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, 2000-2017 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 #include <sys/types.h>
  31
  32 /* Allow the use in C++ code.  */
  33 #ifdef __cplusplus
  34 extern "C" {
  35 #endif
  36
  37 #if !defined _POSIX_C_SOURCE && !defined _POSIX_SOURCE && defined VMS
  38 /* VMS doesn't have `size_t' in <sys/types.h>, even though POSIX says it
  39    should be there.  */
  40 # include <stddef.h>
  41 #endif
  42
  43 /* The following bits are used to determine the regexp syntax we
  44    recognize.  The set/not-set meanings where historically chosen so
  45    that Emacs syntax had the value 0.
  46    The bits are given in alphabetical order, and
  47    the definitions shifted by one from the previous bit; thus, when we
  48    add or remove a bit, only one other definition need change.  */
  49 typedef unsigned long reg_syntax_t;
  50
  51 /* If this bit is not set, then \ inside a bracket expression is literal.
  52    If set, then such a \ quotes the following character.  */
  53 #define RE_BACKSLASH_ESCAPE_IN_LISTS ((unsigned long int) 1)
  54
  55 /* If this bit is not set, then + and ? are operators, and \+ and \? are
  56      literals.
  57    If set, then \+ and \? are operators and + and ? are literals.  */
  58 #define RE_BK_PLUS_QM (RE_BACKSLASH_ESCAPE_IN_LISTS << 1)
  59
  60 /* If this bit is set, then character classes are supported.  They are:
  61      [:alpha:], [:upper:], [:lower:],  [:digit:], [:alnum:], [:xdigit:],
  62      [:space:], [:print:], [:punct:], [:graph:], and [:cntrl:].
  63    If not set, then character classes are not supported.  */
  64 #define RE_CHAR_CLASSES (RE_BK_PLUS_QM << 1)
  65
  66 /* If this bit is set, then ^ and $ are always anchors (outside bracket
  67      expressions, of course).
  68    If this bit is not set, then it depends:
  69         ^  is an anchor if it is at the beginning of a regular
  70            expression or after an open-group or an alternation operator;
  71         $  is an anchor if it is at the end of a regular expression, or
  72            before a close-group or an alternation operator.
  73
  74    This bit could be (re)combined with RE_CONTEXT_INDEP_OPS, because
  75    POSIX draft 11.2 says that * etc. in leading positions is undefined.
  76    We already implemented a previous draft which made those constructs
  77    invalid, though, so we haven't changed the code back.  */
  78 #define RE_CONTEXT_INDEP_ANCHORS (RE_CHAR_CLASSES << 1)
  79
  80 /* If this bit is set, then special characters are always special
  81      regardless of where they are in the pattern.
  82    If this bit is not set, then special characters are special only in
  83      some contexts; otherwise they are ordinary.  Specifically,
  84      * + ? and intervals are only special when not after the beginning,
  85      open-group, or alternation operator.  */
  86 #define RE_CONTEXT_INDEP_OPS (RE_CONTEXT_INDEP_ANCHORS << 1)
  87
  88 /* If this bit is set, then *, +, ?, and { cannot be first in an re or
  89      immediately after an alternation or begin-group operator.  */
  90 #define RE_CONTEXT_INVALID_OPS (RE_CONTEXT_INDEP_OPS << 1)
  91
  92 /* If this bit is set, then . matches newline.
  93    If not set, then it doesn't.  */
  94 #define RE_DOT_NEWLINE (RE_CONTEXT_INVALID_OPS << 1)
  95
  96 /* If this bit is set, then . doesn't match NUL.
  97    If not set, then it does.  */
  98 #define RE_DOT_NOT_NULL (RE_DOT_NEWLINE << 1)
  99
 100 /* If this bit is set, nonmatching lists [^...] do not match newline.
 101    If not set, they do.  */
 102 #define RE_HAT_LISTS_NOT_NEWLINE (RE_DOT_NOT_NULL << 1)
 103
 104 /* If this bit is set, either \{...\} or {...} defines an
 105      interval, depending on RE_NO_BK_BRACES.
