lib/regex.h

   1 /* Definitions for data structures and routines for the regular
   2    expression library.
   3    Copyright (C) 1985, 1989-1993, 1995-1998, 2000-2003, 2005-2017 Free Software
   4    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 /* Define __USE_GNU to declare GNU extensions that violate the
  32    POSIX name space rules.  */
  33 #ifdef _GNU_SOURCE
  34 # define __USE_GNU 1
  35 #endif
  36
  37 #ifdef _REGEX_LARGE_OFFSETS
  38
  39 /* Use types and values that are wide enough to represent signed and
  40    unsigned byte offsets in memory.  This currently works only when
  41    the regex code is used outside of the GNU C library; it is not yet
  42    supported within glibc itself, and glibc users should not define
  43    _REGEX_LARGE_OFFSETS.  */
  44
  45 /* The type of object sizes.  */
  46 typedef size_t __re_size_t;
  47
  48 /* The type of object sizes, in places where the traditional code
  49    uses unsigned long int.  */
  50 typedef size_t __re_long_size_t;
  51
  52 #else
  53
  54 /* The traditional GNU regex implementation mishandles strings longer
  55    than INT_MAX.  */
  56 typedef unsigned int __re_size_t;
  57 typedef unsigned long int __re_long_size_t;
  58
  59 #endif
  60
  61 /* The following two types have to be signed and unsigned integer type
  62    wide enough to hold a value of a pointer.  For most ANSI compilers
  63    ptrdiff_t and size_t should be likely OK.  Still size of these two
  64    types is 2 for Microsoft C.  Ugh... */
  65 typedef long int s_reg_t;
  66 typedef unsigned long int active_reg_t;
  67
  68 /* The following bits are used to determine the regexp syntax we
  69    recognize.  The set/not-set meanings are chosen so that Emacs syntax
  70    remains the value 0.  The bits are given in alphabetical order, and
  71    the definitions shifted by one from the previous bit; thus, when we
  72    add or remove a bit, only one other definition need change.  */
  73 typedef unsigned long int reg_syntax_t;
  74
  75 #ifdef __USE_GNU
  76 /* If this bit is not set, then \ inside a bracket expression is literal.
  77    If set, then such a \ quotes the following character.  */
  78 # define RE_BACKSLASH_ESCAPE_IN_LISTS ((unsigned long int) 1)
  79
  80 /* If this bit is not set, then + and ? are operators, and \+ and \? are
  81      literals.
  82    If set, then \+ and \? are operators and + and ? are literals.  */
  83 # define RE_BK_PLUS_QM (RE_BACKSLASH_ESCAPE_IN_LISTS << 1)
  84
  85 /* If this bit is set, then character classes are supported.  They are:
  86      [:alpha:], [:upper:], [:lower:],  [:digit:], [:alnum:], [:xdigit:],
  87      [:space:], [:print:], [:punct:], [:graph:], and [:cntrl:].
  88    If not set, then character classes are not supported.  */
  89 # define RE_CHAR_CLASSES (RE_BK_PLUS_QM << 1)
  90
  91 /* If this bit is set, then ^ and $ are always anchors (outside bracket
  92      expressions, of course).
  93    If this bit is not set, then it depends:
  94         ^  is an anchor if it is at the beginning of a regular
  95            expression or after an open-group or an alternation operator;
  96         $  is an anchor if it is at the end of a regular expression, or
  97            before a close-group or an alternation operator.
  98
  99    This bit could be (re)combined with RE_CONTEXT_INDEP_OPS, because
 100    POSIX draft 11.2 says that * etc. in leading positions is undefined.
 101    We already implemented a previous draft which made those constructs
 102    invalid, though, so we haven't changed the code back.  */
 103 # define RE_CONTEXT_INDEP_ANCHORS (RE_CHAR_CLASSES << 1)
 104
 105 /* If this bit is set, then special characters are always special
 106      regardless of where they are in the pattern.
