include/regex.h

   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, see
  19    <http://www.gnu.org/licenses/>.  */
  20
  21 #ifndef _REGEX_H
  22 #define _REGEX_H 1
  23
  24 #include <sys/types.h>
  25
  26 __BEGIN_DECLS
  27
  28 /* The following two types have to be signed and unsigned integer type
  29    wide enough to hold a value of a pointer.  For most ANSI compilers
  30    ptrdiff_t and size_t should be likely OK.  Still size of these two
  31    types is 2 for Microsoft C.  Ugh... */
  32 typedef long int s_reg_t;
  33 typedef unsigned long int active_reg_t;
  34
  35 /* The following bits are used to determine the regexp syntax we
  36    recognize.  The set/not-set meanings are chosen so that Emacs syntax
  37    remains the value 0.  The bits are given in alphabetical order, and
  38    the definitions shifted by one from the previous bit; thus, when we
  39    add or remove a bit, only one other definition need change.  */
  40 typedef unsigned long int reg_syntax_t;
  41
  42 #ifdef __USE_GNU
  43 /* If this bit is not set, then \ inside a bracket expression is literal.
  44    If set, then such a \ quotes the following character.  */
  45 # define RE_BACKSLASH_ESCAPE_IN_LISTS ((unsigned long int) 1)
  46
  47 /* If this bit is not set, then + and ? are operators, and \+ and \? are
  48      literals.
  49    If set, then \+ and \? are operators and + and ? are literals.  */
  50 # define RE_BK_PLUS_QM (RE_BACKSLASH_ESCAPE_IN_LISTS << 1)
  51
  52 /* If this bit is set, then character classes are supported.  They are:
  53      [:alpha:], [:upper:], [:lower:],  [:digit:], [:alnum:], [:xdigit:],
  54      [:space:], [:print:], [:punct:], [:graph:], and [:cntrl:].
  55    If not set, then character classes are not supported.  */
  56 # define RE_CHAR_CLASSES (RE_BK_PLUS_QM << 1)
  57
  58 /* If this bit is set, then ^ and $ are always anchors (outside bracket
  59      expressions, of course).
  60    If this bit is not set, then it depends:
  61         ^  is an anchor if it is at the beginning of a regular
  62            expression or after an open-group or an alternation operator;
  63         $  is an anchor if it is at the end of a regular expression, or
  64            before a close-group or an alternation operator.
  65
  66    This bit could be (re)combined with RE_CONTEXT_INDEP_OPS, because
  67    POSIX draft 11.2 says that * etc. in leading positions is undefined.
  68    We already implemented a previous draft which made those constructs
  69    invalid, though, so we haven't changed the code back.  */
  70 # define RE_CONTEXT_INDEP_ANCHORS (RE_CHAR_CLASSES << 1)
  71
  72 /* If this bit is set, then special characters are always special
  73      regardless of where they are in the pattern.
  74    If this bit is not set, then special characters are special only in
  75      some contexts; otherwise they are ordinary.  Specifically,
  76      * + ? and intervals are only special when not after the beginning,
  77      open-group, or alternation operator.  */
  78 # define RE_CONTEXT_INDEP_OPS (RE_CONTEXT_INDEP_ANCHORS << 1)
  79
  80 /* If this bit is set, then *, +, ?, and { cannot be first in an re or
  81      immediately after an alternation or begin-group operator.  */
  82 # define RE_CONTEXT_INVALID_OPS (RE_CONTEXT_INDEP_OPS << 1)
  83
  84 /* If this bit is set, then . matches newline.
  85    If not set, then it doesn't.  */
  86 # define RE_DOT_NEWLINE (RE_CONTEXT_INVALID_OPS << 1)
  87
  88 /* If this bit is set, then . doesn't match NUL.
  89    If not set, then it does.  */
  90 # define RE_DOT_NOT_NULL (RE_DOT_NEWLINE << 1)
  91
  92 /* If this bit is set, nonmatching lists [^...] do not match newline.
