src/regex.h

   1 /* Definitions for data structures and routines for the regular
   2    expression library, version 0.12.
   3
   4    Copyright (C) 1985, 89, 90, 91, 92, 93, 95 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., 59 Temple Place - Suite 330, Boston, MA 02111-1307,
  19    USA.  */
  20
  21 #ifndef __REGEXP_LIBRARY_H__
  22 #define __REGEXP_LIBRARY_H__
  23
  24 /* POSIX says that <sys/types.h> must be included (by the caller) before
  25    <regex.h>.  */
  26
  27 #if !defined (_POSIX_C_SOURCE) && !defined (_POSIX_SOURCE) && defined (VMS)
  28 /* VMS doesn't have `size_t' in <sys/types.h>, even though POSIX says it
  29    should be there.  */
  30 #include <stddef.h>
  31 #endif
  32
  33
  34 /* The following bits are used to determine the regexp syntax we
  35    recognize.  The set/not-set meanings are chosen so that Emacs syntax
  36    remains the value 0.  The bits are given in alphabetical order, and
  37    the definitions shifted by one from the previous bit; thus, when we
  38    add or remove a bit, only one other definition need change.  */
  39 typedef unsigned reg_syntax_t;
  40
  41 /* If this bit is not set, then \ inside a bracket expression is literal.
  42    If set, then such a \ quotes the following character.  */
  43 #define RE_BACKSLASH_ESCAPE_IN_LISTS (1)
  44
  45 /* If this bit is not set, then + and ? are operators, and \+ and \? are
  46      literals.
  47    If set, then \+ and \? are operators and + and ? are literals.  */
  48 #define RE_BK_PLUS_QM (RE_BACKSLASH_ESCAPE_IN_LISTS << 1)
  49
  50 /* If this bit is set, then character classes are supported.  They are:
  51      [:alpha:], [:upper:], [:lower:],  [:digit:], [:alnum:], [:xdigit:],
  52      [:space:], [:print:], [:punct:], [:graph:], and [:cntrl:].
  53    If not set, then character classes are not supported.  */
  54 #define RE_CHAR_CLASSES (RE_BK_PLUS_QM << 1)
  55
  56 /* If this bit is set, then ^ and $ are always anchors (outside bracket
  57      expressions, of course).
  58    If this bit is not set, then it depends:
  59         ^  is an anchor if it is at the beginning of a regular
  60            expression or after an open-group or an alternation operator;
  61         $  is an anchor if it is at the end of a regular expression, or
  62            before a close-group or an alternation operator.
  63
  64    This bit could be (re)combined with RE_CONTEXT_INDEP_OPS, because
  65    POSIX draft 11.2 says that * etc. in leading positions is undefined.
  66    We already implemented a previous draft which made those constructs
  67    invalid, though, so we haven't changed the code back.  */
  68 #define RE_CONTEXT_INDEP_ANCHORS (RE_CHAR_CLASSES << 1)
  69
  70 /* If this bit is set, then special characters are always special
  71      regardless of where they are in the pattern.
  72    If this bit is not set, then special characters are special only in
  73      some contexts; otherwise they are ordinary.  Specifically,
  74      * + ? and intervals are only special when not after the beginning,
  75      open-group, or alternation operator.  */
  76 #define RE_CONTEXT_INDEP_OPS (RE_CONTEXT_INDEP_ANCHORS << 1)
  77
  78 /* If this bit is set, then *, +, ?, and { cannot be first in an re or
  79      immediately after an alternation or begin-group operator.  */
  80 #define RE_CONTEXT_INVALID_OPS (RE_CONTEXT_INDEP_OPS << 1)
  81
  82 /* If this bit is set, then . matches newline.
  83    If not set, then it doesn't.  */
  84 #define RE_DOT_NEWLINE (RE_CONTEXT_INVALID_OPS << 1)
  85
  86 /* If this bit is set, then . doesn't match NUL.
  87    If not set, then it does.  */
  88 #define RE_DOT_NOT_NULL (RE_DOT_NEWLINE << 1)
  89
  90 /* If this bit is set, nonmatching lists [^...] do not match newline.
  91    If not set, they do.  */
  92 #define RE_HAT_LISTS_NOT_NEWLINE (RE_DOT_NOT_NULL << 1)
  93
  94 /* If this bit is set, either \{...\} or {...} defines an
  95      interval, depending on RE_NO_BK_BRACES.
