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, 2000-2013 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 /* Allow the use in C++ code.  */
  24 #ifdef __cplusplus
  25 extern "C" {
  26 #endif
  27
  28 /* POSIX says that <sys/types.h> must be included (by the caller) before
  29    <regex.h>.  */
  30
  31 #if !defined _POSIX_C_SOURCE && !defined _POSIX_SOURCE && defined VMS
  32 /* VMS doesn't have `size_t' in <sys/types.h>, even though POSIX says it
  33    should be there.  */
  34 # include <stddef.h>
  35 #endif
  36
  37 /* The following bits are used to determine the regexp syntax we
  38    recognize.  The set/not-set meanings where historically chosen so
  39    that Emacs syntax had the value 0.
  40    The bits are given in alphabetical order, and
  41    the definitions shifted by one from the previous bit; thus, when we
  42    add or remove a bit, only one other definition need change.  */
  43 typedef unsigned long int reg_syntax_t;
  44
  45 /* If this bit is not set, then \ inside a bracket expression is literal.
  46    If set, then such a \ quotes the following character.  */
  47 #define RE_BACKSLASH_ESCAPE_IN_LISTS ((unsigned long int) 1)
  48
  49 /* If this bit is not set, then + and ? are operators, and \+ and \? are
  50      literals.
  51    If set, then \+ and \? are operators and + and ? are literals.  */
  52 #define RE_BK_PLUS_QM (RE_BACKSLASH_ESCAPE_IN_LISTS << 1)
  53
  54 /* If this bit is set, then character classes are supported.  They are:
  55      [:alpha:], [:upper:], [:lower:],  [:digit:], [:alnum:], [:xdigit:],
  56      [:space:], [:print:], [:punct:], [:graph:], and [:cntrl:].
  57    If not set, then character classes are not supported.  */
  58 #define RE_CHAR_CLASSES (RE_BK_PLUS_QM << 1)
  59
  60 /* If this bit is set, then ^ and $ are always anchors (outside bracket
  61      expressions, of course).
  62    If this bit is not set, then it depends:
  63         ^  is an anchor if it is at the beginning of a regular
  64            expression or after an open-group or an alternation operator;
  65         $  is an anchor if it is at the end of a regular expression, or
  66            before a close-group or an alternation operator.
  67
  68    This bit could be (re)combined with RE_CONTEXT_INDEP_OPS, because
  69    POSIX draft 11.2 says that * etc. in leading positions is undefined.
  70    We already implemented a previous draft which made those constructs
  71    invalid, though, so we haven't changed the code back.  */
  72 #define RE_CONTEXT_INDEP_ANCHORS (RE_CHAR_CLASSES << 1)
  73
  74 /* If this bit is set, then special characters are always special
  75      regardless of where they are in the pattern.
  76    If this bit is not set, then special characters are special only in
  77      some contexts; otherwise they are ordinary.  Specifically,
  78      * + ? and intervals are only special when not after the beginning,
  79      open-group, or alternation operator.  */
  80 #define RE_CONTEXT_INDEP_OPS (RE_CONTEXT_INDEP_ANCHORS << 1)
  81
  82 /* If this bit is set, then *, +, ?, and { cannot be first in an re or
  83      immediately after an alternation or begin-group operator.  */
  84 #define RE_CONTEXT_INVALID_OPS (RE_CONTEXT_INDEP_OPS << 1)
  85
  86 /* If this bit is set, then . matches newline.
  87    If not set, then it doesn't.  */
  88 #define RE_DOT_NEWLINE (RE_CONTEXT_INVALID_OPS << 1)
  89
  90 /* If this bit is set, then . doesn't match NUL.
  91    If not set, then it does.  */
  92 #define RE_DOT_NOT_NULL (RE_DOT_NEWLINE << 1)
  93
  94 /* If this bit is set, nonmatching lists [^...] do not match newline.
  95    If not set, they do.  */
  96 #define RE_HAT_LISTS_NOT_NEWLINE (RE_DOT_NOT_NULL << 1)
  97
  98 /* If this bit is set, either \{...\} or {...} defines an
  99      interval, depending on RE_NO_BK_BRACES.
 100    If not set, \{, \}, {, and } are literals.  */
 101 #define RE_INTERVALS (RE_HAT_LISTS_NOT_NEWLINE << 1)
 102
 103 /* If this bit is set, +, ? and | aren't recognized as operators.
