posix/regex.h

   1 /* Definitions for data structures and routines for the regular
   2    expression library.
   3    Copyright (C) 1985,1989-93,1995-98,2000,2001,2002,2003,2005,2006,2008,2011
   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 /* Allow the use in C++ code.  */
  27 #ifdef __cplusplus
  28 extern "C" {
  29 #endif
  30
  31 /* The following two types have to be signed and unsigned integer type
  32    wide enough to hold a value of a pointer.  For most ANSI compilers
  33    ptrdiff_t and size_t should be likely OK.  Still size of these two
  34    types is 2 for Microsoft C.  Ugh... */
  35 typedef long int s_reg_t;
  36 typedef unsigned long int active_reg_t;
  37
  38 /* The following bits are used to determine the regexp syntax we
  39    recognize.  The set/not-set meanings are chosen so that Emacs syntax
  40    remains the value 0.  The bits are given in alphabetical order, and
  41    the definitions shifted by one from the previous bit; thus, when we
  42    add or remove a bit, only one other definition need change.  */
  43 typedef unsigned long int reg_syntax_t;
  44
  45 #ifdef __USE_GNU
  46 /* If this bit is not set, then \ inside a bracket expression is literal.
  47    If set, then such a \ quotes the following character.  */
  48 # define RE_BACKSLASH_ESCAPE_IN_LISTS ((unsigned long int) 1)
  49
  50 /* If this bit is not set, then + and ? are operators, and \+ and \? are
  51      literals.
  52    If set, then \+ and \? are operators and + and ? are literals.  */
  53 # define RE_BK_PLUS_QM (RE_BACKSLASH_ESCAPE_IN_LISTS << 1)
  54
  55 /* If this bit is set, then character classes are supported.  They are:
  56      [:alpha:], [:upper:], [:lower:],  [:digit:], [:alnum:], [:xdigit:],
  57      [:space:], [:print:], [:punct:], [:graph:], and [:cntrl:].
  58    If not set, then character classes are not supported.  */
  59 # define RE_CHAR_CLASSES (RE_BK_PLUS_QM << 1)
  60
  61 /* If this bit is set, then ^ and $ are always anchors (outside bracket
  62      expressions, of course).
  63    If this bit is not set, then it depends:
  64         ^  is an anchor if it is at the beginning of a regular
  65            expression or after an open-group or an alternation operator;
  66         $  is an anchor if it is at the end of a regular expression, or
  67            before a close-group or an alternation operator.
  68
  69    This bit could be (re)combined with RE_CONTEXT_INDEP_OPS, because
  70    POSIX draft 11.2 says that * etc. in leading positions is undefined.
  71    We already implemented a previous draft which made those constructs
  72    invalid, though, so we haven't changed the code back.  */
  73 # define RE_CONTEXT_INDEP_ANCHORS (RE_CHAR_CLASSES << 1)
  74
  75 /* If this bit is set, then special characters are always special
  76      regardless of where they are in the pattern.
  77    If this bit is not set, then special characters are special only in
  78      some contexts; otherwise they are ordinary.  Specifically,
  79      * + ? and intervals are only special when not after the beginning,
  80      open-group, or alternation operator.  */
  81 # define RE_CONTEXT_INDEP_OPS (RE_CONTEXT_INDEP_ANCHORS << 1)
  82
  83 /* If this bit is set, then *, +, ?, and { cannot be first in an re or
  84      immediately after an alternation or begin-group operator.  */
  85 # define RE_CONTEXT_INVALID_OPS (RE_CONTEXT_INDEP_OPS << 1)
  86
  87 /* If this bit is set, then . matches newline.
  88    If not set, then it doesn't.  */
  89 # define RE_DOT_NEWLINE (RE_CONTEXT_INVALID_OPS << 1)
  90
  91 /* If this bit is set, then . doesn't match NUL.
  92    If not set, then it does.  */
  93 # define RE_DOT_NOT_NULL (RE_DOT_NEWLINE << 1)
  94
  95 /* If this bit is set, nonmatching lists [^...] do not match newline.
