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, 2000-2018 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 <https://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 value is effectively another parameter to re_search_2 and
 185    re_match_2.  No calls into Lisp or thread switches are allowed
 186    before setting re_match_object and calling into the regex search
 187    and match functions.  These functions capture the current value of
 188    re_match_object into gl_state on entry.
 189
 190    TODO: once we get rid of the !emacs case in this code, turn into an
 191    actual function parameter.  */
 192 extern Lisp_Object re_match_object;
 193 #endif
 194
 195 /* Roughly the maximum number of failure points on the stack.  */
 196 extern size_t emacs_re_max_failures;
 197
 198 #ifdef emacs
 199 /* Amount of memory that we can safely stack allocate.  */
 200 extern ptrdiff_t emacs_re_safe_alloca;
 201 #endif
 202
 203 \f
 204 /* Define combinations of the above bits for the standard possibilities.
 205    (The [[[ comments delimit what gets put into the Texinfo file, so
 206    don't delete them!)  */
 207 /* [[[begin syntaxes]]] */
 208 #define RE_SYNTAX_EMACS                                                 \
 209   (RE_CHAR_CLASSES | RE_INTERVALS | RE_SHY_GROUPS | RE_FRUGAL)
 210
 211 #define RE_SYNTAX_AWK                                                   \
 212   (RE_BACKSLASH_ESCAPE_IN_LISTS   | RE_DOT_NOT_NULL                     \
 213    | RE_NO_BK_PARENS              | RE_NO_BK_REFS                       \
 214    | RE_NO_BK_VBAR                | RE_NO_EMPTY_RANGES                  \
 215    | RE_DOT_NEWLINE               | RE_CONTEXT_INDEP_ANCHORS            \
 216    | RE_UNMATCHED_RIGHT_PAREN_ORD | RE_NO_GNU_OPS)
 217
 218 #define RE_SYNTAX_GNU_AWK                                               \
 219   ((RE_SYNTAX_POSIX_EXTENDED | RE_BACKSLASH_ESCAPE_IN_LISTS | RE_DEBUG) \
 220    & ~(RE_DOT_NOT_NULL | RE_INTERVALS | RE_CONTEXT_INDEP_OPS))
 221
 222 #define RE_SYNTAX_POSIX_AWK                                             \
 223   (RE_SYNTAX_POSIX_EXTENDED | RE_BACKSLASH_ESCAPE_IN_LISTS              \
 224    | RE_INTERVALS           | RE_NO_GNU_OPS)
 225
 226 #define RE_SYNTAX_GREP                                                  \
 227   (RE_BK_PLUS_QM              | RE_CHAR_CLASSES                         \
 228    | RE_HAT_LISTS_NOT_NEWLINE | RE_INTERVALS                            \
 229    | RE_NEWLINE_ALT)
 230
 231 #define RE_SYNTAX_EGREP                                                 \
 232   (RE_CHAR_CLASSES        | RE_CONTEXT_INDEP_ANCHORS                    \
 233    | RE_CONTEXT_INDEP_OPS | RE_HAT_LISTS_NOT_NEWLINE                    \
 234    | RE_NEWLINE_ALT       | RE_NO_BK_PARENS                             \
 235    | RE_NO_BK_VBAR)
 236
 237 #define RE_SYNTAX_POSIX_EGREP                                           \
 238   (RE_SYNTAX_EGREP | RE_INTERVALS | RE_NO_BK_BRACES)
 239
 240 /* P1003.2/D11.2, section 4.20.7.1, lines 5078ff.  */
 241 #define RE_SYNTAX_ED RE_SYNTAX_POSIX_BASIC
 242
 243 #define RE_SYNTAX_SED RE_SYNTAX_POSIX_BASIC
 244
 245 /* Syntax bits common to both basic and extended POSIX regex syntax.  */
 246 #define _RE_SYNTAX_POSIX_COMMON                                         \
 247   (RE_CHAR_CLASSES | RE_DOT_NEWLINE      | RE_DOT_NOT_NULL              \
 248    | RE_INTERVALS  | RE_NO_EMPTY_RANGES)
 249
 250 #define RE_SYNTAX_POSIX_BASIC                                           \
 251   (_RE_SYNTAX_POSIX_COMMON | RE_BK_PLUS_QM)
 252
 253 /* Differs from ..._POSIX_BASIC only in that RE_BK_PLUS_QM becomes
 254    RE_LIMITED_OPS, i.e., \? \+ \| are not recognized.  Actually, this
 255    isn't minimal, since other operators, such as \`, aren't disabled.  */
 256 #define RE_SYNTAX_POSIX_MINIMAL_BASIC                                   \
 257   (_RE_SYNTAX_POSIX_COMMON | RE_LIMITED_OPS)
 258
 259 #define RE_SYNTAX_POSIX_EXTENDED                                        \
 260   (_RE_SYNTAX_POSIX_COMMON  | RE_CONTEXT_INDEP_ANCHORS                  \
 261    | RE_CONTEXT_INDEP_OPS   | RE_NO_BK_BRACES                           \
