then . doesn't match NUL. 146 If not set, then it does. */ 147 # define RE_DOT_NOT_NULL (RE_DOT_NEWLINE << 1) 148 149 /* If this bit is set, nonmatching lists [^...] do not match newline. 150 If not set, they do. */ 151 # define RE_HAT_LISTS_NOT_NEWLINE (RE_DOT_NOT_NULL << 1) 152 153 /* If this bit is set, either \{...\} or {...} defines an 154 interval, depending on RE_NO_BK_BRACES. 155 If not set, \{, \}, {, and } are literals. */ 156 # define RE_INTERVALS (RE_HAT_LISTS_NOT_NEWLINE << 1) 157 158 /* If this bit is set, +, ? and | aren't recognized as operators. 159 If not set, they are. */ 160 # define RE_LIMITED_OPS (RE_INTERVALS << 1) 161 162 /* If this bit is set, newline is an alternation operator. 163 If not set, newline is literal. */ 164 # define RE_NEWLINE_ALT (RE_LIMITED_OPS << 1) 165 166 /* If this bit is set, then `{...}' defines an interval, and \{ and \} 167 are literals. 168 If not set, then `\{...\}' defines an interval. */ 169 # define RE_NO_BK_BRACES (RE_NEWLINE_ALT << 1) 170 171 /* If this bit is set, (...) defines a group, and \( and \) are literals. 172 If not set, \(...\) defines a group, and ( and ) are literals. */ 173 # define RE_NO_BK_PARENS (RE_NO_BK_BRACES << 1) 174 175 /* If this bit is set, then \<digit> matches <digit>. 176 If not set, then \<digit> is a back-reference. */ 177 # define RE_NO_BK_REFS (RE_NO_BK_PARENS << 1) 178 179 /* If this bit is set, then | is an alternation operator, and \| is literal. 180 If not set, then \| is an alternation operator, and | is literal. */ 181 # define RE_NO_BK_VBAR (RE_NO_BK_REFS << 1) 182 183 /* If this bit is set, then an ending range point collating higher 184 than the starting range point, as in [z-a], is invalid. 185 If not set, then when ending range point collates higher than the 186 starting range point, the range is ignored. */ 187 # define RE_NO_EMPTY_RANGES (RE_NO_BK_VBAR << 1) 188 189 /* If this bit is set, then an unmatched ) is ordinary. 190 If not set, then an unmatched ) is invalid. */ 191 # define RE_UNMATCHED_RIGHT_PAREN_ORD (RE_NO_EMPTY_RANGES << 1) 192 193 /* If this bit is set, succeed as soon as we match the whole pattern, 194 without further backtracking. */ 195 # define RE_NO_POSIX_BACKTRACKING (RE_UNMATCHED_RIGHT_PAREN_ORD << 1) 196 197 /* If this bit is set, do not process the GNU regex operators. 198 If not set, then the GNU regex operators are recognized. */ 199 # define RE_NO_GNU_OPS (RE_NO_POSIX_BACKTRACKING << 1) 200 201 /* If this bit is set, turn on internal regex debugging. 202 If not set, and debugging was on, turn it off. 203 This only works if regex.c is compiled -DDEBUG. 204 We define this bit always, so that all that's needed to turn on 205 debugging is to recompile regex.c; the calling code can always have 206 this bit set, and it won't affect anything in the normal case. */ 207 # define RE_DEBUG (RE_NO_GNU_OPS << 1) 208 209 /* If this bit is set, a syntactically invalid interval is treated as 210 a string of ordinary characters. For example, the ERE 'a{1' is 211 treated as 'a\{1'. */ 212 # define RE_INVALID_INTERVAL_ORD (RE_DEBUG << 1) 213 214 /* If this bit is set, then ignore case when matching. 