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
   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 /* POSIX says that <sys/types.h> must be included (by the caller) before
  33    <regex.h>.  */
  34
  35 #if !defined _POSIX_C_SOURCE && !defined _POSIX_SOURCE && defined VMS
  36 /* VMS doesn't have `size_t' in <sys/types.h>, even though POSIX says it
  37    should be there.  */
  38 # include <stddef.h>
  39 #endif
  40
  41 /* The following two types have to be signed and unsigned integer type
  42    wide enough to hold a value of a pointer.  For most ANSI compilers
  43    ptrdiff_t and size_t should be likely OK.  Still size of these two
  44    types is 2 for Microsoft C.  Ugh... */
  45 typedef long int s_reg_t;
  46 typedef unsigned long int active_reg_t;
  47
  48 /* The following bits are used to determine the regexp syntax we
  49    recognize.  The set/not-set meanings are chosen so that Emacs syntax
  50    remains the value 0.  The bits are given in alphabetical order, and
  51    the definitions shifted by one from the previous bit; thus, when we
  52    add or remove a bit, only one other definition need change.  */
  53 typedef unsigned long int reg_syntax_t;
  54
  55 /* If this bit is not set, then \ inside a bracket expression is literal.
  56    If set, then such a \ quotes the following character.  */
  57 #define RE_BACKSLASH_ESCAPE_IN_LISTS ((unsigned long int) 1)
  58
  59 /* If this bit is not set, then + and ? are operators, and \+ and \? are
  60      literals.
  61    If set, then \+ and \? are operators and + and ? are literals.  */
  62 #define RE_BK_PLUS_QM (RE_BACKSLASH_ESCAPE_IN_LISTS << 1)
  63
  64 /* If this bit is set, then character classes are supported.  They are:
  65      [:alpha:], [:upper:], [:lower:],  [:digit:], [:alnum:], [:xdigit:],
  66      [:space:], [:print:], [:punct:], [:graph:], and [:cntrl:].
  67    If not set, then character classes are not supported.  */
  68 #define RE_CHAR_CLASSES (RE_BK_PLUS_QM << 1)
  69
  70 /* If this bit is set, then ^ and $ are always anchors (outside bracket
  71      expressions, of course).
  72    If this bit is not set, then it depends:
  73         ^  is an anchor if it is at the beginning of a regular
  74            expression or after an open-group or an alternation operator;
  75         $  is an anchor if it is at the end of a regular expression, or
  76            before a close-group or an alternation operator.
  77
  78    This bit could be (re)combined with RE_CONTEXT_INDEP_OPS, because
  79    POSIX draft 11.2 says that * etc. in leading positions is undefined.
  80    We already implemented a previous draft which made those constructs
  81    invalid, though, so we haven't changed the code back.  */
  82 #define RE_CONTEXT_INDEP_ANCHORS (RE_CHAR_CLASSES << 1)
  83
  84 /* If this bit is set, then special characters are always special
  85      regardless of where they are in the pattern.
  86    If this bit is not set, then special characters are special only in
  87      some contexts; otherwise they are ordinary.  Specifically,
  88      * + ? and intervals are only special when not after the beginning,
  89      open-group, or alternation operator.  */
  90 #define RE_CONTEXT_INDEP_OPS (RE_CONTEXT_INDEP_ANCHORS << 1)
  91
  92 /* If this bit is set, then *, +, ?, and { cannot be first in an re or
  93      immediately after an alternation or begin-group operator.  */
  94 #define RE_CONTEXT_INVALID_OPS (RE_CONTEXT_INDEP_OPS << 1)
  95
  96 /* If this bit is set, then . matches newline.
  97    If not set, then it doesn't.  */
  98 #define RE_DOT_NEWLINE (RE_CONTEXT_INVALID_OPS << 1)
  99
 100 /* If this bit is set, then . doesn't match NUL.
 101    If not set, then it does.  */
 102 #define RE_DOT_NOT_NULL (RE_DOT_NEWLINE << 1)
 103
 104 /* If this bit is set, nonmatching lists [^...] do not match newline.
 105    If not set, they do.  */
 106 #define RE_HAT_LISTS_NOT_NEWLINE (RE_DOT_NOT_NULL << 1)
 107
 108 /* If this bit is set, either \{...\} or {...} defines an
 109      interval, depending on RE_NO_BK_BRACES.
