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., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. */ 19 20 #ifndef __REGEXP_LIBRARY_H__ 21 #define __REGEXP_LIBRARY_H__ 22 23 /* Allow the use in C++ code. */ 24 #ifdef __cplusplus 25 extern "C" { 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; 143 144 /* 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]]] */ 206 207 /* 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 #ifndef RE_DUP_MAX 214 # define RE_DUP_MAX ((1 << 15) - 1) 215 #endif 216 217 218 /* POSIX `cflags' bits (i.e., information for `regcomp'). */ 219 220 /* If this bit is set, then use extended regular expression syntax. 221 If not set, then use basic regular expression syntax. */ 222 #define REG_EXTENDED 1 223 224 /* If this bit is set, then ignore case when matching. 225 If not set, then case is significant. */ 226 #define REG_ICASE (REG_EXTENDED << 1) 227 228 /* If this bit is set, then anchors do not match at newline 229 characters in the string. 230 If not set, then anchors do match at newlines. */ 231 #define REG_NEWLINE (REG_ICASE << 1) 232 233 /* If this bit is set, then report only success or fail in regexec. 234 If not set, then returns differ between not matching and errors. */ 235 #define REG_NOSUB (REG_NEWLINE << 1) 236 237 238 /* POSIX `eflags' bits (i.e., information for regexec). */ 239 240 /* If this bit is set, then the beginning-of-line operator doesn't match 241 the beginning of the string (presumably because it's not the 242 beginning of a line). 243 If not set, then the beginning-of-line operator does match the 244 beginning of the string. */ 245 #define REG_NOTBOL 1 246 247 /* Like REG_NOTBOL, except for the end-of-line. */ 248 #define REG_NOTEOL (1 << 1) 249 250 251 /* If any error codes are removed, changed, or added, update the 252 `re_error_msg' table in regex.c. */ 253 typedef enum 254 { 255 REG_NOERROR = 0, /* Success. */ 256 REG_NOMATCH, /* Didn't find a match (for regexec). */ 257 258 /* POSIX regcomp return error codes. (In the order listed in the 259 standard.) */ 260 REG_BADPAT, /* Invalid pattern. */ 261 REG_ECOLLATE, /* Not implemented. */ 262 REG_ECTYPE, /* Invalid character class name. */ 263 REG_EESCAPE, /* Trailing backslash. */ 264 REG_ESUBREG, /* Invalid back reference. */ 265 REG_EBRACK, /* Unmatched left bracket. */ 266 REG_EPAREN, /* Parenthesis imbalance. */ 267 REG_EBRACE, /* Unmatched \{. */ 268 REG_BADBR, /* Invalid contents of \{\}. */ 269 REG_ERANGE, /* Invalid range end. */ 270 REG_ESPACE, /* Ran out of memory. */ 271 REG_BADRPT, /* No preceding re for repetition op. */ 272 273 /* Error codes we've added. */ 274 REG_EEND, /* Premature end. */ 275 REG_ESIZE, /* Compiled pattern bigger than 2^16 bytes. */ 276 REG_ERPAREN /* Unmatched ) or \); not returned from regcomp. */ 277 } reg_errcode_t; 278 279 /* This data structure represents a compiled pattern. Before calling 280 the pattern compiler, the fields `buffer', `allocated', `fastmap', 281 `translate', and `no_sub' can be set. After the pattern has been 282 compiled, the `re_nsub' field is available. All other fields are 283 private to the regex routines. */ 284 285 struct re_pattern_buffer 286 { 287 /* [[[begin pattern_buffer]]] */ 288 /* Space that holds the compiled pattern. It is declared as 289 `unsigned char *' because its elements are 290 sometimes used as array indexes. */ 291 unsigned char *buffer; 292 293 /* Number of bytes to which `buffer' points. */ 294 unsigned long allocated; 295 296 /* Number of bytes actually used in `buffer'. */ 297 unsigned long used; 298 299 /* Syntax setting