 262    | RE_NO_BK_PARENS        | RE_NO_BK_VBAR                             \
 263    | RE_CONTEXT_INVALID_OPS | RE_UNMATCHED_RIGHT_PAREN_ORD)
 264
 265 /* Differs from ..._POSIX_EXTENDED in that RE_CONTEXT_INDEP_OPS is
 266    removed and RE_NO_BK_REFS is added.  */
 267 #define RE_SYNTAX_POSIX_MINIMAL_EXTENDED                                \
 268   (_RE_SYNTAX_POSIX_COMMON  | RE_CONTEXT_INDEP_ANCHORS                  \
 269    | RE_CONTEXT_INVALID_OPS | RE_NO_BK_BRACES                           \
 270    | RE_NO_BK_PARENS        | RE_NO_BK_REFS                             \
 271    | RE_NO_BK_VBAR          | RE_UNMATCHED_RIGHT_PAREN_ORD)
 272 /* [[[end syntaxes]]] */
 273 \f
 274 /* Maximum number of duplicates an interval can allow.  Some systems
 275    (erroneously) define this in other header files, but we want our
 276    value, so remove any previous define.  */
 277 #ifdef RE_DUP_MAX
 278 # undef RE_DUP_MAX
 279 #endif
 280 /* Repeat counts are stored in opcodes as 2 byte integers.  This was
 281    previously limited to 7fff because the parsing code uses signed
 282    ints.  But Emacs only runs on 32 bit platforms anyway.  */
 283 #define RE_DUP_MAX (0xffff)
 284
 285
 286 /* POSIX `cflags' bits (i.e., information for `regcomp').  */
 287
 288 /* If this bit is set, then use extended regular expression syntax.
 289    If not set, then use basic regular expression syntax.  */
 290 #define REG_EXTENDED 1
 291
 292 /* If this bit is set, then ignore case when matching.
 293    If not set, then case is significant.  */
 294 #define REG_ICASE (REG_EXTENDED << 1)
 295
 296 /* If this bit is set, then anchors do not match at newline
 297      characters in the string.
 298    If not set, then anchors do match at newlines.  */
 299 #define REG_NEWLINE (REG_ICASE << 1)
 300
 301 /* If this bit is set, then report only success or fail in regexec.
 302    If not set, then returns differ between not matching and errors.  */
 303 #define REG_NOSUB (REG_NEWLINE << 1)
 304
 305
 306 /* POSIX `eflags' bits (i.e., information for regexec).  */
 307
 308 /* If this bit is set, then the beginning-of-line operator doesn't match
 309      the beginning of the string (presumably because it's not the
 310      beginning of a line).
 311    If not set, then the beginning-of-line operator does match the
 312      beginning of the string.  */
 313 #define REG_NOTBOL 1
 314
 315 /* Like REG_NOTBOL, except for the end-of-line.  */
 316 #define REG_NOTEOL (1 << 1)
 317
 318
 319 /* If any error codes are removed, changed, or added, update the
 320    `re_error_msg' table in regex.c.  */
 321 typedef enum
 322 {
 323 #ifdef _XOPEN_SOURCE
 324   REG_ENOSYS = -1,      /* This will never happen for this implementation.  */
 325 #endif
 326
 327   REG_NOERROR = 0,      /* Success.  */
 328   REG_NOMATCH,          /* Didn't find a match (for regexec).  */
 329
 330   /* POSIX regcomp return error codes.  (In the order listed in the
 331      standard.)  */
 332   REG_BADPAT,           /* Invalid pattern.  */
 333   REG_ECOLLATE,         /* Not implemented.  */
 334   REG_ECTYPE,           /* Invalid character class name.  */
 335   REG_EESCAPE,          /* Trailing backslash.  */
 336   REG_ESUBREG,          /* Invalid back reference.  */
 337   REG_EBRACK,           /* Unmatched left bracket.  */
 338   REG_EPAREN,           /* Parenthesis imbalance.  */
 339   REG_EBRACE,           /* Unmatched \{.  */
 340   REG_BADBR,            /* Invalid contents of \{\}.  */
 341   REG_ERANGE,           /* Invalid range end.  */
 342   REG_ESPACE,           /* Ran out of memory.  */
 343   REG_BADRPT,           /* No preceding re for repetition op.  */
 344
 345   /* Error codes we've added.  */
 346   REG_EEND,             /* Premature end.  */
 347   REG_ESIZE,            /* Compiled pattern bigger than 2^16 bytes.  */
 348   REG_ERPAREN,          /* Unmatched ) or \); not returned from regcomp.  */
 349   REG_ERANGEX,          /* Range striding over charsets.  */
 350   REG_ESIZEBR           /* n or m too big in \{n,m\} */
 351 } reg_errcode_t;
 352 \f
 353 /* This data structure represents a compiled pattern.  Before calling