 106    If not set, \{, \}, {, and } are literals.  */
 107 #define RE_INTERVALS (RE_HAT_LISTS_NOT_NEWLINE << 1)
 108
 109 /* If this bit is set, +, ? and | aren't recognized as operators.
 110    If not set, they are.  */
 111 #define RE_LIMITED_OPS (RE_INTERVALS << 1)
 112
 113 /* If this bit is set, newline is an alternation operator.
 114    If not set, newline is literal.  */
 115 #define RE_NEWLINE_ALT (RE_LIMITED_OPS << 1)
 116
 117 /* If this bit is set, then `{...}' defines an interval, and \{ and \}
 118      are literals.
 119   If not set, then `\{...\}' defines an interval.  */
 120 #define RE_NO_BK_BRACES (RE_NEWLINE_ALT << 1)
 121
 122 /* If this bit is set, (...) defines a group, and \( and \) are literals.
 123    If not set, \(...\) defines a group, and ( and ) are literals.  */
 124 #define RE_NO_BK_PARENS (RE_NO_BK_BRACES << 1)
 125
 126 /* If this bit is set, then \<digit> matches <digit>.
 127    If not set, then \<digit> is a back-reference.  */
 128 #define RE_NO_BK_REFS (RE_NO_BK_PARENS << 1)
 129
 130 /* If this bit is set, then | is an alternation operator, and \| is literal.
 131    If not set, then \| is an alternation operator, and | is literal.  */
 132 #define RE_NO_BK_VBAR (RE_NO_BK_REFS << 1)
 133
 134 /* If this bit is set, then an ending range point collating higher
 135      than the starting range point, as in [z-a], is invalid.
 136    If not set, then when ending range point collates higher than the
 137      starting range point, the range is ignored.  */
 138 #define RE_NO_EMPTY_RANGES (RE_NO_BK_VBAR << 1)
 139
 140 /* If this bit is set, then an unmatched ) is ordinary.
 141    If not set, then an unmatched ) is invalid.  */
 142 #define RE_UNMATCHED_RIGHT_PAREN_ORD (RE_NO_EMPTY_RANGES << 1)
 143
 144 /* If this bit is set, succeed as soon as we match the whole pattern,
 145    without further backtracking.  */
 146 #define RE_NO_POSIX_BACKTRACKING (RE_UNMATCHED_RIGHT_PAREN_ORD << 1)
 147
 148 /* If this bit is set, do not process the GNU regex operators.
 149    If not set, then the GNU regex operators are recognized. */
 150 #define RE_NO_GNU_OPS (RE_NO_POSIX_BACKTRACKING << 1)
 151
 152 /* If this bit is set, then *?, +? and ?? match non greedily. */
 153 #define RE_FRUGAL (RE_NO_GNU_OPS << 1)
 154
 155 /* If this bit is set, then (?:...) is treated as a shy group.  */
 156 #define RE_SHY_GROUPS (RE_FRUGAL << 1)
 157
 158 /* If this bit is set, ^ and $ only match at beg/end of buffer.  */
 159 #define RE_NO_NEWLINE_ANCHOR (RE_SHY_GROUPS << 1)
 160
 161 /* If this bit is set, turn on internal regex debugging.
 162    If not set, and debugging was on, turn it off.
 163    This only works if regex.c is compiled -DDEBUG.
 164    We define this bit always, so that all that's needed to turn on
 165    debugging is to recompile regex.c; the calling code can always have
 166    this bit set, and it won't affect anything in the normal case. */
 167 #define RE_DEBUG (RE_NO_NEWLINE_ANCHOR << 1)
 168
 169 /* This global variable defines the particular regexp syntax to use (for
 170    some interfaces).  When a regexp is compiled, the syntax used is
 171    stored in the pattern buffer, so changing this does not affect
 172    already-compiled regexps.  */
 173 /* extern reg_syntax_t re_syntax_options; */
 174
 175 #ifdef emacs
 176 # include "lisp.h"
 177 /* In Emacs, this is the string or buffer in which we are matching.
 178    It is used for looking up syntax properties.
 179
 180    If the value is a Lisp string object, we are matching text in that
 181    string; if it's nil, we are matching text in the current buffer; if
 182    it's t, we are matching text in a C string.