 110    If not set, \{, \}, {, and } are literals.  */
 111 #define RE_INTERVALS (RE_HAT_LISTS_NOT_NEWLINE << 1)
 112
 113 /* If this bit is set, +, ? and | aren't recognized as operators.
 114    If not set, they are.  */
 115 #define RE_LIMITED_OPS (RE_INTERVALS << 1)
 116
 117 /* If this bit is set, newline is an alternation operator.
 118    If not set, newline is literal.  */
 119 #define RE_NEWLINE_ALT (RE_LIMITED_OPS << 1)
 120
 121 /* If this bit is set, then `{...}' defines an interval, and \{ and \}
 122      are literals.
 123   If not set, then `\{...\}' defines an interval.  */
 124 #define RE_NO_BK_BRACES (RE_NEWLINE_ALT << 1)
 125
 126 /* If this bit is set, (...) defines a group, and \( and \) are literals.
 127    If not set, \(...\) defines a group, and ( and ) are literals.  */
 128 #define RE_NO_BK_PARENS (RE_NO_BK_BRACES << 1)
 129
 130 /* If this bit is set, then \<digit> matches <digit>.
 131    If not set, then \<digit> is a back-reference.  */
 132 #define RE_NO_BK_REFS (RE_NO_BK_PARENS << 1)
 133
 134 /* If this bit is set, then | is an alternation operator, and \| is literal.
 135    If not set, then \| is an alternation operator, and | is literal.  */
 136 #define RE_NO_BK_VBAR (RE_NO_BK_REFS << 1)
 137
 138 /* If this bit is set, then an ending range point collating higher
 139      than the starting range point, as in [z-a], is invalid.
 140    If not set, then when ending range point collates higher than the
 141      starting range point, the range is ignored.  */
 142 #define RE_NO_EMPTY_RANGES (RE_NO_BK_VBAR << 1)
 143
 144 /* If this bit is set, then an unmatched ) is ordinary.
 145    If not set, then an unmatched ) is invalid.  */
 146 #define RE_UNMATCHED_RIGHT_PAREN_ORD (RE_NO_EMPTY_RANGES << 1)
 147
 148 /* If this bit is set, succeed as soon as we match the whole pattern,
 149    without further backtracking.  */
 150 #define RE_NO_POSIX_BACKTRACKING (RE_UNMATCHED_RIGHT_PAREN_ORD << 1)
 151
 152 /* If this bit is set, do not process the GNU regex operators.
 153    If not set, then the GNU regex operators are recognized. */
 154 #define RE_NO_GNU_OPS (RE_NO_POSIX_BACKTRACKING << 1)
 155
 156 /* If this bit is set, turn on internal regex debugging.
 157    If not set, and debugging was on, turn it off.
 158    This only works if regex.c is compiled -DDEBUG.
 159    We define this bit always, so that all that's needed to turn on
 160    debugging is to recompile regex.c; the calling code can always have
 161    this bit set, and it won't affect anything in the normal case. */
 162 #define RE_DEBUG (RE_NO_GNU_OPS << 1)
 163
 164 /* If this bit is set, a syntactically invalid interval is treated as
 165    a string of ordinary characters.  For example, the ERE 'a{1' is
 166    treated as 'a\{1'.  */
 167 #define RE_INVALID_INTERVAL_ORD (RE_DEBUG << 1)
 168
 169 /* If this bit is set, then ignore case when matching.
 170    If not set, then case is significant.  */
 171 #define RE_ICASE (RE_INVALID_INTERVAL_ORD << 1)
 172
 173 /* This bit is used internally like RE_CONTEXT_INDEP_ANCHORS but only
 174    for ^, because it is difficult to scan the regex backwards to find
 175    whether ^ should be special.  */
 176 #define RE_CARET_ANCHORS_HERE (RE_ICASE << 1)
 177
 178 /* If this bit is set, then \{ cannot be first in an bre or
 179    immediately after an alternation or begin-group operator.  */
 180 #define RE_CONTEXT_INVALID_DUP (RE_CARET_ANCHORS_HERE << 1)
 181
 182 /* This global variable defines the particular regexp syntax to use (for
 183    some interfaces).  When a regexp is compiled, the syntax used is
 184    stored in the pattern buffer, so changing this does not affect
 185    already-compiled regexps.  */
 186 extern reg_syntax_t re_syntax_options;
 187 \f
 188 /* Define combinations of the above bits for the standard possibilities.