  93    If not set, they do.  */
  94 # define RE_HAT_LISTS_NOT_NEWLINE (RE_DOT_NOT_NULL << 1)
  95
  96 /* If this bit is set, either \{...\} or {...} defines an
  97      interval, depending on RE_NO_BK_BRACES.
  98    If not set, \{, \}, {, and } are literals.  */
  99 # define RE_INTERVALS (RE_HAT_LISTS_NOT_NEWLINE << 1)
 100
 101 /* If this bit is set, +, ? and | aren't recognized as operators.
 102    If not set, they are.  */
 103 # define RE_LIMITED_OPS (RE_INTERVALS << 1)
 104
 105 /* If this bit is set, newline is an alternation operator.
 106    If not set, newline is literal.  */
 107 # define RE_NEWLINE_ALT (RE_LIMITED_OPS << 1)
 108
 109 /* If this bit is set, then `{...}' defines an interval, and \{ and \}
 110      are literals.
 111   If not set, then `\{...\}' defines an interval.  */
 112 # define RE_NO_BK_BRACES (RE_NEWLINE_ALT << 1)
 113
 114 /* If this bit is set, (...) defines a group, and \( and \) are literals.
 115    If not set, \(...\) defines a group, and ( and ) are literals.  */
 116 # define RE_NO_BK_PARENS (RE_NO_BK_BRACES << 1)
 117
 118 /* If this bit is set, then \<digit> matches <digit>.
 119    If not set, then \<digit> is a back-reference.  */
 120 # define RE_NO_BK_REFS (RE_NO_BK_PARENS << 1)
 121
 122 /* If this bit is set, then | is an alternation operator, and \| is literal.
 123    If not set, then \| is an alternation operator, and | is literal.  */
 124 # define RE_NO_BK_VBAR (RE_NO_BK_REFS << 1)
 125
 126 /* If this bit is set, then an ending range point collating higher
 127      than the starting range point, as in [z-a], is invalid.
 128    If not set, then when ending range point collates higher than the
 129      starting range point, the range is ignored.  */
 130 # define RE_NO_EMPTY_RANGES (RE_NO_BK_VBAR << 1)
 131
 132 /* If this bit is set, then an unmatched ) is ordinary.
 133    If not set, then an unmatched ) is invalid.  */
 134 # define RE_UNMATCHED_RIGHT_PAREN_ORD (RE_NO_EMPTY_RANGES << 1)
 135
 136 /* If this bit is set, succeed as soon as we match the whole pattern,
 137    without further backtracking.  */
 138 # define RE_NO_POSIX_BACKTRACKING (RE_UNMATCHED_RIGHT_PAREN_ORD << 1)
 139
 140 /* If this bit is set, do not process the GNU regex operators.
 141    If not set, then the GNU regex operators are recognized. */
 142 # define RE_NO_GNU_OPS (RE_NO_POSIX_BACKTRACKING << 1)
 143
 144 /* If this bit is set, turn on internal regex debugging.
 145    If not set, and debugging was on, turn it off.
 146    This only works if regex.c is compiled -DDEBUG.
 147    We define this bit always, so that all that's needed to turn on
 148    debugging is to recompile regex.c; the calling code can always have
 149    this bit set, and it won't affect anything in the normal case. */
 150 # define RE_DEBUG (RE_NO_GNU_OPS << 1)
 151
 152 /* If this bit is set, a syntactically invalid interval is treated as
 153    a string of ordinary characters.  For example, the ERE 'a{1' is
 154    treated as 'a\{1'.  */
 155 # define RE_INVALID_INTERVAL_ORD (RE_DEBUG << 1)
 156
 157 /* If this bit is set, then ignore case when matching.