 183
 184    This is defined as a macro in thread.h, which see.  */
 185 /* extern Lisp_Object re_match_object; */
 186 #endif
 187
 188 /* Roughly the maximum number of failure points on the stack.  */
 189 extern size_t emacs_re_max_failures;
 190
 191 #ifdef emacs
 192 /* Amount of memory that we can safely stack allocate.  */
 193 extern ptrdiff_t emacs_re_safe_alloca;
 194 #endif
 195
 196 \f
 197 /* Define combinations of the above bits for the standard possibilities.
 198    (The [[[ comments delimit what gets put into the Texinfo file, so
 199    don't delete them!)  */
 200 /* [[[begin syntaxes]]] */
 201 #define RE_SYNTAX_EMACS                                                 \
 202   (RE_CHAR_CLASSES | RE_INTERVALS | RE_SHY_GROUPS | RE_FRUGAL)
 203
 204 #define RE_SYNTAX_AWK                                                   \
 205   (RE_BACKSLASH_ESCAPE_IN_LISTS   | RE_DOT_NOT_NULL                     \
 206    | RE_NO_BK_PARENS              | RE_NO_BK_REFS                       \
 207    | RE_NO_BK_VBAR                | RE_NO_EMPTY_RANGES                  \
 208    | RE_DOT_NEWLINE               | RE_CONTEXT_INDEP_ANCHORS            \
 209    | RE_UNMATCHED_RIGHT_PAREN_ORD | RE_NO_GNU_OPS)
 210
 211 #define RE_SYNTAX_GNU_AWK                                               \
 212   ((RE_SYNTAX_POSIX_EXTENDED | RE_BACKSLASH_ESCAPE_IN_LISTS | RE_DEBUG) \
 213    & ~(RE_DOT_NOT_NULL | RE_INTERVALS | RE_CONTEXT_INDEP_OPS))
 214
 215 #define RE_SYNTAX_POSIX_AWK                                             \
 216   (RE_SYNTAX_POSIX_EXTENDED | RE_BACKSLASH_ESCAPE_IN_LISTS              \
 217    | RE_INTERVALS           | RE_NO_GNU_OPS)
 218
 219 #define RE_SYNTAX_GREP                                                  \
 220   (RE_BK_PLUS_QM              | RE_CHAR_CLASSES                         \
 221    | RE_HAT_LISTS_NOT_NEWLINE | RE_INTERVALS                            \
 222    | RE_NEWLINE_ALT)
 223
 224 #define RE_SYNTAX_EGREP                                                 \
 225   (RE_CHAR_CLASSES        | RE_CONTEXT_INDEP_ANCHORS                    \
 226    | RE_CONTEXT_INDEP_OPS | RE_HAT_LISTS_NOT_NEWLINE                    \
 227    | RE_NEWLINE_ALT       | RE_NO_BK_PARENS                             \
 228    | RE_NO_BK_VBAR)
 229
 230 #define RE_SYNTAX_POSIX_EGREP                                           \
 231   (RE_SYNTAX_EGREP | RE_INTERVALS | RE_NO_BK_BRACES)
 232
 233 /* P1003.2/D11.2, section 4.20.7.1, lines 5078ff.  */
 234 #define RE_SYNTAX_ED RE_SYNTAX_POSIX_BASIC
 235
 236 #define RE_SYNTAX_SED RE_SYNTAX_POSIX_BASIC
 237
 238 /* Syntax bits common to both basic and extended POSIX regex syntax.  */
 239 #define _RE_SYNTAX_POSIX_COMMON                                         \
 240   (RE_CHAR_CLASSES | RE_DOT_NEWLINE      | RE_DOT_NOT_NULL              \
 241    | RE_INTERVALS  | RE_NO_EMPTY_RANGES)