 107    If this bit is not set, then special characters are special only in
 108      some contexts; otherwise they are ordinary.  Specifically,
 109      * + ? and intervals are only special when not after the beginning,
 110      open-group, or alternation operator.  */
 111 # define RE_CONTEXT_INDEP_OPS (RE_CONTEXT_INDEP_ANCHORS << 1)
 112
 113 /* If this bit is set, then *, +, ?, and { cannot be first in an re or
 114      immediately after an alternation or begin-group operator.  */
 115 # define RE_CONTEXT_INVALID_OPS (RE_CONTEXT_INDEP_OPS << 1)
 116
 117 /* If this bit is set, then . matches newline.
 118    If not set, then it doesn't.  */
 119 # define RE_DOT_NEWLINE (RE_CONTEXT_INVALID_OPS << 1)
 120
 121 /* If this bit is set, then . doesn't match NUL.
 122    If not set, then it does.  */
 123 # define RE_DOT_NOT_NULL (RE_DOT_NEWLINE << 1)
 124
 125 /* If this bit is set, nonmatching lists [^...] do not match newline.
 126    If not set, they do.  */
 127 # define RE_HAT_LISTS_NOT_NEWLINE (RE_DOT_NOT_NULL << 1)
 128
 129 /* If this bit is set, either \{...\} or {...} defines an
 130      interval, depending on RE_NO_BK_BRACES.
 131    If not set, \{, \}, {, and } are literals.  */
 132 # define RE_INTERVALS (RE_HAT_LISTS_NOT_NEWLINE << 1)
 133
 134 /* If this bit is set, +, ? and | aren't recognized as operators.
 135    If not set, they are.  */
 136 # define RE_LIMITED_OPS (RE_INTERVALS << 1)
 137
 138 /* If this bit is set, newline is an alternation operator.
 139    If not set, newline is literal.  */
 140 # define RE_NEWLINE_ALT (RE_LIMITED_OPS << 1)
 141
 142 /* If this bit is set, then '{...}' defines an interval, and \{ and \}
 143      are literals.
 144   If not set, then '\{...\}' defines an interval.  */
 145 # define RE_NO_BK_BRACES (RE_NEWLINE_ALT << 1)
 146
 147 /* If this bit is set, (...) defines a group, and \( and \) are literals.
 148    If not set, \(...\) defines a group, and ( and ) are literals.  */
 149 # define RE_NO_BK_PARENS (RE_NO_BK_BRACES << 1)
 150
 151 /* If this bit is set, then \<digit> matches <digit>.
 152    If not set, then \<digit> is a back-reference.  */
 153 # define RE_NO_BK_REFS (RE_NO_BK_PARENS << 1)
 154
 155 /* If this bit is set, then | is an alternation operator, and \| is literal.
 156    If not set, then \| is an alternation operator, and | is literal.  */
 157 # define RE_NO_BK_VBAR (RE_NO_BK_REFS << 1)
 158
 159 /* If this bit is set, then an ending range point collating higher
 160      than the starting range point, as in [z-a], is invalid.
 161    If not set, then when ending range point collates higher than the
 162      starting range point, the range is ignored.  */
 163 # define RE_NO_EMPTY_RANGES (RE_NO_BK_VBAR << 1)
 164
 165 /* If this bit is set, then an unmatched ) is ordinary.
 166    If not set, then an unmatched ) is invalid.  */
 167 # define RE_UNMATCHED_RIGHT_PAREN_ORD (RE_NO_EMPTY_RANGES << 1)
 168
 169 /* If this bit is set, succeed as soon as we match the whole pattern,
 170    without further backtracking.  */
 171 # define RE_NO_POSIX_BACKTRACKING (RE_UNMATCHED_RIGHT_PAREN_ORD << 1)
 172
 173 /* If this bit is set, do not process the GNU regex operators.
 174    If not set, then the GNU regex operators are recognized. */
 175 # define RE_NO_GNU_OPS (RE_NO_POSIX_BACKTRACKING << 1)
 176
 177 /* If this bit is set, turn on internal regex debugging.