 104    If not set, they are.  */
 105 #define RE_LIMITED_OPS (RE_INTERVALS << 1)
 106
 107 /* If this bit is set, newline is an alternation operator.
 108    If not set, newline is literal.  */
 109 #define RE_NEWLINE_ALT (RE_LIMITED_OPS << 1)
 110
 111 /* If this bit is set, then `{...}' defines an interval, and \{ and \}
 112      are literals.
 113   If not set, then `\{...\}' defines an interval.  */
 114 #define RE_NO_BK_BRACES (RE_NEWLINE_ALT << 1)
 115
 116 /* If this bit is set, (...) defines a group, and \( and \) are literals.
 117    If not set, \(...\) defines a group, and ( and ) are literals.  */
 118 #define RE_NO_BK_PARENS (RE_NO_BK_BRACES << 1)
 119
 120 /* If this bit is set, then \<digit> matches <digit>.
 121    If not set, then \<digit> is a back-reference.  */
 122 #define RE_NO_BK_REFS (RE_NO_BK_PARENS << 1)
 123
 124 /* If this bit is set, then | is an alternation operator, and \| is literal.
 125    If not set, then \| is an alternation operator, and | is literal.  */
 126 #define RE_NO_BK_VBAR (RE_NO_BK_REFS << 1)
 127
 128 /* If this bit is set, then an ending range point collating higher
 129      than the starting range point, as in [z-a], is invalid.
 130    If not set, then when ending range point collates higher than the
 131      starting range point, the range is ignored.  */
 132 #define RE_NO_EMPTY_RANGES (RE_NO_BK_VBAR << 1)
 133
 134 /* If this bit is set, then an unmatched ) is ordinary.
 135    If not set, then an unmatched ) is invalid.  */
 136 #define RE_UNMATCHED_RIGHT_PAREN_ORD (RE_NO_EMPTY_RANGES << 1)
 137
 138 /* If this bit is set, succeed as soon as we match the whole pattern,
 139    without further backtracking.  */
 140 #define RE_NO_POSIX_BACKTRACKING (RE_UNMATCHED_RIGHT_PAREN_ORD << 1)
 141
 142 /* If this bit is set, do not process the GNU regex operators.
 143    If not set, then the GNU regex operators are recognized. */
 144 #define RE_NO_GNU_OPS (RE_NO_POSIX_BACKTRACKING << 1)
 145
 146 /* If this bit is set, then *?, +? and ?? match non greedily. */
 147 #define RE_FRUGAL (RE_NO_GNU_OPS << 1)
 148
 149 /* If this bit is set, then (?:...) is treated as a shy group.  */
 150 #define RE_SHY_GROUPS (RE_FRUGAL << 1)
 151
 152 /* If this bit is set, ^ and $ only match at beg/end of buffer.  */
 153 #define RE_NO_NEWLINE_ANCHOR (RE_SHY_GROUPS << 1)
 154
 155 /* If this bit is set, turn on internal regex debugging.
 156    If not set, and debugging was on, turn it off.
 157    This only works if regex.c is compiled -DDEBUG.
 158    We define this bit always, so that all that's needed to turn on
 159    debugging is to recompile regex.c; the calling code can always have
 160    this bit set, and it won't affect anything in the normal case. */
 161 #define RE_DEBUG (RE_NO_NEWLINE_ANCHOR << 1)
 162
 163 /* This global variable defines the particular regexp syntax to use (for
 164    some interfaces).  When a regexp is compiled, the syntax used is
 165    stored in the pattern buffer, so changing this does not affect
 166    already-compiled regexps.  */
 167 extern reg_syntax_t re_syntax_options;
 168
 169 #ifdef emacs
 170 /* In Emacs, this is the string or buffer in which we
 171    are matching.  It is used for looking up syntax properties.  */
 172 extern Lisp_Object re_match_object;
 173 #endif
 174
 175 \f
 176 /* Define combinations of the above bits for the standard possibilities.