 242
 243 #define RE_SYNTAX_POSIX_BASIC                                           \
 244   (_RE_SYNTAX_POSIX_COMMON | RE_BK_PLUS_QM)
 245
 246 /* Differs from ..._POSIX_BASIC only in that RE_BK_PLUS_QM becomes
 247    RE_LIMITED_OPS, i.e., \? \+ \| are not recognized.  Actually, this
 248    isn't minimal, since other operators, such as \`, aren't disabled.  */
 249 #define RE_SYNTAX_POSIX_MINIMAL_BASIC                                   \
 250   (_RE_SYNTAX_POSIX_COMMON | RE_LIMITED_OPS)
 251
 252 #define RE_SYNTAX_POSIX_EXTENDED                                        \
 253   (_RE_SYNTAX_POSIX_COMMON  | RE_CONTEXT_INDEP_ANCHORS                  \
 254    | RE_CONTEXT_INDEP_OPS   | RE_NO_BK_BRACES                           \
 255    | RE_NO_BK_PARENS        | RE_NO_BK_VBAR                             \
 256    | RE_CONTEXT_INVALID_OPS | RE_UNMATCHED_RIGHT_PAREN_ORD)
 257
 258 /* Differs from ..._POSIX_EXTENDED in that RE_CONTEXT_INDEP_OPS is
 259    removed and RE_NO_BK_REFS is added.  */
 260 #define RE_SYNTAX_POSIX_MINIMAL_EXTENDED                                \
 261   (_RE_SYNTAX_POSIX_COMMON  | RE_CONTEXT_INDEP_ANCHORS                  \
 262    | RE_CONTEXT_INVALID_OPS | RE_NO_BK_BRACES                           \
 263    | RE_NO_BK_PARENS        | RE_NO_BK_REFS                             \
 264    | RE_NO_BK_VBAR          | RE_UNMATCHED_RIGHT_PAREN_ORD)
 265 /* [[[end syntaxes]]] */
 266 \f
 267 /* Maximum number of duplicates an interval can allow.  Some systems
 268    (erroneously) define this in other header files, but we want our
 269    value, so remove any previous define.  */
 270 #ifdef RE_DUP_MAX
 271 # undef RE_DUP_MAX
 272 #endif
 273 /* If sizeof(int) == 2, then ((1 << 15) - 1) overflows.  */
 274 #define RE_DUP_MAX (0x7fff)
 275
 276
 277 /* POSIX `cflags' bits (i.e., information for `regcomp').  */
 278
 279 /* If this bit is set, then use extended regular expression syntax.
 280    If not set, then use basic regular expression syntax.  */
 281 #define REG_EXTENDED 1
 282
 283 /* If this bit is set, then ignore case when matching.
 284    If not set, then case is significant.  */
 285 #define REG_ICASE (REG_EXTENDED << 1)
 286
 287 /* If this bit is set, then anchors do not match at newline
 288      characters in the string.
 289    If not set, then anchors do match at newlines.  */
 290 #define REG_NEWLINE (REG_ICASE << 1)
 291
 292 /* If this bit is set, then report only success or fail in regexec.
 293    If not set, then returns differ between not matching and errors.  */
 294 #define REG_NOSUB (REG_NEWLINE << 1)
 295
 296
 297 /* POSIX `eflags' bits (i.e., information for regexec).  */
 298
 299 /* If this bit is set, then the beginning-of-line operator doesn't match
 300      the beginning of the string (presumably because it's not the
 301      beginning of a line).
