src/regex.h

   1 /* Definitions for data structures and routines for the regular
   2    expression library, version 0.12.
   3
   4    Copyright (C) 1985, 1989-1993, 1995, 2000-2016 Free Software
   5    Foundation, Inc.
   6
   7    This program is free software; you can redistribute it and/or modify
   8    it under the terms of the GNU General Public License as published by
   9    the Free Software Foundation; either version 3, or (at your option)
  10    any later version.
  11
  12    This program 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
  15    GNU General Public License for more details.
  16
  17    You should have received a copy of the GNU General Public License
  18    along with this program.  If not, see <http://www.gnu.org/licenses/>.  */
  19
  20 #ifndef _REGEX_H
  21 #define _REGEX_H 1
  22
  23 #if defined emacs && (defined _REGEX_RE_COMP || defined _LIBC)
  24 /* We’re not defining re_set_syntax and using a different prototype of
  25    re_compile_pattern when building Emacs so fail compilation early with
  26    a (somewhat helpful) error message when conflict is detected. */
  27 # error "_REGEX_RE_COMP nor _LIBC can be defined if emacs is defined."
  28 #endif
  29
  30 /* Allow the use in C++ code.  */
  31 #ifdef __cplusplus
  32 extern "C" {
  33 #endif
  34
  35 /* POSIX says that <sys/types.h> must be included (by the caller) before
  36    <regex.h>.  */
  37
  38 #if !defined _POSIX_C_SOURCE && !defined _POSIX_SOURCE && defined VMS
  39 /* VMS doesn't have `size_t' in <sys/types.h>, even though POSIX says it
  40    should be there.  */
  41 # include <stddef.h>
  42 #endif
  43
  44 /* The following bits are used to determine the regexp syntax we
  45    recognize.  The set/not-set meanings where historically chosen so
  46    that Emacs syntax had the value 0.
  47    The bits are given in alphabetical order, and
  48    the definitions shifted by one from the previous bit; thus, when we
  49    add or remove a bit, only one other definition need change.  */
  50 typedef unsigned long reg_syntax_t;
  51
  52 /* If this bit is not set, then \ inside a bracket expression is literal.
  53    If set, then such a \ quotes the following character.  */
  54 #define RE_BACKSLASH_ESCAPE_IN_LISTS ((unsigned long int) 1)
  55
  56 /* If this bit is not set, then + and ? are operators, and \+ and \? are
  57      literals.
  58    If set, then \+ and \? are operators and + and ? are literals.  */
  59 #define RE_BK_PLUS_QM (RE_BACKSLASH_ESCAPE_IN_LISTS << 1)
  60
  61 /* If this bit is set, then character classes are supported.  They are:
  62      [:alpha:], [:upper:], [:lower:],  [:digit:], [:alnum:], [:xdigit:],
  63      [:space:], [:print:], [:punct:], [:graph:], and [:cntrl:].
  64    If not set, then character classes are not supported.  */
  65 #define RE_CHAR_CLASSES (RE_BK_PLUS_QM << 1)
  66
  67 /* If this bit is set, then ^ and $ are always anchors (outside bracket
  68      expressions, of course).
  69    If this bit is not set, then it depends:
  70         ^  is an anchor if it is at the beginning of a regular
  71            expression or after an open-group or an alternation operator;
  72         $  is an anchor if it is at the end of a regular expression, or
  73            before a close-group or an alternation operator.
  74
  75    This bit could be (re)combined with RE_CONTEXT_INDEP_OPS, because
  76    POSIX draft 11.2 says that * etc. in leading positions is undefined.
  77    We already implemented a previous draft which made those constructs
  78    invalid, though, so we haven't changed the code back.  */
  79 #define RE_CONTEXT_INDEP_ANCHORS (RE_CHAR_CLASSES << 1)
  80
  81 /* If this bit is set, then special characters are always special
  82      regardless of where they are in the pattern.
  83    If this bit is not set, then special characters are special only in
  84      some contexts; otherwise they are ordinary.  Specifically,
  85      * + ? and intervals are only special when not after the beginning,
  86      open-group, or alternation operator.  */
  87 #define RE_CONTEXT_INDEP_OPS (RE_CONTEXT_INDEP_ANCHORS << 1)
  88
  89 /* If this bit is set, then *, +, ?, and { cannot be first in an re or
  90      immediately after an alternation or begin-group operator.  */
  91 #define RE_CONTEXT_INVALID_OPS (RE_CONTEXT_INDEP_OPS << 1)
  92
  93 /* If this bit is set, then . matches newline.
