windows/regex.h

   1 /* Definitions for data structures and routines for the regular
   2   expression library, version 0.12.
   3
   4   Copyright (C) 1985, 1989, 1990, 1991, 1992, 1993 Free Software Foundation, Inc.
   5
   6   This program is free software; you can redistribute it and/or modify
   7   it under the terms of the GNU General Public License as published by
   8   the Free Software Foundation; either version 2, or (at your option)
   9   any later version.
  10
  11   This program is distributed in the hope that it will be useful,
  12   but WITHOUT ANY WARRANTY; without even the implied warranty of
  13   MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
  14   GNU General Public License for more details.
  15
  16   You should have received a copy of the GNU General Public License
  17   along with this program; if not, write to the Free Software
  18   Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.  */
  19
  20 #ifndef __REGEXP_LIBRARY_H__
  21 #define __REGEXP_LIBRARY_H__
  22
  23 #ifdef __cplusplus
  24 extern "C"
  25 {
  26 #endif
  27
  28     /* POSIX says that <sys/types.h> must be included (by the caller) before
  29        <regex.h>.  */
  30
  31 #ifdef VMS
  32     /* VMS doesn't have `size_t' in <sys/types.h>, even though POSIX says it
  33        should be there.  */
  34 #include <stddef.h>
  35 #endif
  36
  37
  38     /* The following bits are used to determine the regexp syntax we
  39        recognize.  The set/not-set meanings are chosen so that Emacs syntax
  40        remains the value 0.  The bits are given in alphabetical order, and
  41        the definitions shifted by one from the previous bit; thus, when we
  42        add or remove a bit, only one other definition need change.  */
  43     typedef unsigned reg_syntax_t;
  44
  45     /* If this bit is not set, then \ inside a bracket expression is literal.
  46        If set, then such a \ quotes the following character.  */
  47 #define RE_BACKSLASH_ESCAPE_IN_LISTS (1)
  48
  49     /* If this bit is not set, then + and ? are operators, and \+ and \? are
  50          literals.
  51        If set, then \+ and \? are operators and + and ? are literals.  */
  52 #define RE_BK_PLUS_QM (RE_BACKSLASH_ESCAPE_IN_LISTS << 1)
  53
  54     /* If this bit is set, then character classes are supported.  They are:
  55          [:alpha:], [:upper:], [:lower:],  [:digit:], [:alnum:], [:xdigit:],
  56          [:space:], [:print:], [:punct:], [:graph:], and [:cntrl:].
  57        If not set, then character classes are not supported.  */
  58 #define RE_CHAR_CLASSES (RE_BK_PLUS_QM << 1)
  59
  60     /* If this bit is set, then ^ and $ are always anchors (outside bracket
  61          expressions, of course).
  62        If this bit is not set, then it depends:
  63             ^  is an anchor if it is at the beginning of a regular
  64                expression or after an open-group or an alternation operator;
  65             $  is an anchor if it is at the end of a regular expression, or
  66                before a close-group or an alternation operator.
  67
  68        This bit could be (re)combined with RE_CONTEXT_INDEP_OPS, because
  69        POSIX draft 11.2 says that * etc. in leading positions is undefined.
  70        We already implemented a previous draft which made those constructs
  71        invalid, though, so we haven't changed the code back.  */
  72 #define RE_CONTEXT_INDEP_ANCHORS (RE_CHAR_CLASSES << 1)
  73
  74     /* If this bit is set, then special characters are always special
  75          regardless of where they are in the pattern.
  76        If this bit is not set, then special characters are special only in
  77          some contexts; otherwise they are ordinary.  Specifically,
  78          * + ? and intervals are only special when not after the beginning,
  79          open-group, or alternation operator.  */
  80 #define RE_CONTEXT_INDEP_OPS (RE_CONTEXT_INDEP_ANCHORS << 1)
  81
  82     /* If this bit is set, then *, +, ?, and { cannot be first in an re or
  83          immediately after an alternation or begin-group operator.  */
  84 #define RE_CONTEXT_INVALID_OPS (RE_CONTEXT_INDEP_OPS << 1)
  85
  86     /* If this bit is set, then . matches newline.
