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 are matching.
 179    It is used for looking up syntax properties.
 180
 181    If the value is a Lisp string object, we are matching text in that
 182    string; if it's nil, we are matching text in the current buffer; if
 183    it's t, we are matching text in a C string.  */
 184 extern Lisp_Object re_match_object;
 185 #endif
 186
 187 /* Roughly the maximum number of failure points on the stack.  */
 188 extern size_t re_max_failures;
 189
 190 \f
 191 /* Define combinations of the above bits for the standard possibilities.
 192    (The [[[ comments delimit what gets put into the Texinfo file, so
 193    don't delete them!)  */
 194 /* [[[begin syntaxes]]] */
 195 #define RE_SYNTAX_EMACS                                                 \
 196   (RE_CHAR_CLASSES | RE_INTERVALS | RE_SHY_GROUPS | RE_FRUGAL)
 197
 198 #define RE_SYNTAX_AWK                                                   \
 199   (RE_BACKSLASH_ESCAPE_IN_LISTS   | RE_DOT_NOT_NULL                     \
 200    | RE_NO_BK_PARENS              | RE_NO_BK_REFS                       \
 201    | RE_NO_BK_VBAR                | RE_NO_EMPTY_RANGES                  \
 202    | RE_DOT_NEWLINE               | RE_CONTEXT_INDEP_ANCHORS            \
 203    | RE_UNMATCHED_RIGHT_PAREN_ORD | RE_NO_GNU_OPS)
 204
 205 #define RE_SYNTAX_GNU_AWK                                               \
 206   ((RE_SYNTAX_POSIX_EXTENDED | RE_BACKSLASH_ESCAPE_IN_LISTS | RE_DEBUG) \
 207    & ~(RE_DOT_NOT_NULL | RE_INTERVALS | RE_CONTEXT_INDEP_OPS))
 208
 209 #define RE_SYNTAX_POSIX_AWK                                             \
 210   (RE_SYNTAX_POSIX_EXTENDED | RE_BACKSLASH_ESCAPE_IN_LISTS              \
 211    | RE_INTERVALS           | RE_NO_GNU_OPS)
 212
 213 #define RE_SYNTAX_GREP                                                  \
 214   (RE_BK_PLUS_QM              | RE_CHAR_CLASSES                         \
 215    | RE_HAT_LISTS_NOT_NEWLINE | RE_INTERVALS                            \
 216    | RE_NEWLINE_ALT)
 217
 218 #define RE_SYNTAX_EGREP                                                 \
 219   (RE_CHAR_CLASSES        | RE_CONTEXT_INDEP_ANCHORS                    \
 220    | RE_CONTEXT_INDEP_OPS | RE_HAT_LISTS_NOT_NEWLINE                    \
 221    | RE_NEWLINE_ALT       | RE_NO_BK_PARENS                             \
 222    | RE_NO_BK_VBAR)
 223
 224 #define RE_SYNTAX_POSIX_EGREP                                           \
 225   (RE_SYNTAX_EGREP | RE_INTERVALS | RE_NO_BK_BRACES)
 226
 227 /* P1003.2/D11.2, section 4.20.7.1, lines 5078ff.  */
 228 #define RE_SYNTAX_ED RE_SYNTAX_POSIX_BASIC
 229
 230 #define RE_SYNTAX_SED RE_SYNTAX_POSIX_BASIC
 231
 232 /* Syntax bits common to both basic and extended POSIX regex syntax.  */
 233 #define _RE_SYNTAX_POSIX_COMMON                                         \
 234   (RE_CHAR_CLASSES | RE_DOT_NEWLINE      | RE_DOT_NOT_NULL              \
 235    | RE_INTERVALS  | RE_NO_EMPTY_RANGES)
 236
 237 #define RE_SYNTAX_POSIX_BASIC                                           \
 238   (_RE_SYNTAX_POSIX_COMMON | RE_BK_PLUS_QM)
 239
 240 /* Differs from ..._POSIX_BASIC only in that RE_BK_PLUS_QM becomes
 241    RE_LIMITED_OPS, i.e., \? \+ \| are not recognized.  Actually, this
 242    isn't minimal, since other operators, such as \`, aren't disabled.  */
 243 #define RE_SYNTAX_POSIX_MINIMAL_BASIC                                   \
 244   (_RE_SYNTAX_POSIX_COMMON | RE_LIMITED_OPS)
 245
 246 #define RE_SYNTAX_POSIX_EXTENDED                                        \
 247   (_RE_SYNTAX_POSIX_COMMON  | RE_CONTEXT_INDEP_ANCHORS                  \
 248    | RE_CONTEXT_INDEP_OPS   | RE_NO_BK_BRACES                           \
 249    | RE_NO_BK_PARENS        | RE_NO_BK_VBAR                             \
 250    | RE_CONTEXT_INVALID_OPS | RE_UNMATCHED_RIGHT_PAREN_ORD)
 251
 252 /* Differs from ..._POSIX_EXTENDED in that RE_CONTEXT_INDEP_OPS is
 253    removed and RE_NO_BK_REFS is added.  */
 254 #define RE_SYNTAX_POSIX_MINIMAL_EXTENDED                                \
 255   (_RE_SYNTAX_POSIX_COMMON  | RE_CONTEXT_INDEP_ANCHORS                  \
 256    | RE_CONTEXT_INVALID_OPS | RE_NO_BK_BRACES                           \
 257    | RE_NO_BK_PARENS        | RE_NO_BK_REFS                             \
