]>
Commit | Line | Data |
---|---|---|
e9ce8d39 A |
1 | /* |
2 | * Copyright (c) 1999 Apple Computer, Inc. All rights reserved. | |
3 | * | |
4 | * @APPLE_LICENSE_HEADER_START@ | |
5 | * | |
6 | * The contents of this file constitute Original Code as defined in and | |
7 | * are subject to the Apple Public Source License Version 1.1 (the | |
8 | * "License"). You may not use this file except in compliance with the | |
9 | * License. Please obtain a copy of the License at | |
10 | * http://www.apple.com/publicsource and read it before using this file. | |
11 | * | |
12 | * This Original Code and all software distributed under the License are | |
13 | * distributed on an "AS IS" basis, WITHOUT WARRANTY OF ANY KIND, EITHER | |
14 | * EXPRESS OR IMPLIED, AND APPLE HEREBY DISCLAIMS ALL SUCH WARRANTIES, | |
15 | * INCLUDING WITHOUT LIMITATION, ANY WARRANTIES OF MERCHANTABILITY, | |
16 | * FITNESS FOR A PARTICULAR PURPOSE OR NON-INFRINGEMENT. Please see the | |
17 | * License for the specific language governing rights and limitations | |
18 | * under the License. | |
19 | * | |
20 | * @APPLE_LICENSE_HEADER_END@ | |
21 | */ | |
22 | /* | |
23 | * Copyright (c) 1992, 1993, 1994 | |
24 | * The Regents of the University of California. All rights reserved. | |
25 | * | |
26 | * This code is derived from software contributed to Berkeley by | |
27 | * Henry Spencer. | |
28 | * | |
29 | * Redistribution and use in source and binary forms, with or without | |
30 | * modification, are permitted provided that the following conditions | |
31 | * are met: | |
32 | * 1. Redistributions of source code must retain the above copyright | |
33 | * notice, this list of conditions and the following disclaimer. | |
34 | * 2. Redistributions in binary form must reproduce the above copyright | |
35 | * notice, this list of conditions and the following disclaimer in the | |
36 | * documentation and/or other materials provided with the distribution. | |
37 | * 3. All advertising materials mentioning features or use of this software | |
38 | * must display the following acknowledgement: | |
39 | * This product includes software developed by the University of | |
40 | * California, Berkeley and its contributors. | |
41 | * 4. Neither the name of the University nor the names of its contributors | |
42 | * may be used to endorse or promote products derived from this software | |
43 | * without specific prior written permission. | |
44 | * | |
45 | * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND | |
46 | * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE | |
47 | * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE | |
48 | * ARE DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE | |
49 | * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL | |
50 | * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS | |
51 | * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) | |
52 | * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT | |
53 | * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY | |
54 | * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF | |
55 | * SUCH DAMAGE. | |
56 | */ | |
57 | ||
58 | /* | |
59 | * The matching engine and friends. This file is #included by regexec.c | |
60 | * after suitable #defines of a variety of macros used herein, so that | |
61 | * different state representations can be used without duplicating masses | |