with which the pattern was compiled. */ 300 reg_syntax_t syntax; 301 302 /* Pointer to a fastmap, if any, otherwise zero. re_search uses 303 the fastmap, if there is one, to skip over impossible 304 starting points for matches. */ 305 char *fastmap; 306 307 /* Either a translate table to apply to all characters before 308 comparing them, or zero for no translation. The translation 309 is applied to a pattern when it is compiled and to a string 310 when it is matched. */ 311 char *translate; 312 313 /* Number of subexpressions found by the compiler. */ 314 size_t re_nsub; 315 316 /* Zero if this pattern cannot match the empty string, one else. 317 Well, in truth it's used only in `re_search_2', to see 318 whether or not we should use the fastmap, so we don't set 319 this absolutely perfectly; see `re_compile_fastmap' (the 320 `duplicate' case). */ 321 unsigned can_be_null : 1; 322 323 /* If REGS_UNALLOCATED, allocate space in the `regs' structure 324 for `max (RE_NREGS, re_nsub + 1)' groups. 325 If REGS_REALLOCATE, reallocate space if necessary. 326 If REGS_FIXED, use what's there. */ 327 #define REGS_UNALLOCATED 0 328 #define REGS_REALLOCATE 1 329 #define REGS_FIXED 2 330 unsigned regs_allocated : 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 : 1; 335 336 /* If set, `re_match_2' does not return information about 337 subexpressions. */ 338 unsigned no_sub : 1; 339 340 /* If set, a beginning-of-line anchor doesn't match at the 341 beginning of the string. */ 342 unsigned not_bol : 1; 343 344 /* Similarly for an end-of-line anchor. */ 345 unsigned not_eol : 1; 346 347 /* If true, an anchor at a newline matches. */ 348 unsigned newline_anchor : 1; 349 350 /* [[[end pattern_buffer]]] */ 351 }; 352 353 typedef struct re_pattern_buffer regex_t; 354 355 356 /* search.c (search_buffer) in Emacs needs this one opcode value. It is 357 defined both in `regex.c' and here. */ 358 #define RE_EXACTN_VALUE 1 359 360 /* Type for byte offsets within the string. POSIX mandates this. */ 361 typedef int regoff_t; 362 363 364 /* This is the structure we store register match data in. See 365 regex.texinfo for a full description of what registers match. */ 366 struct re_registers 367 { 368 unsigned num_regs; 369 regoff_t *start; 370 regoff_t *end; 371 }; 372 373 374 /* If `regs_allocated' is REGS_UNALLOCATED in the pattern buffer, 375 `re_match_2' returns information about at least this many registers 376 the first time a `regs' structure is passed. */ 377 #ifndef RE_NREGS 378 #define RE_NREGS 30 379 #endif 380 381 382 /* POSIX specification for registers. Aside from the different names than 383 `re_registers', POSIX uses an array of structures, instead of a 384 structure of arrays. */ 385 typedef struct 386 { 387 regoff_t rm_so; /* Byte offset from string's start to substring's start. */ 388 regoff_t rm_eo; /* Byte offset from string's start to substring's end. */ 389 } regmatch_t; 390 391 /* Declarations for routines. */ 392 393 /* To avoid duplicating every routine declaration -- once with a 394 prototype (if we are ANSI), and once without (if we aren't) -- we 395 use the following macro to declare argument types. This 396 unfortunately clutters up the declarations a bit, but I think it's 397 worth it. */ 398 399 #if __STDC__ 400 401 #define _RE_ARGS(args) args 402 403 #else /* not __STDC__ */ 404 405 #define _RE_ARGS(args) () 406 407 #endif /* not __STDC__ */ 408 409 /* Sets the current default syntax to SYNTAX, and return the old syntax. 