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