  96    If not set, they do.  */
  97 # define RE_HAT_LISTS_NOT_NEWLINE (RE_DOT_NOT_NULL << 1)
  98
  99 /* If this bit is set, either \{...\} or {...} defines an
 100      interval, depending on RE_NO_BK_BRACES.
 101    If not set, \{, \}, {, and } are literals.  */
 102 # define RE_INTERVALS (RE_HAT_LISTS_NOT_NEWLINE << 1)
 103
 104 /* If this bit is set, +, ? and | aren't recognized as operators.
 105    If not set, they are.  */
 106 # define RE_LIMITED_OPS (RE_INTERVALS << 1)
 107
 108 /* If this bit is set, newline is an alternation operator.
 109    If not set, newline is literal.  */
 110 # define RE_NEWLINE_ALT (RE_LIMITED_OPS << 1)
 111
 112 /* If this bit is set, then `{...}' defines an interval, and \{ and \}
 113      are literals.
 114   If not set, then `\{...\}' defines an interval.  */
 115 # define RE_NO_BK_BRACES (RE_NEWLINE_ALT << 1)
 116
 117 /* If this bit is set, (...) defines a group, and \( and \) are literals.
 118    If not set, \(...\) defines a group, and ( and ) are literals.  */
 119 # define RE_NO_BK_PARENS (RE_NO_BK_BRACES << 1)
 120
 121 /* If this bit is set, then \<digit> matches <digit>.
 122    If not set, then \<digit> is a back-reference.  */
 123 # define RE_NO_BK_REFS (RE_NO_BK_PARENS << 1)
 124
 125 /* If this bit is set, then | is an alternation operator, and \| is literal.
 126    If not set, then \| is an alternation operator, and | is literal.  */
 127 # define RE_NO_BK_VBAR (RE_NO_BK_REFS << 1)
 128
 129 /* If this bit is set, then an ending range point collating higher
 130      than the starting range point, as in [z-a], is invalid.
 131    If not set, then when ending range point collates higher than the
 132      starting range point, the range is ignored.  */
 133 # define RE_NO_EMPTY_RANGES (RE_NO_BK_VBAR << 1)
 134
 135 /* If this bit is set, then an unmatched ) is ordinary.
 136    If not set, then an unmatched ) is invalid.  */
 137 # define RE_UNMATCHED_RIGHT_PAREN_ORD (RE_NO_EMPTY_RANGES << 1)
 138
 139 /* If this bit is set, succeed as soon as we match the whole pattern,
 140    without further backtracking.  */
 141 # define RE_NO_POSIX_BACKTRACKING (RE_UNMATCHED_RIGHT_PAREN_ORD << 1)
 142
 143 /* If this bit is set, do not process the GNU regex operators.
 144    If not set, then the GNU regex operators are recognized. */
 145 # define RE_NO_GNU_OPS (RE_NO_POSIX_BACKTRACKING << 1)
 146
 147 /* If this bit is set, turn on internal regex debugging.
 148    If not set, and debugging was on, turn it off.
 149    This only works if regex.c is compiled -DDEBUG.
 150    We define this bit always, so that all that's needed to turn on
 151    debugging is to recompile regex.c; the calling code can always have
 152    this bit set, and it won't affect anything in the normal case. */
 153 # define RE_DEBUG (RE_NO_GNU_OPS << 1)
 154
 155 /* If this bit is set, a syntactically invalid interval is treated as
 156    a string of ordinary characters.  For example, the ERE 'a{1' is
 157    treated as 'a\{1'.  */
 158 # define RE_INVALID_INTERVAL_ORD (RE_DEBUG << 1)
 159
 160 /* If this bit is set, then ignore case when matching.