 158    If not set, then case is significant.  */
 159 # define RE_ICASE (RE_INVALID_INTERVAL_ORD << 1)
 160
 161 /* This bit is used internally like RE_CONTEXT_INDEP_ANCHORS but only
 162    for ^, because it is difficult to scan the regex backwards to find
 163    whether ^ should be special.  */
 164 # define RE_CARET_ANCHORS_HERE (RE_ICASE << 1)
 165
 166 /* If this bit is set, then \{ cannot be first in an bre or
 167    immediately after an alternation or begin-group operator.  */
 168 # define RE_CONTEXT_INVALID_DUP (RE_CARET_ANCHORS_HERE << 1)
 169
 170 /* If this bit is set, then no_sub will be set to 1 during
 171    re_compile_pattern.  */
 172 # define RE_NO_SUB (RE_CONTEXT_INVALID_DUP << 1)
 173 #endif
 174
 175 /* This global variable defines the particular regexp syntax to use (for
 176    some interfaces).  When a regexp is compiled, the syntax used is
 177    stored in the pattern buffer, so changing this does not affect
 178    already-compiled regexps.  */
 179 extern reg_syntax_t re_syntax_options;
 180
 181 #ifdef __USE_GNU
 182 /* Define combinations of the above bits for the standard possibilities.
 183    (The [[[ comments delimit what gets put into the Texinfo file, so
 184    don't delete them!)  */
 185 /* [[[begin syntaxes]]] */
 186 #define RE_SYNTAX_EMACS 0
 187
 188 #define RE_SYNTAX_AWK                                                   \
 189   (RE_BACKSLASH_ESCAPE_IN_LISTS   | RE_DOT_NOT_NULL                     \
 190    | RE_NO_BK_PARENS              | RE_NO_BK_REFS                       \
 191    | RE_NO_BK_VBAR                | RE_NO_EMPTY_RANGES                  \
 192    | RE_DOT_NEWLINE               | RE_CONTEXT_INDEP_ANCHORS            \
 193    | RE_UNMATCHED_RIGHT_PAREN_ORD | RE_NO_GNU_OPS)
 194
 195 #define RE_SYNTAX_GNU_AWK                                               \
 196   ((RE_SYNTAX_POSIX_EXTENDED | RE_BACKSLASH_ESCAPE_IN_LISTS | RE_DEBUG) \
 197    & ~(RE_DOT_NOT_NULL | RE_INTERVALS | RE_CONTEXT_INDEP_OPS            \
 198        | RE_CONTEXT_INVALID_OPS ))
 199
 200 #define RE_SYNTAX_POSIX_AWK                                             \
 201   (RE_SYNTAX_POSIX_EXTENDED | RE_BACKSLASH_ESCAPE_IN_LISTS              \
 202    | RE_INTERVALS           | RE_NO_GNU_OPS)
 203
 204 #define RE_SYNTAX_GREP                                                  \
 205   (RE_BK_PLUS_QM              | RE_CHAR_CLASSES                         \
 206    | RE_HAT_LISTS_NOT_NEWLINE | RE_INTERVALS                            \
 207    | RE_NEWLINE_ALT)
 208
 209 #define RE_SYNTAX_EGREP                                                 \
 210   (RE_CHAR_CLASSES        | RE_CONTEXT_INDEP_ANCHORS                    \
 211    | RE_CONTEXT_INDEP_OPS | RE_HAT_LISTS_NOT_NEWLINE                    \
 212    | RE_NEWLINE_ALT       | RE_NO_BK_PARENS                             \
 213    | RE_NO_BK_VBAR)
 214
 215 #define RE_SYNTAX_POSIX_EGREP                                           \
 216   (RE_SYNTAX_EGREP | RE_INTERVALS | RE_NO_BK_BRACES                     \
 217    | RE_INVALID_INTERVAL_ORD)
 218
 219 /* P1003.2/D11.2, section 4.20.7.1, lines 5078ff.  */
 220 #define RE_SYNTAX_ED RE_SYNTAX_POSIX_BASIC
 221