 189    (The [[[ comments delimit what gets put into the Texinfo file, so
 190    don't delete them!)  */
 191 /* [[[begin syntaxes]]] */
 192 #define RE_SYNTAX_EMACS 0
 193
 194 #define RE_SYNTAX_AWK                                                   \
 195   (RE_BACKSLASH_ESCAPE_IN_LISTS   | RE_DOT_NOT_NULL                     \
 196    | RE_NO_BK_PARENS              | RE_NO_BK_REFS                       \
 197    | RE_NO_BK_VBAR                | RE_NO_EMPTY_RANGES                  \
 198    | RE_DOT_NEWLINE               | RE_CONTEXT_INDEP_ANCHORS            \
 199    | RE_UNMATCHED_RIGHT_PAREN_ORD | RE_NO_GNU_OPS)
 200
 201 #define RE_SYNTAX_GNU_AWK                                               \
 202   ((RE_SYNTAX_POSIX_EXTENDED | RE_BACKSLASH_ESCAPE_IN_LISTS | RE_DEBUG) \
 203    & ~(RE_DOT_NOT_NULL | RE_INTERVALS | 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
 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
 268
 269 /* POSIX `cflags' bits (i.e., information for `regcomp').  */
 270
 271 /* If this bit is set, then use extended regular expression syntax.
 272    If not set, then use basic regular expression syntax.  */
 273 #define REG_EXTENDED 1
 274
 275 /* If this bit is set, then ignore case when matching.
 276    If not set, then case is significant.  */
 277 #define REG_ICASE (REG_EXTENDED << 1)
 278
 279 /* If this bit is set, then anchors do not match at newline
 280      characters in the string.
 281    If not set, then anchors do match at newlines.  */
 282 #define REG_NEWLINE (REG_ICASE << 1)
 283
 284 /* If this bit is set, then report only success or fail in regexec.
 285    If not set, then returns differ between not matching and errors.  */
 286 #define REG_NOSUB (REG_NEWLINE << 1)
 287
 288
 289 /* POSIX `eflags' bits (i.e., information for regexec).  */
 290
 291 /* If this bit is set, then the beginning-of-line operator doesn't match
 292      the beginning of the string (presumably because it's not the
 293      beginning of a line).
 294    If not set, then the beginning-of-line operator does match the
 295      beginning of the string.  */
 296 #define REG_NOTBOL 1
 297
 298 /* Like REG_NOTBOL, except for the end-of-line.  */
 299 #define REG_NOTEOL (1 << 1)
 300
 301 /* Use PMATCH[0] to delimit the start and end of the search in the
 302    buffer.  */
 303 #define REG_STARTEND (1 << 2)
 304
 305
 306 /* If any error codes are removed, changed, or added, update the
 307    `re_error_msg' table in regex.c.  */
 308 typedef enum
 309 {
 310 #ifdef _XOPEN_SOURCE
 311   REG_ENOSYS = -1,      /* This will never happen for this implementation.  */
 312 #endif
 313
 314   REG_NOERROR = 0,      /* Success.  */
 315   REG_NOMATCH,          /* Didn't find a match (for regexec).  */
 316
 317   /* POSIX regcomp return error codes.  (In the order listed in the
 318      standard.)  */
 319   REG_BADPAT,           /* Invalid pattern.  */
 320   REG_ECOLLATE,         /* Inalid collating element.  */
 321   REG_ECTYPE,           /* Invalid character class name.  */
 322   REG_EESCAPE,          /* Trailing backslash.  */
 323   REG_ESUBREG,          /* Invalid back reference.  */
 324   REG_EBRACK,           /* Unmatched left bracket.  */
 325   REG_EPAREN,           /* Parenthesis imbalance.  */
 326   REG_EBRACE,           /* Unmatched \{.  */
 327   REG_BADBR,            /* Invalid contents of \{\}.  */
 328   REG_ERANGE,           /* Invalid range end.  */
 329   REG_ESPACE,           /* Ran out of memory.  */
 330   REG_BADRPT,           /* No preceding re for repetition op.  */
 331
 332   /* Error codes we've added.  */
 333   REG_EEND,             /* Premature end.  */
 334   REG_ESIZE,            /* Compiled pattern bigger than 2^16 bytes.  */
 335   REG_ERPAREN           /* Unmatched ) or \); not returned from regcomp.  */
 336 } reg_errcode_t;
 337 \f
 338 /* This data structure represents a compiled pattern.  Before calling
 339    the pattern compiler, the fields `buffer', `allocated', `fastmap',
 340    `translate', and `no_sub' can be set.  After the pattern has been
 341    compiled, the `re_nsub' field is available.  All other fields are
 342    private to the regex routines.  */
 343
 344 #ifndef RE_TRANSLATE_TYPE
 345 # define RE_TRANSLATE_TYPE char *
 346 #endif
 347
 348 struct re_pattern_buffer
 349 {
 350 /* [[[begin pattern_buffer]]] */
