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, 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_CHAR_CLASSES                                                    \
 198    | RE_UNMATCHED_RIGHT_PAREN_ORD | RE_NO_GNU_OPS)
 199
 200 #define RE_SYNTAX_GNU_AWK                                               \
 201   ((RE_SYNTAX_POSIX_EXTENDED | RE_BACKSLASH_ESCAPE_IN_LISTS             \
 202     | RE_INVALID_INTERVAL_ORD)                                          \
 203    & ~(RE_DOT_NOT_NULL | RE_CONTEXT_INDEP_OPS                           \
 204       | RE_CONTEXT_INVALID_OPS ))
 205
 206 #define RE_SYNTAX_POSIX_AWK                                             \
 207   (RE_SYNTAX_POSIX_EXTENDED | RE_BACKSLASH_ESCAPE_IN_LISTS              \
 208    | RE_INTERVALS           | RE_NO_GNU_OPS                             \
 209    | RE_INVALID_INTERVAL_ORD)
 210
 211 #define RE_SYNTAX_GREP                                                  \
 212   (RE_BK_PLUS_QM              | RE_CHAR_CLASSES                         \
 213    | RE_HAT_LISTS_NOT_NEWLINE | RE_INTERVALS                            \
 214    | RE_NEWLINE_ALT)
 215
 216 #define RE_SYNTAX_EGREP                                                 \
 217   (RE_CHAR_CLASSES        | RE_CONTEXT_INDEP_ANCHORS                    \
 218    | RE_CONTEXT_INDEP_OPS | RE_HAT_LISTS_NOT_NEWLINE                    \
 219    | RE_NEWLINE_ALT       | RE_NO_BK_PARENS                             \
 220    | RE_NO_BK_VBAR)
 221
 222 #define RE_SYNTAX_POSIX_EGREP                                           \
 223   (RE_SYNTAX_EGREP | RE_INTERVALS | RE_NO_BK_BRACES                     \
 224    | RE_INVALID_INTERVAL_ORD)
 225
 226 /* P1003.2/D11.2, section 4.20.7.1, lines 5078ff.  */
 227 #define RE_SYNTAX_ED RE_SYNTAX_POSIX_BASIC
 228
 229 #define RE_SYNTAX_SED RE_SYNTAX_POSIX_BASIC
 230
 231 /* Syntax bits common to both basic and extended POSIX regex syntax.  */
 232 #define _RE_SYNTAX_POSIX_COMMON                                         \
 233   (RE_CHAR_CLASSES | RE_DOT_NEWLINE      | RE_DOT_NOT_NULL              \
 234    | RE_INTERVALS  | RE_NO_EMPTY_RANGES)
 235
 236 #define RE_SYNTAX_POSIX_BASIC                                           \
 237   (_RE_SYNTAX_POSIX_COMMON | RE_BK_PLUS_QM | RE_CONTEXT_INVALID_DUP)
 238
 239 /* Differs from ..._POSIX_BASIC only in that RE_BK_PLUS_QM becomes
 240    RE_LIMITED_OPS, i.e., \? \+ \| are not recognized.  Actually, this
 241    isn't minimal, since other operators, such as \`, aren't disabled.  */
 242 #define RE_SYNTAX_POSIX_MINIMAL_BASIC                                   \
 243   (_RE_SYNTAX_POSIX_COMMON | RE_LIMITED_OPS)
 244
 245 #define RE_SYNTAX_POSIX_EXTENDED                                        \
 246   (_RE_SYNTAX_POSIX_COMMON  | RE_CONTEXT_INDEP_ANCHORS                  \
 247    | RE_CONTEXT_INDEP_OPS   | RE_NO_BK_BRACES                           \
 248    | RE_NO_BK_PARENS        | RE_NO_BK_VBAR                             \
 249    | RE_CONTEXT_INVALID_OPS | RE_UNMATCHED_RIGHT_PAREN_ORD)
 250
 251 /* Differs from ..._POSIX_EXTENDED in that RE_CONTEXT_INDEP_OPS is
 252    removed and RE_NO_BK_REFS is added.  */
 253 #define RE_SYNTAX_POSIX_MINIMAL_EXTENDED                                \
 254   (_RE_SYNTAX_POSIX_COMMON  | RE_CONTEXT_INDEP_ANCHORS                  \
 255    | RE_CONTEXT_INVALID_OPS | RE_NO_BK_BRACES                           \
 256    | RE_NO_BK_PARENS        | RE_NO_BK_REFS                             \
 257    | RE_NO_BK_VBAR          | RE_UNMATCHED_RIGHT_PAREN_ORD)
 258 /* [[[end syntaxes]]] */
 259 \f
 260 /* Maximum number of duplicates an interval can allow.  Some systems
 261    (erroneously) define this in other header files, but we want our
 262    value, so remove any previous define.  */
 263 # ifdef RE_DUP_MAX
 264 #  undef RE_DUP_MAX
 265 # endif
 266 /* If sizeof(int) == 2, then ((1 << 15) - 1) overflows.  */
 267 # define RE_DUP_MAX (0x7fff)
 268 #endif
 269
 270
 271 /* POSIX `cflags' bits (i.e., information for `regcomp').  */
 272
 273 /* If this bit is set, then use extended regular expression syntax.