410 You can also simply assign to the `re_syntax_options' variable. */ 411 extern reg_syntax_t re_set_syntax _RE_ARGS ((reg_syntax_t syntax)); 412 413 /* Compile the regular expression PATTERN, with length LENGTH 414 and syntax given by the global `re_syntax_options', into the buffer 415 BUFFER. Return NULL if successful, and an error string if not. */ 416 extern const char *re_compile_pattern 417 _RE_ARGS ((const char *pattern, int length, 418 struct re_pattern_buffer *buffer)); 419 420 421 /* Compile a fastmap for the compiled pattern in BUFFER; used to 422 accelerate searches. Return 0 if successful and -2 if was an 423 internal error. */ 424 extern int re_compile_fastmap _RE_ARGS ((struct re_pattern_buffer *buffer)); 425 426 427 /* Search in the string STRING (with length LENGTH) for the pattern 428 compiled into BUFFER. Start searching at position START, for RANGE 429 characters. Return the starting position of the match, -1 for no 430 match, or -2 for an internal error. Also return register 431 information in REGS (if REGS and BUFFER->no_sub are non-zero). */ 432 extern int re_search 433 _RE_ARGS ((struct re_pattern_buffer *buffer, const char *string, 434 int length, int start, int range, struct re_registers *regs)); 435 436 437 /* Like `re_search', but search in the concatenation of STRING1 and 438 STRING2. Also, stop searching at index START + STOP. */ 439 extern int re_search_2 440 _RE_ARGS ((struct re_pattern_buffer *buffer, const char *string1, 441 int length1, const char *string2, int length2, 442 int start, int range, struct re_registers *regs, int stop)); 443 444 445 /* Like `re_search', but return how many characters in STRING the regexp 446 in BUFFER matched, starting at position START. */ 447 extern int re_match 448 _RE_ARGS ((struct re_pattern_buffer *buffer, const char *string, 449 int length, int start, struct re_registers *regs)); 450 451 452 /* Relates to `re_match' as `re_search_2' relates to `re_search'. */ 453 extern int re_match_2 454 _RE_ARGS ((struct re_pattern_buffer *buffer, const char *string1, 455 int length1, const char *string2, int length2, 456 int start, struct re_registers *regs, int stop)); 457 458 459 /* Set REGS to hold NUM_REGS registers, storing them in STARTS and 460 ENDS. Subsequent matches using BUFFER and REGS will use this memory 461 for recording register information. STARTS and ENDS must be 462 allocated with malloc, and must each be at least `NUM_REGS * sizeof 463 (regoff_t)' bytes long. 464 465 If NUM_REGS == 0, then subsequent matches should allocate their own 466 register data. 467 468 Unless this function is called, the first search or match using 469 PATTERN_BUFFER will allocate its own register data, without 470 freeing the old data. */ 471 extern void re_set_registers 472 _RE_ARGS ((struct re_pattern_buffer *buffer, struct re_registers *regs, 473 unsigned num_regs, regoff_t *starts, regoff_t *ends)); 474 475 /* 4.2 bsd compatibility. */ 476 extern char *re_comp _RE_ARGS ((const char *)); 477 extern int re_exec _RE_ARGS ((const char *)); 478 479 /* POSIX compatibility. */ 480 extern int regcomp _RE_ARGS ((regex_t *preg, const char *pattern, int cflags)); 481 extern int regexec 482 _RE_ARGS ((const regex_t *preg, const char *string, size_t nmatch, 483 regmatch_t pmatch[], int eflags)); 484 extern size_t regerror 485 _RE_ARGS ((int errcode, const regex_t *preg, char *errbuf, 486 size_t errbuf_size)); 487 extern void regfree _RE_ARGS ((regex_t *preg)); 488 489 #ifdef __cplusplus 490 } 491 #endif /* C++ */ 492 493 #endif /* not __REGEXP_LIBRARY_H__ */ 494 495 /* 496 Local variables: 497 make-backup-files: t 498 version-control: t 499 trim-versions-without-asking: nil 500 End: 501 */ 502