62 | * of code. | |
63 | */ | |
64 | ||
65 | #ifdef SNAMES | |
66 | #define matcher smatcher | |
67 | #define fast sfast | |
68 | #define slow sslow | |
69 | #define dissect sdissect | |
70 | #define backref sbackref | |
71 | #define step sstep | |
72 | #define print sprint | |
73 | #define at sat | |
74 | #define match smat | |
75 | #endif | |
76 | #ifdef LNAMES | |
77 | #define matcher lmatcher | |
78 | #define fast lfast | |
79 | #define slow lslow | |
80 | #define dissect ldissect | |
81 | #define backref lbackref | |
82 | #define step lstep | |
83 | #define print lprint | |
84 | #define at lat | |
85 | #define match lmat | |
86 | #endif | |
87 | ||
88 | /* another structure passed up and down to avoid zillions of parameters */ | |
89 | struct match { | |
90 | struct re_guts *g; | |
91 | int eflags; | |
92 | regmatch_t *pmatch; /* [nsub+1] (0 element unused) */ | |
93 | char *offp; /* offsets work from here */ | |
94 | char *beginp; /* start of string -- virtual NUL precedes */ | |
95 | char *endp; /* end of string -- virtual NUL here */ | |
96 | char *coldp; /* can be no match starting before here */ | |
97 | char **lastpos; /* [nplus+1] */ | |
98 | STATEVARS; | |
99 | states st; /* current states */ | |
100 | states fresh; /* states for a fresh start */ | |
101 | states tmp; /* temporary */ | |
102 | states empty; /* empty set of states */ | |
103 | }; | |
104 | ||
105 | /* ========= begin header generated by ./mkh ========= */ | |
106 | #ifdef __cplusplus | |
107 | extern "C" { | |
108 | #endif | |
109 | ||
110 | /* === engine.c === */ | |
111 | static int matcher __P((struct re_guts *g, char *string, size_t nmatch, regmatch_t pmatch[], int eflags)); | |
112 | static char *dissect __P((struct match *m, char *start, char *stop, sopno startst, sopno stopst)); | |
113 | static char *backref __P((struct match *m, char *start, char *stop, sopno startst, sopno stopst, sopno lev)); | |
114 | static char *fast __P((struct match *m, char *start, char *stop, sopno startst, sopno stopst)); | |
115 | static char *slow __P((struct match *m, char *start, char *stop, sopno startst, sopno stopst)); | |
116 | static states step __P((struct re_guts *g, sopno start, sopno stop, states bef, int ch, states aft)); | |
117 | #define BOL (OUT+1) | |
118 | #define EOL (BOL+1) | |
119 | #define BOLEOL (BOL+2) | |
120 | #define NOTHING (BOL+3) | |
121 | #define BOW (BOL+4) | |
122 | #define EOW (BOL+5) | |
123 | #define CODEMAX (BOL+5) /* highest code used */ | |
124 | #define NONCHAR(c) ((c) > CHAR_MAX) | |
125 | #define NNONCHAR (CODEMAX-CHAR_MAX) | |
126 | #ifdef REDEBUG | |
127 | static void print __P((struct match *m, char *caption, states st, int ch, FILE *d)); | |
128 | #endif | |
129 | #ifdef REDEBUG | |
130 | static void at __P((struct match *m, char *title, char *start, char *stop, sopno startst, sopno stopst)); | |
131 | #endif | |
132 | #ifdef REDEBUG | |
133 | static char *pchar __P((int ch)); | |
134 | #endif | |
135 | ||
136 | #ifdef __cplusplus | |
137 | } | |
138 | #endif | |
139 | /* ========= end header generated by ./mkh ========= */ | |
140 | ||
141 | #ifdef REDEBUG | |
142 | #define SP(t, s, c) print(m, t, s, c, stdout) | |
143 | #define AT(t, p1, p2, s1, s2) at(m, t, p1, p2, s1, s2) | |
144 | #define NOTE(str) { if (m->eflags®_TRACE) printf("=%s\n", (str)); } | |
145 | #else | |
146 | #define SP(t, s, c) /* nothing */ | |
