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, 2000-2015 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 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 /* Roughly the maximum number of failure points on the stack.  */
 176 extern size_t re_max_failures;
 177
 178 \f
 179 /* Define combinations of the above bits for the standard possibilities.
 180    (The [[[ comments delimit what gets put into the Texinfo file, so
 181    don't delete them!)  */
 182 /* [[[begin syntaxes]]] */
 183 #define RE_SYNTAX_EMACS                                                 \
 184   (RE_CHAR_CLASSES | RE_INTERVALS | RE_SHY_GROUPS | RE_FRUGAL)
 185
 186 #define RE_SYNTAX_AWK                                                   \
 187   (RE_BACKSLASH_ESCAPE_IN_LISTS   | RE_DOT_NOT_NULL                     \
 188    | RE_NO_BK_PARENS              | RE_NO_BK_REFS                       \
 189    | RE_NO_BK_VBAR                | RE_NO_EMPTY_RANGES                  \
 190    | RE_DOT_NEWLINE               | RE_CONTEXT_INDEP_ANCHORS            \
 191    | RE_UNMATCHED_RIGHT_PAREN_ORD | RE_NO_GNU_OPS)
 192
 193 #define RE_SYNTAX_GNU_AWK                                               \
 194   ((RE_SYNTAX_POSIX_EXTENDED | RE_BACKSLASH_ESCAPE_IN_LISTS | RE_DEBUG) \
 195    & ~(RE_DOT_NOT_NULL | RE_INTERVALS | RE_CONTEXT_INDEP_OPS))
 196
 197 #define RE_SYNTAX_POSIX_AWK                                             \
 198   (RE_SYNTAX_POSIX_EXTENDED | RE_BACKSLASH_ESCAPE_IN_LISTS              \
 199    | RE_INTERVALS           | RE_NO_GNU_OPS)
 200
 201 #define RE_SYNTAX_GREP                                                  \
 202   (RE_BK_PLUS_QM              | RE_CHAR_CLASSES                         \
 203    | RE_HAT_LISTS_NOT_NEWLINE | RE_INTERVALS                            \
 204    | RE_NEWLINE_ALT)
 205
 206 #define RE_SYNTAX_EGREP                                                 \
 207   (RE_CHAR_CLASSES        | RE_CONTEXT_INDEP_ANCHORS                    \
 208    | RE_CONTEXT_INDEP_OPS | RE_HAT_LISTS_NOT_NEWLINE                    \
 209    | RE_NEWLINE_ALT       | RE_NO_BK_PARENS                             \
 210    | RE_NO_BK_VBAR)
 211
 212 #define RE_SYNTAX_POSIX_EGREP                                           \
 213   (RE_SYNTAX_EGREP | RE_INTERVALS | RE_NO_BK_BRACES)
 214
 215 /* P1003.2/D11.2, section 4.20.7.1, lines 5078ff.  */
 216 #define RE_SYNTAX_ED RE_SYNTAX_POSIX_BASIC
 217
 218 #define RE_SYNTAX_SED RE_SYNTAX_POSIX_BASIC
 219
 220 /* Syntax bits common to both basic and extended POSIX regex syntax.  */
 221 #define _RE_SYNTAX_POSIX_COMMON                                         \
 222   (RE_CHAR_CLASSES | RE_DOT_NEWLINE      | RE_DOT_NOT_NULL              \
 223    | RE_INTERVALS  | RE_NO_EMPTY_RANGES)
 224
 225 #define RE_SYNTAX_POSIX_BASIC                                           \
 226   (_RE_SYNTAX_POSIX_COMMON | RE_BK_PLUS_QM)
 227
 228 /* Differs from ..._POSIX_BASIC only in that RE_BK_PLUS_QM becomes
 229    RE_LIMITED_OPS, i.e., \? \+ \| are not recognized.  Actually, this
 230    isn't minimal, since other operators, such as \`, aren't disabled.  */
 231 #define RE_SYNTAX_POSIX_MINIMAL_BASIC                                   \
 232   (_RE_SYNTAX_POSIX_COMMON | RE_LIMITED_OPS)
 233
 234 #define RE_SYNTAX_POSIX_EXTENDED                                        \
 235   (_RE_SYNTAX_POSIX_COMMON  | RE_CONTEXT_INDEP_ANCHORS                  \
 236    | RE_CONTEXT_INDEP_OPS   | RE_NO_BK_BRACES                           \
 237    | RE_NO_BK_PARENS        | RE_NO_BK_VBAR                             \
 238    | RE_CONTEXT_INVALID_OPS | RE_UNMATCHED_RIGHT_PAREN_ORD)
 239
 240 /* Differs from ..._POSIX_EXTENDED in that RE_CONTEXT_INDEP_OPS is
 241    removed and RE_NO_BK_REFS is added.  */
 242 #define RE_SYNTAX_POSIX_MINIMAL_EXTENDED                                \