215 If not set, then case is significant. */ 216 # define RE_ICASE (RE_INVALID_INTERVAL_ORD << 1) 217 218 /* This bit is used internally like RE_CONTEXT_INDEP_ANCHORS but only 219 for ^, because it is difficult to scan the regex backwards to find 220 whether ^ should be special. */ 221 # define RE_CARET_ANCHORS_HERE (RE_ICASE << 1) 222 223 /* If this bit is set, then \{ cannot be first in an bre or 224 immediately after an alternation or begin-group operator. */ 225 # define RE_CONTEXT_INVALID_DUP (RE_CARET_ANCHORS_HERE << 1) 226 227 /* If this bit is set, then no_sub will be set to 1 during 228 re_compile_pattern. */ 229 # define RE_NO_SUB (RE_CONTEXT_INVALID_DUP << 1) 230 231 #endif /* defined __USE_GNU_REGEX */ 232 233 /* This global variable defines the particular regexp syntax to use (for 234 some interfaces). When a regexp is compiled, the syntax used is 235 stored in the pattern buffer, so changing this does not affect 236 already-compiled regexps. */ 237 extern reg_syntax_t re_syntax_options; 238 239 #ifdef __USE_GNU_REGEX 241 /* Define combinations of the above bits for the standard possibilities. 242 (The [[[ comments delimit what gets put into the Texinfo file, so 243 don't delete them!) */ 244 /* [[[begin syntaxes]]] */ 245 # define RE_SYNTAX_EMACS 0 246 247 # define RE_SYNTAX_AWK \ 248 (RE_BACKSLASH_ESCAPE_IN_LISTS | RE_DOT_NOT_NULL \ 249 | RE_NO_BK_PARENS | RE_NO_BK_REFS \ 250 | RE_NO_BK_VBAR | RE_NO_EMPTY_RANGES \ 251 | RE_DOT_NEWLINE | RE_CONTEXT_INDEP_ANCHORS \ 252 | RE_UNMATCHED_RIGHT_PAREN_ORD | RE_NO_GNU_OPS) 253 254 # define RE_SYNTAX_GNU_AWK \ 255 ((RE_SYNTAX_POSIX_EXTENDED | RE_BACKSLASH_ESCAPE_IN_LISTS | RE_DEBUG) \ 256 & ~(RE_DOT_NOT_NULL | RE_INTERVALS | RE_CONTEXT_INDEP_OPS \ 257 | RE_CONTEXT_INVALID_OPS )) 258 259 # define RE_SYNTAX_POSIX_AWK \ 260 (RE_SYNTAX_POSIX_EXTENDED | RE_BACKSLASH_ESCAPE_IN_LISTS \ 261 | RE_INTERVALS | RE_NO_GNU_OPS) 262 263 # define RE_SYNTAX_GREP \ 264 (RE_BK_PLUS_QM | RE_CHAR_CLASSES \ 265 | RE_HAT_LISTS_NOT_NEWLINE | RE_INTERVALS \ 266 | RE_NEWLINE_ALT) 267 268 # define RE_SYNTAX_EGREP \ 269 (RE_CHAR_CLASSES | RE_CONTEXT_INDEP_ANCHORS \ 270 | RE_CONTEXT_INDEP_OPS | RE_HAT_LISTS_NOT_NEWLINE \ 271 | RE_NEWLINE_ALT | RE_NO_BK_PARENS \ 272 | RE_NO_BK_VBAR) 273 274 # define RE_SYNTAX_POSIX_EGREP \ 275 (RE_SYNTAX_EGREP | RE_INTERVALS | RE_NO_BK_BRACES \ 276 | RE_INVALID_INTERVAL_ORD) 277 278 /* P1003.2/D11.2, section 4.20.7.1, lines 5078ff. */ 279 # define RE_SYNTAX_ED RE_SYNTAX_POSIX_BASIC 280 281 # define RE_SYNTAX_SED RE_SYNTAX_POSIX_BASIC 282 283 /* Syntax bits common to both basic and extended POSIX regex syntax. */ 284 # define _RE_SYNTAX_POSIX_COMMON \ 285 (RE_CHAR_CLASSES | RE_DOT_NEWLINE | RE_DOT_NOT_NULL \ 286 | RE_INTERVALS | RE_NO_EMPTY_RANGES) 287 288 # define RE_SYNTAX_POSIX_BASIC \ 289 (_RE_SYNTAX_POSIX_COMMON | RE_BK_PLUS_QM | RE_CONTEXT_INVALID_DUP) 290 291 /* Differs from ..._POSIX_BASIC only in that RE_BK_PLUS_QM becomes 292 RE_LIMITED_OPS, i.e., \? \+ \| are not recognized. Actually, this 293 isn't minimal, since other operators, such as \`, aren't disabled. */ 294 # define RE_SYNTAX_POSIX_MINIMAL_BASIC \ 295 (_RE_SYNTAX_POSIX_COMMON | RE_LIMITED_OPS) 296 297 # define RE_SYNTAX_POSIX_EXTENDED \ 298 (_RE_SYNTAX_POSIX_COMMON | RE_CONTEXT_INDEP_ANCHORS \ 299 | RE_CONTEXT_INDEP_OPS | RE_NO_BK_BRACES \ 300 | RE_NO_BK_PARENS | RE_NO_BK_VBAR \ 301 | RE_CONTEXT_INVALID_OPS | RE_UNMATCHED_RIGHT_PAREN_ORD) 302 303 /* Differs from ..._POSIX_EXTENDED in that RE_CONTEXT_INDEP_OPS is 304 removed and RE_NO_BK_REFS is added. */ 305 # define RE_SYNTAX_POSIX_MINIMAL_EXTENDED \ 306 (_RE_SYNTAX_POSIX_COMMON | RE_CONTEXT_INDEP_ANCHORS \ 307 | RE_CONTEXT_INVALID_OPS | RE_NO_BK_BRACES \ 308 | RE_NO_BK_PARENS | RE_NO_BK_REFS \ 309 | RE_NO_BK_VBAR | RE_UNMATCHED_RIGHT_PAREN_ORD) 310 /* [[[end syntaxes]]] */ 311 312 #endif /* defined __USE_GNU_REGEX */ 313 314 #ifdef __USE_GNU_REGEX 316 317 /* Maximum number of duplicates an interval can allow. POSIX-conforming 318 systems might define this in <limits.h>, but we want our 319 value, so remove any previous define. */ 320 # ifdef RE_DUP_MAX 321 # undef RE_DUP_MAX 322 # endif 323 324 /* RE_DUP_MAX is 2**15 - 1 because an earlier implementation stored 325 the counter as a 2-byte signed integer. This is no longer true, so 326 RE_DUP_MAX could be increased to (INT_MAX / 10 - 1), or to 327 ((SIZE_MAX - 2) / 10 - 1) if _REGEX_LARGE_OFFSETS is defined. 328 However, there would be a huge performance problem if someone 329 actually used a pattern like a\{214748363\}, so RE_DUP_MAX retains 330 its historical value. */ 331 # define RE_DUP_MAX (0x7fff) 332 333 #endif /* defined __USE_GNU_REGEX */ 334 335 336 /* POSIX `cflags' bits (i.e., information for `regcomp'). */ 337 338 /* If this bit is set, then use extended regular expression syntax. 339 If not set, then use basic regular expression syntax. */ 340 #define REG_EXTENDED 1 341 342 /* If this bit is set, then ignore case when matching. 343 If not set, then case is significant. */ 344 #define REG_ICASE (1 << 1) 345 346 /* If this bit is set, then anchors do not match at newline 347 characters in the string. 348 If not set, then anchors do match at newlines. */ 349 #define REG_NEWLINE (1 << 2) 350 351 /* If this bit is set, then report only success or fail in regexec. 352 If not set, then returns differ between not matching and errors. */ 353 #define REG_NOSUB (1 << 3) 354 355 356 /* POSIX `eflags' bits (i.e., information for regexec). */ 357 358 /* If this bit is set, then the beginning-of-line operator doesn't match 359 the beginning of the string (presumably because it's not the 360 beginning of a line). 361 If not set, then the beginning-of-line operator does match the 362 beginning of the string. */ 363 #define REG_NOTBOL 1 364 365 /* Like REG_NOTBOL, except for the end-of-line. */ 366 #define REG_NOTEOL (1 << 1) 367 368 /* Use PMATCH[0] to delimit the start and end of the search in the 369 buffer. */ 370 #define REG_STARTEND (1 << 2) 371 372 373 /* If any error codes are removed, changed, or added, update the 374 `__re_error_msgid' table in regcomp.c. */ 375 376 typedef enum 377 { 378 _REG_ENOSYS = -1, /* This will never happen for this implementation. */ 379 _REG_NOERROR = 0, /* Success. */ 380 _REG_NOMATCH, /* Didn't find a match (for regexec). */ 381 382 /* POSIX regcomp return error codes. (In the order listed in the 383 standard.) */ 384 _REG_BADPAT, /* Invalid pattern. */ 385 _REG_ECOLLATE, /* Invalid collating element. */ 386 _REG_ECTYPE, /* Invalid character class name. */ 387 _REG_EESCAPE, /* Trailing backslash. */ 388 _REG_ESUBREG, /* Invalid back reference. */ 389 _REG_EBRACK, /* Unmatched left bracket. */ 390 _REG_EPAREN, /* Parenthesis imbalance. */ 391 _REG_EBRACE, /* Unmatched \{. */ 392 _REG_BADBR, /* Invalid contents of \{\}. */ 393 _REG_ERANGE, /* Invalid range end. */ 394 _REG_ESPACE, /* Ran out of memory. */ 395 _REG_BADRPT, /* No preceding re for repetition op. */ 396 397 /* Error codes we've added. */ 398 _REG_EEND, /* Premature end. */ 399 _REG_ESIZE, /* Compiled pattern bigger than 2^16 bytes. */ 400 _REG_ERPAREN /* Unmatched ) or \); not returned from regcomp. */ 401 } reg_errcode_t; 