 258    | RE_NO_BK_VBAR          | RE_UNMATCHED_RIGHT_PAREN_ORD)
 259 /* [[[end syntaxes]]] */
 260 \f
 261 /* Maximum number of duplicates an interval can allow.  Some systems
 262    (erroneously) define this in other header files, but we want our
 263    value, so remove any previous define.  */
 264 #ifdef RE_DUP_MAX
 265 # undef RE_DUP_MAX
 266 #endif
 267 /* If sizeof(int) == 2, then ((1 << 15) - 1) overflows.  */
 268 #define RE_DUP_MAX (0x7fff)
 269
 270
 271 /* POSIX `cflags' bits (i.e., information for `regcomp').  */
 272
 273 /* If this bit is set, then use extended regular expression syntax.
 274    If not set, then use basic regular expression syntax.  */
 275 #define REG_EXTENDED 1
 276
 277 /* If this bit is set, then ignore case when matching.
 278    If not set, then case is significant.  */
 279 #define REG_ICASE (REG_EXTENDED << 1)
 280
 281 /* If this bit is set, then anchors do not match at newline
 282      characters in the string.
 283    If not set, then anchors do match at newlines.  */
 284 #define REG_NEWLINE (REG_ICASE << 1)
 285
 286 /* If this bit is set, then report only success or fail in regexec.
 287    If not set, then returns differ between not matching and errors.  */
 288 #define REG_NOSUB (REG_NEWLINE << 1)
 289
 290
 291 /* POSIX `eflags' bits (i.e., information for regexec).  */
 292
 293 /* If this bit is set, then the beginning-of-line operator doesn't match
 294      the beginning of the string (presumably because it's not the
 295      beginning of a line).
 296    If not set, then the beginning-of-line operator does match the
 297      beginning of the string.  */
 298 #define REG_NOTBOL 1
 299
 300 /* Like REG_NOTBOL, except for the end-of-line.  */
 301 #define REG_NOTEOL (1 << 1)
 302
 303
 304 /* If any error codes are removed, changed, or added, update the
 305    `re_error_msg' table in regex.c.  */
 306 typedef enum
 307 {
 308 #ifdef _XOPEN_SOURCE
 309   REG_ENOSYS = -1,      /* This will never happen for this implementation.  */
 310 #endif
 311
 312   REG_NOERROR = 0,      /* Success.  */
 313   REG_NOMATCH,          /* Didn't find a match (for regexec).  */
 314
 315   /* POSIX regcomp return error codes.  (In the order listed in the
 316      standard.)  */
 317   REG_BADPAT,           /* Invalid pattern.  */
 318   REG_ECOLLATE,         /* Not implemented.  */
 319   REG_ECTYPE,           /* Invalid character class name.  */
 320   REG_EESCAPE,          /* Trailing backslash.  */
 321   REG_ESUBREG,          /* Invalid back reference.  */
 322   REG_EBRACK,           /* Unmatched left bracket.  */
 323   REG_EPAREN,           /* Parenthesis imbalance.  */
 324   REG_EBRACE,           /* Unmatched \{.  */
 325   REG_BADBR,            /* Invalid contents of \{\}.  */
 326   REG_ERANGE,           /* Invalid range end.  */
 327   REG_ESPACE,           /* Ran out of memory.  */
 328   REG_BADRPT,           /* No preceding re for repetition op.  */
 329
 330   /* Error codes we've added.  */
 331   REG_EEND,             /* Premature end.  */
 332   REG_ESIZE,            /* Compiled pattern bigger than 2^16 bytes.  */
 333   REG_ERPAREN,          /* Unmatched ) or \); not returned from regcomp.  */
 334   REG_ERANGEX           /* Range striding over charsets.  */
 335 } reg_errcode_t;
 336 \f
 337 /* This data structure represents a compiled pattern.  Before calling
 338    the pattern compiler, the fields `buffer', `allocated', `fastmap',
 339    `translate', and `no_sub' can be set.  After the pattern has been
 340    compiled, the `re_nsub' field is available.  All other fields are
 341    private to the regex routines.  */
 342
 343 #ifndef RE_TRANSLATE_TYPE
 344 # define RE_TRANSLATE_TYPE char *
 345 #endif
 346
 347 struct re_pattern_buffer
 348 {
 349 /* [[[begin pattern_buffer]]] */
 350         /* Space that holds the compiled pattern.  It is declared as
 351           `unsigned char *' because its elements are
 352            sometimes used as array indexes.  */
 353   unsigned char *buffer;
 354
 355         /* Number of bytes to which `buffer' points.  */
 356   size_t allocated;
 357
 358         /* Number of bytes actually used in `buffer'.  */
 359   size_t used;
 360
 361 #ifndef emacs