 177    (The [[[ comments delimit what gets put into the Texinfo file, so
 178    don't delete them!)  */
 179 /* [[[begin syntaxes]]] */
 180 #define RE_SYNTAX_EMACS                                                 \
 181   (RE_CHAR_CLASSES | RE_INTERVALS | RE_SHY_GROUPS | RE_FRUGAL)
 182
 183 #define RE_SYNTAX_AWK                                                   \
 184   (RE_BACKSLASH_ESCAPE_IN_LISTS   | RE_DOT_NOT_NULL                     \
 185    | RE_NO_BK_PARENS              | RE_NO_BK_REFS                       \
 186    | RE_NO_BK_VBAR                | RE_NO_EMPTY_RANGES                  \
 187    | RE_DOT_NEWLINE               | RE_CONTEXT_INDEP_ANCHORS            \
 188    | RE_UNMATCHED_RIGHT_PAREN_ORD | RE_NO_GNU_OPS)
 189
 190 #define RE_SYNTAX_GNU_AWK                                               \
 191   ((RE_SYNTAX_POSIX_EXTENDED | RE_BACKSLASH_ESCAPE_IN_LISTS | RE_DEBUG) \
 192    & ~(RE_DOT_NOT_NULL | RE_INTERVALS | RE_CONTEXT_INDEP_OPS))
 193
 194 #define RE_SYNTAX_POSIX_AWK                                             \
 195   (RE_SYNTAX_POSIX_EXTENDED | RE_BACKSLASH_ESCAPE_IN_LISTS              \
 196    | RE_INTERVALS           | RE_NO_GNU_OPS)
 197
 198 #define RE_SYNTAX_GREP                                                  \
 199   (RE_BK_PLUS_QM              | RE_CHAR_CLASSES                         \
 200    | RE_HAT_LISTS_NOT_NEWLINE | RE_INTERVALS                            \
 201    | RE_NEWLINE_ALT)
 202
 203 #define RE_SYNTAX_EGREP                                                 \
 204   (RE_CHAR_CLASSES        | RE_CONTEXT_INDEP_ANCHORS                    \
 205    | RE_CONTEXT_INDEP_OPS | RE_HAT_LISTS_NOT_NEWLINE                    \
 206    | RE_NEWLINE_ALT       | RE_NO_BK_PARENS                             \
 207    | RE_NO_BK_VBAR)
 208
 209 #define RE_SYNTAX_POSIX_EGREP                                           \
 210   (RE_SYNTAX_EGREP | RE_INTERVALS | RE_NO_BK_BRACES)
 211
 212 /* P1003.2/D11.2, section 4.20.7.1, lines 5078ff.  */
 213 #define RE_SYNTAX_ED RE_SYNTAX_POSIX_BASIC
 214
 215 #define RE_SYNTAX_SED RE_SYNTAX_POSIX_BASIC
 216
 217 /* Syntax bits common to both basic and extended POSIX regex syntax.  */
 218 #define _RE_SYNTAX_POSIX_COMMON                                         \
 219   (RE_CHAR_CLASSES | RE_DOT_NEWLINE      | RE_DOT_NOT_NULL              \
 220    | RE_INTERVALS  | RE_NO_EMPTY_RANGES)
 221
 222 #define RE_SYNTAX_POSIX_BASIC                                           \
 223   (_RE_SYNTAX_POSIX_COMMON | RE_BK_PLUS_QM)
 224
 225 /* Differs from ..._POSIX_BASIC only in that RE_BK_PLUS_QM becomes
 226    RE_LIMITED_OPS, i.e., \? \+ \| are not recognized.  Actually, this
 227    isn't minimal, since other operators, such as \`, aren't disabled.  */
 228 #define RE_SYNTAX_POSIX_MINIMAL_BASIC                                   \
 229   (_RE_SYNTAX_POSIX_COMMON | RE_LIMITED_OPS)
 230
 231 #define RE_SYNTAX_POSIX_EXTENDED                                        \
 232   (_RE_SYNTAX_POSIX_COMMON  | RE_CONTEXT_INDEP_ANCHORS                  \
 233    | RE_CONTEXT_INDEP_OPS   | RE_NO_BK_BRACES                           \
 234    | RE_NO_BK_PARENS        | RE_NO_BK_VBAR                             \
 235    | RE_CONTEXT_INVALID_OPS | RE_UNMATCHED_RIGHT_PAREN_ORD)
 236
 237 /* Differs from ..._POSIX_EXTENDED in that RE_CONTEXT_INDEP_OPS is
 238    removed and RE_NO_BK_REFS is added.  */
 239 #define RE_SYNTAX_POSIX_MINIMAL_EXTENDED                                \
 240   (_RE_SYNTAX_POSIX_COMMON  | RE_CONTEXT_INDEP_ANCHORS                  \
 241    | RE_CONTEXT_INVALID_OPS | RE_NO_BK_BRACES                           \
 242    | RE_NO_BK_PARENS        | RE_NO_BK_REFS                             \
 243    | RE_NO_BK_VBAR          | RE_UNMATCHED_RIGHT_PAREN_ORD)
 244 /* [[[end syntaxes]]] */
 245 \f
 246 /* Maximum number of duplicates an interval can allow.  Some systems
 247    (erroneously) define this in other header files, but we want our
 248    value, so remove any previous define.  */
 249 #ifdef RE_DUP_MAX
 250 # undef RE_DUP_MAX
 251 #endif
 252 /* If sizeof(int) == 2, then ((1 << 15) - 1) overflows.  */
 253 #define RE_DUP_MAX (0x7fff)
 254
 255
 256 /* POSIX `cflags' bits (i.e., information for `regcomp').  */
 257
 258 /* If this bit is set, then use extended regular expression syntax.