 274    If not set, then use basic regular expression syntax.  */
 275 #define REG_EXTENDED 1
 276
 277 /* If this bit is set, then ignore case when matching.
 278    If not set, then case is significant.  */
 279 #define REG_ICASE (REG_EXTENDED << 1)
 280
 281 /* If this bit is set, then anchors do not match at newline
 282      characters in the string.
 283    If not set, then anchors do match at newlines.  */
 284 #define REG_NEWLINE (REG_ICASE << 1)
 285
 286 /* If this bit is set, then report only success or fail in regexec.
 287    If not set, then returns differ between not matching and errors.  */
 288 #define REG_NOSUB (REG_NEWLINE << 1)
 289
 290
 291 /* POSIX `eflags' bits (i.e., information for regexec).  */
 292
 293 /* If this bit is set, then the beginning-of-line operator doesn't match
 294      the beginning of the string (presumably because it's not the
 295      beginning of a line).
 296    If not set, then the beginning-of-line operator does match the
 297      beginning of the string.  */
 298 #define REG_NOTBOL 1
 299
 300 /* Like REG_NOTBOL, except for the end-of-line.  */
 301 #define REG_NOTEOL (1 << 1)
 302
 303 /* Use PMATCH[0] to delimit the start and end of the search in the
 304    buffer.  */
 305 #define REG_STARTEND (1 << 2)
 306
 307
 308 /* If any error codes are removed, changed, or added, update the
 309    `re_error_msg' table in regex.c.  */
 310 typedef enum
 311 {
 312 #if defined _XOPEN_SOURCE || defined __USE_XOPEN2K
 313   REG_ENOSYS = -1,      /* This will never happen for this implementation.  */
 314 #endif
 315
 316   REG_NOERROR = 0,      /* Success.  */
 317   REG_NOMATCH,          /* Didn't find a match (for regexec).  */
 318
 319   /* POSIX regcomp return error codes.  (In the order listed in the
 320      standard.)  */
 321   REG_BADPAT,           /* Invalid pattern.  */
 322   REG_ECOLLATE,         /* Inalid collating element.  */
 323   REG_ECTYPE,           /* Invalid character class name.  */
 324   REG_EESCAPE,          /* Trailing backslash.  */
 325   REG_ESUBREG,          /* Invalid back reference.  */
 326   REG_EBRACK,           /* Unmatched left bracket.  */
 327   REG_EPAREN,           /* Parenthesis imbalance.  */
 328   REG_EBRACE,           /* Unmatched \{.  */
 329   REG_BADBR,            /* Invalid contents of \{\}.  */
 330   REG_ERANGE,           /* Invalid range end.  */
 331   REG_ESPACE,           /* Ran out of memory.  */
 332   REG_BADRPT,           /* No preceding re for repetition op.  */
 333
 334   /* Error codes we've added.  */
 335   REG_EEND,             /* Premature end.  */
 336   REG_ESIZE,            /* Compiled pattern bigger than 2^16 bytes.  */
 337   REG_ERPAREN           /* Unmatched ) or \); not returned from regcomp.  */
 338 } reg_errcode_t;
 339 \f
 340 /* This data structure represents a compiled pattern.  Before calling
 341    the pattern compiler, the fields `buffer', `allocated', `fastmap',
 342    and `translate' can be set.  After the pattern has been compiled,
 343    the fields `re_nsub', `not_bol' and `not_eol' are available.  All
 344    other fields are private to the regex routines.  */
 345
 346 #ifndef RE_TRANSLATE_TYPE
 347 # define __RE_TRANSLATE_TYPE unsigned char *
 348 # ifdef __USE_GNU
 349 #  define RE_TRANSLATE_TYPE __RE_TRANSLATE_TYPE
 350 # endif
 351 #endif
 352
 353 #ifdef __USE_GNU
 354 # define __REPB_PREFIX(name) name
 355 #else
 356 # define __REPB_PREFIX(name) __##name
 357 #endif
 358
 359 struct re_pattern_buffer
 360 {
 361   /* Space that holds the compiled pattern.  It is declared as
 362      `unsigned char *' because its elements are sometimes used as
 363      array indexes.  */
 364   unsigned char *__REPB_PREFIX(buffer);
 365
 366   /* Number of bytes to which `buffer' points.  */
 367   unsigned long int __REPB_PREFIX(allocated);
 368
 369   /* Number of bytes actually used in `buffer'.  */
 370   unsigned long int __REPB_PREFIX(used);
 371
 372   /* Syntax setting with which the pattern was compiled.  */
 373   reg_syntax_t __REPB_PREFIX(syntax);
 374
 375   /* Pointer to a fastmap, if any, otherwise zero.  re_search uses the
 376      fastmap, if there is one, to skip over impossible starting points
 377      for matches.  */
 378   char *__REPB_PREFIX(fastmap);
 379
 380   /* Either a translate table to apply to all characters before
 381      comparing them, or zero for no translation.  The translation is
 382      applied to a pattern when it is compiled and to a string when it
 383      is matched.  */
 384   __RE_TRANSLATE_TYPE __REPB_PREFIX(translate);
 385
 386   /* Number of subexpressions found by the compiler.  */
 387   size_t re_nsub;
 388
 389   /* Zero if this pattern cannot match the empty string, one else.