 243   (_RE_SYNTAX_POSIX_COMMON  | RE_CONTEXT_INDEP_ANCHORS                  \
 244    | RE_CONTEXT_INVALID_OPS | RE_NO_BK_BRACES                           \
 245    | RE_NO_BK_PARENS        | RE_NO_BK_REFS                             \
 246    | RE_NO_BK_VBAR          | RE_UNMATCHED_RIGHT_PAREN_ORD)
 247 /* [[[end syntaxes]]] */
 248 \f
 249 /* Maximum number of duplicates an interval can allow.  Some systems
 250    (erroneously) define this in other header files, but we want our
 251    value, so remove any previous define.  */
 252 #ifdef RE_DUP_MAX
 253 # undef RE_DUP_MAX
 254 #endif
 255 /* If sizeof(int) == 2, then ((1 << 15) - 1) overflows.  */
 256 #define RE_DUP_MAX (0x7fff)
 257
 258
 259 /* POSIX `cflags' bits (i.e., information for `regcomp').  */
 260
 261 /* If this bit is set, then use extended regular expression syntax.
 262    If not set, then use basic regular expression syntax.  */
 263 #define REG_EXTENDED 1
 264
 265 /* If this bit is set, then ignore case when matching.
 266    If not set, then case is significant.  */
 267 #define REG_ICASE (REG_EXTENDED << 1)
 268
 269 /* If this bit is set, then anchors do not match at newline
 270      characters in the string.
 271    If not set, then anchors do match at newlines.  */
 272 #define REG_NEWLINE (REG_ICASE << 1)
 273
 274 /* If this bit is set, then report only success or fail in regexec.
 275    If not set, then returns differ between not matching and errors.  */
 276 #define REG_NOSUB (REG_NEWLINE << 1)
 277
 278
 279 /* POSIX `eflags' bits (i.e., information for regexec).  */
 280
 281 /* If this bit is set, then the beginning-of-line operator doesn't match
 282      the beginning of the string (presumably because it's not the
 283      beginning of a line).
 284    If not set, then the beginning-of-line operator does match the
 285      beginning of the string.  */
 286 #define REG_NOTBOL 1
 287
 288 /* Like REG_NOTBOL, except for the end-of-line.  */
 289 #define REG_NOTEOL (1 << 1)
 290
 291
 292 /* If any error codes are removed, changed, or added, update the
 293    `re_error_msg' table in regex.c.  */
 294 typedef enum
 295 {
 296 #ifdef _XOPEN_SOURCE
 297   REG_ENOSYS = -1,      /* This will never happen for this implementation.  */
 298 #endif
 299
 300   REG_NOERROR = 0,      /* Success.  */
 301   REG_NOMATCH,          /* Didn't find a match (for regexec).  */
 302
 303   /* POSIX regcomp return error codes.  (In the order listed in the
 304      standard.)  */
 305   REG_BADPAT,           /* Invalid pattern.  */
 306   REG_ECOLLATE,         /* Not implemented.  */
 307   REG_ECTYPE,           /* Invalid character class name.  */
 308   REG_EESCAPE,          /* Trailing backslash.  */
 309   REG_ESUBREG,          /* Invalid back reference.  */
 310   REG_EBRACK,           /* Unmatched left bracket.  */
 311   REG_EPAREN,           /* Parenthesis imbalance.  */
 312   REG_EBRACE,           /* Unmatched \{.  */
 313   REG_BADBR,            /* Invalid contents of \{\}.  */
 314   REG_ERANGE,           /* Invalid range end.  */
 315   REG_ESPACE,           /* Ran out of memory.  */
 316   REG_BADRPT,           /* No preceding re for repetition op.  */
 317
 318   /* Error codes we've added.  */
 319   REG_EEND,             /* Premature end.  */
 320   REG_ESIZE,            /* Compiled pattern bigger than 2^16 bytes.  */
 321   REG_ERPAREN,          /* Unmatched ) or \); not returned from regcomp.  */
 322   REG_ERANGEX           /* Range striding over charsets.  */
 323 } reg_errcode_t;
 324 \f
 325 /* This data structure represents a compiled pattern.  Before calling
 326    the pattern compiler, the fields `buffer', `allocated', `fastmap',
 327    `translate', and `no_sub' can be set.  After the pattern has been
 328    compiled, the `re_nsub' field is available.  All other fields are