  87        If not set, then it doesn't.  */
  88 #define RE_DOT_NEWLINE (RE_CONTEXT_INVALID_OPS << 1)
  89
  90     /* If this bit is set, then . doesn't match NUL.
  91        If not set, then it does.  */
  92 #define RE_DOT_NOT_NULL (RE_DOT_NEWLINE << 1)
  93
  94     /* If this bit is set, nonmatching lists [^...] do not match newline.
  95        If not set, they do.  */
  96 #define RE_HAT_LISTS_NOT_NEWLINE (RE_DOT_NOT_NULL << 1)
  97
  98     /* If this bit is set, either \{...\} or {...} defines an
  99          interval, depending on RE_NO_BK_BRACES.
 100        If not set, \{, \}, {, and } are literals.  */
 101 #define RE_INTERVALS (RE_HAT_LISTS_NOT_NEWLINE << 1)
 102
 103     /* If this bit is set, +, ? and | aren't recognized as operators.
 104        If not set, they are.  */
 105 #define RE_LIMITED_OPS (RE_INTERVALS << 1)
 106
 107     /* If this bit is set, newline is an alternation operator.
 108        If not set, newline is literal.  */
 109 #define RE_NEWLINE_ALT (RE_LIMITED_OPS << 1)
 110
 111     /* If this bit is set, then `{...}' defines an interval, and \{ and \}
 112          are literals.
 113       If not set, then `\{...\}' defines an interval.  */
 114 #define RE_NO_BK_BRACES (RE_NEWLINE_ALT << 1)
 115
 116     /* If this bit is set, (...) defines a group, and \( and \) are literals.
 117        If not set, \(...\) defines a group, and ( and ) are literals.  */
 118 #define RE_NO_BK_PARENS (RE_NO_BK_BRACES << 1)
 119
 120     /* If this bit is set, then \<digit> matches <digit>.
 121        If not set, then \<digit> is a back-reference.  */
 122 #define RE_NO_BK_REFS (RE_NO_BK_PARENS << 1)
 123
 124     /* If this bit is set, then | is an alternation operator, and \| is literal.
 125        If not set, then \| is an alternation operator, and | is literal.  */
 126 #define RE_NO_BK_VBAR (RE_NO_BK_REFS << 1)
 127
 128     /* If this bit is set, then an ending range point collating higher
 129          than the starting range point, as in [z-a], is invalid.
 130        If not set, then when ending range point collates higher than the
 131          starting range point, the range is ignored.  */
 132 #define RE_NO_EMPTY_RANGES (RE_NO_BK_VBAR << 1)
 133
 134     /* If this bit is set, then an unmatched ) is ordinary.
 135        If not set, then an unmatched ) is invalid.  */
 136 #define RE_UNMATCHED_RIGHT_PAREN_ORD (RE_NO_EMPTY_RANGES << 1)
 137
 138     /* This global variable defines the particular regexp syntax to use (for
 139        some interfaces).  When a regexp is compiled, the syntax used is
 140        stored in the pattern buffer, so changing this does not affect
 141        already-compiled regexps.  */
 142     extern reg_syntax_t re_syntax_options;
 143     \f
 144     /* Define combinations of the above bits for the standard possibilities.