 259    If not set, then use basic regular expression syntax.  */
 260 #define REG_EXTENDED 1
 261
 262 /* If this bit is set, then ignore case when matching.
 263    If not set, then case is significant.  */
 264 #define REG_ICASE (REG_EXTENDED << 1)
 265
 266 /* If this bit is set, then anchors do not match at newline
 267      characters in the string.
 268    If not set, then anchors do match at newlines.  */
 269 #define REG_NEWLINE (REG_ICASE << 1)
 270
 271 /* If this bit is set, then report only success or fail in regexec.
 272    If not set, then returns differ between not matching and errors.  */
 273 #define REG_NOSUB (REG_NEWLINE << 1)
 274
 275
 276 /* POSIX `eflags' bits (i.e., information for regexec).  */
 277
 278 /* If this bit is set, then the beginning-of-line operator doesn't match
 279      the beginning of the string (presumably because it's not the
 280      beginning of a line).
 281    If not set, then the beginning-of-line operator does match the
 282      beginning of the string.  */
 283 #define REG_NOTBOL 1
 284
 285 /* Like REG_NOTBOL, except for the end-of-line.  */
 286 #define REG_NOTEOL (1 << 1)
 287
 288
 289 /* If any error codes are removed, changed, or added, update the
 290    `re_error_msg' table in regex.c.  */
 291 typedef enum
 292 {
 293 #ifdef _XOPEN_SOURCE
 294   REG_ENOSYS = -1,      /* This will never happen for this implementation.  */
 295 #endif
 296
 297   REG_NOERROR = 0,      /* Success.  */
 298   REG_NOMATCH,          /* Didn't find a match (for regexec).  */
 299
 300   /* POSIX regcomp return error codes.  (In the order listed in the
 301      standard.)  */
 302   REG_BADPAT,           /* Invalid pattern.  */
 303   REG_ECOLLATE,         /* Not implemented.  */
 304   REG_ECTYPE,           /* Invalid character class name.  */
 305   REG_EESCAPE,          /* Trailing backslash.  */
 306   REG_ESUBREG,          /* Invalid back reference.  */
 307   REG_EBRACK,           /* Unmatched left bracket.  */
 308   REG_EPAREN,           /* Parenthesis imbalance.  */
 309   REG_EBRACE,           /* Unmatched \{.  */
 310   REG_BADBR,            /* Invalid contents of \{\}.  */
 311   REG_ERANGE,           /* Invalid range end.  */
 312   REG_ESPACE,           /* Ran out of memory.  */
 313   REG_BADRPT,           /* No preceding re for repetition op.  */
 314
 315   /* Error codes we've added.  */
 316   REG_EEND,             /* Premature end.  */
 317   REG_ESIZE,            /* Compiled pattern bigger than 2^16 bytes.  */
 318   REG_ERPAREN,          /* Unmatched ) or \); not returned from regcomp.  */
 319   REG_ERANGEX           /* Range striding over charsets.  */
 320 } reg_errcode_t;
 321 \f
 322 /* This data structure represents a compiled pattern.  Before calling
 323    the pattern compiler, the fields `buffer', `allocated', `fastmap',
 324    `translate', and `no_sub' can be set.  After the pattern has been
 325    compiled, the `re_nsub' field is available.  All other fields are
 326    private to the regex routines.  */
 327
 328 #ifndef RE_TRANSLATE_TYPE
 329 # define RE_TRANSLATE_TYPE char *
 330 #endif
 331
 332 struct re_pattern_buffer
 333 {
 334 /* [[[begin pattern_buffer]]] */
 335         /* Space that holds the compiled pattern.  It is declared as
 336           `unsigned char *' because its elements are
 337            sometimes used as array indexes.  */
 338   unsigned char *buffer;
 339
 340         /* Number of bytes to which `buffer' points.  */
 341   size_t allocated;
 342
 343         /* Number of bytes actually used in `buffer'.  */
 344   size_t used;
 345
 346         /* Syntax setting with which the pattern was compiled.  */
 347   reg_syntax_t syntax;
 348
 349         /* Pointer to a fastmap, if any, otherwise zero.  re_search uses
 350            the fastmap, if there is one, to skip over impossible
 351            starting points for matches.  */
 352   char *fastmap;
 353