 178    If not set, and debugging was on, turn it off.
 179    This only works if regex.c is compiled -DDEBUG.
 180    We define this bit always, so that all that's needed to turn on
 181    debugging is to recompile regex.c; the calling code can always have
 182    this bit set, and it won't affect anything in the normal case. */
 183 # define RE_DEBUG (RE_NO_GNU_OPS << 1)
 184
 185 /* If this bit is set, a syntactically invalid interval is treated as
 186    a string of ordinary characters.  For example, the ERE 'a{1' is
 187    treated as 'a\{1'.  */
 188 # define RE_INVALID_INTERVAL_ORD (RE_DEBUG << 1)
 189
 190 /* If this bit is set, then ignore case when matching.
 191    If not set, then case is significant.  */
 192 # define RE_ICASE (RE_INVALID_INTERVAL_ORD << 1)
 193
 194 /* This bit is used internally like RE_CONTEXT_INDEP_ANCHORS but only
 195    for ^, because it is difficult to scan the regex backwards to find
 196    whether ^ should be special.  */
 197 # define RE_CARET_ANCHORS_HERE (RE_ICASE << 1)
 198
 199 /* If this bit is set, then \{ cannot be first in a regex or
 200    immediately after an alternation, open-group or \} operator.  */
 201 # define RE_CONTEXT_INVALID_DUP (RE_CARET_ANCHORS_HERE << 1)
 202
 203 /* If this bit is set, then no_sub will be set to 1 during
 204    re_compile_pattern.  */
 205 # define RE_NO_SUB (RE_CONTEXT_INVALID_DUP << 1)
 206 #endif
 207
 208 /* This global variable defines the particular regexp syntax to use (for
 209    some interfaces).  When a regexp is compiled, the syntax used is
 210    stored in the pattern buffer, so changing this does not affect
 211    already-compiled regexps.  */
 212 extern reg_syntax_t re_syntax_options;
 213 \f
 214 #ifdef __USE_GNU
 215 /* Define combinations of the above bits for the standard possibilities.
 216    (The [[[ comments delimit what gets put into the Texinfo file, so
 217    don't delete them!)  */
 218 /* [[[begin syntaxes]]] */
 219 # define RE_SYNTAX_EMACS 0
 220
 221 # define RE_SYNTAX_AWK                                                  \
 222   (RE_BACKSLASH_ESCAPE_IN_LISTS   | RE_DOT_NOT_NULL                     \
 223    | RE_NO_BK_PARENS              | RE_NO_BK_REFS                       \
 224    | RE_NO_BK_VBAR                | RE_NO_EMPTY_RANGES                  \
 225    | RE_DOT_NEWLINE               | RE_CONTEXT_INDEP_ANCHORS            \
 226    | RE_CHAR_CLASSES                                                    \
 227    | RE_UNMATCHED_RIGHT_PAREN_ORD | RE_NO_GNU_OPS)
 228
 229 # define RE_SYNTAX_GNU_AWK                                              \
 230   ((RE_SYNTAX_POSIX_EXTENDED | RE_BACKSLASH_ESCAPE_IN_LISTS             \
 231     | RE_INVALID_INTERVAL_ORD)                                          \
 232    & ~(RE_DOT_NOT_NULL | RE_CONTEXT_INDEP_OPS                           \
 233       | RE_CONTEXT_INVALID_OPS ))
 234
 235 # define RE_SYNTAX_POSIX_AWK                                            \
 236   (RE_SYNTAX_POSIX_EXTENDED | RE_BACKSLASH_ESCAPE_IN_LISTS              \
 237    | RE_INTERVALS           | RE_NO_GNU_OPS                             \
 238    | RE_INVALID_INTERVAL_ORD)
 239
 240 # define RE_SYNTAX_GREP                                                 \
 241   ((RE_SYNTAX_POSIX_BASIC | RE_NEWLINE_ALT)                             \