  94    If not set, then it doesn't.  */
  95 #define RE_DOT_NEWLINE (RE_CONTEXT_INVALID_OPS << 1)
  96
  97 /* If this bit is set, then . doesn't match NUL.
  98    If not set, then it does.  */
  99 #define RE_DOT_NOT_NULL (RE_DOT_NEWLINE << 1)
 100
 101 /* If this bit is set, nonmatching lists [^...] do not match newline.
 102    If not set, they do.  */
 103 #define RE_HAT_LISTS_NOT_NEWLINE (RE_DOT_NOT_NULL << 1)
 104
 105 /* If this bit is set, either \{...\} or {...} defines an
 106      interval, depending on RE_NO_BK_BRACES.
 107    If not set, \{, \}, {, and } are literals.  */
 108 #define RE_INTERVALS (RE_HAT_LISTS_NOT_NEWLINE << 1)
 109
 110 /* If this bit is set, +, ? and | aren't recognized as operators.
 111    If not set, they are.  */
 112 #define RE_LIMITED_OPS (RE_INTERVALS << 1)
 113
 114 /* If this bit is set, newline is an alternation operator.
 115    If not set, newline is literal.  */
 116 #define RE_NEWLINE_ALT (RE_LIMITED_OPS << 1)
 117
 118 /* If this bit is set, then `{...}' defines an interval, and \{ and \}
 119      are literals.
 120   If not set, then `\{...\}' defines an interval.  */
 121 #define RE_NO_BK_BRACES (RE_NEWLINE_ALT << 1)
 122
 123 /* If this bit is set, (...) defines a group, and \( and \) are literals.
 124    If not set, \(...\) defines a group, and ( and ) are literals.  */
 125 #define RE_NO_BK_PARENS (RE_NO_BK_BRACES << 1)
 126
 127 /* If this bit is set, then \<digit> matches <digit>.
 128    If not set, then \<digit> is a back-reference.  */
 129 #define RE_NO_BK_REFS (RE_NO_BK_PARENS << 1)
 130
 131 /* If this bit is set, then | is an alternation operator, and \| is literal.
 132    If not set, then \| is an alternation operator, and | is literal.  */
 133 #define RE_NO_BK_VBAR (RE_NO_BK_REFS << 1)
 134
 135 /* If this bit is set, then an ending range point collating higher
 136      than the starting range point, as in [z-a], is invalid.
 137    If not set, then when ending range point collates higher than the
 138      starting range point, the range is ignored.  */
 139 #define RE_NO_EMPTY_RANGES (RE_NO_BK_VBAR << 1)
 140
 141 /* If this bit is set, then an unmatched ) is ordinary.
 142    If not set, then an unmatched ) is invalid.  */
 143 #define RE_UNMATCHED_RIGHT_PAREN_ORD (RE_NO_EMPTY_RANGES << 1)
 144
 145 /* If this bit is set, succeed as soon as we match the whole pattern,
 146    without further backtracking.  */
 147 #define RE_NO_POSIX_BACKTRACKING (RE_UNMATCHED_RIGHT_PAREN_ORD << 1)
 148
 149 /* If this bit is set, do not process the GNU regex operators.
 150    If not set, then the GNU regex operators are recognized. */
 151 #define RE_NO_GNU_OPS (RE_NO_POSIX_BACKTRACKING << 1)
 152
 153 /* If this bit is set, then *?, +? and ?? match non greedily. */
 154 #define RE_FRUGAL (RE_NO_GNU_OPS << 1)
 155
 156 /* If this bit is set, then (?:...) is treated as a shy group.  */
 157 #define RE_SHY_GROUPS (RE_FRUGAL << 1)
 158
 159 /* If this bit is set, ^ and $ only match at beg/end of buffer.  */
 160 #define RE_NO_NEWLINE_ANCHOR (RE_SHY_GROUPS << 1)
 161
 162 /* If this bit is set, turn on internal regex debugging.