 161    If not set, then case is significant.  */
 162 # define RE_ICASE (RE_INVALID_INTERVAL_ORD << 1)
 163
 164 /* This bit is used internally like RE_CONTEXT_INDEP_ANCHORS but only
 165    for ^, because it is difficult to scan the regex backwards to find
 166    whether ^ should be special.  */
 167 # define RE_CARET_ANCHORS_HERE (RE_ICASE << 1)
 168
 169 /* If this bit is set, then \{ cannot be first in an bre or
 170    immediately after an alternation or begin-group operator.  */
 171 # define RE_CONTEXT_INVALID_DUP (RE_CARET_ANCHORS_HERE << 1)
 172
 173 /* If this bit is set, then no_sub will be set to 1 during
 174    re_compile_pattern.  */
 175 # define RE_NO_SUB (RE_CONTEXT_INVALID_DUP << 1)
 176 #endif
 177
 178 /* This global variable defines the particular regexp syntax to use (for
 179    some interfaces).  When a regexp is compiled, the syntax used is
 180    stored in the pattern buffer, so changing this does not affect
 181    already-compiled regexps.  */
 182 extern reg_syntax_t re_syntax_options;
 183 \f
 184 #ifdef __USE_GNU
 185 /* Define combinations of the above bits for the standard possibilities.
 186    (The [[[ comments delimit what gets put into the Texinfo file, so
 187    don't delete them!)  */
 188 /* [[[begin syntaxes]]] */
 189 #define RE_SYNTAX_EMACS 0
 190
 191 #define RE_SYNTAX_AWK                                                   \
 192   (RE_BACKSLASH_ESCAPE_IN_LISTS   | RE_DOT_NOT_NULL                     \
 193    | RE_NO_BK_PARENS              | RE_NO_BK_REFS                       \
 194    | RE_NO_BK_VBAR                | RE_NO_EMPTY_RANGES                  \
 195    | RE_DOT_NEWLINE               | RE_CONTEXT_INDEP_ANCHORS            \
 196    | RE_CHAR_CLASSES                                                    \
 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             \
 201     | RE_INVALID_INTERVAL_ORD)                                          \
 202    & ~(RE_DOT_NOT_NULL | RE_CONTEXT_INDEP_OPS                           \
 203       | RE_CONTEXT_INVALID_OPS ))
 204
 205 #define RE_SYNTAX_POSIX_AWK                                             \
 206   (RE_SYNTAX_POSIX_EXTENDED | RE_BACKSLASH_ESCAPE_IN_LISTS              \
 207    | RE_INTERVALS           | RE_NO_GNU_OPS                             \
 208    | RE_INVALID_INTERVAL_ORD)
 209
 210 #define RE_SYNTAX_GREP                                                  \
 211   (RE_BK_PLUS_QM              | RE_CHAR_CLASSES                         \
 212    | RE_HAT_LISTS_NOT_NEWLINE | RE_INTERVALS                            \
 213    | RE_NEWLINE_ALT)
 214
 215 #define RE_SYNTAX_EGREP                                                 \
 216   (RE_CHAR_CLASSES        | RE_CONTEXT_INDEP_ANCHORS                    \
 217    | RE_CONTEXT_INDEP_OPS | RE_HAT_LISTS_NOT_NEWLINE                    \
 218    | RE_NEWLINE_ALT       | RE_NO_BK_PARENS                             \
 219    | RE_NO_BK_VBAR)
 220
 221 #define RE_SYNTAX_POSIX_EGREP                                           \
 222   (RE_SYNTAX_EGREP | RE_INTERVALS | RE_NO_BK_BRACES                     \
 223    | RE_INVALID_INTERVAL_ORD)
 224
 225 /* P1003.2/D11.2, section 4.20.7.1, lines 5078ff.  */
 226 #define RE_SYNTAX_ED RE_SYNTAX_POSIX_BASIC
