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