 145        (The [[[ comments delimit what gets put into the Texinfo file, so
 146        don't delete them!)  */
 147     /* [[[begin syntaxes]]] */
 148 #define RE_SYNTAX_EMACS 0
 149
 150 #define RE_SYNTAX_AWK                                                   \
 151   (RE_BACKSLASH_ESCAPE_IN_LISTS | RE_DOT_NOT_NULL                       \
 152    | RE_NO_BK_PARENS            | RE_NO_BK_REFS                         \
 153    | RE_NO_BK_VBAR               | RE_NO_EMPTY_RANGES                   \
 154    | RE_UNMATCHED_RIGHT_PAREN_ORD)
 155
 156 #define RE_SYNTAX_POSIX_AWK                                             \
 157   (RE_SYNTAX_POSIX_EXTENDED | RE_BACKSLASH_ESCAPE_IN_LISTS)
 158
 159 #define RE_SYNTAX_GREP                                                  \
 160   (RE_BK_PLUS_QM              | RE_CHAR_CLASSES                         \
 161    | RE_HAT_LISTS_NOT_NEWLINE | RE_INTERVALS                            \
 162    | RE_NEWLINE_ALT)
 163
 164 #define RE_SYNTAX_EGREP                                                 \
 165   (RE_CHAR_CLASSES        | RE_CONTEXT_INDEP_ANCHORS                    \
 166    | RE_CONTEXT_INDEP_OPS | RE_HAT_LISTS_NOT_NEWLINE                    \
 167    | RE_NEWLINE_ALT       | RE_NO_BK_PARENS                             \
 168    | RE_NO_BK_VBAR)
 169
 170 #define RE_SYNTAX_POSIX_EGREP                                           \
 171   (RE_SYNTAX_EGREP | RE_INTERVALS | RE_NO_BK_BRACES)
 172
 173     /* P1003.2/D11.2, section 4.20.7.1, lines 5078ff.  */
 174 #define RE_SYNTAX_ED RE_SYNTAX_POSIX_BASIC
 175
 176 #define RE_SYNTAX_SED RE_SYNTAX_POSIX_BASIC
 177
 178     /* Syntax bits common to both basic and extended POSIX regex syntax.  */
 179 #define _RE_SYNTAX_POSIX_COMMON                                         \
 180   (RE_CHAR_CLASSES | RE_DOT_NEWLINE      | RE_DOT_NOT_NULL              \
 181    | RE_INTERVALS  | RE_NO_EMPTY_RANGES)
 182
 183 #define RE_SYNTAX_POSIX_BASIC                                           \
 184   (_RE_SYNTAX_POSIX_COMMON | RE_BK_PLUS_QM)
 185
 186     /* Differs from ..._POSIX_BASIC only in that RE_BK_PLUS_QM becomes
 187        RE_LIMITED_OPS, i.e., \? \+ \| are not recognized.  Actually, this
 188        isn't minimal, since other operators, such as \`, aren't disabled.  */
 189 #define RE_SYNTAX_POSIX_MINIMAL_BASIC                                   \
 190   (_RE_SYNTAX_POSIX_COMMON | RE_LIMITED_OPS)
 191
 192 #define RE_SYNTAX_POSIX_EXTENDED                                        \
 193   (_RE_SYNTAX_POSIX_COMMON | RE_CONTEXT_INDEP_ANCHORS                   \
 194    | RE_CONTEXT_INDEP_OPS  | RE_NO_BK_BRACES                            \
 195    | RE_NO_BK_PARENS       | RE_NO_BK_VBAR                              \
 196    | RE_UNMATCHED_RIGHT_PAREN_ORD)
 197
 198     /* Differs from ..._POSIX_EXTENDED in that RE_CONTEXT_INVALID_OPS
 199        replaces RE_CONTEXT_INDEP_OPS and RE_NO_BK_REFS is added.  */
 200 #define RE_SYNTAX_POSIX_MINIMAL_EXTENDED                                \
 201   (_RE_SYNTAX_POSIX_COMMON  | RE_CONTEXT_INDEP_ANCHORS                  \
 202    | RE_CONTEXT_INVALID_OPS | RE_NO_BK_BRACES                           \
 203    | RE_NO_BK_PARENS        | RE_NO_BK_REFS                             \
 204    | RE_NO_BK_VBAR          | RE_UNMATCHED_RIGHT_PAREN_ORD)
 205     /* [[[end syntaxes]]] */
 206     \f
 207     /* Maximum number of duplicates an interval can allow.  Some systems
 208        (erroneously) define this in other header files, but we want our
 209        value, so remove any previous define.  */
 210 #ifdef RE_DUP_MAX
 211 #undef RE_DUP_MAX
 212 #endif
 213 #define RE_DUP_MAX ((1 << 15) - 1)
 214
 215
 216     /* POSIX `cflags' bits (i.e., information for `regcomp').  */
 217
 218     /* If this bit is set, then use extended regular expression syntax.