 354    the pattern compiler, the fields `buffer', `allocated', `fastmap',
 355    `translate', and `no_sub' can be set.  After the pattern has been
 356    compiled, the `re_nsub' field is available.  All other fields are
 357    private to the regex routines.  */
 358
 359 #ifndef RE_TRANSLATE_TYPE
 360 # define RE_TRANSLATE_TYPE char *
 361 #endif
 362
 363 struct re_pattern_buffer
 364 {
 365 /* [[[begin pattern_buffer]]] */
 366         /* Space that holds the compiled pattern.  It is declared as
 367           `unsigned char *' because its elements are
 368            sometimes used as array indexes.  */
 369   unsigned char *buffer;
 370
 371         /* Number of bytes to which `buffer' points.  */
 372   size_t allocated;
 373
 374         /* Number of bytes actually used in `buffer'.  */
 375   size_t used;
 376
 377 #ifdef emacs
 378         /* Charset of unibyte characters at compiling time. */
 379   int charset_unibyte;
 380 #else
 381         /* Syntax setting with which the pattern was compiled.  */
 382   reg_syntax_t syntax;
 383 #endif
 384         /* Pointer to a fastmap, if any, otherwise zero.  re_search uses
 385            the fastmap, if there is one, to skip over impossible
 386            starting points for matches.  */
 387   char *fastmap;
 388
 389         /* Either a translate table to apply to all characters before
 390            comparing them, or zero for no translation.  The translation
 391            is applied to a pattern when it is compiled and to a string
 392            when it is matched.  */
 393   RE_TRANSLATE_TYPE translate;
 394
 395         /* Number of subexpressions found by the compiler.  */
 396   size_t re_nsub;
 397
 398         /* Zero if this pattern cannot match the empty string, one else.
 399            Well, in truth it's used only in `re_search_2', to see
 400            whether or not we should use the fastmap, so we don't set
 401            this absolutely perfectly; see `re_compile_fastmap'.  */
 402   unsigned can_be_null : 1;
 403
 404         /* If REGS_UNALLOCATED, allocate space in the `regs' structure
 405              for `max (RE_NREGS, re_nsub + 1)' groups.
 406            If REGS_REALLOCATE, reallocate space if necessary.
 407            If REGS_FIXED, use what's there.  */
 408 #define REGS_UNALLOCATED 0
 409 #define REGS_REALLOCATE 1
 410 #define REGS_FIXED 2
 411   unsigned regs_allocated : 2;
 412
 413         /* Set to zero when `regex_compile' compiles a pattern; set to one
 414            by `re_compile_fastmap' if it updates the fastmap.  */
 415   unsigned fastmap_accurate : 1;
 416
 417         /* If set, `re_match_2' does not return information about
 418            subexpressions.  */
 419   unsigned no_sub : 1;
 420
 421         /* If set, a beginning-of-line anchor doesn't match at the
 422            beginning of the string.  */
 423   unsigned not_bol : 1;
 424
 425         /* Similarly for an end-of-line anchor.  */
 426   unsigned not_eol : 1;
 427
 428   /* If true, the compilation of the pattern had to look up the syntax table,
 429      so the compiled pattern is only valid for the current syntax table.  */
 430   unsigned used_syntax : 1;
 431
 432 #ifdef emacs
 433   /* If true, multi-byte form in the regexp pattern should be
 434      recognized as a multibyte character.  */
 435   unsigned multibyte : 1;
 436
 437   /* If true, multi-byte form in the target of match should be
 438      recognized as a multibyte character.  */
 439   unsigned target_multibyte : 1;
 440 #endif
 441
 442 /* [[[end pattern_buffer]]] */
 443 };
 444
 445 typedef struct re_pattern_buffer regex_t;
 446 \f
 447 /* POSIX 1003.1-2008 requires that regoff_t be at least as wide as
 448    ptrdiff_t and ssize_t.  We don't know of any hosts where ptrdiff_t
 449    is wider than ssize_t, so ssize_t is safe.  ptrdiff_t is not
 450    necessarily visible here, so use ssize_t.  */
 451 typedef ssize_t regoff_t;
 452
 453
 454 /* This is the structure we store register match data in.  See
 455    regex.texinfo for a full description of what registers match.  */
 456 struct re_registers
 457 {
 458   unsigned num_regs;
 459   regoff_t *start;