  96    If not set, \{, \}, {, and } are literals.  */
  97 #define RE_INTERVALS (RE_HAT_LISTS_NOT_NEWLINE << 1)
  98
  99 /* If this bit is set, +, ? and | aren't recognized as operators.
 100    If not set, they are.  */
 101 #define RE_LIMITED_OPS (RE_INTERVALS << 1)
 102
 103 /* If this bit is set, newline is an alternation operator.
 104    If not set, newline is literal.  */
 105 #define RE_NEWLINE_ALT (RE_LIMITED_OPS << 1)
 106
 107 /* If this bit is set, then `{...}' defines an interval, and \{ and \}
 108      are literals.
 109   If not set, then `\{...\}' defines an interval.  */
 110 #define RE_NO_BK_BRACES (RE_NEWLINE_ALT << 1)
 111
 112 /* If this bit is set, (...) defines a group, and \( and \) are literals.
 113    If not set, \(...\) defines a group, and ( and ) are literals.  */
 114 #define RE_NO_BK_PARENS (RE_NO_BK_BRACES << 1)
 115
 116 /* If this bit is set, then \<digit> matches <digit>.
 117    If not set, then \<digit> is a back-reference.  */
 118 #define RE_NO_BK_REFS (RE_NO_BK_PARENS << 1)
 119
 120 /* If this bit is set, then | is an alternation operator, and \| is literal.
 121    If not set, then \| is an alternation operator, and | is literal.  */
 122 #define RE_NO_BK_VBAR (RE_NO_BK_REFS << 1)
 123
 124 /* If this bit is set, then an ending range point collating higher
 125      than the starting range point, as in [z-a], is invalid.
 126    If not set, then when ending range point collates higher than the
 127      starting range point, the range is ignored.  */
 128 #define RE_NO_EMPTY_RANGES (RE_NO_BK_VBAR << 1)
 129
 130 /* If this bit is set, then an unmatched ) is ordinary.
 131    If not set, then an unmatched ) is invalid.  */
 132 #define RE_UNMATCHED_RIGHT_PAREN_ORD (RE_NO_EMPTY_RANGES << 1)
 133
 134 /* If this bit is set, succeed as soon as we match the whole pattern,
 135    without further backtracking.  */
 136 #define RE_NO_POSIX_BACKTRACKING (RE_UNMATCHED_RIGHT_PAREN_ORD << 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 {
 253   REG_NOERROR = 0,      /* Success.  */
 254   REG_NOMATCH,          /* Didn't find a match (for regexec).  */
 255
 256   /* POSIX regcomp return error codes.  (In the order listed in the
 257      standard.)  */
 258   REG_BADPAT,           /* Invalid pattern.  */
 259   REG_ECOLLATE,         /* Not implemented.  */
 260   REG_ECTYPE,           /* Invalid character class name.  */
 261   REG_EESCAPE,          /* Trailing backslash.  */
 262   REG_ESUBREG,          /* Invalid back reference.  */
 263   REG_EBRACK,           /* Unmatched left bracket.  */
 264   REG_EPAREN,           /* Parenthesis imbalance.  */
 265   REG_EBRACE,           /* Unmatched \{.  */
 266   REG_BADBR,            /* Invalid contents of \{\}.  */
 267   REG_ERANGE,           /* Invalid range end.  */
 268   REG_ESPACE,           /* Ran out of memory.  */
 269   REG_BADRPT,           /* No preceding re for repetition op.  */
 270
 271   /* Error codes we've added.  */
 272   REG_EEND,             /* Premature end.  */
 273   REG_ESIZE,            /* Compiled pattern bigger than 2^16 bytes.  */
 274   REG_ERPAREN           /* Unmatched ) or \); not returned from regcomp.  */
 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 #ifndef RE_TRANSLATE_TYPE
 284 #define RE_TRANSLATE_TYPE char *
 285 #endif
 286
 287 struct re_pattern_buffer
 288 {
 289 /* [[[begin pattern_buffer]]] */
 290         /* Space that holds the compiled pattern.  It is declared as
 291           `unsigned char *' because its elements are
 292            sometimes used as array indexes.  */
 293   unsigned char *buffer;
 294
 295         /* Number of bytes to which `buffer' points.  */
 296   unsigned long allocated;
 297
 298         /* Number of bytes actually used in `buffer'.  */
 299   unsigned long used;
 300
 301         /* Syntax setting with which the pattern was compiled.  */
 302   reg_syntax_t syntax;
 303
 304         /* Pointer to a fastmap, if any, otherwise zero.  re_search uses
 305            the fastmap, if there is one, to skip over impossible
 306            starting points for matches.  */
 307   char *fastmap;
 308
 309         /* Either a translate table to apply to all characters before
 310            comparing them, or zero for no translation.  The translation
 311            is applied to a pattern when it is compiled and to a string
 312            when it is matched.  */
 313   RE_TRANSLATE_TYPE translate;
 314
 315         /* Number of subexpressions found by the compiler.  */
 316   size_t re_nsub;
 317
 318         /* Zero if this pattern cannot match the empty string, one else.