147 | #define AT(t, p1, p2, s1, s2) /* nothing */ | |
148 | #define NOTE(s) /* nothing */ | |
149 | #endif | |
150 | ||
151 | /* | |
152 | - matcher - the actual matching engine | |
153 | == static int matcher(register struct re_guts *g, char *string, \ | |
154 | == size_t nmatch, regmatch_t pmatch[], int eflags); | |
155 | */ | |
156 | static int /* 0 success, REG_NOMATCH failure */ | |
157 | matcher(g, string, nmatch, pmatch, eflags) | |
158 | register struct re_guts *g; | |
159 | char *string; | |
160 | size_t nmatch; | |
161 | regmatch_t pmatch[]; | |
162 | int eflags; | |
163 | { | |
164 | register char *endp; | |
165 | register int i; | |
166 | struct match mv; | |
167 | register struct match *m = &mv; | |
168 | register char *dp; | |
169 | const register sopno gf = g->firststate+1; /* +1 for OEND */ | |
170 | const register sopno gl = g->laststate; | |
171 | char *start; | |
172 | char *stop; | |
173 | ||
174 | /* simplify the situation where possible */ | |
175 | if (g->cflags®_NOSUB) | |
176 | nmatch = 0; | |
177 | if (eflags®_STARTEND) { | |
178 | start = string + pmatch[0].rm_so; | |
179 | stop = string + pmatch[0].rm_eo; | |
180 | } else { | |
181 | start = string; | |
182 | stop = start + strlen(start); | |
183 | } | |
184 | if (stop < start) | |
185 | return(REG_INVARG); | |
186 | ||
187 | /* prescreening; this does wonders for this rather slow code */ | |
188 | if (g->must != NULL) { | |
189 | for (dp = start; dp < stop; dp++) | |
190 | if (*dp == g->must[0] && stop - dp >= g->mlen && | |
191 | memcmp(dp, g->must, (size_t)g->mlen) == 0) | |
192 | break; | |
193 | if (dp == stop) /* we didn't find g->must */ | |
194 | return(REG_NOMATCH); | |
195 | } | |
196 | ||
197 | /* match struct setup */ | |
198 | m->g = g; | |
199 | m->eflags = eflags; | |
200 | m->pmatch = NULL; | |
201 | m->lastpos = NULL; | |
202 | m->offp = string; | |
203 | m->beginp = start; | |
204 | m->endp = stop; | |
205 | STATESETUP(m, 4); | |
206 | SETUP(m->st); | |
207 | SETUP(m->fresh); | |
208 | SETUP(m->tmp); | |
209 | SETUP(m->empty); | |
210 | CLEAR(m->empty); | |
211 | ||
212 | /* this loop does only one repetition except for backrefs */ | |
213 | for (;;) { | |
214 | endp = fast(m, start, stop, gf, gl); | |
215 | if (endp == NULL) { /* a miss */ | |
216 | STATETEARDOWN(m); | |
217 | return(REG_NOMATCH); | |
218 | } | |
219 | if (nmatch == 0 && !g->backrefs) | |
220 | break; /* no further info needed */ | |
221 | ||
222 | /* where? */ | |
223 | assert(m->coldp != NULL); | |
224 | for (;;) { | |
225 | NOTE("finding start"); | |
226 | endp = slow(m, m->coldp, stop, gf, gl); | |
227 | if (endp != NULL) | |
228 | break; | |
229 | assert(m->coldp < m->endp); | |
230 | m->coldp++; | |
231 | } | |
232 | if (nmatch == 1 && !g->backrefs) | |
233 | break; /* no further info needed */ | |
234 | ||
235 | /* oh my, he wants the subexpressions... */ | |
236 | if (m->pmatch == NULL) | |
237 | m->pmatch = (regmatch_t *)malloc((m->g->nsub + 1) * | |
238 | sizeof(regmatch_t)); | |
239 | if (m->pmatch == NULL) { | |
240 | STATETEARDOWN(m); | |
241 | return(REG_ESPACE); | |
242 | } | |
243 | for (i = 1; i <= m->g->nsub; i++) | |
244 | m->pmatch[i].rm_so = m->pmatch[i].rm_eo = -1; | |
245 | if (!g->backrefs && !(m->eflags®_BACKR)) { | |
246 | NOTE("dissecting"); | |
247 | dp = dissect(m, m->coldp, endp, gf, gl); | |
248 | } else { | |
249 | if (g->nplus > 0 && m->lastpos == NULL) | |