 460   regoff_t *end;
 461 };
 462
 463
 464 /* If `regs_allocated' is REGS_UNALLOCATED in the pattern buffer,
 465    `re_match_2' returns information about at least this many registers
 466    the first time a `regs' structure is passed.  */
 467 #ifndef RE_NREGS
 468 # define RE_NREGS 30
 469 #endif
 470
 471
 472 /* POSIX specification for registers.  Aside from the different names than
 473    `re_registers', POSIX uses an array of structures, instead of a
 474    structure of arrays.  */
 475 typedef struct
 476 {
 477   regoff_t rm_so;  /* Byte offset from string's start to substring's start.  */
 478   regoff_t rm_eo;  /* Byte offset from string's start to substring's end.  */
 479 } regmatch_t;
 480 \f
 481 /* Declarations for routines.  */
 482
 483 #ifndef emacs
 484
 485 /* Sets the current default syntax to SYNTAX, and return the old syntax.
 486    You can also simply assign to the `re_syntax_options' variable.  */
 487 extern reg_syntax_t re_set_syntax (reg_syntax_t __syntax);
 488
 489 #endif
 490
 491 /* Compile the regular expression PATTERN, with length LENGTH
 492    and syntax given by the global `re_syntax_options', into the buffer
 493    BUFFER.  Return NULL if successful, and an error string if not.  */
 494 extern const char *re_compile_pattern (const char *__pattern, size_t __length,
 495 #ifdef emacs
 496                                        bool posix_backtracking,
 497                                        const char *whitespace_regexp,
 498 #endif
 499                                        struct re_pattern_buffer *__buffer);
 500
 501
 502 /* Compile a fastmap for the compiled pattern in BUFFER; used to
 503    accelerate searches.  Return 0 if successful and -2 if was an
 504    internal error.  */
 505 extern int re_compile_fastmap (struct re_pattern_buffer *__buffer);
 506
 507
 508 /* Search in the string STRING (with length LENGTH) for the pattern
 509    compiled into BUFFER.  Start searching at position START, for RANGE
 510    characters.  Return the starting position of the match, -1 for no
 511    match, or -2 for an internal error.  Also return register
 512    information in REGS (if REGS and BUFFER->no_sub are nonzero).  */
 513 extern regoff_t re_search (struct re_pattern_buffer *__buffer,
 514                            const char *__string, size_t __length,
 515                            ssize_t __start, ssize_t __range,
 516                            struct re_registers *__regs);
 517
 518
 519 /* Like `re_search', but search in the concatenation of STRING1 and
 520    STRING2.  Also, stop searching at index START + STOP.  */
 521 extern regoff_t re_search_2 (struct re_pattern_buffer *__buffer,
 522                              const char *__string1, size_t __length1,
 523                              const char *__string2, size_t __length2,
 524                              ssize_t __start, ssize_t __range,
 525                              struct re_registers *__regs,
 526                              ssize_t __stop);
 527
 528
 529 /* Like `re_search', but return how many characters in STRING the regexp
 530    in BUFFER matched, starting at position START.  */
 531 extern regoff_t re_match (struct re_pattern_buffer *__buffer,
 532                           const char *__string, size_t __length,
 533                           ssize_t __start, struct re_registers *__regs);
 534
 535
 536 /* Relates to `re_match' as `re_search_2' relates to `re_search'.  */
 537 extern regoff_t re_match_2 (struct re_pattern_buffer *__buffer,
 538                             const char *__string1, size_t __length1,
 539                             const char *__string2, size_t __length2,
 540                             ssize_t __start, struct re_registers *__regs,
 541                             ssize_t __stop);
 542
 543
 544 /* Set REGS to hold NUM_REGS registers, storing them in STARTS and
 545    ENDS.  Subsequent matches using BUFFER and REGS will use this memory
 546    for recording register information.  STARTS and ENDS must be
 547    allocated with malloc, and must each be at least `NUM_REGS * sizeof
 548    (regoff_t)' bytes long.