 329    private to the regex routines.  */
 330
 331 #ifndef RE_TRANSLATE_TYPE
 332 # define RE_TRANSLATE_TYPE char *
 333 #endif
 334
 335 struct re_pattern_buffer
 336 {
 337 /* [[[begin pattern_buffer]]] */
 338         /* Space that holds the compiled pattern.  It is declared as
 339           `unsigned char *' because its elements are
 340            sometimes used as array indexes.  */
 341   unsigned char *buffer;
 342
 343         /* Number of bytes to which `buffer' points.  */
 344   size_t allocated;
 345
 346         /* Number of bytes actually used in `buffer'.  */
 347   size_t used;
 348
 349         /* Syntax setting with which the pattern was compiled.  */
 350   reg_syntax_t syntax;
 351
 352         /* Pointer to a fastmap, if any, otherwise zero.  re_search uses
 353            the fastmap, if there is one, to skip over impossible
 354            starting points for matches.  */
 355   char *fastmap;
 356
 357         /* Either a translate table to apply to all characters before
 358            comparing them, or zero for no translation.  The translation
 359            is applied to a pattern when it is compiled and to a string
 360            when it is matched.  */
 361   RE_TRANSLATE_TYPE translate;
 362
 363         /* Number of subexpressions found by the compiler.  */
 364   size_t re_nsub;
 365
 366         /* Zero if this pattern cannot match the empty string, one else.
 367            Well, in truth it's used only in `re_search_2', to see
 368            whether or not we should use the fastmap, so we don't set
 369            this absolutely perfectly; see `re_compile_fastmap'.  */
 370   unsigned can_be_null : 1;
 371
 372         /* If REGS_UNALLOCATED, allocate space in the `regs' structure
 373              for `max (RE_NREGS, re_nsub + 1)' groups.
 374            If REGS_REALLOCATE, reallocate space if necessary.
 375            If REGS_FIXED, use what's there.  */
 376 #define REGS_UNALLOCATED 0
 377 #define REGS_REALLOCATE 1
 378 #define REGS_FIXED 2
 379   unsigned regs_allocated : 2;
 380
 381         /* Set to zero when `regex_compile' compiles a pattern; set to one
 382            by `re_compile_fastmap' if it updates the fastmap.  */
 383   unsigned fastmap_accurate : 1;
 384
 385         /* If set, `re_match_2' does not return information about
 386            subexpressions.  */
 387   unsigned no_sub : 1;
 388
 389         /* If set, a beginning-of-line anchor doesn't match at the
 390            beginning of the string.  */
 391   unsigned not_bol : 1;
 392
 393         /* Similarly for an end-of-line anchor.  */
 394   unsigned not_eol : 1;
 395
 396   /* If true, the compilation of the pattern had to look up the syntax table,
 397      so the compiled pattern is only valid for the current syntax table.  */
 398   unsigned used_syntax : 1;
 399
 400 #ifdef emacs
 401   /* If true, multi-byte form in the regexp pattern should be
 402      recognized as a multibyte character.  */
 403   unsigned multibyte : 1;
 404
 405   /* If true, multi-byte form in the target of match should be
 406      recognized as a multibyte character.  */
 407   unsigned target_multibyte : 1;
 408
 409   /* Charset of unibyte characters at compiling time. */
 410   int charset_unibyte;
 411 #endif
 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 to be an int,
 419    but the Open Group has signaled its intention to change the requirement to
 420    be that regoff_t be at least as wide as ptrdiff_t and ssize_t.  Current
 421    gnulib sources also use ssize_t, and we need this for supporting buffers and
 422    strings > 2GB on 64-bit hosts.  */
 423 typedef ssize_t 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 /* Sets the current default syntax to SYNTAX, and return the old syntax.