 227
 228 #define RE_SYNTAX_SED RE_SYNTAX_POSIX_BASIC
 229
 230 /* Syntax bits common to both basic and extended POSIX regex syntax.  */
 231 #define _RE_SYNTAX_POSIX_COMMON                                         \
 232   (RE_CHAR_CLASSES | RE_DOT_NEWLINE      | RE_DOT_NOT_NULL              \
 233    | RE_INTERVALS  | RE_NO_EMPTY_RANGES)
 234
 235 #define RE_SYNTAX_POSIX_BASIC                                           \
 236   (_RE_SYNTAX_POSIX_COMMON | RE_BK_PLUS_QM | RE_CONTEXT_INVALID_DUP)
 237
 238 /* Differs from ..._POSIX_BASIC only in that RE_BK_PLUS_QM becomes
 239    RE_LIMITED_OPS, i.e., \? \+ \| are not recognized.  Actually, this
 240    isn't minimal, since other operators, such as \`, aren't disabled.  */
 241 #define RE_SYNTAX_POSIX_MINIMAL_BASIC                                   \
 242   (_RE_SYNTAX_POSIX_COMMON | RE_LIMITED_OPS)
 243
 244 #define RE_SYNTAX_POSIX_EXTENDED                                        \
 245   (_RE_SYNTAX_POSIX_COMMON  | RE_CONTEXT_INDEP_ANCHORS                  \
 246    | RE_CONTEXT_INDEP_OPS   | RE_NO_BK_BRACES                           \
 247    | RE_NO_BK_PARENS        | RE_NO_BK_VBAR                             \
 248    | RE_CONTEXT_INVALID_OPS | RE_UNMATCHED_RIGHT_PAREN_ORD)
 249
 250 /* Differs from ..._POSIX_EXTENDED in that RE_CONTEXT_INDEP_OPS is
 251    removed and RE_NO_BK_REFS is added.  */
 252 #define RE_SYNTAX_POSIX_MINIMAL_EXTENDED                                \
 253   (_RE_SYNTAX_POSIX_COMMON  | RE_CONTEXT_INDEP_ANCHORS                  \
 254    | RE_CONTEXT_INVALID_OPS | RE_NO_BK_BRACES                           \
 255    | RE_NO_BK_PARENS        | RE_NO_BK_REFS                             \
 256    | RE_NO_BK_VBAR          | RE_UNMATCHED_RIGHT_PAREN_ORD)
 257 /* [[[end syntaxes]]] */
 258 \f
 259 /* Maximum number of duplicates an interval can allow.  Some systems
 260    (erroneously) define this in other header files, but we want our
 261    value, so remove any previous define.  */
 262 # ifdef RE_DUP_MAX
 263 #  undef RE_DUP_MAX
 264 # endif
 265 /* If sizeof(int) == 2, then ((1 << 15) - 1) overflows.  */
 266 # define RE_DUP_MAX (0x7fff)
 267 #endif
 268
 269
 270 /* POSIX `cflags' bits (i.e., information for `regcomp').  */
 271
 272 /* If this bit is set, then use extended regular expression syntax.
 273    If not set, then use basic regular expression syntax.  */
 274 #define REG_EXTENDED 1
 275
 276 /* If this bit is set, then ignore case when matching.
 277    If not set, then case is significant.  */
 278 #define REG_ICASE (REG_EXTENDED << 1)
 279
 280 /* If this bit is set, then anchors do not match at newline
 281      characters in the string.
 282    If not set, then anchors do match at newlines.  */
 283 #define REG_NEWLINE (REG_ICASE << 1)
 284
 285 /* If this bit is set, then report only success or fail in regexec.
 286    If not set, then returns differ between not matching and errors.  */
 287 #define REG_NOSUB (REG_NEWLINE << 1)
 288
 289
 290 /* POSIX `eflags' bits (i.e., information for regexec).  */
 291
 292 /* If this bit is set, then the beginning-of-line operator doesn't match
 293      the beginning of the string (presumably because it's not the
 294      beginning of a line).