 163    If not set, and debugging was on, turn it off.
 164    This only works if regex.c is compiled -DDEBUG.
 165    We define this bit always, so that all that's needed to turn on
 166    debugging is to recompile regex.c; the calling code can always have
 167    this bit set, and it won't affect anything in the normal case. */
 168 #define RE_DEBUG (RE_NO_NEWLINE_ANCHOR << 1)
 169
 170 /* This global variable defines the particular regexp syntax to use (for
 171    some interfaces).  When a regexp is compiled, the syntax used is
 172    stored in the pattern buffer, so changing this does not affect
 173    already-compiled regexps.  */
 174 extern reg_syntax_t re_syntax_options;
 175
 176 #ifdef emacs
 177 # include "lisp.h"
 178 /* In Emacs, this is the string or buffer in which we
 179    are matching.  It is used for looking up syntax properties.  */
 180 extern Lisp_Object re_match_object;
 181 #endif
 182
 183 /* Roughly the maximum number of failure points on the stack.  */
 184 extern size_t re_max_failures;
 185
 186 \f
 187 /* Define combinations of the above bits for the standard possibilities.
 188    (The [[[ comments delimit what gets put into the Texinfo file, so
 189    don't delete them!)  */
 190 /* [[[begin syntaxes]]] */
 191 #define RE_SYNTAX_EMACS                                                 \
 192   (RE_CHAR_CLASSES | RE_INTERVALS | RE_SHY_GROUPS | RE_FRUGAL)
 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
 205 #define RE_SYNTAX_POSIX_AWK                                             \
 206   (RE_SYNTAX_POSIX_EXTENDED | RE_BACKSLASH_ESCAPE_IN_LISTS              \
 207    | RE_INTERVALS           | RE_NO_GNU_OPS)
 208
 209 #define RE_SYNTAX_GREP                                                  \
 210   (RE_BK_PLUS_QM              | RE_CHAR_CLASSES                         \
 211    | RE_HAT_LISTS_NOT_NEWLINE | RE_INTERVALS                            \
 212    | RE_NEWLINE_ALT)
 213
 214 #define RE_SYNTAX_EGREP                                                 \
 215   (RE_CHAR_CLASSES        | RE_CONTEXT_INDEP_ANCHORS                    \
 216    | RE_CONTEXT_INDEP_OPS | RE_HAT_LISTS_NOT_NEWLINE                    \
 217    | RE_NEWLINE_ALT       | RE_NO_BK_PARENS                             \
 218    | RE_NO_BK_VBAR)
 219
 220 #define RE_SYNTAX_POSIX_EGREP                                           \
 221   (RE_SYNTAX_EGREP | RE_INTERVALS | RE_NO_BK_BRACES)
 222
 223 /* P1003.2/D11.2, section 4.20.7.1, lines 5078ff.  */
 224 #define RE_SYNTAX_ED RE_SYNTAX_POSIX_BASIC
 225
 226 #define RE_SYNTAX_SED RE_SYNTAX_POSIX_BASIC
 227
 228 /* Syntax bits common to both basic and extended POSIX regex syntax.  */
 229 #define _RE_SYNTAX_POSIX_COMMON                                         \
 230   (RE_CHAR_CLASSES | RE_DOT_NEWLINE      | RE_DOT_NOT_NULL              \
 231    | RE_INTERVALS  | RE_NO_EMPTY_RANGES)
 232
 233 #define RE_SYNTAX_POSIX_BASIC                                           \
 234   (_RE_SYNTAX_POSIX_COMMON | RE_BK_PLUS_QM)
 235
 236 /* Differs from ..._POSIX_BASIC only in that RE_BK_PLUS_QM becomes
 237    RE_LIMITED_OPS, i.e., \? \+ \| are not recognized.  Actually, this
 238    isn't minimal, since other operators, such as \`, aren't disabled.  */
 239 #define RE_SYNTAX_POSIX_MINIMAL_BASIC                                   \
 240   (_RE_SYNTAX_POSIX_COMMON | RE_LIMITED_OPS)
 241
 242 #define RE_SYNTAX_POSIX_EXTENDED                                        \
 243   (_RE_SYNTAX_POSIX_COMMON  | RE_CONTEXT_INDEP_ANCHORS                  \
 244    | RE_CONTEXT_INDEP_OPS   | RE_NO_BK_BRACES                           \
 245    | RE_NO_BK_PARENS        | RE_NO_BK_VBAR                             \
 246    | RE_CONTEXT_INVALID_OPS | RE_UNMATCHED_RIGHT_PAREN_ORD)
 247
 248 /* Differs from ..._POSIX_EXTENDED in that RE_CONTEXT_INDEP_OPS is
 249    removed and RE_NO_BK_REFS is added.  */