402 403 #ifdef _XOPEN_SOURCE 404 # define REG_ENOSYS _REG_ENOSYS 405 #endif 406 #define REG_NOERROR _REG_NOERROR 407 #define REG_NOMATCH _REG_NOMATCH 408 #define REG_BADPAT _REG_BADPAT 409 #define REG_ECOLLATE _REG_ECOLLATE 410 #define REG_ECTYPE _REG_ECTYPE 411 #define REG_EESCAPE _REG_EESCAPE 412 #define REG_ESUBREG _REG_ESUBREG 413 #define REG_EBRACK _REG_EBRACK 414 #define REG_EPAREN _REG_EPAREN 415 #define REG_EBRACE _REG_EBRACE 416 #define REG_BADBR _REG_BADBR 417 #define REG_ERANGE _REG_ERANGE 418 #define REG_ESPACE _REG_ESPACE 419 #define REG_BADRPT _REG_BADRPT 420 #define REG_EEND _REG_EEND 421 #define REG_ESIZE _REG_ESIZE 422 #define REG_ERPAREN _REG_ERPAREN 423 424 /* struct re_pattern_buffer normally uses member names like `buffer' 426 that POSIX does not allow. In POSIX mode these members have names 427 with leading `re_' (e.g., `re_buffer'). */ 428 #ifdef __USE_GNU_REGEX 429 # define _REG_RE_NAME(id) id 430 # define _REG_RM_NAME(id) id 431 #else 432 # define _REG_RE_NAME(id) re_##id 433 # define _REG_RM_NAME(id) rm_##id 434 #endif 435 436 /* The user can specify the type of the re_translate member by 437 defining the macro RE_TRANSLATE_TYPE, which defaults to unsigned 438 char *. This pollutes the POSIX name space, so in POSIX mode just 439 use unsigned char *. */ 440 #ifdef __USE_GNU_REGEX 441 # ifndef RE_TRANSLATE_TYPE 442 # define RE_TRANSLATE_TYPE unsigned char * 443 # endif 444 # define REG_TRANSLATE_TYPE RE_TRANSLATE_TYPE 445 #else 446 # define REG_TRANSLATE_TYPE unsigned char * 447 #endif 448 449 /* This data structure represents a compiled pattern. Before calling 450 the pattern compiler, the fields `buffer', `allocated', `fastmap', 451 `translate', and `no_sub' can be set. After the pattern has been 452 compiled, the `re_nsub' field is available. All other fields are 453 private to the regex routines. */ 454 455 struct re_pattern_buffer 456 { 457 /* Space that holds the compiled pattern. It is declared as 458 `unsigned char *' because its elements are sometimes used as 459 array indexes. */ 460 unsigned char *_REG_RE_NAME (buffer); 461 462 /* Number of bytes to which `buffer' points. */ 463 __re_long_size_t _REG_RE_NAME (allocated); 464 465 /* Number of bytes actually used in `buffer'. */ 466 __re_long_size_t _REG_RE_NAME (used); 467 468 /* Syntax setting with which the pattern was compiled. */ 469 reg_syntax_t _REG_RE_NAME (syntax); 470 471 /* Pointer to a fastmap, if any, otherwise zero. re_search uses the 472 fastmap, if there is one, to skip over impossible starting points 473 for matches. */ 474 char *_REG_RE_NAME (fastmap); 475 476 /* Either a translate table to apply to all characters before 477 comparing them, or zero for no translation. The translation is 478 applied to a pattern when it is compiled and to a string when it 479 is matched. */ 480 REG_TRANSLATE_TYPE _REG_RE_NAME (translate); 481 482 /* Number of subexpressions found by the compiler. */ 483 size_t re_nsub; 484 485 /* Zero if this pattern cannot match the empty string, one else. 486 Well, in truth it's used only in `re_search_2', to see whether or 487 not we should use the fastmap, so we don't set this absolutely 488 perfectly; see `re_compile_fastmap' (the `duplicate' case). */ 489 unsigned int _REG_RE_NAME (can_be_null) : 1; 490 491 /* If REGS_UNALLOCATED, allocate space in the `regs' structure 492 for `max (RE_NREGS, re_nsub + 1)' groups. 493 If REGS_REALLOCATE, reallocate space if necessary. 