 362         /* Syntax setting with which the pattern was compiled.  */
 363   reg_syntax_t syntax;
 364 #endif
 365         /* Pointer to a fastmap, if any, otherwise zero.  re_search uses
 366            the fastmap, if there is one, to skip over impossible
 367            starting points for matches.  */
 368   char *fastmap;
 369
 370         /* Either a translate table to apply to all characters before
 371            comparing them, or zero for no translation.  The translation
 372            is applied to a pattern when it is compiled and to a string
 373            when it is matched.  */
 374   RE_TRANSLATE_TYPE translate;
 375
 376         /* Number of subexpressions found by the compiler.  */
 377   size_t re_nsub;
 378
 379         /* Zero if this pattern cannot match the empty string, one else.
 380            Well, in truth it's used only in `re_search_2', to see
 381            whether or not we should use the fastmap, so we don't set
 382            this absolutely perfectly; see `re_compile_fastmap'.  */
 383   unsigned can_be_null : 1;
 384
 385         /* If REGS_UNALLOCATED, allocate space in the `regs' structure
 386              for `max (RE_NREGS, re_nsub + 1)' groups.
 387            If REGS_REALLOCATE, reallocate space if necessary.
 388            If REGS_FIXED, use what's there.  */
 389 #define REGS_UNALLOCATED 0
 390 #define REGS_REALLOCATE 1
 391 #define REGS_FIXED 2
 392   unsigned regs_allocated : 2;
 393
 394         /* Set to zero when `regex_compile' compiles a pattern; set to one
 395            by `re_compile_fastmap' if it updates the fastmap.  */
 396   unsigned fastmap_accurate : 1;
 397
 398         /* If set, `re_match_2' does not return information about
 399            subexpressions.  */
 400   unsigned no_sub : 1;
 401
 402         /* If set, a beginning-of-line anchor doesn't match at the
 403            beginning of the string.  */
 404   unsigned not_bol : 1;
 405
 406         /* Similarly for an end-of-line anchor.  */
 407   unsigned not_eol : 1;
 408
 409   /* If true, the compilation of the pattern had to look up the syntax table,
 410      so the compiled pattern is only valid for the current syntax table.  */
 411   unsigned used_syntax : 1;
 412
 413 #ifdef emacs
 414   /* If true, multi-byte form in the regexp pattern should be
 415      recognized as a multibyte character.  */
 416   unsigned multibyte : 1;
 417
 418   /* If true, multi-byte form in the target of match should be
 419      recognized as a multibyte character.  */
 420   unsigned target_multibyte : 1;
 421
 422   /* Charset of unibyte characters at compiling time. */
 423   int charset_unibyte;
 424 #endif
 425
 426 /* [[[end pattern_buffer]]] */
 427 };
 428
 429 typedef struct re_pattern_buffer regex_t;
 430 \f
 431 /* Type for byte offsets within the string.  POSIX mandates this to be an int,
 432    but the Open Group has signaled its intention to change the requirement to
 433    be that regoff_t be at least as wide as ptrdiff_t and ssize_t.  Current
 434    gnulib sources also use ssize_t, and we need this for supporting buffers and
 435    strings > 2GB on 64-bit hosts.  */
 436 typedef ssize_t regoff_t;
 437
 438
 439 /* This is the structure we store register match data in.  See
 440    regex.texinfo for a full description of what registers match.  */
 441 struct re_registers
 442 {
 443   unsigned num_regs;
 444   regoff_t *start;
 445   regoff_t *end;
 446 };
 447
 448
 449 /* If `regs_allocated' is REGS_UNALLOCATED in the pattern buffer,
 450    `re_match_2' returns information about at least this many registers
 451    the first time a `regs' structure is passed.  */
 452 #ifndef RE_NREGS
 453 # define RE_NREGS 30
 454 #endif
 455
 456
 457 /* POSIX specification for registers.  Aside from the different names than
 458    `re_registers', POSIX uses an array of structures, instead of a
 459    structure of arrays.  */
 460 typedef struct
 461 {
 462   regoff_t rm_so;  /* Byte offset from string's start to substring's start.  */
 463   regoff_t rm_eo;  /* Byte offset from string's start to substring's end.  */
 464 } regmatch_t;
 465 \f
 466 /* Declarations for routines.  */
 467
 468 #ifndef emacs
 469
 470 /* Sets the current default syntax to SYNTAX, and return the old syntax.