 351         /* Space that holds the compiled pattern.  It is declared as
 352           `unsigned char *' because its elements are
 353            sometimes used as array indexes.  */
 354   unsigned char *buffer;
 355
 356         /* Number of bytes to which `buffer' points.  */
 357   unsigned long int allocated;
 358
 359         /* Number of bytes actually used in `buffer'.  */
 360   unsigned long int used;
 361
 362         /* Syntax setting with which the pattern was compiled.  */
 363   reg_syntax_t syntax;
 364
 365         /* Pointer to a fastmap, if any, otherwise zero.  re_search uses
 366            the fastmap, if there is one, to skip over impossible
 367            starting points for matches.  */
 368   char *fastmap;
 369
 370         /* Either a translate table to apply to all characters before
 371            comparing them, or zero for no translation.  The translation
 372            is applied to a pattern when it is compiled and to a string
 373            when it is matched.  */
 374   RE_TRANSLATE_TYPE translate;
 375
 376         /* Number of subexpressions found by the compiler.  */
 377   size_t re_nsub;
 378
 379         /* Zero if this pattern cannot match the empty string, one else.
 380            Well, in truth it's used only in `re_search_2', to see
 381            whether or not we should use the fastmap, so we don't set
 382            this absolutely perfectly; see `re_compile_fastmap' (the
 383            `duplicate' case).  */
 384   unsigned can_be_null : 1;
 385
 386         /* If REGS_UNALLOCATED, allocate space in the `regs' structure
 387              for `max (RE_NREGS, re_nsub + 1)' groups.
 388            If REGS_REALLOCATE, reallocate space if necessary.
 389            If REGS_FIXED, use what's there.  */
 390 #define REGS_UNALLOCATED 0
 391 #define REGS_REALLOCATE 1
 392 #define REGS_FIXED 2
 393   unsigned regs_allocated : 2;
 394
 395         /* Set to zero when `regex_compile' compiles a pattern; set to one
 396            by `re_compile_fastmap' if it updates the fastmap.  */
 397   unsigned fastmap_accurate : 1;
 398
 399         /* If set, `re_match_2' does not return information about
 400            subexpressions.  */
 401   unsigned no_sub : 1;
 402
 403         /* If set, a beginning-of-line anchor doesn't match at the
 404            beginning of the string.  */
 405   unsigned not_bol : 1;
 406
 407         /* Similarly for an end-of-line anchor.  */
 408   unsigned not_eol : 1;
 409
 410         /* If true, an anchor at a newline matches.  */
 411   unsigned newline_anchor : 1;
 412
 413 /* [[[end pattern_buffer]]] */
 414 };
 415
 416 typedef struct re_pattern_buffer regex_t;
 417 \f
 418 /* Type for byte offsets within the string.  POSIX mandates this.  */
 419 typedef int regoff_t;
 420
 421
 422 /* This is the structure we store register match data in.  See
 423    regex.texinfo for a full description of what registers match.  */
 424 struct re_registers
 425 {
 426   unsigned num_regs;
 427   regoff_t *start;
 428   regoff_t *end;
 429 };
 430
 431
 432 /* If `regs_allocated' is REGS_UNALLOCATED in the pattern buffer,
 433    `re_match_2' returns information about at least this many registers
 434    the first time a `regs' structure is passed.  */
 435 #ifndef RE_NREGS
 436 # define RE_NREGS 30
 437 #endif
 438
 439
 440 /* POSIX specification for registers.  Aside from the different names than
 441    `re_registers', POSIX uses an array of structures, instead of a
 442    structure of arrays.  */
 443 typedef struct
 444 {
 445   regoff_t rm_so;  /* Byte offset from string's start to substring's start.  */
 446   regoff_t rm_eo;  /* Byte offset from string's start to substring's end.  */
 447 } regmatch_t;
 448 \f
 449 /* Declarations for routines.  */
 450
 451 /* To avoid duplicating every routine declaration -- once with a
 452    prototype (if we are ANSI), and once without (if we aren't) -- we
 453    use the following macro to declare argument types.  This
 454    unfortunately clutters up the declarations a bit, but I think it's
 455    worth it.  */
 456
 457 #if __STDC__
 458
 459 # define _RE_ARGS(args) args
 460
 461 #else /* not __STDC__ */
 462
 463 # define _RE_ARGS(args) ()
 464
 465 #endif /* not __STDC__ */
 466
 467 /* Sets the current default syntax to SYNTAX, and return the old syntax.
