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 /* If this bit is set, then no_sub will be set to 1 during
 183    re_compile_pattern.  */
 184 #define RE_NO_SUB (RE_CONTEXT_INVALID_DUP << 1)
 185
 186 /* This global variable defines the particular regexp syntax to use (for
 187    some interfaces).  When a regexp is compiled, the syntax used is
 188    stored in the pattern buffer, so changing this does not affect
 189    already-compiled regexps.  */
 190 extern reg_syntax_t re_syntax_options;
 191 \f
 192 /* Define combinations of the above bits for the standard possibilities.
 193    (The [[[ comments delimit what gets put into the Texinfo file, so
 194    don't delete them!)  */
 195 /* [[[begin syntaxes]]] */
 196 #define RE_SYNTAX_EMACS 0
 197
 198 #define RE_SYNTAX_AWK                                                   \
 199   (RE_BACKSLASH_ESCAPE_IN_LISTS   | RE_DOT_NOT_NULL                     \
 200    | RE_NO_BK_PARENS              | RE_NO_BK_REFS                       \
 201    | RE_NO_BK_VBAR                | RE_NO_EMPTY_RANGES                  \
 202    | RE_DOT_NEWLINE               | RE_CONTEXT_INDEP_ANCHORS            \
 203    | RE_UNMATCHED_RIGHT_PAREN_ORD | RE_NO_GNU_OPS)
 204
 205 #define RE_SYNTAX_GNU_AWK                                               \
 206   ((RE_SYNTAX_POSIX_EXTENDED | RE_BACKSLASH_ESCAPE_IN_LISTS | RE_DEBUG) \
 207    & ~(RE_DOT_NOT_NULL | RE_INTERVALS | RE_CONTEXT_INDEP_OPS            \
 208        | RE_CONTEXT_INVALID_OPS ))
 209
 210 #define RE_SYNTAX_POSIX_AWK                                             \
 211   (RE_SYNTAX_POSIX_EXTENDED | RE_BACKSLASH_ESCAPE_IN_LISTS              \
 212    | RE_INTERVALS           | RE_NO_GNU_OPS)
 213
 214 #define RE_SYNTAX_GREP                                                  \
 215   (RE_BK_PLUS_QM              | RE_CHAR_CLASSES                         \
 216    | RE_HAT_LISTS_NOT_NEWLINE | RE_INTERVALS                            \
 217    | RE_NEWLINE_ALT)
 218
 219 #define RE_SYNTAX_EGREP                                                 \
 220   (RE_CHAR_CLASSES        | RE_CONTEXT_INDEP_ANCHORS                    \
 221    | RE_CONTEXT_INDEP_OPS | RE_HAT_LISTS_NOT_NEWLINE                    \
 222    | RE_NEWLINE_ALT       | RE_NO_BK_PARENS                             \
 223    | RE_NO_BK_VBAR)
 224
 225 #define RE_SYNTAX_POSIX_EGREP                                           \
 226   (RE_SYNTAX_EGREP | RE_INTERVALS | RE_NO_BK_BRACES                     \
 227    | RE_INVALID_INTERVAL_ORD)
 228
 229 /* P1003.2/D11.2, section 4.20.7.1, lines 5078ff.  */
 230 #define RE_SYNTAX_ED RE_SYNTAX_POSIX_BASIC
 231
 232 #define RE_SYNTAX_SED RE_SYNTAX_POSIX_BASIC
 233
 234 /* Syntax bits common to both basic and extended POSIX regex syntax.  */
 235 #define _RE_SYNTAX_POSIX_COMMON                                         \
 236   (RE_CHAR_CLASSES | RE_DOT_NEWLINE      | RE_DOT_NOT_NULL              \