 295    If not set, then the beginning-of-line operator does match the
 296      beginning of the string.  */
 297 #define REG_NOTBOL 1
 298
 299 /* Like REG_NOTBOL, except for the end-of-line.  */
 300 #define REG_NOTEOL (1 << 1)
 301
 302 /* Use PMATCH[0] to delimit the start and end of the search in the
 303    buffer.  */
 304 #define REG_STARTEND (1 << 2)
 305
 306
 307 /* If any error codes are removed, changed, or added, update the
 308    `re_error_msg' table in regex.c.  */
 309 typedef enum
 310 {
 311 #if defined _XOPEN_SOURCE || defined __USE_XOPEN2K
 312   REG_ENOSYS = -1,      /* This will never happen for this implementation.  */
 313 #endif
 314
 315   REG_NOERROR = 0,      /* Success.  */
 316   REG_NOMATCH,          /* Didn't find a match (for regexec).  */
 317
 318   /* POSIX regcomp return error codes.  (In the order listed in the
 319      standard.)  */
 320   REG_BADPAT,           /* Invalid pattern.  */
 321   REG_ECOLLATE,         /* Inalid collating element.  */
 322   REG_ECTYPE,           /* Invalid character class name.  */
 323   REG_EESCAPE,          /* Trailing backslash.  */
 324   REG_ESUBREG,          /* Invalid back reference.  */
 325   REG_EBRACK,           /* Unmatched left bracket.  */
 326   REG_EPAREN,           /* Parenthesis imbalance.  */
 327   REG_EBRACE,           /* Unmatched \{.  */
 328   REG_BADBR,            /* Invalid contents of \{\}.  */
 329   REG_ERANGE,           /* Invalid range end.  */
 330   REG_ESPACE,           /* Ran out of memory.  */
 331   REG_BADRPT,           /* No preceding re for repetition op.  */
 332
 333   /* Error codes we've added.  */
 334   REG_EEND,             /* Premature end.  */
 335   REG_ESIZE,            /* Compiled pattern bigger than 2^16 bytes.  */
 336   REG_ERPAREN           /* Unmatched ) or \); not returned from regcomp.  */
 337 } reg_errcode_t;
 338 \f
 339 /* This data structure represents a compiled pattern.  Before calling
 340    the pattern compiler, the fields `buffer', `allocated', `fastmap',
 341    and `translate' can be set.  After the pattern has been compiled,
 342    the fields `re_nsub', `not_bol' and `not_eol' are available.  All
 343    other fields are private to the regex routines.  */
 344
 345 #ifndef RE_TRANSLATE_TYPE
 346 # define __RE_TRANSLATE_TYPE unsigned char *
 347 # ifdef __USE_GNU
 348 #  define RE_TRANSLATE_TYPE __RE_TRANSLATE_TYPE
 349 # endif
 350 #endif
 351
 352 #ifdef __USE_GNU
 353 # define __REPB_PREFIX(name) name
 354 #else
 355 # define __REPB_PREFIX(name) __##name
 356 #endif
 357
 358 struct re_pattern_buffer
 359 {
 360   /* Space that holds the compiled pattern.  It is declared as
 361      `unsigned char *' because its elements are sometimes used as
 362      array indexes.  */
 363   unsigned char *__REPB_PREFIX(buffer);
 364
 365   /* Number of bytes to which `buffer' points.  */
 366   unsigned long int __REPB_PREFIX(allocated);
 367
 368   /* Number of bytes actually used in `buffer'.  */
 369   unsigned long int __REPB_PREFIX(used);
 370
 371   /* Syntax setting with which the pattern was compiled.  */
 372   reg_syntax_t __REPB_PREFIX(syntax);
 373
 374   /* Pointer to a fastmap, if any, otherwise zero.  re_search uses the
 375      fastmap, if there is one, to skip over impossible starting points
 376      for matches.  */
 377   char *__REPB_PREFIX(fastmap);
 378
 379   /* Either a translate table to apply to all characters before
 380      comparing them, or zero for no translation.  The translation is
 381      applied to a pattern when it is compiled and to a string when it
 382      is matched.  */
 383   __RE_TRANSLATE_TYPE __REPB_PREFIX(translate);
 384
 385   /* Number of subexpressions found by the compiler.  */
 386   size_t re_nsub;
 387
 388   /* Zero if this pattern cannot match the empty string, one else.
 389      Well, in truth it's used only in `re_search_2', to see whether or
 390      not we should use the fastmap, so we don't set this absolutely
 391      perfectly; see `re_compile_fastmap' (the `duplicate' case).  */
 392   unsigned __REPB_PREFIX(can_be_null) : 1;
 393
 394   /* If REGS_UNALLOCATED, allocate space in the `regs' structure
 395      for `max (RE_NREGS, re_nsub + 1)' groups.