 302    If not set, then the beginning-of-line operator does match the
 303      beginning of the string.  */
 304 #define REG_NOTBOL 1
 305
 306 /* Like REG_NOTBOL, except for the end-of-line.  */
 307 #define REG_NOTEOL (1 << 1)
 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,         /* Not implemented.  */
 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_ERANGEX           /* Range striding over charsets.  */
 341 } reg_errcode_t;
 342 \f
 343 /* This data structure represents a compiled pattern.  Before calling
 344    the pattern compiler, the fields `buffer', `allocated', `fastmap',
 345    `translate', and `no_sub' can be set.  After the pattern has been
 346    compiled, the `re_nsub' field is available.  All other fields are
 347    private to the regex routines.  */
 348
 349 #ifndef RE_TRANSLATE_TYPE
 350 # define RE_TRANSLATE_TYPE char *
 351 #endif
 352
 353 struct re_pattern_buffer
 354 {
 355 /* [[[begin pattern_buffer]]] */
 356         /* Space that holds the compiled pattern.  It is declared as
 357           `unsigned char *' because its elements are
 358            sometimes used as array indexes.  */
 359   unsigned char *buffer;
 360
 361         /* Number of bytes to which `buffer' points.  */
 362   size_t allocated;
 363
 364         /* Number of bytes actually used in `buffer'.  */
 365   size_t used;
 366
 367 #ifndef emacs
 368         /* Syntax setting with which the pattern was compiled.  */
 369   reg_syntax_t syntax;
 370 #endif
 371         /* Pointer to a fastmap, if any, otherwise zero.  re_search uses
 372            the fastmap, if there is one, to skip over impossible
 373            starting points for matches.  */
 374   char *fastmap;
 375
 376         /* Either a translate table to apply to all characters before
 377            comparing them, or zero for no translation.  The translation
 378            is applied to a pattern when it is compiled and to a string
 379            when it is matched.  */
 380   RE_TRANSLATE_TYPE translate;
 381
 382         /* Number of subexpressions found by the compiler.  */
 383   size_t re_nsub;
 384
 385         /* Zero if this pattern cannot match the empty string, one else.
 386            Well, in truth it's used only in `re_search_2', to see
 387            whether or not we should use the fastmap, so we don't set
 388            this absolutely perfectly; see `re_compile_fastmap'.  */
 389   unsigned can_be_null : 1;
 390
 391         /* If REGS_UNALLOCATED, allocate space in the `regs' structure
 392              for `max (RE_NREGS, re_nsub + 1)' groups.
 393            If REGS_REALLOCATE, reallocate space if necessary.
 394            If REGS_FIXED, use what's there.  */
 395 #define REGS_UNALLOCATED 0
 396 #define REGS_REALLOCATE 1
 397 #define REGS_FIXED 2
 398   unsigned regs_allocated : 2;
 399
 400         /* Set to zero when `regex_compile' compiles a pattern; set to one
 401            by `re_compile_fastmap' if it updates the fastmap.  */
 402   unsigned fastmap_accurate : 1;
 403
 404         /* If set, `re_match_2' does not return information about
 405            subexpressions.  */
 406   unsigned no_sub : 1;
 407
 408         /* If set, a beginning-of-line anchor doesn't match at the
 409            beginning of the string.  */
 410   unsigned not_bol : 1;
 411
 412         /* Similarly for an end-of-line anchor.  */
 413   unsigned not_eol : 1;
 414
 415   /* If true, the compilation of the pattern had to look up the syntax table,
 416      so the compiled pattern is only valid for the current syntax table.  */
 417   unsigned used_syntax : 1;
 418
 419 #ifdef emacs
 420   /* If true, multi-byte form in the regexp pattern should be
 421      recognized as a multibyte character.  */
 422   unsigned multibyte : 1;
 423
 424   /* If true, multi-byte form in the target of match should be
 425      recognized as a multibyte character.  */
 426   unsigned target_multibyte : 1;
 427
 428   /* Charset of unibyte characters at compiling time. */
 429   int charset_unibyte;
 430 #endif
 431
 432 /* [[[end pattern_buffer]]] */
 433 };
 434
 435 typedef struct re_pattern_buffer regex_t;
 436 \f
 437 /* POSIX 1003.1-2008 requires that regoff_t be at least as wide as
 438    ptrdiff_t and ssize_t.  We don't know of any hosts where ptrdiff_t
 439    is wider than ssize_t, so ssize_t is safe.  ptrdiff_t is not
 440    necessarily visible here, so use ssize_t.  */
 441 typedef ssize_t regoff_t;
 442
 443
 444 /* This is the structure we store register match data in.  See
 445    regex.texinfo for a full description of what registers match.  */
 446 struct re_registers
 447 {
 448   unsigned num_regs;
 449   regoff_t *start;
 450   regoff_t *end;
 451 };
 452
 453
 454 /* If `regs_allocated' is REGS_UNALLOCATED in the pattern buffer,
 455    `re_match_2' returns information about at least this many registers
 456    the first time a `regs' structure is passed.  */
 457 #ifndef RE_NREGS
 458 # define RE_NREGS 30
 459 #endif
 460
 461
 462 /* POSIX specification for registers.  Aside from the different names than
 463    `re_registers', POSIX uses an array of structures, instead of a
 464    structure of arrays.  */
 465 typedef struct
 466 {
 467   regoff_t rm_so;  /* Byte offset from string's start to substring's start.  */
 468   regoff_t rm_eo;  /* Byte offset from string's start to substring's end.  */
 469 } regmatch_t;
 470 \f
 471 /* Declarations for routines.  */
 472
 473 #ifndef emacs
 474
 475 /* Sets the current default syntax to SYNTAX, and return the old syntax.