 242    & ~(RE_CONTEXT_INVALID_DUP | RE_DOT_NOT_NULL))
 243
 244 # define RE_SYNTAX_EGREP                                                \
 245   ((RE_SYNTAX_POSIX_EXTENDED | RE_INVALID_INTERVAL_ORD | RE_NEWLINE_ALT) \
 246    & ~(RE_CONTEXT_INVALID_OPS | RE_DOT_NOT_NULL))
 247
 248 /* POSIX grep -E behavior is no longer incompatible with GNU.  */
 249 # define RE_SYNTAX_POSIX_EGREP                                          \
 250   RE_SYNTAX_EGREP
 251
 252 /* P1003.2/D11.2, section 4.20.7.1, lines 5078ff.  */
 253 # define RE_SYNTAX_ED RE_SYNTAX_POSIX_BASIC
 254
 255 # define RE_SYNTAX_SED RE_SYNTAX_POSIX_BASIC
 256
 257 /* Syntax bits common to both basic and extended POSIX regex syntax.  */
 258 # define _RE_SYNTAX_POSIX_COMMON                                        \
 259   (RE_CHAR_CLASSES | RE_DOT_NEWLINE      | RE_DOT_NOT_NULL              \
 260    | RE_INTERVALS  | RE_NO_EMPTY_RANGES)
 261
 262 # define RE_SYNTAX_POSIX_BASIC                                          \
 263   (_RE_SYNTAX_POSIX_COMMON | RE_BK_PLUS_QM | RE_CONTEXT_INVALID_DUP)
 264
 265 /* Differs from ..._POSIX_BASIC only in that RE_BK_PLUS_QM becomes
 266    RE_LIMITED_OPS, i.e., \? \+ \| are not recognized.  Actually, this
 267    isn't minimal, since other operators, such as \`, aren't disabled.  */
 268 # define RE_SYNTAX_POSIX_MINIMAL_BASIC                                  \
 269   (_RE_SYNTAX_POSIX_COMMON | RE_LIMITED_OPS)
 270
 271 # define RE_SYNTAX_POSIX_EXTENDED                                       \
 272   (_RE_SYNTAX_POSIX_COMMON  | RE_CONTEXT_INDEP_ANCHORS                  \
 273    | RE_CONTEXT_INDEP_OPS   | RE_NO_BK_BRACES                           \
 274    | RE_NO_BK_PARENS        | RE_NO_BK_VBAR                             \
 275    | RE_CONTEXT_INVALID_OPS | RE_UNMATCHED_RIGHT_PAREN_ORD)
 276
 277 /* Differs from ..._POSIX_EXTENDED in that RE_CONTEXT_INDEP_OPS is
 278    removed and RE_NO_BK_REFS is added.  */
 279 # define RE_SYNTAX_POSIX_MINIMAL_EXTENDED                               \
 280   (_RE_SYNTAX_POSIX_COMMON  | RE_CONTEXT_INDEP_ANCHORS                  \
 281    | RE_CONTEXT_INVALID_OPS | RE_NO_BK_BRACES                           \
 282    | RE_NO_BK_PARENS        | RE_NO_BK_REFS                             \
 283    | RE_NO_BK_VBAR          | RE_UNMATCHED_RIGHT_PAREN_ORD)
 284 /* [[[end syntaxes]]] */
 285
 286 /* Maximum number of duplicates an interval can allow.  POSIX-conforming
 287    systems might define this in <limits.h>, but we want our
 288    value, so remove any previous define.  */
 289 # ifdef _REGEX_INCLUDE_LIMITS_H
 290 #  include <limits.h>
 291 # endif
 292 # ifdef RE_DUP_MAX
 293 #  undef RE_DUP_MAX
 294 # endif
 295
 296 /* RE_DUP_MAX is 2**15 - 1 because an earlier implementation stored
 297    the counter as a 2-byte signed integer.  This is no longer true, so
 298    RE_DUP_MAX could be increased to (INT_MAX / 10 - 1), or to
 299    ((SIZE_MAX - 9) / 10) if _REGEX_LARGE_OFFSETS is defined.
 300    However, there would be a huge performance problem if someone
 301    actually used a pattern like a\{214748363\}, so RE_DUP_MAX retains
 302    its historical value.  */
 303 # define RE_DUP_MAX (0x7fff)
 304 #endif
 305
 306
 307 /* POSIX 'cflags' bits (i.e., information for 'regcomp').  */
 308
 309 /* If this bit is set, then use extended regular expression syntax.