 319            Well, in truth it's used only in `re_search_2', to see
 320            whether or not we should use the fastmap, so we don't set
 321            this absolutely perfectly; see `re_compile_fastmap' (the
 322            `duplicate' case).  */
 323   unsigned can_be_null : 1;
 324
 325         /* If REGS_UNALLOCATED, allocate space in the `regs' structure
 326              for `max (RE_NREGS, re_nsub + 1)' groups.
 327            If REGS_REALLOCATE, reallocate space if necessary.
 328            If REGS_FIXED, use what's there.  */
 329 #define REGS_UNALLOCATED 0
 330 #define REGS_REALLOCATE 1
 331 #define REGS_FIXED 2
 332   unsigned regs_allocated : 2;
 333
 334         /* Set to zero when `regex_compile' compiles a pattern; set to one
 335            by `re_compile_fastmap' if it updates the fastmap.  */
 336   unsigned fastmap_accurate : 1;
 337
 338         /* If set, `re_match_2' does not return information about
 339            subexpressions.  */
 340   unsigned no_sub : 1;
 341
 342         /* If set, a beginning-of-line anchor doesn't match at the
 343            beginning of the string.  */
 344   unsigned not_bol : 1;
 345
 346         /* Similarly for an end-of-line anchor.  */
 347   unsigned not_eol : 1;
 348
 349         /* If true, an anchor at a newline matches.  */
 350   unsigned newline_anchor : 1;
 351
 352 /* [[[end pattern_buffer]]] */
 353 };
 354
 355 typedef struct re_pattern_buffer regex_t;
 356 \f
 357 /* Type for byte offsets within the string.  POSIX mandates this.  */
 358 typedef int regoff_t;
 359
 360
 361 /* This is the structure we store register match data in.  See
 362    regex.texinfo for a full description of what registers match.  */
 363 struct re_registers
 364 {
 365   unsigned num_regs;
 366   regoff_t *start;
 367   regoff_t *end;
 368 };
 369
 370
 371 /* If `regs_allocated' is REGS_UNALLOCATED in the pattern buffer,
 372    `re_match_2' returns information about at least this many registers
 373    the first time a `regs' structure is passed.  */
 374 #ifndef RE_NREGS
 375 #define RE_NREGS 30
 376 #endif
 377
 378
 379 /* POSIX specification for registers.  Aside from the different names than
 380    `re_registers', POSIX uses an array of structures, instead of a
 381    structure of arrays.  */
 382 typedef struct
 383 {
 384   regoff_t rm_so;  /* Byte offset from string's start to substring's start.  */
 385   regoff_t rm_eo;  /* Byte offset from string's start to substring's end.  */
 386 } regmatch_t;
 387 \f
 388 /* Declarations for routines.  */
 389
 390 /* To avoid duplicating every routine declaration -- once with a
 391    prototype (if we are ANSI), and once without (if we aren't) -- we
 392    use the following macro to declare argument types.  This
 393    unfortunately clutters up the declarations a bit, but I think it's
 394    worth it.  */
 395
 396 #if __STDC__
 397
 398 #define _RE_ARGS(args) args
 399
 400 #else /* not __STDC__ */
 401
 402 #define _RE_ARGS(args) ()
 403
 404 #endif /* not __STDC__ */
 405
 406 /* Sets the current default syntax to SYNTAX, and return the old syntax.