250 | m->lastpos = (char **)malloc((g->nplus+1) * | |
251 | sizeof(char *)); | |
252 | if (g->nplus > 0 && m->lastpos == NULL) { | |
253 | free(m->pmatch); | |
254 | STATETEARDOWN(m); | |
255 | return(REG_ESPACE); | |
256 | } | |
257 | NOTE("backref dissect"); | |
258 | dp = backref(m, m->coldp, endp, gf, gl, (sopno)0); | |
259 | } | |
260 | if (dp != NULL) | |
261 | break; | |
262 | ||
263 | /* uh-oh... we couldn't find a subexpression-level match */ | |
264 | assert(g->backrefs); /* must be back references doing it */ | |
265 | assert(g->nplus == 0 || m->lastpos != NULL); | |
266 | for (;;) { | |
267 | if (dp != NULL || endp <= m->coldp) | |
268 | break; /* defeat */ | |
269 | NOTE("backoff"); | |
270 | endp = slow(m, m->coldp, endp-1, gf, gl); | |
271 | if (endp == NULL) | |
272 | break; /* defeat */ | |
273 | /* try it on a shorter possibility */ | |
274 | #ifndef NDEBUG | |
275 | for (i = 1; i <= m->g->nsub; i++) { | |
276 | assert(m->pmatch[i].rm_so == -1); | |
277 | assert(m->pmatch[i].rm_eo == -1); | |
278 | } | |
279 | #endif | |
280 | NOTE("backoff dissect"); | |
281 | dp = backref(m, m->coldp, endp, gf, gl, (sopno)0); | |
282 | } | |
283 | assert(dp == NULL || dp == endp); | |
284 | if (dp != NULL) /* found a shorter one */ | |
285 | break; | |
286 | ||
287 | /* despite initial appearances, there is no match here */ | |
288 | NOTE("false alarm"); | |
289 | start = m->coldp + 1; /* recycle starting later */ | |
290 | assert(start <= stop); | |
291 | } | |
292 | ||
293 | /* fill in the details if requested */ | |
294 | if (nmatch > 0) { | |
295 | pmatch[0].rm_so = m->coldp - m->offp; | |
296 | pmatch[0].rm_eo = endp - m->offp; | |
297 | } | |
298 | if (nmatch > 1) { | |
299 | assert(m->pmatch != NULL); | |
300 | for (i = 1; i < nmatch; i++) | |
301 | if (i <= m->g->nsub) | |
302 | pmatch[i] = m->pmatch[i]; | |
303 | else { | |
304 | pmatch[i].rm_so = -1; | |
305 | pmatch[i].rm_eo = -1; | |
306 | } | |
307 | } | |
308 | ||
309 | if (m->pmatch != NULL) | |
310 | free((char *)m->pmatch); | |
311 | if (m->lastpos != NULL) | |
312 | free((char *)m->lastpos); | |
313 | STATETEARDOWN(m); | |
314 | return(0); | |
315 | } | |
316 | ||
317 | /* | |
318 | - dissect - figure out what matched what, no back references | |
319 | == static char *dissect(register struct match *m, char *start, \ | |
320 | == char *stop, sopno startst, sopno stopst); | |
321 | */ | |
322 | static char * /* == stop (success) always */ | |
323 | dissect(m, start, stop, startst, stopst) | |
324 | register struct match *m; | |
325 | char *start; | |
326 | char *stop; | |
327 | sopno startst; | |
328 | sopno stopst; | |
329 | { | |
330 | register int i; | |
331 | register sopno ss; /* start sop of current subRE */ | |
332 | register sopno es; /* end sop of current subRE */ | |
333 | register char *sp; /* start of string matched by it */ | |
334 | register char *stp; /* string matched by it cannot pass here */ | |
335 | register char *rest; /* start of rest of string */ | |
336 | register char *tail; /* string unmatched by rest of RE */ | |
337 | register sopno ssub; /* start sop of subsubRE */ | |
338 | register sopno esub; /* end sop of subsubRE */ | |
339 | register char *ssp; /* start of string matched by subsubRE */ | |
340 | register char *sep; /* end of string matched by subsubRE */ | |
341 | register char *oldssp; /* previous ssp */ | |
342 | register char *dp; | |