 222 #define RE_SYNTAX_SED RE_SYNTAX_POSIX_BASIC
 223
 224 /* Syntax bits common to both basic and extended POSIX regex syntax.  */
 225 #define _RE_SYNTAX_POSIX_COMMON                                         \
 226   (RE_CHAR_CLASSES | RE_DOT_NEWLINE      | RE_DOT_NOT_NULL              \
 227    | RE_INTERVALS  | RE_NO_EMPTY_RANGES)
 228
 229 #define RE_SYNTAX_POSIX_BASIC                                           \
 230   (_RE_SYNTAX_POSIX_COMMON | RE_BK_PLUS_QM | RE_CONTEXT_INVALID_DUP)
 231
 232 /* Differs from ..._POSIX_BASIC only in that RE_BK_PLUS_QM becomes
 233    RE_LIMITED_OPS, i.e., \? \+ \| are not recognized.  Actually, this
 234    isn't minimal, since other operators, such as \`, aren't disabled.  */
 235 #define RE_SYNTAX_POSIX_MINIMAL_BASIC                                   \
 236   (_RE_SYNTAX_POSIX_COMMON | RE_LIMITED_OPS)
 237
 238 #define RE_SYNTAX_POSIX_EXTENDED                                        \
 239   (_RE_SYNTAX_POSIX_COMMON  | RE_CONTEXT_INDEP_ANCHORS                  \
 240    | RE_CONTEXT_INDEP_OPS   | RE_NO_BK_BRACES                           \
 241    | RE_NO_BK_PARENS        | RE_NO_BK_VBAR                             \
 242    | RE_CONTEXT_INVALID_OPS | RE_UNMATCHED_RIGHT_PAREN_ORD)
 243
 244 /* Differs from ..._POSIX_EXTENDED in that RE_CONTEXT_INDEP_OPS is
 245    removed and RE_NO_BK_REFS is added.  */
 246 #define RE_SYNTAX_POSIX_MINIMAL_EXTENDED                                \
 247   (_RE_SYNTAX_POSIX_COMMON  | RE_CONTEXT_INDEP_ANCHORS                  \
 248    | RE_CONTEXT_INVALID_OPS | RE_NO_BK_BRACES                           \
 249    | RE_NO_BK_PARENS        | RE_NO_BK_REFS                             \
 250    | RE_NO_BK_VBAR          | RE_UNMATCHED_RIGHT_PAREN_ORD)
 251 /* [[[end syntaxes]]] */
 252
 253 /* Maximum number of duplicates an interval can allow.  Some systems
 254    (erroneously) define this in other header files, but we want our
 255    value, so remove any previous define.  */
 256 # ifdef RE_DUP_MAX
 257 #  undef RE_DUP_MAX
 258 # endif
 259 /* If sizeof(int) == 2, then ((1 << 15) - 1) overflows.  */
 260 # define RE_DUP_MAX (0x7fff)
 261 #endif
 262
 263
 264 /* POSIX `cflags' bits (i.e., information for `regcomp').  */
 265
 266 /* If this bit is set, then use extended regular expression syntax.
 267    If not set, then use basic regular expression syntax.  */
 268 #define REG_EXTENDED 1
 269
 270 /* If this bit is set, then ignore case when matching.
 271    If not set, then case is significant.  */
 272 #define REG_ICASE (REG_EXTENDED << 1)
 273
 274 /* If this bit is set, then anchors do not match at newline
 275      characters in the string.
 276    If not set, then anchors do match at newlines.  */
 277 #define REG_NEWLINE (REG_ICASE << 1)
 278
 279 /* If this bit is set, then report only success or fail in regexec.
 280    If not set, then returns differ between not matching and errors.  */
 281 #define REG_NOSUB (REG_NEWLINE << 1)
 282
 283
 284 /* POSIX `eflags' bits (i.e., information for regexec).  */
 285
 286 /* If this bit is set, then the beginning-of-line operator doesn't match
 287      the beginning of the string (presumably because it's not the
 288      beginning of a line).