 310    If not set, then use basic regular expression syntax.  */
 311 #define REG_EXTENDED 1
 312
 313 /* If this bit is set, then ignore case when matching.
 314    If not set, then case is significant.  */
 315 #define REG_ICASE (1 << 1)
 316
 317 /* If this bit is set, then anchors do not match at newline
 318      characters in the string.
 319    If not set, then anchors do match at newlines.  */
 320 #define REG_NEWLINE (1 << 2)
 321
 322 /* If this bit is set, then report only success or fail in regexec.
 323    If not set, then returns differ between not matching and errors.  */
 324 #define REG_NOSUB (1 << 3)
 325
 326
 327 /* POSIX 'eflags' bits (i.e., information for regexec).  */
 328
 329 /* If this bit is set, then the beginning-of-line operator doesn't match
 330      the beginning of the string (presumably because it's not the
 331      beginning of a line).
 332    If not set, then the beginning-of-line operator does match the
 333      beginning of the string.  */
 334 #define REG_NOTBOL 1
 335
 336 /* Like REG_NOTBOL, except for the end-of-line.  */
 337 #define REG_NOTEOL (1 << 1)
 338
 339 /* Use PMATCH[0] to delimit the start and end of the search in the
 340    buffer.  */
 341 #define REG_STARTEND (1 << 2)
 342
 343
 344 /* If any error codes are removed, changed, or added, update the
 345    '__re_error_msgid' table in regcomp.c.  */
 346
 347 typedef enum
 348 {
 349   _REG_ENOSYS = -1,     /* This will never happen for this implementation.  */
 350   _REG_NOERROR = 0,     /* Success.  */
 351   _REG_NOMATCH,         /* Didn't find a match (for regexec).  */
 352
 353   /* POSIX regcomp return error codes.  (In the order listed in the
 354      standard.)  */
 355   _REG_BADPAT,          /* Invalid pattern.  */
 356   _REG_ECOLLATE,        /* Invalid collating element.  */
 357   _REG_ECTYPE,          /* Invalid character class name.  */
 358   _REG_EESCAPE,         /* Trailing backslash.  */
 359   _REG_ESUBREG,         /* Invalid back reference.  */
 360   _REG_EBRACK,          /* Unmatched left bracket.  */
 361   _REG_EPAREN,          /* Parenthesis imbalance.  */
 362   _REG_EBRACE,          /* Unmatched \{.  */
 363   _REG_BADBR,           /* Invalid contents of \{\}.  */
 364   _REG_ERANGE,          /* Invalid range end.  */
 365   _REG_ESPACE,          /* Ran out of memory.  */
 366   _REG_BADRPT,          /* No preceding re for repetition op.  */
 367
 368   /* Error codes we've added.  */
 369   _REG_EEND,            /* Premature end.  */
 370   _REG_ESIZE,           /* Too large (e.g., repeat count too large).  */
 371   _REG_ERPAREN          /* Unmatched ) or \); not returned from regcomp.  */
 372 } reg_errcode_t;
 373
 374 #if defined _XOPEN_SOURCE || defined __USE_XOPEN2K
 375 # define REG_ENOSYS     _REG_ENOSYS
 376 #endif
 377 #define REG_NOERROR     _REG_NOERROR
 378 #define REG_NOMATCH     _REG_NOMATCH