 354         /* Either a translate table to apply to all characters before
 355            comparing them, or zero for no translation.  The translation
 356            is applied to a pattern when it is compiled and to a string
 357            when it is matched.  */
 358   RE_TRANSLATE_TYPE translate;
 359
 360         /* Number of subexpressions found by the compiler.  */
 361   size_t re_nsub;
 362
 363         /* Zero if this pattern cannot match the empty string, one else.
 364            Well, in truth it's used only in `re_search_2', to see
 365            whether or not we should use the fastmap, so we don't set
 366            this absolutely perfectly; see `re_compile_fastmap'.  */
 367   unsigned can_be_null : 1;
 368
 369         /* If REGS_UNALLOCATED, allocate space in the `regs' structure
 370              for `max (RE_NREGS, re_nsub + 1)' groups.
 371            If REGS_REALLOCATE, reallocate space if necessary.
 372            If REGS_FIXED, use what's there.  */
 373 #define REGS_UNALLOCATED 0
 374 #define REGS_REALLOCATE 1
 375 #define REGS_FIXED 2
 376   unsigned regs_allocated : 2;
 377
 378         /* Set to zero when `regex_compile' compiles a pattern; set to one
 379            by `re_compile_fastmap' if it updates the fastmap.  */
 380   unsigned fastmap_accurate : 1;
 381
 382         /* If set, `re_match_2' does not return information about
 383            subexpressions.  */
 384   unsigned no_sub : 1;
 385
 386         /* If set, a beginning-of-line anchor doesn't match at the
 387            beginning of the string.  */
 388   unsigned not_bol : 1;
 389
 390         /* Similarly for an end-of-line anchor.  */
 391   unsigned not_eol : 1;
 392
 393   /* If true, the compilation of the pattern had to look up the syntax table,
 394      so the compiled pattern is only valid for the current syntax table.  */
 395   unsigned used_syntax : 1;
 396
 397 #ifdef emacs
 398   /* If true, multi-byte form in the regexp pattern should be
 399      recognized as a multibyte character.  */
 400   unsigned multibyte : 1;
 401
 402   /* If true, multi-byte form in the target of match should be
 403      recognized as a multibyte character.  */
 404   unsigned target_multibyte : 1;
 405
 406   /* Charset of unibyte characters at compiling time. */
 407   int charset_unibyte;
 408 #endif
 409
 410 /* [[[end pattern_buffer]]] */
 411 };
 412
 413 typedef struct re_pattern_buffer regex_t;
 414 \f
 415 /* Type for byte offsets within the string.  POSIX mandates this to be an int,
 416    but the Open Group has signaled its intention to change the requirement to
 417    be that regoff_t be at least as wide as ptrdiff_t and ssize_t.  Current
 418    gnulib sources also use ssize_t, and we need this for supporting buffers and
 419    strings > 2GB on 64-bit hosts.  */
 420 typedef ssize_t regoff_t;
 421
 422
 423 /* This is the structure we store register match data in.  See
 424    regex.texinfo for a full description of what registers match.  */
 425 struct re_registers
 426 {
 427   unsigned num_regs;
 428   regoff_t *start;
 429   regoff_t *end;
 430 };
 431
 432
 433 /* If `regs_allocated' is REGS_UNALLOCATED in the pattern buffer,
 434    `re_match_2' returns information about at least this many registers
 435    the first time a `regs' structure is passed.  */
 436 #ifndef RE_NREGS
 437 # define RE_NREGS 30
 438 #endif
 439
 440
 441 /* POSIX specification for registers.  Aside from the different names than
 442    `re_registers', POSIX uses an array of structures, instead of a
 443    structure of arrays.  */
 444 typedef struct
 445 {
 446   regoff_t rm_so;  /* Byte offset from string's start to substring's start.  */
 447   regoff_t rm_eo;  /* Byte offset from string's start to substring's end.  */
 448 } regmatch_t;
 449 \f
 450 /* Declarations for routines.  */
 451
 452 /* Sets the current default syntax to SYNTAX, and return the old syntax.