 456    You can also simply assign to the `re_syntax_options' variable.  */
 457 extern reg_syntax_t re_set_syntax (reg_syntax_t __syntax);
 458
 459 /* Compile the regular expression PATTERN, with length LENGTH
 460    and syntax given by the global `re_syntax_options', into the buffer
 461    BUFFER.  Return NULL if successful, and an error string if not.  */
 462 extern const char *re_compile_pattern (const char *__pattern, size_t __length,
 463                                        struct re_pattern_buffer *__buffer);
 464
 465
 466 /* Compile a fastmap for the compiled pattern in BUFFER; used to
 467    accelerate searches.  Return 0 if successful and -2 if was an
 468    internal error.  */
 469 extern int re_compile_fastmap (struct re_pattern_buffer *__buffer);
 470
 471
 472 /* Search in the string STRING (with length LENGTH) for the pattern
 473    compiled into BUFFER.  Start searching at position START, for RANGE
 474    characters.  Return the starting position of the match, -1 for no
 475    match, or -2 for an internal error.  Also return register
 476    information in REGS (if REGS and BUFFER->no_sub are nonzero).  */
 477 extern regoff_t re_search (struct re_pattern_buffer *__buffer,
 478                            const char *__string, size_t __length,
 479                            ssize_t __start, ssize_t __range,
 480                            struct re_registers *__regs);
 481
 482
 483 /* Like `re_search', but search in the concatenation of STRING1 and
 484    STRING2.  Also, stop searching at index START + STOP.  */
 485 extern regoff_t re_search_2 (struct re_pattern_buffer *__buffer,
 486                              const char *__string1, size_t __length1,
 487                              const char *__string2, size_t __length2,
 488                              ssize_t __start, ssize_t __range,
 489                              struct re_registers *__regs,
 490                              ssize_t __stop);
 491
 492
 493 /* Like `re_search', but return how many characters in STRING the regexp
 494    in BUFFER matched, starting at position START.  */
 495 extern regoff_t re_match (struct re_pattern_buffer *__buffer,
 496                           const char *__string, size_t __length,
 497                           ssize_t __start, struct re_registers *__regs);
 498
 499
 500 /* Relates to `re_match' as `re_search_2' relates to `re_search'.  */
 501 extern regoff_t re_match_2 (struct re_pattern_buffer *__buffer,
 502                             const char *__string1, size_t __length1,
 503                             const char *__string2, size_t __length2,
 504                             ssize_t __start, struct re_registers *__regs,
 505                             ssize_t __stop);
 506
 507
 508 /* Set REGS to hold NUM_REGS registers, storing them in STARTS and
 509    ENDS.  Subsequent matches using BUFFER and REGS will use this memory
 510    for recording register information.  STARTS and ENDS must be
 511    allocated with malloc, and must each be at least `NUM_REGS * sizeof
 512    (regoff_t)' bytes long.
 513
 514    If NUM_REGS == 0, then subsequent matches should allocate their own
 515    register data.
 516
 517    Unless this function is called, the first search or match using
 518    PATTERN_BUFFER will allocate its own register data, without
 519    freeing the old data.  */
 520 extern void re_set_registers (struct re_pattern_buffer *__buffer,
 521                               struct re_registers *__regs,
 522                               unsigned __num_regs,
 523                               regoff_t *__starts, regoff_t *__ends);
 524
 525 #if defined _REGEX_RE_COMP || defined _LIBC
 526 # ifndef _CRAY
 527 /* 4.2 bsd compatibility.  */
 528 extern char *re_comp (const char *);
 529 extern int re_exec (const char *);
 530 # endif
 531 #endif
 532
 533 /* GCC 2.95 and later have "__restrict"; C99 compilers have
 534    "restrict", and "configure" may have defined "restrict".