 379 #define REG_BADPAT      _REG_BADPAT
 380 #define REG_ECOLLATE    _REG_ECOLLATE
 381 #define REG_ECTYPE      _REG_ECTYPE
 382 #define REG_EESCAPE     _REG_EESCAPE
 383 #define REG_ESUBREG     _REG_ESUBREG
 384 #define REG_EBRACK      _REG_EBRACK
 385 #define REG_EPAREN      _REG_EPAREN
 386 #define REG_EBRACE      _REG_EBRACE
 387 #define REG_BADBR       _REG_BADBR
 388 #define REG_ERANGE      _REG_ERANGE
 389 #define REG_ESPACE      _REG_ESPACE
 390 #define REG_BADRPT      _REG_BADRPT
 391 #define REG_EEND        _REG_EEND
 392 #define REG_ESIZE       _REG_ESIZE
 393 #define REG_ERPAREN     _REG_ERPAREN
 394 \f
 395 /* This data structure represents a compiled pattern.  Before calling
 396    the pattern compiler, the fields 'buffer', 'allocated', 'fastmap',
 397    and 'translate' can be set.  After the pattern has been compiled,
 398    the fields 're_nsub', 'not_bol' and 'not_eol' are available.  All
 399    other fields are private to the regex routines.  */
 400
 401 #ifndef RE_TRANSLATE_TYPE
 402 # define __RE_TRANSLATE_TYPE unsigned char *
 403 # ifdef __USE_GNU
 404 #  define RE_TRANSLATE_TYPE __RE_TRANSLATE_TYPE
 405 # endif
 406 #endif
 407
 408 #ifdef __USE_GNU
 409 # define __REPB_PREFIX(name) name
 410 #else
 411 # define __REPB_PREFIX(name) __##name
 412 #endif
 413
 414 struct re_pattern_buffer
 415 {
 416   /* Space that holds the compiled pattern.  The type
 417      'struct re_dfa_t' is private and is not declared here.  */
 418   struct re_dfa_t *__REPB_PREFIX(buffer);
 419
 420   /* Number of bytes to which 'buffer' points.  */
 421   __re_long_size_t __REPB_PREFIX(allocated);
 422
 423   /* Number of bytes actually used in 'buffer'.  */
 424   __re_long_size_t __REPB_PREFIX(used);
 425
 426   /* Syntax setting with which the pattern was compiled.  */
 427   reg_syntax_t __REPB_PREFIX(syntax);
 428
 429   /* Pointer to a fastmap, if any, otherwise zero.  re_search uses the
 430      fastmap, if there is one, to skip over impossible starting points
 431      for matches.  */
 432   char *__REPB_PREFIX(fastmap);
 433
 434   /* Either a translate table to apply to all characters before
 435      comparing them, or zero for no translation.  The translation is
 436      applied to a pattern when it is compiled and to a string when it
 437      is matched.  */
 438   __RE_TRANSLATE_TYPE __REPB_PREFIX(translate);
 439
 440   /* Number of subexpressions found by the compiler.  */
 441   size_t re_nsub;
 442
 443   /* Zero if this pattern cannot match the empty string, one else.
 444      Well, in truth it's used only in 're_search_2', to see whether or
 445      not we should use the fastmap, so we don't set this absolutely
 446      perfectly; see 're_compile_fastmap' (the "duplicate" case).  */
 447   unsigned __REPB_PREFIX(can_be_null) : 1;
 448
 449   /* If REGS_UNALLOCATED, allocate space in the 'regs' structure
 450      for 'max (RE_NREGS, re_nsub + 1)' groups.