 237    | RE_INTERVALS  | RE_NO_EMPTY_RANGES)
 238
 239 #define RE_SYNTAX_POSIX_BASIC                                           \
 240   (_RE_SYNTAX_POSIX_COMMON | RE_BK_PLUS_QM | RE_CONTEXT_INVALID_DUP)
 241
 242 /* Differs from ..._POSIX_BASIC only in that RE_BK_PLUS_QM becomes
 243    RE_LIMITED_OPS, i.e., \? \+ \| are not recognized.  Actually, this
 244    isn't minimal, since other operators, such as \`, aren't disabled.  */
 245 #define RE_SYNTAX_POSIX_MINIMAL_BASIC                                   \
 246   (_RE_SYNTAX_POSIX_COMMON | RE_LIMITED_OPS)
 247
 248 #define RE_SYNTAX_POSIX_EXTENDED                                        \
 249   (_RE_SYNTAX_POSIX_COMMON  | RE_CONTEXT_INDEP_ANCHORS                  \
 250    | RE_CONTEXT_INDEP_OPS   | RE_NO_BK_BRACES                           \
 251    | RE_NO_BK_PARENS        | RE_NO_BK_VBAR                             \
 252    | RE_CONTEXT_INVALID_OPS | RE_UNMATCHED_RIGHT_PAREN_ORD)
 253
 254 /* Differs from ..._POSIX_EXTENDED in that RE_CONTEXT_INDEP_OPS is
 255    removed and RE_NO_BK_REFS is added.  */
 256 #define RE_SYNTAX_POSIX_MINIMAL_EXTENDED                                \
 257   (_RE_SYNTAX_POSIX_COMMON  | RE_CONTEXT_INDEP_ANCHORS                  \
 258    | RE_CONTEXT_INVALID_OPS | RE_NO_BK_BRACES                           \
 259    | RE_NO_BK_PARENS        | RE_NO_BK_REFS                             \
 260    | RE_NO_BK_VBAR          | RE_UNMATCHED_RIGHT_PAREN_ORD)
 261 /* [[[end syntaxes]]] */
 262 \f
 263 /* Maximum number of duplicates an interval can allow.  Some systems
 264    (erroneously) define this in other header files, but we want our
 265    value, so remove any previous define.  */
 266 #ifdef RE_DUP_MAX
 267 # undef RE_DUP_MAX
 268 #endif
 269 /* If sizeof(int) == 2, then ((1 << 15) - 1) overflows.  */
 270 #define RE_DUP_MAX (0x7fff)
 271
 272
 273 /* POSIX `cflags' bits (i.e., information for `regcomp').  */
 274
 275 /* If this bit is set, then use extended regular expression syntax.
 276    If not set, then use basic regular expression syntax.  */
 277 #define REG_EXTENDED 1
 278
 279 /* If this bit is set, then ignore case when matching.
 280    If not set, then case is significant.  */
 281 #define REG_ICASE (REG_EXTENDED << 1)
 282
 283 /* If this bit is set, then anchors do not match at newline
 284      characters in the string.
 285    If not set, then anchors do match at newlines.  */
 286 #define REG_NEWLINE (REG_ICASE << 1)
 287
 288 /* If this bit is set, then report only success or fail in regexec.
 289    If not set, then returns differ between not matching and errors.  */
 290 #define REG_NOSUB (REG_NEWLINE << 1)
 291
 292
 293 /* POSIX `eflags' bits (i.e., information for regexec).  */
 294
 295 /* If this bit is set, then the beginning-of-line operator doesn't match
 296      the beginning of the string (presumably because it's not the
 297      beginning of a line).