 219        If not set, then use basic regular expression syntax.  */
 220 #define REG_EXTENDED 1
 221
 222     /* If this bit is set, then ignore case when matching.
 223        If not set, then case is significant.  */
 224 #define REG_ICASE (REG_EXTENDED << 1)
 225
 226     /* If this bit is set, then anchors do not match at newline
 227          characters in the string.
 228        If not set, then anchors do match at newlines.  */
 229 #define REG_NEWLINE (REG_ICASE << 1)
 230
 231     /* If this bit is set, then report only success or fail in regexec.
 232        If not set, then returns differ between not matching and errors.  */
 233 #define REG_NOSUB (REG_NEWLINE << 1)
 234
 235
 236     /* POSIX `eflags' bits (i.e., information for regexec).  */
 237
 238     /* If this bit is set, then the beginning-of-line operator doesn't match
 239          the beginning of the string (presumably because it's not the
 240          beginning of a line).
 241        If not set, then the beginning-of-line operator does match the
 242          beginning of the string.  */
 243 #define REG_NOTBOL 1
 244
 245     /* Like REG_NOTBOL, except for the end-of-line.  */
 246 #define REG_NOTEOL (1 << 1)
 247
 248
 249     /* If any error codes are removed, changed, or added, update the
 250        `re_error_msg' table in regex.c.  */
 251     typedef enum {
 252         REG_NOERROR = 0,        /* Success.  */
 253         REG_NOMATCH,            /* Didn't find a match (for regexec).  */
 254
 255         /* POSIX regcomp return error codes.  (In the order listed in the
 256            standard.)  */
 257         REG_BADPAT,             /* Invalid pattern.  */
 258         REG_ECOLLATE,           /* Not implemented.  */
 259         REG_ECTYPE,             /* Invalid character class name.  */
 260         REG_EESCAPE,            /* Trailing backslash.  */
 261         REG_ESUBREG,            /* Invalid back reference.  */
 262         REG_EBRACK,             /* Unmatched left bracket.  */
 263         REG_EPAREN,             /* Parenthesis imbalance.  */
 264         REG_EBRACE,             /* Unmatched \{.  */
 265         REG_BADBR,              /* Invalid contents of \{\}.  */
 266         REG_ERANGE,             /* Invalid range end.  */
 267         REG_ESPACE,             /* Ran out of memory.  */
 268         REG_BADRPT,             /* No preceding re for repetition op.  */
 269
 270         /* Error codes we've added.  */
 271         REG_EEND,               /* Premature end.  */
 272         REG_ESIZE,              /* Compiled pattern bigger than 2^16 bytes.  */
 273         REG_ERPAREN             /* Unmatched ) or \); not returned from regcomp.  */
 274     }
 275     reg_errcode_t;
 276     \f
 277     /* This data structure represents a compiled pattern.  Before calling
 278        the pattern compiler, the fields `buffer', `allocated', `fastmap',
 279        `translate', and `no_sub' can be set.  After the pattern has been
 280        compiled, the `re_nsub' field is available.  All other fields are
 281        private to the regex routines.  */
 282
 283     struct re_pattern_buffer {
 284         /* [[[begin pattern_buffer]]] */
 285         /* Space that holds the compiled pattern.  It is declared as
 286                  `unsigned char *' because its elements are
 287                   sometimes used as array indexes.  */
 288         unsigned char *buffer;
 289
 290         /* Number of bytes to which `buffer' points.  */
 291         unsigned long allocated;
 292
 293         /* Number of bytes actually used in `buffer'.  */
 294         unsigned long used;
 295
 296         /* Syntax setting with which the pattern was compiled.  */
 297         reg_syntax_t syntax;
 298
 299         /* Pointer to a fastmap, if any, otherwise zero.  re_search uses
 300            the fastmap, if there is one, to skip over impossible
 301            starting points for matches.  */
 302         char *fastmap;
 303
 304         /* Either a translate table to apply to all characters before
 305            comparing them, or zero for no translation.  The translation
 306            is applied to a pattern when it is compiled and to a string
 307            when it is matched.  */
 308         char *translate;
 309
 310         /* Number of subexpressions found by the compiler.  */
 311         size_t re_nsub;
 312
 313         /* Zero if this pattern cannot match the empty string, one else.