 250 #define RE_SYNTAX_POSIX_MINIMAL_EXTENDED                                \
 251   (_RE_SYNTAX_POSIX_COMMON  | RE_CONTEXT_INDEP_ANCHORS                  \
 252    | RE_CONTEXT_INVALID_OPS | RE_NO_BK_BRACES                           \
 253    | RE_NO_BK_PARENS        | RE_NO_BK_REFS                             \
 254    | RE_NO_BK_VBAR          | RE_UNMATCHED_RIGHT_PAREN_ORD)
 255 /* [[[end syntaxes]]] */
 256 \f
 257 /* Maximum number of duplicates an interval can allow.  Some systems
 258    (erroneously) define this in other header files, but we want our
 259    value, so remove any previous define.  */
 260 #ifdef RE_DUP_MAX
 261 # undef RE_DUP_MAX
 262 #endif
 263 /* If sizeof(int) == 2, then ((1 << 15) - 1) overflows.  */
 264 #define RE_DUP_MAX (0x7fff)
 265
 266
 267 /* POSIX `cflags' bits (i.e., information for `regcomp').  */
 268
 269 /* If this bit is set, then use extended regular expression syntax.
 270    If not set, then use basic regular expression syntax.  */
 271 #define REG_EXTENDED 1
 272
 273 /* If this bit is set, then ignore case when matching.
 274    If not set, then case is significant.  */
 275 #define REG_ICASE (REG_EXTENDED << 1)
 276
 277 /* If this bit is set, then anchors do not match at newline
 278      characters in the string.
 279    If not set, then anchors do match at newlines.  */
 280 #define REG_NEWLINE (REG_ICASE << 1)
 281
 282 /* If this bit is set, then report only success or fail in regexec.
 283    If not set, then returns differ between not matching and errors.  */
 284 #define REG_NOSUB (REG_NEWLINE << 1)
 285
 286
 287 /* POSIX `eflags' bits (i.e., information for regexec).  */
 288
 289 /* If this bit is set, then the beginning-of-line operator doesn't match
 290      the beginning of the string (presumably because it's not the
 291      beginning of a line).
 292    If not set, then the beginning-of-line operator does match the
 293      beginning of the string.  */
 294 #define REG_NOTBOL 1
 295
 296 /* Like REG_NOTBOL, except for the end-of-line.  */
 297 #define REG_NOTEOL (1 << 1)
 298
 299
 300 /* If any error codes are removed, changed, or added, update the
 301    `re_error_msg' table in regex.c.  */
 302 typedef enum
 303 {
 304 #ifdef _XOPEN_SOURCE
 305   REG_ENOSYS = -1,      /* This will never happen for this implementation.  */
 306 #endif
 307
 308   REG_NOERROR = 0,      /* Success.  */
 309   REG_NOMATCH,          /* Didn't find a match (for regexec).  */
 310
 311   /* POSIX regcomp return error codes.  (In the order listed in the
 312      standard.)  */
 313   REG_BADPAT,           /* Invalid pattern.  */
 314   REG_ECOLLATE,         /* Not implemented.  */
 315   REG_ECTYPE,           /* Invalid character class name.  */
 316   REG_EESCAPE,          /* Trailing backslash.  */
 317   REG_ESUBREG,          /* Invalid back reference.  */
 318   REG_EBRACK,           /* Unmatched left bracket.  */
 319   REG_EPAREN,           /* Parenthesis imbalance.  */
 320   REG_EBRACE,           /* Unmatched \{.  */
 321   REG_BADBR,            /* Invalid contents of \{\}.  */
 322   REG_ERANGE,           /* Invalid range end.  */
 323   REG_ESPACE,           /* Ran out of memory.  */
 324   REG_BADRPT,           /* No preceding re for repetition op.  */
 325
 326   /* Error codes we've added.  */
 327   REG_EEND,             /* Premature end.  */
 328   REG_ESIZE,            /* Compiled pattern bigger than 2^16 bytes.  */
 329   REG_ERPAREN,          /* Unmatched ) or \); not returned from regcomp.  */
 330   REG_ERANGEX           /* Range striding over charsets.  */
 331 } reg_errcode_t;
 332 \f
 333 /* This data structure represents a compiled pattern.  Before calling
 334    the pattern compiler, the fields `buffer', `allocated', `fastmap',
 335    `translate', and `no_sub' can be set.  After the pattern has been