 453    You can also simply assign to the `re_syntax_options' variable.  */
 454 extern reg_syntax_t re_set_syntax (reg_syntax_t __syntax);
 455
 456 /* Compile the regular expression PATTERN, with length LENGTH
 457    and syntax given by the global `re_syntax_options', into the buffer
 458    BUFFER.  Return NULL if successful, and an error string if not.  */
 459 extern const char *re_compile_pattern (const char *__pattern, size_t __length,
 460                                        struct re_pattern_buffer *__buffer);
 461
 462
 463 /* Compile a fastmap for the compiled pattern in BUFFER; used to
 464    accelerate searches.  Return 0 if successful and -2 if was an
 465    internal error.  */
 466 extern int re_compile_fastmap (struct re_pattern_buffer *__buffer);
 467
 468
 469 /* Search in the string STRING (with length LENGTH) for the pattern
 470    compiled into BUFFER.  Start searching at position START, for RANGE
 471    characters.  Return the starting position of the match, -1 for no
 472    match, or -2 for an internal error.  Also return register
 473    information in REGS (if REGS and BUFFER->no_sub are nonzero).  */
 474 extern regoff_t re_search (struct re_pattern_buffer *__buffer,
 475                            const char *__string, size_t __length,
 476                            ssize_t __start, ssize_t __range,
 477                            struct re_registers *__regs);
 478
 479
 480 /* Like `re_search', but search in the concatenation of STRING1 and
 481    STRING2.  Also, stop searching at index START + STOP.  */
 482 extern regoff_t re_search_2 (struct re_pattern_buffer *__buffer,
 483                              const char *__string1, size_t __length1,
 484                              const char *__string2, size_t __length2,
 485                              ssize_t __start, ssize_t __range,
 486                              struct re_registers *__regs,
 487                              ssize_t __stop);
 488
 489
 490 /* Like `re_search', but return how many characters in STRING the regexp
 491    in BUFFER matched, starting at position START.  */
 492 extern regoff_t re_match (struct re_pattern_buffer *__buffer,
 493                           const char *__string, size_t __length,
 494                           ssize_t __start, struct re_registers *__regs);
 495
 496
 497 /* Relates to `re_match' as `re_search_2' relates to `re_search'.  */
 498 extern regoff_t re_match_2 (struct re_pattern_buffer *__buffer,
 499                             const char *__string1, size_t __length1,
 500                             const char *__string2, size_t __length2,
 501                             ssize_t __start, struct re_registers *__regs,
 502                             ssize_t __stop);
 503
 504
 505 /* Set REGS to hold NUM_REGS registers, storing them in STARTS and
 506    ENDS.  Subsequent matches using BUFFER and REGS will use this memory
 507    for recording register information.  STARTS and ENDS must be
 508    allocated with malloc, and must each be at least `NUM_REGS * sizeof
 509    (regoff_t)' bytes long.
 510
 511    If NUM_REGS == 0, then subsequent matches should allocate their own
 512    register data.
 513
 514    Unless this function is called, the first search or match using
 515    PATTERN_BUFFER will allocate its own register data, without
 516    freeing the old data.  */
 517 extern void re_set_registers (struct re_pattern_buffer *__buffer,
 518                               struct re_registers *__regs,
 519                               unsigned __num_regs,
 520                               regoff_t *__starts, regoff_t *__ends);
 521
 522 #if defined _REGEX_RE_COMP || defined _LIBC
 523 # ifndef _CRAY
 524 /* 4.2 bsd compatibility.  */
 525 extern char *re_comp (const char *);
 526 extern int re_exec (const char *);
 527 # endif
 528 #endif
 529
 530 /* GCC 2.95 and later have "__restrict"; C99 compilers have
 531    "restrict", and "configure" may have defined "restrict".