 396      If REGS_REALLOCATE, reallocate space if necessary.
 397      If REGS_FIXED, use what's there.  */
 398 #ifdef __USE_GNU
 399 # define REGS_UNALLOCATED 0
 400 # define REGS_REALLOCATE 1
 401 # define REGS_FIXED 2
 402 #endif
 403   unsigned __REPB_PREFIX(regs_allocated) : 2;
 404
 405   /* Set to zero when `regex_compile' compiles a pattern; set to one
 406      by `re_compile_fastmap' if it updates the fastmap.  */
 407   unsigned __REPB_PREFIX(fastmap_accurate) : 1;
 408
 409   /* If set, `re_match_2' does not return information about
 410      subexpressions.  */
 411   unsigned __REPB_PREFIX(no_sub) : 1;
 412
 413   /* If set, a beginning-of-line anchor doesn't match at the beginning
 414      of the string.  */
 415   unsigned __REPB_PREFIX(not_bol) : 1;
 416
 417   /* Similarly for an end-of-line anchor.  */
 418   unsigned __REPB_PREFIX(not_eol) : 1;
 419
 420   /* If true, an anchor at a newline matches.  */
 421   unsigned __REPB_PREFIX(newline_anchor) : 1;
 422 };
 423
 424 typedef struct re_pattern_buffer regex_t;
 425 \f
 426 /* Type for byte offsets within the string.  POSIX mandates this.  */
 427 typedef int regoff_t;
 428
 429
 430 #ifdef __USE_GNU
 431 /* This is the structure we store register match data in.  See
 432    regex.texinfo for a full description of what registers match.  */
 433 struct re_registers
 434 {
 435   unsigned num_regs;
 436   regoff_t *start;
 437   regoff_t *end;
 438 };
 439
 440
 441 /* If `regs_allocated' is REGS_UNALLOCATED in the pattern buffer,
 442    `re_match_2' returns information about at least this many registers
 443    the first time a `regs' structure is passed.  */
 444 # ifndef RE_NREGS
 445 #  define RE_NREGS 30
 446 # endif
 447 #endif
 448
 449
 450 /* POSIX specification for registers.  Aside from the different names than
 451    `re_registers', POSIX uses an array of structures, instead of a
 452    structure of arrays.  */
 453 typedef struct
 454 {
 455   regoff_t rm_so;  /* Byte offset from string's start to substring's start.  */
 456   regoff_t rm_eo;  /* Byte offset from string's start to substring's end.  */
 457 } regmatch_t;
 458 \f
 459 /* Declarations for routines.  */
 460
 461 #ifdef __USE_GNU
 462 /* Sets the current default syntax to SYNTAX, and return the old syntax.
 463    You can also simply assign to the `re_syntax_options' variable.  */
 464 extern reg_syntax_t re_set_syntax (reg_syntax_t __syntax);