 468    You can also simply assign to the `re_syntax_options' variable.  */
 469 extern reg_syntax_t re_set_syntax _RE_ARGS ((reg_syntax_t syntax));
 470
 471 /* Compile the regular expression PATTERN, with length LENGTH
 472    and syntax given by the global `re_syntax_options', into the buffer
 473    BUFFER.  Return NULL if successful, and an error string if not.  */
 474 extern const char *re_compile_pattern
 475   _RE_ARGS ((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 _RE_ARGS ((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
 491   _RE_ARGS ((struct re_pattern_buffer *buffer, const char *string,
 492             int length, int start, int range, 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
 498   _RE_ARGS ((struct re_pattern_buffer *buffer, const char *string1,
 499              int length1, const char *string2, int length2,
 500              int start, 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
 506   _RE_ARGS ((struct re_pattern_buffer *buffer, const char *string,
 507              int length, int start, struct re_registers *regs));
 508
 509
 510 /* Relates to `re_match' as `re_search_2' relates to `re_search'.  */
 511 extern int re_match_2
 512   _RE_ARGS ((struct re_pattern_buffer *buffer, const char *string1,
 513              int length1, const char *string2, int length2,
 514              int start, struct re_registers *regs, int stop));
 515
 516
 517 /* Set REGS to hold NUM_REGS registers, storing them in STARTS and
 518    ENDS.  Subsequent matches using BUFFER and REGS will use this memory
 519    for recording register information.  STARTS and ENDS must be
 520    allocated with malloc, and must each be at least `NUM_REGS * sizeof
 521    (regoff_t)' bytes long.
 522
 523    If NUM_REGS == 0, then subsequent matches should allocate their own
 524    register data.
 525
 526    Unless this function is called, the first search or match using
 527    PATTERN_BUFFER will allocate its own register data, without
 528    freeing the old data.  */
 529 extern void re_set_registers
 530   _RE_ARGS ((struct re_pattern_buffer *buffer, struct re_registers *regs,
 531              unsigned num_regs, regoff_t *starts, regoff_t *ends));
 532
 533 #if defined _REGEX_RE_COMP || defined _LIBC
 534 # ifndef _CRAY
 535 /* 4.2 bsd compatibility.  */
 536 extern char *re_comp _RE_ARGS ((const char *));
 537 extern int re_exec _RE_ARGS ((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 #  define __restrict_arr __restrict
 556 # else
 557 #  define __restrict_arr
 558 # endif
 559 #endif
 560
 561 /* POSIX compatibility.  */
 562 extern int regcomp _RE_ARGS ((regex_t *__restrict __preg,
 563                               const char *__restrict __pattern,
 564                               int __cflags));
 565
 566 extern int regexec _RE_ARGS ((const regex_t *__restrict __preg,
 567                               const char *__restrict __string, size_t __nmatch,
 568                               regmatch_t __pmatch[__restrict_arr],
 569                               int __eflags));
 570
 571 extern size_t regerror _RE_ARGS ((int __errcode, const regex_t *__preg,
 572                                   char *__errbuf, size_t __errbuf_size));
 573
 574 extern void regfree _RE_ARGS ((regex_t *__preg));
 575
 576
 577 #ifdef __cplusplus
 578 }
 579 #endif  /* C++ */
 580
 581 #endif /* regex.h */
 582 \f
 583 /*
 584 Local variables:
 585 make-backup-files: t
 586 version-control: t
 587 trim-versions-without-asking: nil
 588 End:
 589 */