 390      Well, in truth it's used only in `re_search_2', to see whether or
 391      not we should use the fastmap, so we don't set this absolutely
 392      perfectly; see `re_compile_fastmap' (the `duplicate' case).  */
 393   unsigned __REPB_PREFIX(can_be_null) : 1;
 394
 395   /* If REGS_UNALLOCATED, allocate space in the `regs' structure
 396      for `max (RE_NREGS, re_nsub + 1)' groups.
 397      If REGS_REALLOCATE, reallocate space if necessary.
 398      If REGS_FIXED, use what's there.  */
 399 #ifdef __USE_GNU
 400 # define REGS_UNALLOCATED 0
 401 # define REGS_REALLOCATE 1
 402 # define REGS_FIXED 2
 403 #endif
 404   unsigned __REPB_PREFIX(regs_allocated) : 2;
 405
 406   /* Set to zero when `regex_compile' compiles a pattern; set to one
 407      by `re_compile_fastmap' if it updates the fastmap.  */
 408   unsigned __REPB_PREFIX(fastmap_accurate) : 1;
 409
 410   /* If set, `re_match_2' does not return information about
 411      subexpressions.  */
 412   unsigned __REPB_PREFIX(no_sub) : 1;
 413
 414   /* If set, a beginning-of-line anchor doesn't match at the beginning
 415      of the string.  */
 416   unsigned __REPB_PREFIX(not_bol) : 1;
 417
 418   /* Similarly for an end-of-line anchor.  */
 419   unsigned __REPB_PREFIX(not_eol) : 1;
 420
 421   /* If true, an anchor at a newline matches.  */
 422   unsigned __REPB_PREFIX(newline_anchor) : 1;
 423 };
 424
 425 typedef struct re_pattern_buffer regex_t;
 426 \f
 427 /* Type for byte offsets within the string.  POSIX mandates this.  */
 428 typedef int regoff_t;
 429
 430
 431 #ifdef __USE_GNU
 432 /* This is the structure we store register match data in.  See
 433    regex.texinfo for a full description of what registers match.  */
 434 struct re_registers
 435 {
 436   unsigned num_regs;
 437   regoff_t *start;
 438   regoff_t *end;
 439 };
 440
 441
 442 /* If `regs_allocated' is REGS_UNALLOCATED in the pattern buffer,
 443    `re_match_2' returns information about at least this many registers
 444    the first time a `regs' structure is passed.  */
 445 # ifndef RE_NREGS
 446 #  define RE_NREGS 30
 447 # endif
 448 #endif
 449
 450
 451 /* POSIX specification for registers.  Aside from the different names than
 452    `re_registers', POSIX uses an array of structures, instead of a
 453    structure of arrays.  */
 454 typedef struct
 455 {
 456   regoff_t rm_so;  /* Byte offset from string's start to substring's start.  */
 457   regoff_t rm_eo;  /* Byte offset from string's start to substring's end.  */
 458 } regmatch_t;
 459 \f
 460 /* Declarations for routines.  */
 461
 462 #ifdef __USE_GNU
 463 /* Sets the current default syntax to SYNTAX, and return the old syntax.