 298    If not set, then the beginning-of-line operator does match the
 299      beginning of the string.  */
 300 #define REG_NOTBOL 1
 301
 302 /* Like REG_NOTBOL, except for the end-of-line.  */
 303 #define REG_NOTEOL (1 << 1)
 304
 305 /* Use PMATCH[0] to delimit the start and end of the search in the
 306    buffer.  */
 307 #define REG_STARTEND (1 << 2)
 308
 309
 310 /* If any error codes are removed, changed, or added, update the
 311    `re_error_msg' table in regex.c.  */
 312 typedef enum
 313 {
 314 #ifdef _XOPEN_SOURCE
 315   REG_ENOSYS = -1,      /* This will never happen for this implementation.  */
 316 #endif
 317
 318   REG_NOERROR = 0,      /* Success.  */
 319   REG_NOMATCH,          /* Didn't find a match (for regexec).  */
 320
 321   /* POSIX regcomp return error codes.  (In the order listed in the
 322      standard.)  */
 323   REG_BADPAT,           /* Invalid pattern.  */
 324   REG_ECOLLATE,         /* Inalid collating element.  */
 325   REG_ECTYPE,           /* Invalid character class name.  */
 326   REG_EESCAPE,          /* Trailing backslash.  */
 327   REG_ESUBREG,          /* Invalid back reference.  */
 328   REG_EBRACK,           /* Unmatched left bracket.  */
 329   REG_EPAREN,           /* Parenthesis imbalance.  */
 330   REG_EBRACE,           /* Unmatched \{.  */
 331   REG_BADBR,            /* Invalid contents of \{\}.  */
 332   REG_ERANGE,           /* Invalid range end.  */
 333   REG_ESPACE,           /* Ran out of memory.  */
 334   REG_BADRPT,           /* No preceding re for repetition op.  */
 335
 336   /* Error codes we've added.  */
 337   REG_EEND,             /* Premature end.  */
 338   REG_ESIZE,            /* Compiled pattern bigger than 2^16 bytes.  */
 339   REG_ERPAREN           /* Unmatched ) or \); not returned from regcomp.  */
 340 } reg_errcode_t;
 341 \f
 342 /* This data structure represents a compiled pattern.  Before calling
 343    the pattern compiler, the fields `buffer', `allocated', `fastmap',
 344    `translate', and `no_sub' can be set.  After the pattern has been
 345    compiled, the `re_nsub' field is available.  All other fields are
 346    private to the regex routines.  */
 347
 348 #ifndef RE_TRANSLATE_TYPE
 349 # define RE_TRANSLATE_TYPE char *
 350 #endif
 351
 352 struct re_pattern_buffer
 353 {
 354 /* [[[begin pattern_buffer]]] */
 355         /* Space that holds the compiled pattern.  It is declared as
 356           `unsigned char *' because its elements are
 357            sometimes used as array indexes.  */
 358   unsigned char *buffer;
 359
 360         /* Number of bytes to which `buffer' points.  */
 361   unsigned long int allocated;
 362
 363         /* Number of bytes actually used in `buffer'.  */
 364   unsigned long int used;
 365
 366         /* Syntax setting with which the pattern was compiled.  */
 367   reg_syntax_t syntax;
 368
 369         /* Pointer to a fastmap, if any, otherwise zero.  re_search uses
 370            the fastmap, if there is one, to skip over impossible
 371            starting points for matches.  */
 372   char *fastmap;
 373
 374         /* Either a translate table to apply to all characters before
 375            comparing them, or zero for no translation.  The translation
 376            is applied to a pattern when it is compiled and to a string
 377            when it is matched.  */
 378   RE_TRANSLATE_TYPE translate;
 379
 380         /* Number of subexpressions found by the compiler.  */
 381   size_t re_nsub;
 382
 383         /* Zero if this pattern cannot match the empty string, one else.
 384            Well, in truth it's used only in `re_search_2', to see
 385            whether or not we should use the fastmap, so we don't set
 386            this absolutely perfectly; see `re_compile_fastmap' (the
 387            `duplicate' case).  */
 388   unsigned can_be_null : 1;
 389
 390         /* If REGS_UNALLOCATED, allocate space in the `regs' structure
 391              for `max (RE_NREGS, re_nsub + 1)' groups.