 407    You can also simply assign to the `re_syntax_options' variable.  */
 408 extern reg_syntax_t re_set_syntax _RE_ARGS ((reg_syntax_t syntax));
 409
 410 /* Compile the regular expression PATTERN, with length LENGTH
 411    and syntax given by the global `re_syntax_options', into the buffer
 412    BUFFER.  Return NULL if successful, and an error string if not.  */
 413 extern const char *re_compile_pattern
 414   _RE_ARGS ((const char *pattern, int length,
 415              struct re_pattern_buffer *buffer));
 416
 417
 418 /* Compile a fastmap for the compiled pattern in BUFFER; used to
 419    accelerate searches.  Return 0 if successful and -2 if was an
 420    internal error.  */
 421 extern int re_compile_fastmap _RE_ARGS ((struct re_pattern_buffer *buffer));
 422
 423
 424 /* Search in the string STRING (with length LENGTH) for the pattern
 425    compiled into BUFFER.  Start searching at position START, for RANGE
 426    characters.  Return the starting position of the match, -1 for no
 427    match, or -2 for an internal error.  Also return register
 428    information in REGS (if REGS and BUFFER->no_sub are nonzero).  */
 429 extern int re_search
 430   _RE_ARGS ((struct re_pattern_buffer *buffer, const char *string,
 431             int length, int start, int range, struct re_registers *regs));
 432
 433
 434 /* Like `re_search', but search in the concatenation of STRING1 and
 435    STRING2.  Also, stop searching at index START + STOP.  */
 436 extern int re_search_2
 437   _RE_ARGS ((struct re_pattern_buffer *buffer, const char *string1,
 438              int length1, const char *string2, int length2,
 439              int start, int range, struct re_registers *regs, int stop));
 440
 441
 442 /* Like `re_search', but return how many characters in STRING the regexp
 443    in BUFFER matched, starting at position START.  */
 444 extern int re_match
 445   _RE_ARGS ((struct re_pattern_buffer *buffer, const char *string,
 446              int length, int start, struct re_registers *regs));
 447
 448
 449 /* Relates to `re_match' as `re_search_2' relates to `re_search'.  */
 450 extern int re_match_2
 451   _RE_ARGS ((struct re_pattern_buffer *buffer, const char *string1,
 452              int length1, const char *string2, int length2,
 453              int start, struct re_registers *regs, int stop));
 454
 455
 456 /* Set REGS to hold NUM_REGS registers, storing them in STARTS and
 457    ENDS.  Subsequent matches using BUFFER and REGS will use this memory
 458    for recording register information.  STARTS and ENDS must be
 459    allocated with malloc, and must each be at least `NUM_REGS * sizeof
 460    (regoff_t)' bytes long.
 461
 462    If NUM_REGS == 0, then subsequent matches should allocate their own
 463    register data.
 464
 465    Unless this function is called, the first search or match using
 466    PATTERN_BUFFER will allocate its own register data, without
 467    freeing the old data.  */
 468 extern void re_set_registers
 469   _RE_ARGS ((struct re_pattern_buffer *buffer, struct re_registers *regs,
 470              unsigned num_regs, regoff_t *starts, regoff_t *ends));
 471
 472 #ifdef _REGEX_RE_COMP
 473 /* 4.2 bsd compatibility.  */
 474 extern char *re_comp _RE_ARGS ((const char *));
 475 extern int re_exec _RE_ARGS ((const char *));
 476 #endif
 477
 478 /* POSIX compatibility.  */
 479 extern int regcomp _RE_ARGS ((regex_t *preg, const char *pattern, int cflags));
 480 extern int regexec
 481   _RE_ARGS ((const regex_t *preg, const char *string, size_t nmatch,
 482              regmatch_t pmatch[], int eflags));
 483 extern size_t regerror
 484   _RE_ARGS ((int errcode, const regex_t *preg, char *errbuf,
 485              size_t errbuf_size));
 486 extern void regfree _RE_ARGS ((regex_t *preg));
 487
 488 #endif /* not __REGEXP_LIBRARY_H__ */
 489 \f
 490 /*
 491 Local variables:
 492 make-backup-files: t
 493 version-control: t
 494 trim-versions-without-asking: nil
 495 End:
 496 */