 336    compiled, the `re_nsub' field is available.  All other fields are
 337    private to the regex routines.  */
 338
 339 #ifndef RE_TRANSLATE_TYPE
 340 # define RE_TRANSLATE_TYPE char *
 341 #endif
 342
 343 struct re_pattern_buffer
 344 {
 345 /* [[[begin pattern_buffer]]] */
 346         /* Space that holds the compiled pattern.  It is declared as
 347           `unsigned char *' because its elements are
 348            sometimes used as array indexes.  */
 349   unsigned char *buffer;
 350
 351         /* Number of bytes to which `buffer' points.  */
 352   size_t allocated;
 353
 354         /* Number of bytes actually used in `buffer'.  */
 355   size_t used;
 356
 357 #ifndef emacs
 358         /* Syntax setting with which the pattern was compiled.  */
 359   reg_syntax_t syntax;
 360 #endif
 361         /* Pointer to a fastmap, if any, otherwise zero.  re_search uses
 362            the fastmap, if there is one, to skip over impossible
 363            starting points for matches.  */
 364   char *fastmap;
 365
 366         /* Either a translate table to apply to all characters before
 367            comparing them, or zero for no translation.  The translation
 368            is applied to a pattern when it is compiled and to a string
 369            when it is matched.  */
 370   RE_TRANSLATE_TYPE translate;
 371
 372         /* Number of subexpressions found by the compiler.  */
 373   size_t re_nsub;
 374
 375         /* Zero if this pattern cannot match the empty string, one else.
 376            Well, in truth it's used only in `re_search_2', to see
 377            whether or not we should use the fastmap, so we don't set
 378            this absolutely perfectly; see `re_compile_fastmap'.  */
 379   unsigned can_be_null : 1;
 380
 381         /* If REGS_UNALLOCATED, allocate space in the `regs' structure
 382              for `max (RE_NREGS, re_nsub + 1)' groups.
 383            If REGS_REALLOCATE, reallocate space if necessary.
 384            If REGS_FIXED, use what's there.  */
 385 #define REGS_UNALLOCATED 0
 386 #define REGS_REALLOCATE 1
 387 #define REGS_FIXED 2
 388   unsigned regs_allocated : 2;
 389
 390         /* Set to zero when `regex_compile' compiles a pattern; set to one
 391            by `re_compile_fastmap' if it updates the fastmap.  */
 392   unsigned fastmap_accurate : 1;
 393
 394         /* If set, `re_match_2' does not return information about
 395            subexpressions.  */
 396   unsigned no_sub : 1;
 397
 398         /* If set, a beginning-of-line anchor doesn't match at the
 399            beginning of the string.  */
 400   unsigned not_bol : 1;
 401
 402         /* Similarly for an end-of-line anchor.  */
 403   unsigned not_eol : 1;
 404
 405   /* If true, the compilation of the pattern had to look up the syntax table,
 406      so the compiled pattern is only valid for the current syntax table.  */
 407   unsigned used_syntax : 1;
 408
 409 #ifdef emacs
 410   /* If true, multi-byte form in the regexp pattern should be
 411      recognized as a multibyte character.  */
 412   unsigned multibyte : 1;
 413
 414   /* If true, multi-byte form in the target of match should be
 415      recognized as a multibyte character.  */
 416   unsigned target_multibyte : 1;
 417
 418   /* Charset of unibyte characters at compiling time. */
 419   int charset_unibyte;
 420 #endif
 421
 422 /* [[[end pattern_buffer]]] */
 423 };
 424
 425 typedef struct re_pattern_buffer regex_t;
 426 \f
 427 /* Type for byte offsets within the string.  POSIX mandates this to be an int,
 428    but the Open Group has signaled its intention to change the requirement to
 429    be that regoff_t be at least as wide as ptrdiff_t and ssize_t.  Current
 430    gnulib sources also use ssize_t, and we need this for supporting buffers and
 431    strings > 2GB on 64-bit hosts.  */
 432 typedef ssize_t regoff_t;
 433
 434