 535    Other compilers use __restrict, __restrict__, and _Restrict, and
 536    'configure' might #define 'restrict' to those words, so pick a
 537    different name.  */
 538 #ifndef _Restrict_
 539 # if 199901L <= __STDC_VERSION__
 540 #  define _Restrict_ restrict
 541 # elif 2 < __GNUC__ || (2 == __GNUC__ && 95 <= __GNUC_MINOR__)
 542 #  define _Restrict_ __restrict
 543 # else
 544 #  define _Restrict_
 545 # endif
 546 #endif
 547 /* gcc 3.1 and up support the [restrict] syntax.  Don't trust
 548    sys/cdefs.h's definition of __restrict_arr, though, as it
 549    mishandles gcc -ansi -pedantic.  */
 550 #ifndef _Restrict_arr_
 551 # if ((199901L <= __STDC_VERSION__                                      \
 552        || ((3 < __GNUC__ || (3 == __GNUC__ && 1 <= __GNUC_MINOR__))     \
 553            && !defined __STRICT_ANSI__))                                        \
 554       && !defined __GNUG__)
 555 #  define _Restrict_arr_ _Restrict_
 556 # else
 557 #  define _Restrict_arr_
 558 # endif
 559 #endif
 560
 561 /* POSIX compatibility.  */
 562 extern reg_errcode_t regcomp (regex_t *_Restrict_ __preg,
 563                               const char *_Restrict_ __pattern,
 564                               int __cflags);
 565
 566 extern reg_errcode_t regexec (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 (int __errcode, const regex_t * __preg,
 572                         char *__errbuf, size_t __errbuf_size);
 573
 574 extern void regfree (regex_t *__preg);
 575
 576
 577 #ifdef __cplusplus
 578 }
 579 #endif  /* C++ */
 580
 581 /* For platform which support the ISO C amendment 1 functionality we
 582    support user defined character classes.  */
 583 #if WIDE_CHAR_SUPPORT
 584 /* Solaris 2.5 has a bug: <wchar.h> must be included before <wctype.h>.  */
 585 # include <wchar.h>
 586 # include <wctype.h>
 587 #endif
 588
 589 #if WIDE_CHAR_SUPPORT
 590 /* The GNU C library provides support for user-defined character classes
 591    and the functions from ISO C amendment 1.  */
 592 # ifdef CHARCLASS_NAME_MAX
 593 #  define CHAR_CLASS_MAX_LENGTH CHARCLASS_NAME_MAX
 594 # else
 595 /* This shouldn't happen but some implementation might still have this
 596    problem.  Use a reasonable default value.  */
 597 #  define CHAR_CLASS_MAX_LENGTH 256
 598 # endif
 599 typedef wctype_t re_wctype_t;
 600 typedef wchar_t re_wchar_t;
 601 # define re_wctype wctype
 602 # define re_iswctype iswctype
 603 # define re_wctype_to_bit(cc) 0
 604 #else
 605 # define CHAR_CLASS_MAX_LENGTH  9 /* Namely, `multibyte'.  */
 606 # ifndef emacs
 607 #  define btowc(c) c
 608 # endif
 609
 610 /* Character classes.  */
 611 typedef enum { RECC_ERROR = 0,
 612                RECC_ALNUM, RECC_ALPHA, RECC_WORD,
 613                RECC_GRAPH, RECC_PRINT,
 614                RECC_LOWER, RECC_UPPER,
 615                RECC_PUNCT, RECC_CNTRL,
 616                RECC_DIGIT, RECC_XDIGIT,
 617                RECC_BLANK, RECC_SPACE,
 618                RECC_MULTIBYTE, RECC_NONASCII,
 619                RECC_ASCII, RECC_UNIBYTE
 620 } re_wctype_t;
 621
 622 extern char re_iswctype (int ch,    re_wctype_t cc);
 623 extern re_wctype_t re_wctype (const unsigned char* str);
 624
 625 typedef int re_wchar_t;
 626
 627 extern void re_set_whitespace_regexp (const char *regexp);
 628
 629 #endif /* not WIDE_CHAR_SUPPORT */
 630
 631 #endif /* regex.h */
 632 \f