 392            If REGS_REALLOCATE, reallocate space if necessary.
 393            If REGS_FIXED, use what's there.  */
 394 #define REGS_UNALLOCATED 0
 395 #define REGS_REALLOCATE 1
 396 #define REGS_FIXED 2
 397   unsigned regs_allocated : 2;
 398
 399         /* Set to zero when `regex_compile' compiles a pattern; set to one
 400            by `re_compile_fastmap' if it updates the fastmap.  */
 401   unsigned fastmap_accurate : 1;
 402
 403         /* If set, `re_match_2' does not return information about
 404            subexpressions.  */
 405   unsigned no_sub : 1;
 406
 407         /* If set, a beginning-of-line anchor doesn't match at the
 408            beginning of the string.  */
 409   unsigned not_bol : 1;
 410
 411         /* Similarly for an end-of-line anchor.  */
 412   unsigned not_eol : 1;
 413
 414         /* If true, an anchor at a newline matches.  */
 415   unsigned newline_anchor : 1;
 416
 417 /* [[[end pattern_buffer]]] */
 418 };
 419
 420 typedef struct re_pattern_buffer regex_t;
 421 \f
 422 /* Type for byte offsets within the string.  POSIX mandates this.  */
 423 typedef int regoff_t;
 424
 425
 426 /* This is the structure we store register match data in.  See
 427    regex.texinfo for a full description of what registers match.  */
 428 struct re_registers
 429 {
 430   unsigned num_regs;
 431   regoff_t *start;
 432   regoff_t *end;
 433 };
 434
 435
 436 /* If `regs_allocated' is REGS_UNALLOCATED in the pattern buffer,
 437    `re_match_2' returns information about at least this many registers
 438    the first time a `regs' structure is passed.  */
 439 #ifndef RE_NREGS
 440 # define RE_NREGS 30
 441 #endif
 442
 443
 444 /* POSIX specification for registers.  Aside from the different names than
 445    `re_registers', POSIX uses an array of structures, instead of a
 446    structure of arrays.  */
 447 typedef struct
 448 {
 449   regoff_t rm_so;  /* Byte offset from string's start to substring's start.  */
 450   regoff_t rm_eo;  /* Byte offset from string's start to substring's end.  */
 451 } regmatch_t;
 452 \f
 453 /* Declarations for routines.  */
 454
 455 /* To avoid duplicating every routine declaration -- once with a
 456    prototype (if we are ANSI), and once without (if we aren't) -- we
 457    use the following macro to declare argument types.  This
 458    unfortunately clutters up the declarations a bit, but I think it's
 459    worth it.  */
 460
 461 #if __STDC__
 462
 463 # define _RE_ARGS(args) args
 464
 465 #else /* not __STDC__ */
 466
 467 # define _RE_ARGS(args) ()
 468
 469 #endif /* not __STDC__ */
 470
 471 /* Sets the current default syntax to SYNTAX, and return the old syntax.
 472    You can also simply assign to the `re_syntax_options' variable.  */
 473 extern reg_syntax_t re_set_syntax _RE_ARGS ((reg_syntax_t syntax));
 474
 475 /* Compile the regular expression PATTERN, with length LENGTH
 476    and syntax given by the global `re_syntax_options', into the buffer
 477    BUFFER.  Return NULL if successful, and an error string if not.  */
 478 extern const char *re_compile_pattern
 479   _RE_ARGS ((const char *pattern, size_t length,
 480              struct re_pattern_buffer *buffer));
 481
 482
 483 /* Compile a fastmap for the compiled pattern in BUFFER; used to
 484    accelerate searches.  Return 0 if successful and -2 if was an
 485    internal error.  */
 486 extern int re_compile_fastmap _RE_ARGS ((struct re_pattern_buffer *buffer));
 487
 488
 489 /* Search in the string STRING (with length LENGTH) for the pattern
 490    compiled into BUFFER.  Start searching at position START, for RANGE
 491    characters.  Return the starting position of the match, -1 for no
 492    match, or -2 for an internal error.  Also return register
 493    information in REGS (if REGS and BUFFER->no_sub are nonzero).  */
 494 extern int re_search