343 | ||
344 | AT("diss", start, stop, startst, stopst); | |
345 | sp = start; | |
346 | for (ss = startst; ss < stopst; ss = es) { | |
347 | /* identify end of subRE */ | |
348 | es = ss; | |
349 | switch (OP(m->g->strip[es])) { | |
350 | case OPLUS_: | |
351 | case OQUEST_: | |
352 | es += OPND(m->g->strip[es]); | |
353 | break; | |
354 | case OCH_: | |
355 | while (OP(m->g->strip[es]) != O_CH) | |
356 | es += OPND(m->g->strip[es]); | |
357 | break; | |
358 | } | |
359 | es++; | |
360 | ||
361 | /* figure out what it matched */ | |
362 | switch (OP(m->g->strip[ss])) { | |
363 | case OEND: | |
364 | assert(nope); | |
365 | break; | |
366 | case OCHAR: | |
367 | sp++; | |
368 | break; | |
369 | case OBOL: | |
370 | case OEOL: | |
371 | case OBOW: | |
372 | case OEOW: | |
373 | break; | |
374 | case OANY: | |
375 | case OANYOF: | |
376 | sp++; | |
377 | break; | |
378 | case OBACK_: | |
379 | case O_BACK: | |
380 | assert(nope); | |
381 | break; | |
382 | /* cases where length of match is hard to find */ | |
383 | case OQUEST_: | |
384 | stp = stop; | |
385 | for (;;) { | |
386 | /* how long could this one be? */ | |
387 | rest = slow(m, sp, stp, ss, es); | |
388 | assert(rest != NULL); /* it did match */ | |
389 | /* could the rest match the rest? */ | |
390 | tail = slow(m, rest, stop, es, stopst); | |
391 | if (tail == stop) | |
392 | break; /* yes! */ | |
393 | /* no -- try a shorter match for this one */ | |
394 | stp = rest - 1; | |
395 | assert(stp >= sp); /* it did work */ | |
396 | } | |
397 | ssub = ss + 1; | |
398 | esub = es - 1; | |
399 | /* did innards match? */ | |
400 | if (slow(m, sp, rest, ssub, esub) != NULL) { | |
401 | dp = dissect(m, sp, rest, ssub, esub); | |
402 | assert(dp == rest); | |
403 | } else /* no */ | |
404 | assert(sp == rest); | |
405 | sp = rest; | |
406 | break; | |
407 | case OPLUS_: | |
408 | stp = stop; | |
409 | for (;;) { | |
410 | /* how long could this one be? */ | |
411 | rest = slow(m, sp, stp, ss, es); | |
412 | assert(rest != NULL); /* it did match */ | |
413 | /* could the rest match the rest? */ | |
414 | tail = slow(m, rest, stop, es, stopst); | |
415 | if (tail == stop) | |
416 | break; /* yes! */ | |
417 | /* no -- try a shorter match for this one */ | |
418 | stp = rest - 1; | |
419 | assert(stp >= sp); /* it did work */ | |
420 | } | |
421 | ssub = ss + 1; | |
422 | esub = es - 1; | |
423 | ssp = sp; | |
424 | oldssp = ssp; | |
425 | for (;;) { /* find last match of innards */ | |
426 | sep = slow(m, ssp, rest, ssub, esub); | |
427 | if (sep == NULL || sep == ssp) | |
428 | break; /* failed or matched null */ | |
429 | oldssp = ssp; /* on to next try */ | |
430 | ssp = sep; | |
431 | } | |
432 | if (sep == NULL) { | |
433 | /* last successful match */ | |
434 | sep = ssp; | |
435 | ssp = oldssp; | |
436 | } | |
437 | assert(sep == rest); /* must exhaust substring */ | |
438 | assert(slow(m, ssp, sep, ssub, esub) == rest); | |
439 | dp = dissect(m, ssp, sep, ssub, esub); | |
440 | assert(dp == sep); | |
441 | sp = rest; | |
442 | break; | |
443 | case OCH_: | |
444 | stp = stop; | |
445 | for (;;) { | |
446 | /* how long could this one be? */ | |
447 | rest = slow(m, sp, stp, ss, es); | |
448 | assert(rest != NULL); /* it did match */ | |
449 | /* could the rest match the rest? */ | |
450 | tail = slow(m, rest, stop, es, stopst); | |
451 | if (tail == stop) | |
452 | break; /* yes! */ | |