 314            Well, in truth it's used only in `re_search_2', to see
 315            whether or not we should use the fastmap, so we don't set
 316            this absolutely perfectly; see `re_compile_fastmap' (the
 317            `duplicate' case).  */
 318 unsigned can_be_null :
 319         1;
 320
 321         /* If REGS_UNALLOCATED, allocate space in the `regs' structure
 322              for `max (RE_NREGS, re_nsub + 1)' groups.
 323            If REGS_REALLOCATE, reallocate space if necessary.
 324            If REGS_FIXED, use what's there.  */
 325 #define REGS_UNALLOCATED 0
 326 #define REGS_REALLOCATE 1
 327 #define REGS_FIXED 2
 328
 329 unsigned regs_allocated :
 330         2;
 331
 332         /* Set to zero when `regex_compile' compiles a pattern; set to one
 333            by `re_compile_fastmap' if it updates the fastmap.  */
 334 unsigned fastmap_accurate :
 335         1;
 336
 337         /* If set, `re_match_2' does not return information about
 338            subexpressions.  */
 339 unsigned no_sub :
 340         1;
 341
 342         /* If set, a beginning-of-line anchor doesn't match at the
 343            beginning of the string.  */
 344 unsigned not_bol :
 345         1;
 346
 347         /* Similarly for an end-of-line anchor.  */
 348 unsigned not_eol :
 349         1;
 350
 351         /* If true, an anchor at a newline matches.  */
 352 unsigned newline_anchor :
 353         1;
 354
 355         /* [[[end pattern_buffer]]] */
 356     };
 357
 358     typedef struct re_pattern_buffer regex_t;
 359
 360
 361     /* search.c (search_buffer) in Emacs needs this one opcode value.  It is
 362        defined both in `regex.c' and here.  */
 363 #define RE_EXACTN_VALUE 1
 364     \f
 365     /* Type for byte offsets within the string.  POSIX mandates this.  */
 366     typedef int regoff_t;
 367
 368
 369     /* This is the structure we store register match data in.  See
 370        regex.texinfo for a full description of what registers match.  */
 371     struct re_registers {
 372         unsigned num_regs;
 373         regoff_t *start;
 374         regoff_t *end;
 375     };
 376
 377
 378     /* If `regs_allocated' is REGS_UNALLOCATED in the pattern buffer,
 379        `re_match_2' returns information about at least this many registers
 380        the first time a `regs' structure is passed.  */
 381 #ifndef RE_NREGS
 382 #define RE_NREGS 30
 383 #endif
 384
 385
 386     /* POSIX specification for registers.  Aside from the different names than
 387        `re_registers', POSIX uses an array of structures, instead of a
 388        structure of arrays.  */
 389     typedef struct {
 390         regoff_t rm_so;  /* Byte offset from string's start to substring's start.  */
 391         regoff_t rm_eo;  /* Byte offset from string's start to substring's end.  */
 392     }
 393     regmatch_t;
 394     \f
 395     /* Declarations for routines.  */
 396
 397     /* To avoid duplicating every routine declaration -- once with a
 398        prototype (if we are ANSI), and once without (if we aren't) -- we
 399        use the following macro to declare argument types.  This
 400        unfortunately clutters up the declarations a bit, but I think it's
 401        worth it.  */
 402
 403 #if __STDC__
 404
 405 #define _RE_ARGS(args) args
 406
 407 #else /* not __STDC__ */
 408
 409 #define _RE_ARGS(args) ()
 410
 411 #endif /* not __STDC__ */
 412
 413     /* Sets the current default syntax to SYNTAX, and return the old syntax.