 465
 466 /* Compile the regular expression PATTERN, with length LENGTH
 467    and syntax given by the global `re_syntax_options', into the buffer
 468    BUFFER.  Return NULL if successful, and an error string if not.
 469
 470    To free the allocated storage, you must call `regfree' on BUFFER.
 471    Note that the translate table must either have been initialised by
 472    `regcomp', with a malloc'ed value, or set to NULL before calling
 473    `regfree'.  */
 474 extern const char *re_compile_pattern (const char *__pattern, size_t __length,
 475                                        struct re_pattern_buffer *__buffer);
 476
 477
 478 /* Compile a fastmap for the compiled pattern in BUFFER; used to
 479    accelerate searches.  Return 0 if successful and -2 if was an
 480    internal error.  */
 481 extern int re_compile_fastmap (struct re_pattern_buffer *__buffer);
 482
 483
 484 /* Search in the string STRING (with length LENGTH) for the pattern
 485    compiled into BUFFER.  Start searching at position START, for RANGE
 486    characters.  Return the starting position of the match, -1 for no
 487    match, or -2 for an internal error.  Also return register
 488    information in REGS (if REGS and BUFFER->no_sub are nonzero).  */
 489 extern int re_search (struct re_pattern_buffer *__buffer, const char *__string,
 490                       int __length, int __start, int __range,
 491                       struct re_registers *__regs);
 492
 493
 494 /* Like `re_search', but search in the concatenation of STRING1 and
 495    STRING2.  Also, stop searching at index START + STOP.  */
 496 extern int re_search_2 (struct re_pattern_buffer *__buffer,
 497                         const char *__string1, int __length1,
 498                         const char *__string2, int __length2, int __start,
 499                         int __range, struct re_registers *__regs, int __stop);
 500
 501
 502 /* Like `re_search', but return how many characters in STRING the regexp
 503    in BUFFER matched, starting at position START.  */
 504 extern int re_match (struct re_pattern_buffer *__buffer, const char *__string,
 505                      int __length, int __start, struct re_registers *__regs);
 506
 507
 508 /* Relates to `re_match' as `re_search_2' relates to `re_search'.  */
 509 extern int re_match_2 (struct re_pattern_buffer *__buffer,
 510                        const char *__string1, int __length1,
 511                        const char *__string2, int __length2, int __start,
 512                        struct re_registers *__regs, int __stop);
 513
 514
 515 /* Set REGS to hold NUM_REGS registers, storing them in STARTS and
 516    ENDS.  Subsequent matches using BUFFER and REGS will use this memory
 517    for recording register information.  STARTS and ENDS must be
 518    allocated with malloc, and must each be at least `NUM_REGS * sizeof
 519    (regoff_t)' bytes long.
 520
 521    If NUM_REGS == 0, then subsequent matches should allocate their own
 522    register data.
 523
 524    Unless this function is called, the first search or match using
 525    PATTERN_BUFFER will allocate its own register data, without
 526    freeing the old data.  */
 527 extern void re_set_registers (struct re_pattern_buffer *__buffer,
 528                               struct re_registers *__regs,
 529                               unsigned int __num_regs,
 530                               regoff_t *__starts, regoff_t *__ends);
 531 #endif  /* Use GNU */
 532
 533 #if defined _REGEX_RE_COMP || (defined _LIBC && defined __USE_BSD)
 534 # ifndef _CRAY
 535 /* 4.2 bsd compatibility.  */
 536 extern char *re_comp (const char *);
 537 extern int re_exec (const char *);
 538 # endif
 539 #endif
 540
 541 /* GCC 2.95 and later have "__restrict"; C99 compilers have
 542    "restrict", and "configure" may have defined "restrict".  */
 543 #ifndef __restrict
 544 # if ! (2 < __GNUC__ || (2 == __GNUC__ && 95 <= __GNUC_MINOR__))
 545 #  if defined restrict || 199901L <= __STDC_VERSION__
 546 #   define __restrict restrict
 547 #  else
 548 #   define __restrict
 549 #  endif
 550 # endif
 551 #endif
 552 /* gcc 3.1 and up support the [restrict] syntax.  */
 553 #ifndef __restrict_arr
 554 # if (__GNUC__ > 3 || (__GNUC__ == 3 && __GNUC_MINOR__ >= 1)) \
 555      && !defined __GNUG__
 556 #  define __restrict_arr __restrict
 557 # else
 558 #  define __restrict_arr
 559 # endif
 560 #endif
 561
 562 /* POSIX compatibility.  */
 563 extern int regcomp (regex_t *__restrict __preg,
 564                     const char *__restrict __pattern,
 565                     int __cflags);
 566
 567 extern int regexec (const regex_t *__restrict __preg,
 568                     const char *__restrict __string, size_t __nmatch,
 569                     regmatch_t __pmatch[__restrict_arr],
 570                     int __eflags);
 571
 572 extern size_t regerror (int __errcode, const regex_t *__restrict __preg,
 573                         char *__restrict __errbuf, size_t __errbuf_size);
 574
 575 extern void regfree (regex_t *__preg);
 576
 577
 578 #ifdef __cplusplus
 579 }
 580 #endif  /* C++ */
 581
 582 #endif /* regex.h */