 495   _RE_ARGS ((struct re_pattern_buffer *buffer, const char *string,
 496             int length, int start, int range, struct re_registers *regs));
 497
 498
 499 /* Like `re_search', but search in the concatenation of STRING1 and
 500    STRING2.  Also, stop searching at index START + STOP.  */
 501 extern int re_search_2
 502   _RE_ARGS ((struct re_pattern_buffer *buffer, const char *string1,
 503              int length1, const char *string2, int length2,
 504              int start, int range, struct re_registers *regs, int stop));
 505
 506
 507 /* Like `re_search', but return how many characters in STRING the regexp
 508    in BUFFER matched, starting at position START.  */
 509 extern int re_match
 510   _RE_ARGS ((struct re_pattern_buffer *buffer, const char *string,
 511              int length, int start, struct re_registers *regs));
 512
 513
 514 /* Relates to `re_match' as `re_search_2' relates to `re_search'.  */
 515 extern int re_match_2
 516   _RE_ARGS ((struct re_pattern_buffer *buffer, const char *string1,
 517              int length1, const char *string2, int length2,
 518              int start, struct re_registers *regs, int stop));
 519
 520
 521 /* Set REGS to hold NUM_REGS registers, storing them in STARTS and
 522    ENDS.  Subsequent matches using BUFFER and REGS will use this memory
 523    for recording register information.  STARTS and ENDS must be
 524    allocated with malloc, and must each be at least `NUM_REGS * sizeof
 525    (regoff_t)' bytes long.
 526
 527    If NUM_REGS == 0, then subsequent matches should allocate their own
 528    register data.
 529
 530    Unless this function is called, the first search or match using
 531    PATTERN_BUFFER will allocate its own register data, without
 532    freeing the old data.  */
 533 extern void re_set_registers
 534   _RE_ARGS ((struct re_pattern_buffer *buffer, struct re_registers *regs,
 535              unsigned num_regs, regoff_t *starts, regoff_t *ends));
 536
 537 #if defined _REGEX_RE_COMP || defined _LIBC
 538 # ifndef _CRAY
 539 /* 4.2 bsd compatibility.  */
 540 extern char *re_comp _RE_ARGS ((const char *));
 541 extern int re_exec _RE_ARGS ((const char *));
 542 # endif
 543 #endif
 544
 545 /* GCC 2.95 and later have "__restrict"; C99 compilers have
 546    "restrict", and "configure" may have defined "restrict".  */
 547 #ifndef __restrict
 548 # if ! (2 < __GNUC__ || (2 == __GNUC__ && 95 <= __GNUC_MINOR__))
 549 #  if defined restrict || 199901L <= __STDC_VERSION__
 550 #   define __restrict restrict
 551 #  else
 552 #   define __restrict
 553 #  endif
 554 # endif
 555 #endif
 556 /* gcc 3.1 and up support the [restrict] syntax.  */
 557 #ifndef __restrict_arr
 558 # if __GNUC__ > 3 || (__GNUC__ == 3 && __GNUC_MINOR__ >= 1)
 559 #  define __restrict_arr __restrict
 560 # else
 561 #  define __restrict_arr
 562 # endif
 563 #endif
 564
 565 /* POSIX compatibility.  */
 566 extern int regcomp _RE_ARGS ((regex_t *__restrict __preg,
 567                               const char *__restrict __pattern,
 568                               int __cflags));
 569
 570 extern int regexec _RE_ARGS ((const regex_t *__restrict __preg,
 571                               const char *__restrict __string, size_t __nmatch,
 572                               regmatch_t __pmatch[__restrict_arr],
 573                               int __eflags));
 574
 575 extern size_t regerror _RE_ARGS ((int __errcode, const regex_t *__preg,
 576                                   char *__errbuf, size_t __errbuf_size));
 577
 578 extern void regfree _RE_ARGS ((regex_t *__preg));
 579
 580
 581 #ifdef __cplusplus
 582 }
 583 #endif  /* C++ */
 584
 585 #endif /* regex.h */
 586 \f
 587 /*
 588 Local variables:
 589 make-backup-files: t
 590 version-control: t
 591 trim-versions-without-asking: nil
 592 End:
 593 */