 289    If not set, then the beginning-of-line operator does match the
 290      beginning of the string.  */
 291 #define REG_NOTBOL 1
 292
 293 /* Like REG_NOTBOL, except for the end-of-line.  */
 294 #define REG_NOTEOL (1 << 1)
 295
 296 /* Use PMATCH[0] to delimit the start and end of the search in the
 297    buffer.  */
 298 #define REG_STARTEND (1 << 2)
 299
 300
 301 /* If any error codes are removed, changed, or added, update the
 302    `re_error_msg' table in regex.c.  */
 303 typedef enum
 304 {
 305 #if defined _XOPEN_SOURCE || defined __USE_XOPEN2K
 306   REG_ENOSYS = -1,      /* This will never happen for this implementation.  */
 307 #endif
 308
 309   REG_NOERROR = 0,      /* Success.  */
 310   REG_NOMATCH,          /* Didn't find a match (for regexec).  */
 311
 312   /* POSIX regcomp return error codes.  (In the order listed in the
 313      standard.)  */
 314   REG_BADPAT,           /* Invalid pattern.  */
 315   REG_ECOLLATE,         /* Invalid collating element.  */
 316   REG_ECTYPE,           /* Invalid character class name.  */
 317   REG_EESCAPE,          /* Trailing backslash.  */
 318   REG_ESUBREG,          /* Invalid back reference.  */
 319   REG_EBRACK,           /* Unmatched left bracket.  */
 320   REG_EPAREN,           /* Parenthesis imbalance.  */
 321   REG_EBRACE,           /* Unmatched \{.  */
 322   REG_BADBR,            /* Invalid contents of \{\}.  */
 323   REG_ERANGE,           /* Invalid range end.  */
 324   REG_ESPACE,           /* Ran out of memory.  */
 325   REG_BADRPT,           /* No preceding re for repetition op.  */
 326
 327   /* Error codes we've added.  */
 328   REG_EEND,             /* Premature end.  */
 329   REG_ESIZE,            /* Compiled pattern bigger than 2^16 bytes.  */
 330   REG_ERPAREN           /* Unmatched ) or \); not returned from regcomp.  */
 331 } reg_errcode_t;
 332
 333 /* This data structure represents a compiled pattern.  Before calling
 334    the pattern compiler, the fields `buffer', `allocated', `fastmap',
 335    `translate', and `no_sub' can be set.  After the pattern has been
 336    compiled, the `re_nsub' field is available.  All other fields are
 337    private to the regex routines.  */
 338
 339 #ifndef RE_TRANSLATE_TYPE
 340 # define __RE_TRANSLATE_TYPE unsigned char *
 341 # ifdef __USE_GNU
 342 #  define RE_TRANSLATE_TYPE __RE_TRANSLATE_TYPE
 343 # endif
 344 #endif
 345
 346 #ifdef __USE_GNU
 347 # define __REPB_PREFIX(name) name
 348 #else
 349 # define __REPB_PREFIX(name) __##name
 350 #endif
 351
 352 struct re_pattern_buffer
 353 {
 354   /* Space that holds the compiled pattern.  It is declared as
 355      `unsigned char *' because its elements are sometimes used as
 356      array indexes.  */
 357   unsigned char *__REPB_PREFIX(buffer);
 358
 359   /* Number of bytes to which `buffer' points.  */
 360   unsigned long int __REPB_PREFIX(allocated);
 361
 362   /* Number of bytes actually used in `buffer'.  */
 363   unsigned long int __REPB_PREFIX(used);
 364
 365   /* Syntax setting with which the pattern was compiled.  */
 366   reg_syntax_t __REPB_PREFIX(syntax);
 367
 368   /* Pointer to a fastmap, if any, otherwise zero.  re_search uses the
 369      fastmap, if there is one, to skip over impossible starting points
 370      for matches.  */
 371   char *__REPB_PREFIX(fastmap);
 372
 373   /* Either a translate table to apply to all characters before
 374      comparing them, or zero for no translation.  The translation is
 375      applied to a pattern when it is compiled and to a string when it
 376      is matched.  */
 377   __RE_TRANSLATE_TYPE __REPB_PREFIX(translate);
 378
 379   /* Number of subexpressions found by the compiler.  */
 380   size_t re_nsub;
 381
 382   /* Zero if this pattern cannot match the empty string, one else.
 383      Well, in truth it's used only in `re_search_2', to see whether or
 384      not we should use the fastmap, so we don't set this absolutely
 385      perfectly; see `re_compile_fastmap' (the `duplicate' case).  */
 386   unsigned __REPB_PREFIX(can_be_null) : 1;
 387
 388   /* If REGS_UNALLOCATED, allocate space in the `regs' structure
 389      for `max (RE_NREGS, re_nsub + 1)' groups.