 532    Other compilers use __restrict, __restrict__, and _Restrict, and
 533    'configure' might #define 'restrict' to those words, so pick a
 534    different name.  */
 535 #ifndef _Restrict_
 536 # if 199901L <= __STDC_VERSION__
 537 #  define _Restrict_ restrict
 538 # elif 2 < __GNUC__ || (2 == __GNUC__ && 95 <= __GNUC_MINOR__)
 539 #  define _Restrict_ __restrict
 540 # else
 541 #  define _Restrict_
 542 # endif
 543 #endif
 544 /* gcc 3.1 and up support the [restrict] syntax.  Don't trust
 545    sys/cdefs.h's definition of __restrict_arr, though, as it
 546    mishandles gcc -ansi -pedantic.  */
 547 #ifndef _Restrict_arr_
 548 # if ((199901L <= __STDC_VERSION__                                      \
 549        || ((3 < __GNUC__ || (3 == __GNUC__ && 1 <= __GNUC_MINOR__))     \
 550            && !defined __STRICT_ANSI__))                                        \
 551       && !defined __GNUG__)
 552 #  define _Restrict_arr_ _Restrict_
 553 # else
 554 #  define _Restrict_arr_
 555 # endif
 556 #endif
 557
 558 /* POSIX compatibility.  */
 559 extern reg_errcode_t regcomp (regex_t *_Restrict_ __preg,
 560                               const char *_Restrict_ __pattern,
 561                               int __cflags);
 562
 563 extern reg_errcode_t regexec (const regex_t *_Restrict_ __preg,
 564                               const char *_Restrict_ __string, size_t __nmatch,
 565                               regmatch_t __pmatch[_Restrict_arr_],
 566                               int __eflags);
 567
 568 extern size_t regerror (int __errcode, const regex_t * __preg,
 569                         char *__errbuf, size_t __errbuf_size);
 570
 571 extern void regfree (regex_t *__preg);
 572
 573
 574 #ifdef __cplusplus
 575 }
 576 #endif  /* C++ */
 577
 578 /* For platform which support the ISO C amendment 1 functionality we
 579    support user defined character classes.  */
 580 #if WIDE_CHAR_SUPPORT
 581 /* Solaris 2.5 has a bug: <wchar.h> must be included before <wctype.h>.  */
 582 # include <wchar.h>
 583 # include <wctype.h>
 584 #endif
 585
 586 #if WIDE_CHAR_SUPPORT
 587 /* The GNU C library provides support for user-defined character classes
 588    and the functions from ISO C amendment 1.  */
 589 # ifdef CHARCLASS_NAME_MAX
 590 #  define CHAR_CLASS_MAX_LENGTH CHARCLASS_NAME_MAX
 591 # else
 592 /* This shouldn't happen but some implementation might still have this
 593    problem.  Use a reasonable default value.  */
 594 #  define CHAR_CLASS_MAX_LENGTH 256
 595 # endif
 596 typedef wctype_t re_wctype_t;
 597 typedef wchar_t re_wchar_t;
 598 # define re_wctype wctype
 599 # define re_iswctype iswctype
 600 # define re_wctype_to_bit(cc) 0
 601 #else
 602 # define CHAR_CLASS_MAX_LENGTH  9 /* Namely, `multibyte'.  */
 603 # define btowc(c) c
 604
 605 /* Character classes.  */
 606 typedef enum { RECC_ERROR = 0,
 607                RECC_ALNUM, RECC_ALPHA, RECC_WORD,
 608                RECC_GRAPH, RECC_PRINT,
 609                RECC_LOWER, RECC_UPPER,
 610                RECC_PUNCT, RECC_CNTRL,
 611                RECC_DIGIT, RECC_XDIGIT,
 612                RECC_BLANK, RECC_SPACE,
 613                RECC_MULTIBYTE, RECC_NONASCII,
 614                RECC_ASCII, RECC_UNIBYTE
 615 } re_wctype_t;
 616
 617 extern char re_iswctype (int ch,    re_wctype_t cc);
 618 extern re_wctype_t re_wctype (const unsigned char* str);
 619
 620 typedef int re_wchar_t;
 621
 622 extern void re_set_whitespace_regexp (const char *regexp);
 623
 624 #endif /* not WIDE_CHAR_SUPPORT */
 625
 626 #endif /* regex.h */
 627 \f