 549
 550    If NUM_REGS == 0, then subsequent matches should allocate their own
 551    register data.
 552
 553    Unless this function is called, the first search or match using
 554    PATTERN_BUFFER will allocate its own register data, without
 555    freeing the old data.  */
 556 extern void re_set_registers (struct re_pattern_buffer *__buffer,
 557                               struct re_registers *__regs,
 558                               unsigned __num_regs,
 559                               regoff_t *__starts, regoff_t *__ends);
 560
 561 #if defined _REGEX_RE_COMP || defined _LIBC
 562 # ifndef _CRAY
 563 /* 4.2 bsd compatibility.  */
 564 extern char *re_comp (const char *);
 565 extern int re_exec (const char *);
 566 # endif
 567 #endif
 568
 569 /* GCC 2.95 and later have "__restrict"; C99 compilers have
 570    "restrict", and "configure" may have defined "restrict".
 571    Other compilers use __restrict, __restrict__, and _Restrict, and
 572    'configure' might #define 'restrict' to those words, so pick a
 573    different name.  */
 574 #ifndef _Restrict_
 575 # if 199901L <= __STDC_VERSION__
 576 #  define _Restrict_ restrict
 577 # elif 2 < __GNUC__ || (2 == __GNUC__ && 95 <= __GNUC_MINOR__)
 578 #  define _Restrict_ __restrict
 579 # else
 580 #  define _Restrict_
 581 # endif
 582 #endif
 583 /* gcc 3.1 and up support the [restrict] syntax.  Don't trust
 584    sys/cdefs.h's definition of __restrict_arr, though, as it
 585    mishandles gcc -ansi -pedantic.  */
 586 #ifndef _Restrict_arr_
 587 # if ((199901L <= __STDC_VERSION__                                      \
 588        || ((3 < __GNUC__ || (3 == __GNUC__ && 1 <= __GNUC_MINOR__))     \
 589            && !defined __STRICT_ANSI__))                                        \
 590       && !defined __GNUG__)
 591 #  define _Restrict_arr_ _Restrict_
 592 # else
 593 #  define _Restrict_arr_
 594 # endif
 595 #endif
 596
 597 /* POSIX compatibility.  */
 598 extern reg_errcode_t regcomp (regex_t *_Restrict_ __preg,
 599                               const char *_Restrict_ __pattern,
 600                               int __cflags);
 601
 602 extern reg_errcode_t regexec (const regex_t *_Restrict_ __preg,
 603                               const char *_Restrict_ __string, size_t __nmatch,
 604                               regmatch_t __pmatch[_Restrict_arr_],
 605                               int __eflags);
 606
 607 extern size_t regerror (int __errcode, const regex_t * __preg,
 608                         char *__errbuf, size_t __errbuf_size);
 609
 610 extern void regfree (regex_t *__preg);
 611
 612
 613 #ifdef __cplusplus
 614 }
 615 #endif  /* C++ */
 616
 617 /* For platform which support the ISO C amendment 1 functionality we
 618    support user defined character classes.  */
 619 #if WIDE_CHAR_SUPPORT
 620 /* Solaris 2.5 has a bug: <wchar.h> must be included before <wctype.h>.  */
 621 # include <wchar.h>
 622 # include <wctype.h>
 623
 624 typedef wctype_t re_wctype_t;
 625 typedef wchar_t re_wchar_t;
 626 # define re_wctype wctype
 627 # define re_iswctype iswctype
 628 # define re_wctype_to_bit(cc) 0
 629 #else
 630 # ifndef emacs
 631 #  define btowc(c) c
 632 # endif
 633
 634 /* Character classes.  */
 635 typedef enum { RECC_ERROR = 0,
 636                RECC_ALNUM, RECC_ALPHA, RECC_WORD,
 637                RECC_GRAPH, RECC_PRINT,
 638                RECC_LOWER, RECC_UPPER,
 639                RECC_PUNCT, RECC_CNTRL,
 640                RECC_DIGIT, RECC_XDIGIT,
 641                RECC_BLANK, RECC_SPACE,
 642                RECC_MULTIBYTE, RECC_NONASCII,
 643                RECC_ASCII, RECC_UNIBYTE
 644 } re_wctype_t;
 645
 646 extern char re_iswctype (int ch,    re_wctype_t cc);
 647 extern re_wctype_t re_wctype_parse (const unsigned char **strp, unsigned limit);
 648
 649 typedef int re_wchar_t;
 650
 651 #endif /* not WIDE_CHAR_SUPPORT */
 652
 653 #endif /* regex.h */
 654 \f