 435 /* This is the structure we store register match data in.  See
 436    regex.texinfo for a full description of what registers match.  */
 437 struct re_registers
 438 {
 439   unsigned num_regs;
 440   regoff_t *start;
 441   regoff_t *end;
 442 };
 443
 444
 445 /* If `regs_allocated' is REGS_UNALLOCATED in the pattern buffer,
 446    `re_match_2' returns information about at least this many registers
 447    the first time a `regs' structure is passed.  */
 448 #ifndef RE_NREGS
 449 # define RE_NREGS 30
 450 #endif
 451
 452
 453 /* POSIX specification for registers.  Aside from the different names than
 454    `re_registers', POSIX uses an array of structures, instead of a
 455    structure of arrays.  */
 456 typedef struct
 457 {
 458   regoff_t rm_so;  /* Byte offset from string's start to substring's start.  */
 459   regoff_t rm_eo;  /* Byte offset from string's start to substring's end.  */
 460 } regmatch_t;
 461 \f
 462 /* Declarations for routines.  */
 463
 464 #ifndef emacs
 465
 466 /* Sets the current default syntax to SYNTAX, and return the old syntax.
 467    You can also simply assign to the `re_syntax_options' variable.  */
 468 extern reg_syntax_t re_set_syntax (reg_syntax_t __syntax);
 469
 470 #endif
 471
 472 /* Compile the regular expression PATTERN, with length LENGTH
 473    and syntax given by the global `re_syntax_options', into the buffer
 474    BUFFER.  Return NULL if successful, and an error string if not.  */
 475 extern const char *re_compile_pattern (const char *__pattern, size_t __length,
 476 #ifdef emacs
 477                                        bool posix_backtracking,
 478                                        const char *whitespace_regexp,
 479 #endif
 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 (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 regoff_t re_search (struct re_pattern_buffer *__buffer,
 495                            const char *__string, size_t __length,
 496                            ssize_t __start, ssize_t __range,
 497                            struct re_registers *__regs);
 498
 499
 500 /* Like `re_search', but search in the concatenation of STRING1 and
 501    STRING2.  Also, stop searching at index START + STOP.  */
 502 extern regoff_t re_search_2 (struct re_pattern_buffer *__buffer,
 503                              const char *__string1, size_t __length1,
 504                              const char *__string2, size_t __length2,
 505                              ssize_t __start, ssize_t __range,
 506                              struct re_registers *__regs,
 507                              ssize_t __stop);
 508
 509
 510 /* Like `re_search', but return how many characters in STRING the regexp
 511    in BUFFER matched, starting at position START.  */
 512 extern regoff_t re_match (struct re_pattern_buffer *__buffer,
 513                           const char *__string, size_t __length,
 514                           ssize_t __start, struct re_registers *__regs);
 515
 516
 517 /* Relates to `re_match' as `re_search_2' relates to `re_search'.  */
 518 extern regoff_t re_match_2 (struct re_pattern_buffer *__buffer,
 519                             const char *__string1, size_t __length1,
 520                             const char *__string2, size_t __length2,
 521                             ssize_t __start, struct re_registers *__regs,
 522                             ssize_t __stop);
 523
 524
 525 /* Set REGS to hold NUM_REGS registers, storing them in STARTS and
 526    ENDS.  Subsequent matches using BUFFER and REGS will use this memory
 527    for recording register information.  STARTS and ENDS must be
 528    allocated with malloc, and must each be at least `NUM_REGS * sizeof
 529    (regoff_t)' bytes long.
 530
 531    If NUM_REGS == 0, then subsequent matches should allocate their own
 532    register data.