 390      If REGS_REALLOCATE, reallocate space if necessary.
 391      If REGS_FIXED, use what's there.  */
 392 #ifdef __USE_GNU
 393 # define REGS_UNALLOCATED 0
 394 # define REGS_REALLOCATE 1
 395 # define REGS_FIXED 2
 396 #endif
 397   unsigned __REPB_PREFIX(regs_allocated) : 2;
 398
 399   /* Set to zero when `regex_compile' compiles a pattern; set to one
 400      by `re_compile_fastmap' if it updates the fastmap.  */
 401   unsigned __REPB_PREFIX(fastmap_accurate) : 1;
 402
 403   /* If set, `re_match_2' does not return information about
 404      subexpressions.  */
 405   unsigned __REPB_PREFIX(no_sub) : 1;
 406
 407   /* If set, a beginning-of-line anchor doesn't match at the beginning
 408      of the string.  */
 409   unsigned __REPB_PREFIX(not_bol) : 1;
 410
 411   /* Similarly for an end-of-line anchor.  */
 412   unsigned __REPB_PREFIX(not_eol) : 1;
 413
 414   /* If true, an anchor at a newline matches.  */
 415   unsigned __REPB_PREFIX(newline_anchor) : 1;
 416 };
 417
 418 typedef struct re_pattern_buffer regex_t;
 419
 420 /* Type for byte offsets within the string.  POSIX mandates this.  */
 421 typedef int regoff_t;
 422
 423
 424 #ifdef __USE_GNU
 425 /* This is the structure we store register match data in.  See
 426    regex.texinfo for a full description of what registers match.  */
 427 struct re_registers
 428 {
 429   unsigned num_regs;
 430   regoff_t *start;
 431   regoff_t *end;
 432 };
 433
 434
 435 /* If `regs_allocated' is REGS_UNALLOCATED in the pattern buffer,
 436    `re_match_2' returns information about at least this many registers
 437    the first time a `regs' structure is passed.  */
 438 # ifndef RE_NREGS
 439 #  define RE_NREGS 30
 440 # endif
 441 #endif
 442
 443
 444 /* POSIX specification for registers.  Aside from the different names than
 445    `re_registers', POSIX uses an array of structures, instead of a
 446    structure of arrays.  */
 447 typedef struct
 448 {
 449   regoff_t rm_so;  /* Byte offset from string's start to substring's start.  */
 450   regoff_t rm_eo;  /* Byte offset from string's start to substring's end.  */
 451 } regmatch_t;
 452
 453 /* Declarations for routines.  */
 454
 455 #ifdef __USE_GNU
 456 /* Sets the current default syntax to SYNTAX, and return the old syntax.
 457    You can also simply assign to the `re_syntax_options' variable.  */
 458 extern reg_syntax_t re_set_syntax (reg_syntax_t __syntax);
 459
 460 /* Compile the regular expression PATTERN, with length LENGTH
 461    and syntax given by the global `re_syntax_options', into the buffer
 462    BUFFER.  Return NULL if successful, and an error string if not.  */
 463 extern const char *re_compile_pattern (const char *__pattern, size_t __length,
 464                                        struct re_pattern_buffer *__buffer);
 465
 466
 467 /* Compile a fastmap for the compiled pattern in BUFFER; used to
 468    accelerate searches.  Return 0 if successful and -2 if was an
 469    internal error.  */
 470 extern int re_compile_fastmap (struct re_pattern_buffer *__buffer);
 471 libc_hidden_proto(re_compile_fastmap)
 472
 473
 474 /* Search in the string STRING (with length LENGTH) for the pattern
 475    compiled into BUFFER.  Start searching at position START, for RANGE
 476    characters.  Return the starting position of the match, -1 for no
 477    match, or -2 for an internal error.  Also return register
 478    information in REGS (if REGS and BUFFER->no_sub are nonzero).  */
 479 extern int re_search (struct re_pattern_buffer *__buffer, const char *__string,
 480                       int __length, int __start, int __range,
 481                       struct re_registers *__regs);
 482 libc_hidden_proto(re_search)
 483
 484