 471    You can also simply assign to the `re_syntax_options' variable.  */
 472 extern reg_syntax_t re_set_syntax (reg_syntax_t __syntax);
 473
 474 #endif
 475
 476 /* Compile the regular expression PATTERN, with length LENGTH
 477    and syntax given by the global `re_syntax_options', into the buffer
 478    BUFFER.  Return NULL if successful, and an error string if not.  */
 479 extern const char *re_compile_pattern (const char *__pattern, size_t __length,
 480 #ifdef emacs
 481                                        bool posix_backtracking,
 482                                        const char *whitespace_regexp,
 483 #endif
 484                                        struct re_pattern_buffer *__buffer);
 485
 486
 487 /* Compile a fastmap for the compiled pattern in BUFFER; used to
 488    accelerate searches.  Return 0 if successful and -2 if was an
 489    internal error.  */
 490 extern int re_compile_fastmap (struct re_pattern_buffer *__buffer);
 491
 492
 493 /* Search in the string STRING (with length LENGTH) for the pattern
 494    compiled into BUFFER.  Start searching at position START, for RANGE
 495    characters.  Return the starting position of the match, -1 for no
 496    match, or -2 for an internal error.  Also return register
 497    information in REGS (if REGS and BUFFER->no_sub are nonzero).  */
 498 extern regoff_t re_search (struct re_pattern_buffer *__buffer,
 499                            const char *__string, size_t __length,
 500                            ssize_t __start, ssize_t __range,
 501                            struct re_registers *__regs);
 502
 503
 504 /* Like `re_search', but search in the concatenation of STRING1 and
 505    STRING2.  Also, stop searching at index START + STOP.  */
 506 extern regoff_t re_search_2 (struct re_pattern_buffer *__buffer,
 507                              const char *__string1, size_t __length1,
 508                              const char *__string2, size_t __length2,
 509                              ssize_t __start, ssize_t __range,
 510                              struct re_registers *__regs,
 511                              ssize_t __stop);
 512
 513
 514 /* Like `re_search', but return how many characters in STRING the regexp
 515    in BUFFER matched, starting at position START.  */
 516 extern regoff_t re_match (struct re_pattern_buffer *__buffer,
 517                           const char *__string, size_t __length,
 518                           ssize_t __start, struct re_registers *__regs);
 519
 520
 521 /* Relates to `re_match' as `re_search_2' relates to `re_search'.  */
 522 extern regoff_t re_match_2 (struct re_pattern_buffer *__buffer,
 523                             const char *__string1, size_t __length1,
 524                             const char *__string2, size_t __length2,
 525                             ssize_t __start, struct re_registers *__regs,
 526                             ssize_t __stop);
 527
 528
 529 /* Set REGS to hold NUM_REGS registers, storing them in STARTS and
 530    ENDS.  Subsequent matches using BUFFER and REGS will use this memory
 531    for recording register information.  STARTS and ENDS must be
 532    allocated with malloc, and must each be at least `NUM_REGS * sizeof
 533    (regoff_t)' bytes long.
 534
 535    If NUM_REGS == 0, then subsequent matches should allocate their own
 536    register data.