 414        You can also simply assign to the `re_syntax_options' variable.  */
 415     extern reg_syntax_t re_set_syntax _RE_ARGS ((reg_syntax_t syntax));
 416
 417     /* Compile the regular expression PATTERN, with length LENGTH
 418        and syntax given by the global `re_syntax_options', into the buffer
 419        BUFFER.  Return NULL if successful, and an error string if not.  */
 420     extern const char *re_compile_pattern
 421         _RE_ARGS ((const char *pattern, int length,
 422                    struct re_pattern_buffer *buffer));
 423
 424
 425     /* Compile a fastmap for the compiled pattern in BUFFER; used to
 426        accelerate searches.  Return 0 if successful and -2 if was an
 427        internal error.  */
 428     extern int re_compile_fastmap _RE_ARGS ((struct re_pattern_buffer *buffer));
 429
 430
 431     /* Search in the string STRING (with length LENGTH) for the pattern
 432        compiled into BUFFER.  Start searching at position START, for RANGE
 433        characters.  Return the starting position of the match, -1 for no
 434        match, or -2 for an internal error.  Also return register
 435        information in REGS (if REGS and BUFFER->no_sub are nonzero).  */
 436     extern int re_search
 437         _RE_ARGS ((struct re_pattern_buffer *buffer, const char *string,
 438                    int length, int start, int range, struct re_registers *regs));
 439
 440
 441     /* Like `re_search', but search in the concatenation of STRING1 and
 442        STRING2.  Also, stop searching at index START + STOP.  */
 443     extern int re_search_2
 444         _RE_ARGS ((struct re_pattern_buffer *buffer, const char *string1,
 445                    int length1, const char *string2, int length2,
 446                    int start, int range, struct re_registers *regs, int stop));
 447
 448
 449     /* Like `re_search', but return how many characters in STRING the regexp
 450        in BUFFER matched, starting at position START.  */
 451     extern int re_match
 452         _RE_ARGS ((struct re_pattern_buffer *buffer, const char *string,
 453                    int length, int start, struct re_registers *regs));
 454
 455
 456     /* Relates to `re_match' as `re_search_2' relates to `re_search'.  */
 457     extern int re_match_2
 458         _RE_ARGS ((struct re_pattern_buffer *buffer, const char *string1,
 459                    int length1, const char *string2, int length2,
 460                    int start, struct re_registers *regs, int stop));
 461
 462
 463     /* Set REGS to hold NUM_REGS registers, storing them in STARTS and
 464        ENDS.  Subsequent matches using BUFFER and REGS will use this memory
 465        for recording register information.  STARTS and ENDS must be
 466        allocated with malloc, and must each be at least `NUM_REGS * sizeof
 467        (regoff_t)' bytes long.
 468
 469        If NUM_REGS == 0, then subsequent matches should allocate their own
 470        register data.
 471
 472        Unless this function is called, the first search or match using
 473        PATTERN_BUFFER will allocate its own register data, without
 474        freeing the old data.  */
 475     extern void re_set_registers
 476         _RE_ARGS ((struct re_pattern_buffer *buffer, struct re_registers *regs,
 477                    unsigned num_regs, regoff_t *starts, regoff_t *ends));
 478
 479     /* 4.2 bsd compatibility.  */
 480     extern char *re_comp _RE_ARGS ((const char *));
 481     extern int re_exec _RE_ARGS ((const char *));
 482
 483     /* POSIX compatibility.  */
 484     extern int regcomp _RE_ARGS ((regex_t *preg, const char *pattern, int cflags));
 485     extern int regexec
 486         _RE_ARGS ((const regex_t *preg, const char *string, size_t nmatch,
 487                    regmatch_t pmatch[], int eflags));
 488     extern size_t regerror
 489         _RE_ARGS ((int errcode, const regex_t *preg, char *errbuf,
 490                    size_t errbuf_size));
 491     extern void regfree _RE_ARGS ((regex_t *preg));
 492
 493 #endif /* not __REGEXP_LIBRARY_H__ */
 494
 495 #ifdef __cplusplus
 496 }
 497 #endif
 498
 499 \f
 500 /*
 501 Local variables:
 502 make-backup-files: t
 503 version-control: t
 504 trim-versions-without-asking: nil
 505 End:
 506 */