494 If REGS_FIXED, use what's there. */ 495 #ifdef __USE_GNU_REGEX 496 # define REGS_UNALLOCATED 0 497 # define REGS_REALLOCATE 1 498 # define REGS_FIXED 2 499 #endif 500 unsigned int _REG_RE_NAME (regs_allocated) : 2; 501 502 /* Set to zero when `regex_compile' compiles a pattern; set to one 503 by `re_compile_fastmap' if it updates the fastmap. */ 504 unsigned int _REG_RE_NAME (fastmap_accurate) : 1; 505 506 /* If set, `re_match_2' does not return information about 507 subexpressions. */ 508 unsigned int _REG_RE_NAME (no_sub) : 1; 509 510 /* If set, a beginning-of-line anchor doesn't match at the beginning 511 of the string. */ 512 unsigned int _REG_RE_NAME (not_bol) : 1; 513 514 /* Similarly for an end-of-line anchor. */ 515 unsigned int _REG_RE_NAME (not_eol) : 1; 516 517 /* If true, an anchor at a newline matches. */ 518 unsigned int _REG_RE_NAME (newline_anchor) : 1; 519 520 /* [[[end pattern_buffer]]] */ 521 }; 522 523 typedef struct re_pattern_buffer regex_t; 524 525 /* This is the structure we store register match data in. See 527 regex.texinfo for a full description of what registers match. */ 528 struct re_registers 529 { 530 __re_size_t _REG_RM_NAME (num_regs); 531 regoff_t *_REG_RM_NAME (start); 532 regoff_t *_REG_RM_NAME (end); 533 }; 534 535 536 /* If `regs_allocated' is REGS_UNALLOCATED in the pattern buffer, 537 `re_match_2' returns information about at least this many registers 538 the first time a `regs' structure is passed. */ 539 #if !defined RE_NREGS && defined __USE_GNU_REGEX 540 # define RE_NREGS 30 541 #endif 542 543 544 /* POSIX specification for registers. Aside from the different names than 545 `re_registers', POSIX uses an array of structures, instead of a 546 structure of arrays. */ 547 typedef struct 548 { 549 regoff_t rm_so; /* Byte offset from string's start to substring's start. */ 550 regoff_t rm_eo; /* Byte offset from string's start to substring's end. */ 551 } regmatch_t; 552 553 /* Declarations for routines. */ 555 556 /* Sets the current default syntax to SYNTAX, and return the old syntax. 557 You can also simply assign to the `re_syntax_options' variable. */ 558 extern reg_syntax_t re_set_syntax (reg_syntax_t __syntax); 559 560 /* Compile the regular expression PATTERN, with length LENGTH 561 and syntax given by the global `re_syntax_options', into the buffer 562 BUFFER. Return NULL if successful, and an error string if not. */ 563 extern const char *re_compile_pattern (const char *__pattern, size_t __length, 564 struct re_pattern_buffer *__buffer); 565 566 567 /* Compile a fastmap for the compiled pattern in BUFFER; used to 568 accelerate searches. Return 0 if successful and -2 if was an 569 internal error. */ 570 extern int re_compile_fastmap (struct re_pattern_buffer *__buffer); 571 572 573 /* Search in the string STRING (with length LENGTH) for the pattern 574 compiled into BUFFER. Start searching at position START, for RANGE 575 characters. Return the starting position of the match, -1 for no 576 match, or -2 for an internal error. Also return register 577 information in REGS (if REGS and BUFFER->no_sub are nonzero). */ 578 extern regoff_t re_search (struct re_pattern_buffer *__buffer, 579 const char *__string, __re_idx_t __length, 580 __re_idx_t __start, regoff_t __range, 581 struct re_registers *__regs); 582 583 584 /* Like `re_search', but search in the concatenation of STRING1 and 585 STRING2. Also, stop searching at index START + STOP. */ 586 extern regoff_t re_search_2 (struct re_pattern_buffer *__buffer, 587 const char *__string1, __re_idx_t __length1, 588 const char *__string2, __re_idx_t __length2, 589 __re_idx_t __start, regoff_t __range, 590 struct re_registers *__regs, 591 __re_idx_t __stop); 592 593 594 /* Like `re_search', but return how many characters in STRING the regexp 595 in BUFFER matched, starting at position START. */ 596 extern regoff_t re_match (struct re_pattern_buffer *__buffer, 597 const char *__string, __re_idx_t __length, 598 __re_idx_t __start, struct re_registers *__regs); 599 600 601 /* Relates to `re_match' as `re_search_2' relates to `re_search'. */ 602 extern regoff_t re_match_2 (struct re_pattern_buffer *__buffer, 603 const char *__string1, __re_idx_t __length1, 604 const char *__string2, __re_idx_t __length2, 605 __re_idx_t __start, struct re_registers *__regs, 606 __re_idx_t __stop); 607 608 609 /* Set REGS to hold NUM_REGS registers, storing them in STARTS and 610 ENDS. Subsequent matches using BUFFER and REGS will use this memory 611 for recording register information. STARTS and ENDS must be 612 allocated with malloc, and must each be at least `NUM_REGS * sizeof 613 (regoff_t)' bytes long. 614 615 If NUM_REGS == 0, then subsequent matches should allocate their own 616 register data. 617 618 Unless this function is called, the first search or match using 619 PATTERN_BUFFER will allocate its own register data, without 620 freeing the old data. */ 621 extern void re_set_registers (struct re_pattern_buffer *__buffer, 622 struct re_registers *__regs, 623 __re_size_t __num_regs, 624 regoff_t *__starts, regoff_t *__ends); 625 626 #if defined _REGEX_RE_COMP || defined _LIBC 627 # ifndef _CRAY 628 /* 4.2 bsd compatibility. */ 629 extern char *re_comp (const char *); 630 extern int re_exec (const char *); 631 # endif 632 #endif 633 634 /* GCC 2.95 and later have "__restrict"; C99 compilers have 635 "restrict", and "configure" may have defined "restrict". 636 Other compilers use __restrict, __restrict__, and _Restrict, and 637 'configure' might #define 'restrict' to those words, so pick a 638 different name. */ 639 #ifndef _Restrict_ 640 # if 199901L <= __STDC_VERSION__ 641 # define _Restrict_ restrict 642 # elif 2 < __GNUC__ || (2 == __GNUC__ && 95 <= __GNUC_MINOR__) 643 # define _Restrict_ __restrict 644 # else 645 # define _Restrict_ 646 # endif 647 #endif 648 /* gcc 3.1 and up support the [restrict] syntax. Don't trust 649 sys/cdefs.h's definition of __restrict_arr, though, as it 650 mishandles gcc -ansi -pedantic. */ 651 #ifndef _Restrict_arr_ 652 # if ((199901L <= __STDC_VERSION__ \ 653 || ((3 < __GNUC__ || (3 == __GNUC__ && 1 <= __GNUC_MINOR__)) \ 654 && !__STRICT_ANSI__)) \ 655 && !defined __GNUG__) 656 # define _Restrict_arr_ _Restrict_ 657 # else 658 # define _Restrict_arr_ 659 # endif 660 #endif 661 662 /* POSIX compatibility. */ 663 extern int regcomp (regex_t *_Restrict_ __preg, 664 const char *_Restrict_ __pattern, 665 int __cflags); 666 667 extern int regexec (const regex_t *_Restrict_ __preg, 668 const char *_Restrict_ __string, size_t __nmatch, 669 regmatch_t __pmatch[_Restrict_arr_], 670 int __eflags); 671 672 extern size_t regerror (int __errcode, const regex_t *_Restrict_ __preg, 673 char *_Restrict_ __errbuf, size_t __errbuf_size); 674 675 extern void regfree (regex_t *__preg); 676 677 678 #ifdef __cplusplus 679 } 680 #endif /* C++ */ 681 682 #endif /* regex.h */ 683