 476    You can also simply assign to the `re_syntax_options' variable.  */
 477 extern reg_syntax_t re_set_syntax (reg_syntax_t __syntax);
 478
 479 #endif
 480
 481 /* Compile the regular expression PATTERN, with length LENGTH
 482    and syntax given by the global `re_syntax_options', into the buffer
 483    BUFFER.  Return NULL if successful, and an error string if not.  */
 484 extern const char *re_compile_pattern (const char *__pattern, size_t __length,
 485 #ifdef emacs
 486                                        bool posix_backtracking,
 487                                        const char *whitespace_regexp,
 488 #endif
 489                                        struct re_pattern_buffer *__buffer);
 490
 491
 492 /* Compile a fastmap for the compiled pattern in BUFFER; used to
 493    accelerate searches.  Return 0 if successful and -2 if was an
 494    internal error.  */
 495 extern int re_compile_fastmap (struct re_pattern_buffer *__buffer);
 496
 497
 498 /* Search in the string STRING (with length LENGTH) for the pattern
 499    compiled into BUFFER.  Start searching at position START, for RANGE
 500    characters.  Return the starting position of the match, -1 for no
 501    match, or -2 for an internal error.  Also return register
 502    information in REGS (if REGS and BUFFER->no_sub are nonzero).  */
 503 extern regoff_t re_search (struct re_pattern_buffer *__buffer,
 504                            const char *__string, size_t __length,
 505                            ssize_t __start, ssize_t __range,
 506                            struct re_registers *__regs);
 507
 508
 509 /* Like `re_search', but search in the concatenation of STRING1 and
 510    STRING2.  Also, stop searching at index START + STOP.  */
 511 extern regoff_t re_search_2 (struct re_pattern_buffer *__buffer,
 512                              const char *__string1, size_t __length1,
 513                              const char *__string2, size_t __length2,
 514                              ssize_t __start, ssize_t __range,
 515                              struct re_registers *__regs,
 516                              ssize_t __stop);
 517
 518
 519 /* Like `re_search', but return how many characters in STRING the regexp
 520    in BUFFER matched, starting at position START.  */
 521 extern regoff_t re_match (struct re_pattern_buffer *__buffer,
 522                           const char *__string, size_t __length,
 523                           ssize_t __start, struct re_registers *__regs);
 524
 525
 526 /* Relates to `re_match' as `re_search_2' relates to `re_search'.  */
 527 extern regoff_t re_match_2 (struct re_pattern_buffer *__buffer,
 528                             const char *__string1, size_t __length1,
 529                             const char *__string2, size_t __length2,
 530                             ssize_t __start, struct re_registers *__regs,
 531                             ssize_t __stop);
 532
 533
 534 /* Set REGS to hold NUM_REGS registers, storing them in STARTS and
 535    ENDS.  Subsequent matches using BUFFER and REGS will use this memory
 536    for recording register information.  STARTS and ENDS must be
 537    allocated with malloc, and must each be at least `NUM_REGS * sizeof
 538    (regoff_t)' bytes long.
 539
 540    If NUM_REGS == 0, then subsequent matches should allocate their own
 541    register data.