 537
 538    Unless this function is called, the first search or match using
 539    PATTERN_BUFFER will allocate its own register data, without
 540    freeing the old data.  */
 541 extern void re_set_registers (struct re_pattern_buffer *__buffer,
 542                               struct re_registers *__regs,
 543                               unsigned __num_regs,
 544                               regoff_t *__starts, regoff_t *__ends);
 545
 546 #if defined _REGEX_RE_COMP || defined _LIBC
 547 # ifndef _CRAY
 548 /* 4.2 bsd compatibility.  */
 549 extern char *re_comp (const char *);
 550 extern int re_exec (const char *);
 551 # endif
 552 #endif
 553
 554 /* GCC 2.95 and later have "__restrict"; C99 compilers have
 555    "restrict", and "configure" may have defined "restrict".
 556    Other compilers use __restrict, __restrict__, and _Restrict, and
 557    'configure' might #define 'restrict' to those words, so pick a
 558    different name.  */
 559 #ifndef _Restrict_
 560 # if 199901L <= __STDC_VERSION__
 561 #  define _Restrict_ restrict
 562 # elif 2 < __GNUC__ || (2 == __GNUC__ && 95 <= __GNUC_MINOR__)
 563 #  define _Restrict_ __restrict
 564 # else
 565 #  define _Restrict_
 566 # endif
 567 #endif
 568 /* gcc 3.1 and up support the [restrict] syntax.  Don't trust
 569    sys/cdefs.h's definition of __restrict_arr, though, as it
 570    mishandles gcc -ansi -pedantic.  */
 571 #ifndef _Restrict_arr_
 572 # if ((199901L <= __STDC_VERSION__                                      \
 573        || ((3 < __GNUC__ || (3 == __GNUC__ && 1 <= __GNUC_MINOR__))     \
 574            && !defined __STRICT_ANSI__))                                        \
 575       && !defined __GNUG__)
 576 #  define _Restrict_arr_ _Restrict_
 577 # else
 578 #  define _Restrict_arr_
 579 # endif
 580 #endif
 581
 582 /* POSIX compatibility.  */
 583 extern reg_errcode_t regcomp (regex_t *_Restrict_ __preg,
 584                               const char *_Restrict_ __pattern,
 585                               int __cflags);
 586
 587 extern reg_errcode_t regexec (const regex_t *_Restrict_ __preg,
 588                               const char *_Restrict_ __string, size_t __nmatch,
 589                               regmatch_t __pmatch[_Restrict_arr_],
 590                               int __eflags);
 591
 592 extern size_t regerror (int __errcode, const regex_t * __preg,
 593                         char *__errbuf, size_t __errbuf_size);
 594
 595 extern void regfree (regex_t *__preg);
 596
 597
 598 #ifdef __cplusplus
 599 }
 600 #endif  /* C++ */
 601
 602 /* For platform which support the ISO C amendment 1 functionality we
 603    support user defined character classes.  */
 604 #if WIDE_CHAR_SUPPORT
 605 /* Solaris 2.5 has a bug: <wchar.h> must be included before <wctype.h>.  */
 606 # include <wchar.h>
 607 # include <wctype.h>
 608
 609 typedef wctype_t re_wctype_t;
 610 typedef wchar_t re_wchar_t;
 611 # define re_wctype wctype
 612 # define re_iswctype iswctype
 613 # define re_wctype_to_bit(cc) 0
 614 #else
 615 # ifndef emacs
 616 #  define btowc(c) c
 617 # endif
 618
 619 /* Character classes.  */
 620 typedef enum { RECC_ERROR = 0,
 621                RECC_ALNUM, RECC_ALPHA, RECC_WORD,
 622                RECC_GRAPH, RECC_PRINT,
 623                RECC_LOWER, RECC_UPPER,
 624                RECC_PUNCT, RECC_CNTRL,
 625                RECC_DIGIT, RECC_XDIGIT,
 626                RECC_BLANK, RECC_SPACE,
 627                RECC_MULTIBYTE, RECC_NONASCII,
 628                RECC_ASCII, RECC_UNIBYTE
 629 } re_wctype_t;
 630
 631 extern char re_iswctype (int ch,    re_wctype_t cc);
 632 extern re_wctype_t re_wctype_parse (const unsigned char **strp, unsigned limit);
 633
 634 typedef int re_wchar_t;
 635
 636 #endif /* not WIDE_CHAR_SUPPORT */
 637
 638 #endif /* regex.h */
 639 \f