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