 542
 543    Unless this function is called, the first search or match using
 544    PATTERN_BUFFER will allocate its own register data, without
 545    freeing the old data.  */
 546 extern void re_set_registers (struct re_pattern_buffer *__buffer,
 547                               struct re_registers *__regs,
 548                               unsigned __num_regs,
 549                               regoff_t *__starts, regoff_t *__ends);
 550
 551 #if defined _REGEX_RE_COMP || defined _LIBC
 552 # ifndef _CRAY
 553 /* 4.2 bsd compatibility.  */
 554 extern char *re_comp (const char *);
 555 extern int re_exec (const char *);
 556 # endif
 557 #endif
 558
 559 /* GCC 2.95 and later have "__restrict"; C99 compilers have
 560    "restrict", and "configure" may have defined "restrict".
 561    Other compilers use __restrict, __restrict__, and _Restrict, and
 562    'configure' might #define 'restrict' to those words, so pick a
 563    different name.  */
 564 #ifndef _Restrict_
 565 # if 199901L <= __STDC_VERSION__
 566 #  define _Restrict_ restrict
 567 # elif 2 < __GNUC__ || (2 == __GNUC__ && 95 <= __GNUC_MINOR__)
 568 #  define _Restrict_ __restrict
 569 # else
 570 #  define _Restrict_
 571 # endif
 572 #endif
 573 /* gcc 3.1 and up support the [restrict] syntax.  Don't trust
 574    sys/cdefs.h's definition of __restrict_arr, though, as it
 575    mishandles gcc -ansi -pedantic.  */
 576 #ifndef _Restrict_arr_
 577 # if ((199901L <= __STDC_VERSION__                                      \
 578        || ((3 < __GNUC__ || (3 == __GNUC__ && 1 <= __GNUC_MINOR__))     \
 579            && !defined __STRICT_ANSI__))                                        \
 580       && !defined __GNUG__)
 581 #  define _Restrict_arr_ _Restrict_
 582 # else
 583 #  define _Restrict_arr_
 584 # endif
 585 #endif
 586
 587 /* POSIX compatibility.  */
 588 extern reg_errcode_t regcomp (regex_t *_Restrict_ __preg,
 589                               const char *_Restrict_ __pattern,
 590                               int __cflags);
 591
 592 extern reg_errcode_t regexec (const regex_t *_Restrict_ __preg,
 593                               const char *_Restrict_ __string, size_t __nmatch,
 594                               regmatch_t __pmatch[_Restrict_arr_],
 595                               int __eflags);
 596
 597 extern size_t regerror (int __errcode, const regex_t * __preg,
 598                         char *__errbuf, size_t __errbuf_size);
 599
 600 extern void regfree (regex_t *__preg);
 601
 602
 603 #ifdef __cplusplus
 604 }
 605 #endif  /* C++ */
 606
 607 /* For platform which support the ISO C amendment 1 functionality we
 608    support user defined character classes.  */
 609 #if WIDE_CHAR_SUPPORT
 610 /* Solaris 2.5 has a bug: <wchar.h> must be included before <wctype.h>.  */
 611 # include <wchar.h>
 612 # include <wctype.h>
 613
 614 typedef wctype_t re_wctype_t;
 615 typedef wchar_t re_wchar_t;
 616 # define re_wctype wctype
 617 # define re_iswctype iswctype
 618 # define re_wctype_to_bit(cc) 0
 619 #else
 620 # ifndef emacs
 621 #  define btowc(c) c
 622 # endif
 623
 624 /* Character classes.  */
 625 typedef enum { RECC_ERROR = 0,
 626                RECC_ALNUM, RECC_ALPHA, RECC_WORD,
 627                RECC_GRAPH, RECC_PRINT,
 628                RECC_LOWER, RECC_UPPER,
 629                RECC_PUNCT, RECC_CNTRL,
 630                RECC_DIGIT, RECC_XDIGIT,
 631                RECC_BLANK, RECC_SPACE,
 632                RECC_MULTIBYTE, RECC_NONASCII,
 633                RECC_ASCII, RECC_UNIBYTE
 634 } re_wctype_t;
 635
 636 extern char re_iswctype (int ch,    re_wctype_t cc);
 637 extern re_wctype_t re_wctype_parse (const unsigned char **strp, unsigned limit);
 638
 639 typedef int re_wchar_t;
 640
 641 #endif /* not WIDE_CHAR_SUPPORT */
 642
 643 #endif /* regex.h */
 644 \f