 533
 534    Unless this function is called, the first search or match using
 535    PATTERN_BUFFER will allocate its own register data, without
 536    freeing the old data.  */
 537 extern void re_set_registers (struct re_pattern_buffer *__buffer,
 538                               struct re_registers *__regs,
 539                               unsigned __num_regs,
 540                               regoff_t *__starts, regoff_t *__ends);
 541
 542 #if defined _REGEX_RE_COMP || defined _LIBC
 543 # ifndef _CRAY
 544 /* 4.2 bsd compatibility.  */
 545 extern char *re_comp (const char *);
 546 extern int re_exec (const char *);
 547 # endif
 548 #endif
 549
 550 /* GCC 2.95 and later have "__restrict"; C99 compilers have
 551    "restrict", and "configure" may have defined "restrict".
 552    Other compilers use __restrict, __restrict__, and _Restrict, and
 553    'configure' might #define 'restrict' to those words, so pick a
 554    different name.  */
 555 #ifndef _Restrict_
 556 # if 199901L <= __STDC_VERSION__
 557 #  define _Restrict_ restrict
 558 # elif 2 < __GNUC__ || (2 == __GNUC__ && 95 <= __GNUC_MINOR__)
 559 #  define _Restrict_ __restrict
 560 # else
 561 #  define _Restrict_
 562 # endif
 563 #endif
 564 /* gcc 3.1 and up support the [restrict] syntax.  Don't trust
 565    sys/cdefs.h's definition of __restrict_arr, though, as it
 566    mishandles gcc -ansi -pedantic.  */
 567 #ifndef _Restrict_arr_
 568 # if ((199901L <= __STDC_VERSION__                                      \
 569        || ((3 < __GNUC__ || (3 == __GNUC__ && 1 <= __GNUC_MINOR__))     \
 570            && !defined __STRICT_ANSI__))                                        \
 571       && !defined __GNUG__)
 572 #  define _Restrict_arr_ _Restrict_
 573 # else
 574 #  define _Restrict_arr_
 575 # endif
 576 #endif
 577
 578 /* POSIX compatibility.  */
 579 extern reg_errcode_t regcomp (regex_t *_Restrict_ __preg,
 580                               const char *_Restrict_ __pattern,
 581                               int __cflags);
 582
 583 extern reg_errcode_t regexec (const regex_t *_Restrict_ __preg,
 584                               const char *_Restrict_ __string, size_t __nmatch,
 585                               regmatch_t __pmatch[_Restrict_arr_],
 586                               int __eflags);
 587
 588 extern size_t regerror (int __errcode, const regex_t * __preg,
 589                         char *__errbuf, size_t __errbuf_size);
 590
 591 extern void regfree (regex_t *__preg);
 592
 593
 594 #ifdef __cplusplus
 595 }
 596 #endif  /* C++ */
 597
 598 /* For platform which support the ISO C amendment 1 functionality we
 599    support user defined character classes.  */
 600 #if WIDE_CHAR_SUPPORT
 601 /* Solaris 2.5 has a bug: <wchar.h> must be included before <wctype.h>.  */
 602 # include <wchar.h>
 603 # include <wctype.h>
 604
 605 typedef wctype_t re_wctype_t;
 606 typedef wchar_t re_wchar_t;
 607 # define re_wctype wctype
 608 # define re_iswctype iswctype
 609 # define re_wctype_to_bit(cc) 0
 610 #else
 611 # ifndef emacs
 612 #  define btowc(c) c
 613 # endif
 614
 615 /* Character classes.  */
 616 typedef enum { RECC_ERROR = 0,
 617                RECC_ALNUM, RECC_ALPHA, RECC_WORD,
 618                RECC_GRAPH, RECC_PRINT,
 619                RECC_LOWER, RECC_UPPER,
 620                RECC_PUNCT, RECC_CNTRL,
 621                RECC_DIGIT, RECC_XDIGIT,
 622                RECC_BLANK, RECC_SPACE,
 623                RECC_MULTIBYTE, RECC_NONASCII,
 624                RECC_ASCII, RECC_UNIBYTE
 625 } re_wctype_t;
 626
 627 extern char re_iswctype (int ch,    re_wctype_t cc);
 628 extern re_wctype_t re_wctype_parse (const unsigned char **strp, unsigned limit);
 629
 630 typedef int re_wchar_t;
 631
 632 #endif /* not WIDE_CHAR_SUPPORT */
 633
 634 #endif /* regex.h */
 635 \f