 464    You can also simply assign to the `re_syntax_options' variable.  */
 465 extern reg_syntax_t re_set_syntax (reg_syntax_t __syntax);
 466
 467 /* Compile the regular expression PATTERN, with length LENGTH
 468    and syntax given by the global `re_syntax_options', into the buffer
 469    BUFFER.  Return NULL if successful, and an error string if not.
 470
 471    To free the allocated storage, you must call `regfree' on BUFFER.
 472    Note that the translate table must either have been initialised by
 473    `regcomp', with a malloc'ed value, or set to NULL before calling
 474    `regfree'.  */
 475 extern const char *re_compile_pattern (const char *__pattern, size_t __length,
 476                                        struct re_pattern_buffer *__buffer);
 477
 478
 479 /* Compile a fastmap for the compiled pattern in BUFFER; used to
 480    accelerate searches.  Return 0 if successful and -2 if was an
 481    internal error.  */
 482 extern int re_compile_fastmap (struct re_pattern_buffer *__buffer);
 483
 484
 485 /* Search in the string STRING (with length LENGTH) for the pattern
 486    compiled into BUFFER.  Start searching at position START, for RANGE
 487    characters.  Return the starting position of the match, -1 for no
 488    match, or -2 for an internal error.  Also return register
 489    information in REGS (if REGS and BUFFER->no_sub are nonzero).  */
 490 extern int re_search (struct re_pattern_buffer *__buffer, const char *__string,
 491                       int __length, int __start, int __range,
 492                       struct re_registers *__regs);
 493
 494
 495 /* Like `re_search', but search in the concatenation of STRING1 and
 496    STRING2.  Also, stop searching at index START + STOP.  */
 497 extern int re_search_2 (struct re_pattern_buffer *__buffer,
 498                         const char *__string1, int __length1,
 499                         const char *__string2, int __length2, int __start,
 500                         int __range, struct re_registers *__regs, int __stop);
 501
 502
 503 /* Like `re_search', but return how many characters in STRING the regexp
 504    in BUFFER matched, starting at position START.  */
 505 extern int re_match (struct re_pattern_buffer *__buffer, const char *__string,
 506                      int __length, int __start, struct re_registers *__regs);
 507
 508
 509 /* Relates to `re_match' as `re_search_2' relates to `re_search'.  */
 510 extern int re_match_2 (struct re_pattern_buffer *__buffer,
 511                        const char *__string1, int __length1,
 512                        const char *__string2, int __length2, int __start,
 513                        struct re_registers *__regs, int __stop);
 514
 515
 516 /* Set REGS to hold NUM_REGS registers, storing them in STARTS and
 517    ENDS.  Subsequent matches using BUFFER and REGS will use this memory
 518    for recording register information.  STARTS and ENDS must be
 519    allocated with malloc, and must each be at least `NUM_REGS * sizeof
 520    (regoff_t)' bytes long.
 521
 522    If NUM_REGS == 0, then subsequent matches should allocate their own
 523    register data.
 524
 525    Unless this function is called, the first search or match using
 526    PATTERN_BUFFER will allocate its own register data, without
 527    freeing the old data.  */
 528 extern void re_set_registers (struct re_pattern_buffer *__buffer,
 529                               struct re_registers *__regs,
 530                               unsigned int __num_regs,
 531                               regoff_t *__starts, regoff_t *__ends);
 532 #endif  /* Use GNU */
 533
 534 #if defined _REGEX_RE_COMP || (defined _LIBC && defined __USE_BSD)
 535 # ifndef _CRAY
 536 /* 4.2 bsd compatibility.  */
 537 extern char *re_comp (const char *);
 538 extern int re_exec (const char *);
 539 # endif
 540 #endif
 541
 542 /* GCC 2.95 and later have "__restrict"; C99 compilers have
 543    "restrict", and "configure" may have defined "restrict".  */
 544 #ifndef __restrict
 545 # if ! (2 < __GNUC__ || (2 == __GNUC__ && 95 <= __GNUC_MINOR__))
 546 #  if defined restrict || 199901L <= __STDC_VERSION__
 547 #   define __restrict restrict
 548 #  else
 549 #   define __restrict
 550 #  endif
 551 # endif
 552 #endif
 553 /* gcc 3.1 and up support the [restrict] syntax.  */
 554 #ifndef __restrict_arr
 555 # if (__GNUC__ > 3 || (__GNUC__ == 3 && __GNUC_MINOR__ >= 1)) \
 556      && !defined __GNUG__
 557 #  define __restrict_arr __restrict
 558 # else
 559 #  define __restrict_arr
 560 # endif
 561 #endif
 562
 563 /* POSIX compatibility.  */
 564 extern int regcomp (regex_t *__restrict __preg,
 565                     const char *__restrict __pattern,
 566                     int __cflags);
 567
 568 extern int regexec (const regex_t *__restrict __preg,
 569                     const char *__restrict __string, size_t __nmatch,
 570                     regmatch_t __pmatch[__restrict_arr],
 571                     int __eflags);
 572
 573 extern size_t regerror (int __errcode, const regex_t *__restrict __preg,
 574                         char *__restrict __errbuf, size_t __errbuf_size);
 575
 576 extern void regfree (regex_t *__preg);
 577
 578
 579 #ifdef __cplusplus
 580 }
 581 #endif  /* C++ */
 582
 583 #endif /* regex.h */