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