 451      If REGS_REALLOCATE, reallocate space if necessary.
 452      If REGS_FIXED, use what's there.  */
 453 #ifdef __USE_GNU
 454 # define REGS_UNALLOCATED 0
 455 # define REGS_REALLOCATE 1
 456 # define REGS_FIXED 2
 457 #endif
 458   unsigned __REPB_PREFIX(regs_allocated) : 2;
 459
 460   /* Set to zero when 're_compile_pattern' compiles a pattern; set to
 461      one by 're_compile_fastmap' if it updates the fastmap.  */
 462   unsigned __REPB_PREFIX(fastmap_accurate) : 1;
 463
 464   /* If set, 're_match_2' does not return information about
 465      subexpressions.  */
 466   unsigned __REPB_PREFIX(no_sub) : 1;
 467
 468   /* If set, a beginning-of-line anchor doesn't match at the beginning
 469      of the string.  */
 470   unsigned __REPB_PREFIX(not_bol) : 1;
 471
 472   /* Similarly for an end-of-line anchor.  */
 473   unsigned __REPB_PREFIX(not_eol) : 1;
 474
 475   /* If true, an anchor at a newline matches.  */
 476   unsigned __REPB_PREFIX(newline_anchor) : 1;
 477 };
 478
 479 typedef struct re_pattern_buffer regex_t;
 480 \f
 481 /* Type for byte offsets within the string.  POSIX mandates this.  */
 482 #ifdef _REGEX_LARGE_OFFSETS
 483 /* POSIX 1003.1-2008 requires that regoff_t be at least as wide as
 484    ptrdiff_t and ssize_t.  We don't know of any hosts where ptrdiff_t
 485    is wider than ssize_t, so ssize_t is safe.  ptrdiff_t is not
 486    visible here, so use ssize_t.  */
 487 typedef ssize_t regoff_t;
 488 #else
 489 /* The traditional GNU regex implementation mishandles strings longer
 490    than INT_MAX.  */
 491 typedef int regoff_t;
 492 #endif
 493
 494
 495 #ifdef __USE_GNU
 496 /* This is the structure we store register match data in.  See
 497    regex.texinfo for a full description of what registers match.  */
 498 struct re_registers
 499 {
 500   __re_size_t num_regs;
 501   regoff_t *start;
 502   regoff_t *end;
 503 };
 504
 505
 506 /* If 'regs_allocated' is REGS_UNALLOCATED in the pattern buffer,
 507    're_match_2' returns information about at least this many registers
 508    the first time a 'regs' structure is passed.  */
 509 # ifndef RE_NREGS
 510 #  define RE_NREGS 30
 511 # endif
 512 #endif
 513
 514
 515 /* POSIX specification for registers.  Aside from the different names than
 516    're_registers', POSIX uses an array of structures, instead of a
 517    structure of arrays.  */
 518 typedef struct
 519 {
 520   regoff_t rm_so;  /* Byte offset from string's start to substring's start.  */
 521   regoff_t rm_eo;  /* Byte offset from string's start to substring's end.  */
 522 } regmatch_t;
 523 \f
 524 /* Declarations for routines.  */
 525
 526 #ifdef __USE_GNU
 527 /* Sets the current default syntax to SYNTAX, and return the old syntax.
 528    You can also simply assign to the 're_syntax_options' variable.  */
 529 extern reg_syntax_t re_set_syntax (reg_syntax_t __syntax);
 530
 531 /* Compile the regular expression PATTERN, with length LENGTH
 532    and syntax given by the global 're_syntax_options', into the buffer
 533    BUFFER.  Return NULL if successful, and an error string if not.
 534
 535    To free the allocated storage, you must call 'regfree' on BUFFER.
 536    Note that the translate table must either have been initialized by
 537    'regcomp', with a malloc'ed value, or set to NULL before calling
 538    'regfree'.  */
 539 extern const char *re_compile_pattern (const char *__pattern, size_t __length,
 540                                        struct re_pattern_buffer *__buffer);
 541
 542
 543 /* Compile a fastmap for the compiled pattern in BUFFER; used to
 544    accelerate searches.  Return 0 if successful and -2 if was an
 545    internal error.  */
 546 extern int re_compile_fastmap (struct re_pattern_buffer *__buffer);
 547
 548
 549 /* Search in the string STRING (with length LENGTH) for the pattern
 550    compiled into BUFFER.  Start searching at position START, for RANGE
 551    characters.  Return the starting position of the match, -1 for no
 552    match, or -2 for an internal error.  Also return register
 553    information in REGS (if REGS and BUFFER->no_sub are nonzero).  */
 554 extern regoff_t re_search (struct re_pattern_buffer *__buffer,
 555                            const char *__String, regoff_t __length,
 556                            regoff_t __start, regoff_t __range,
 557                            struct re_registers *__regs);
 558
 559
 560 /* Like 're_search', but search in the concatenation of STRING1 and
 561    STRING2.  Also, stop searching at index START + STOP.  */
 562 extern regoff_t re_search_2 (struct re_pattern_buffer *__buffer,
 563                              const char *__string1, regoff_t __length1,
 564                              const char *__string2, regoff_t __length2,
 565                              regoff_t __start, regoff_t __range,
 566                              struct re_registers *__regs,
 567                              regoff_t __stop);
 568
 569
 570 /* Like 're_search', but return how many characters in STRING the regexp
 571    in BUFFER matched, starting at position START.  */
 572 extern regoff_t re_match (struct re_pattern_buffer *__buffer,
 573                           const char *__String, regoff_t __length,
 574                           regoff_t __start, struct re_registers *__regs);
 575
 576
 577 /* Relates to 're_match' as 're_search_2' relates to 're_search'.  */
 578 extern regoff_t re_match_2 (struct re_pattern_buffer *__buffer,
 579                             const char *__string1, regoff_t __length1,
 580                             const char *__string2, regoff_t __length2,
 581                             regoff_t __start, struct re_registers *__regs,
 582                             regoff_t __stop);
 583
 584
 585 /* Set REGS to hold NUM_REGS registers, storing them in STARTS and
 586    ENDS.  Subsequent matches using BUFFER and REGS will use this memory
 587    for recording register information.  STARTS and ENDS must be
 588    allocated with malloc, and must each be at least 'NUM_REGS * sizeof
 589    (regoff_t)' bytes long.