 485 /* Like `re_search', but search in the concatenation of STRING1 and
 486    STRING2.  Also, stop searching at index START + STOP.  */
 487 extern int re_search_2 (struct re_pattern_buffer *__buffer,
 488                         const char *__string1, int __length1,
 489                         const char *__string2, int __length2, int __start,
 490                         int __range, struct re_registers *__regs, int __stop);
 491 libc_hidden_proto(re_search_2)
 492
 493
 494 /* Like `re_search', but return how many characters in STRING the regexp
 495    in BUFFER matched, starting at position START.  */
 496 extern int re_match (struct re_pattern_buffer *__buffer, const char *__string,
 497                      int __length, int __start, struct re_registers *__regs);
 498
 499
 500 /* Relates to `re_match' as `re_search_2' relates to `re_search'.  */
 501 extern int re_match_2 (struct re_pattern_buffer *__buffer,
 502                        const char *__string1, int __length1,
 503                        const char *__string2, int __length2, int __start,
 504                        struct re_registers *__regs, int __stop);
 505
 506
 507 /* Set REGS to hold NUM_REGS registers, storing them in STARTS and
 508    ENDS.  Subsequent matches using BUFFER and REGS will use this memory
 509    for recording register information.  STARTS and ENDS must be
 510    allocated with malloc, and must each be at least `NUM_REGS * sizeof
 511    (regoff_t)' bytes long.
 512
 513    If NUM_REGS == 0, then subsequent matches should allocate their own
 514    register data.
 515
 516    Unless this function is called, the first search or match using
 517    PATTERN_BUFFER will allocate its own register data, without
 518    freeing the old data.  */
 519 extern void re_set_registers (struct re_pattern_buffer *__buffer,
 520                               struct re_registers *__regs,
 521                               unsigned int __num_regs,
 522                               regoff_t *__starts, regoff_t *__ends);
 523 #endif /* Use GNU */
 524
 525 #if defined _REGEX_RE_COMP || (defined _LIBC && defined __USE_BSD)
 526 # ifndef _CRAY
 527 /* 4.2 bsd compatibility.  */
 528 extern char *re_comp (const char *);
 529 extern int re_exec (const char *);
 530 # endif
 531 #endif
 532
 533 /* GCC 2.95 and later have "__restrict"; C99 compilers have
 534    "restrict", and "configure" may have defined "restrict".  */
 535 #ifndef __restrict
 536 # if ! (2 < __GNUC__ || (2 == __GNUC__ && 95 <= __GNUC_MINOR__))
 537 #  if defined restrict || 199901L <= __STDC_VERSION__
 538 #   define __restrict restrict
 539 #  else
 540 #   define __restrict
 541 #  endif
 542 # endif
 543 #endif
 544 /* gcc 3.1 and up support the [restrict] syntax.  */
 545 #ifndef __restrict_arr
 546 # if (__GNUC__ > 3 || (__GNUC__ == 3 && __GNUC_MINOR__ >= 1)) \
 547      && !defined __GNUG__
 548 #  define __restrict_arr __restrict
 549 # else
 550 #  define __restrict_arr
 551 # endif
 552 #endif
 553
 554 /* POSIX compatibility.  */
 555 extern int regcomp (regex_t *__restrict __preg,
 556                     const char *__restrict __pattern,
 557                     int __cflags);
 558
 559 extern int regexec (const regex_t *__restrict __preg,
 560                     const char *__restrict __string, size_t __nmatch,
 561                     regmatch_t __pmatch[__restrict_arr],
 562                     int __eflags);
 563 libc_hidden_proto(regexec)
 564
 565 extern size_t regerror (int __errcode, const regex_t *__restrict __preg,
 566                         char *__restrict __errbuf, size_t __errbuf_size);
 567
 568 extern void regfree (regex_t *__preg);
 569 libc_hidden_proto(regfree)
 570
 571
 572 __END_DECLS
 573
 574 #endif /* regex.h */