 590
 591    If NUM_REGS == 0, then subsequent matches should allocate their own
 592    register data.
 593
 594    Unless this function is called, the first search or match using
 595    BUFFER will allocate its own register data, without
 596    freeing the old data.  */
 597 extern void re_set_registers (struct re_pattern_buffer *__buffer,
 598                               struct re_registers *__regs,
 599                               __re_size_t __num_regs,
 600                               regoff_t *__starts, regoff_t *__ends);
 601 #endif  /* Use GNU */
 602
 603 #if defined _REGEX_RE_COMP || (defined _LIBC && defined __USE_MISC)
 604 # ifndef _CRAY
 605 /* 4.2 bsd compatibility.  */
 606 extern char *re_comp (const char *);
 607 extern int re_exec (const char *);
 608 # endif
 609 #endif
 610
 611 /* GCC 2.95 and later have "__restrict"; C99 compilers have
 612    "restrict", and "configure" may have defined "restrict".
 613    Other compilers use __restrict, __restrict__, and _Restrict, and
 614    'configure' might #define 'restrict' to those words, so pick a
 615    different name.  */
 616 #ifndef _Restrict_
 617 # if 199901L <= __STDC_VERSION__
 618 #  define _Restrict_ restrict
 619 # elif 2 < __GNUC__ || (2 == __GNUC__ && 95 <= __GNUC_MINOR__)
 620 #  define _Restrict_ __restrict
 621 # else
 622 #  define _Restrict_
 623 # endif
 624 #endif
 625 /* gcc 3.1 and up support the [restrict] syntax.  Don't trust
 626    sys/cdefs.h's definition of __restrict_arr, though, as it
 627    mishandles gcc -ansi -pedantic.  */
 628 #ifndef _Restrict_arr_
 629 # if ((199901L <= __STDC_VERSION__                                      \
 630        || ((3 < __GNUC__ || (3 == __GNUC__ && 1 <= __GNUC_MINOR__))     \
 631            && !defined __STRICT_ANSI__))                                        \
 632       && !defined __GNUG__)
 633 #  define _Restrict_arr_ _Restrict_
 634 # else
 635 #  define _Restrict_arr_
 636 # endif
 637 #endif
 638
 639 /* POSIX compatibility.  */
 640 extern int regcomp (regex_t *_Restrict_ __preg,
 641                     const char *_Restrict_ __pattern,
 642                     int __cflags);
 643
 644 extern int regexec (const regex_t *_Restrict_ __preg,
 645                     const char *_Restrict_ __String, size_t __nmatch,
 646                     regmatch_t __pmatch[_Restrict_arr_],
 647                     int __eflags);
 648
 649 extern size_t regerror (int __errcode, const regex_t *_Restrict_ __preg,
 650                         char *_Restrict_ __errbuf, size_t __errbuf_size);
 651
 652 extern void regfree (regex_t *__preg);
 653
 654
 655 #ifdef __cplusplus
 656 }
 657 #endif  /* C++ */
 658
 659 #endif /* regex.h */