453 | /* no -- try a shorter match for this one */ | |
454 | stp = rest - 1; | |
455 | assert(stp >= sp); /* it did work */ | |
456 | } | |
457 | ssub = ss + 1; | |
458 | esub = ss + OPND(m->g->strip[ss]) - 1; | |
459 | assert(OP(m->g->strip[esub]) == OOR1); | |
460 | for (;;) { /* find first matching branch */ | |
461 | if (slow(m, sp, rest, ssub, esub) == rest) | |
462 | break; /* it matched all of it */ | |
463 | /* that one missed, try next one */ | |
464 | assert(OP(m->g->strip[esub]) == OOR1); | |
465 | esub++; | |
466 | assert(OP(m->g->strip[esub]) == OOR2); | |
467 | ssub = esub + 1; | |
468 | esub += OPND(m->g->strip[esub]); | |
469 | if (OP(m->g->strip[esub]) == OOR2) | |
470 | esub--; | |
471 | else | |
472 | assert(OP(m->g->strip[esub]) == O_CH); | |
473 | } | |
474 | dp = dissect(m, sp, rest, ssub, esub); | |
475 | assert(dp == rest); | |
476 | sp = rest; | |
477 | break; | |
478 | case O_PLUS: | |
479 | case O_QUEST: | |
480 | case OOR1: | |
481 | case OOR2: | |
482 | case O_CH: | |
483 | assert(nope); | |
484 | break; | |
485 | case OLPAREN: | |
486 | i = OPND(m->g->strip[ss]); | |
487 | assert(0 < i && i <= m->g->nsub); | |
488 | m->pmatch[i].rm_so = sp - m->offp; | |
489 | break; | |
490 | case ORPAREN: | |
491 | i = OPND(m->g->strip[ss]); | |
492 | assert(0 < i && i <= m->g->nsub); | |
493 | m->pmatch[i].rm_eo = sp - m->offp; | |
494 | break; | |
495 | default: /* uh oh */ | |
496 | assert(nope); | |
497 | break; | |
498 | } | |
499 | } | |
500 | ||
501 | assert(sp == stop); | |
502 | return(sp); | |
503 | } | |
504 | ||
505 | /* | |
506 | - backref - figure out what matched what, figuring in back references | |
507 | == static char *backref(register struct match *m, char *start, \ | |
508 | == char *stop, sopno startst, sopno stopst, sopno lev); | |
509 | */ | |
510 | static char * /* == stop (success) or NULL (failure) */ | |
511 | backref(m, start, stop, startst, stopst, lev) | |
512 | register struct match *m; | |
513 | char *start; | |
514 | char *stop; | |
515 | sopno startst; | |
516 | sopno stopst; | |
517 | sopno lev; /* PLUS nesting level */ | |
518 | { | |
519 | register int i; | |
520 | register sopno ss; /* start sop of current subRE */ | |
521 | register char *sp; /* start of string matched by it */ | |
522 | register sopno ssub; /* start sop of subsubRE */ | |
523 | register sopno esub; /* end sop of subsubRE */ | |
524 | register char *ssp; /* start of string matched by subsubRE */ | |
525 | register char *dp; | |
526 | register size_t len; | |
527 | register int hard; | |
528 | register sop s; | |
529 | register regoff_t offsave; | |
530 | register cset *cs; | |
531 | ||
532 | AT("back", start, stop, startst, stopst); | |
533 | sp = start; | |
534 | ||
535 | /* get as far as we can with easy stuff */ | |
536 | hard = 0; | |
537 | for (ss = startst; !hard && ss < stopst; ss++) | |
538 | switch (OP(s = m->g->strip[ss])) { | |
539 | case OCHAR: | |
540 | if (sp == stop || *sp++ != (char)OPND(s)) | |
541 | return(NULL); | |
542 | break; | |
543 | case OANY: | |
544 | if (sp == stop) | |
545 | return(NULL); | |
546 | sp++; | |
547 | break; | |
548 | case OANYOF: | |
549 | cs = &m->g->sets[OPND(s)]; | |
550 | if (sp == stop || !CHIN(cs, *sp++)) | |
551 | return(NULL); | |
552 | break; | |
553 | case OBOL: | |
554 | if ( (sp == m->beginp && !(m->eflags®_NOTBOL)) || | |
555 | (sp < m->endp && *(sp-1) == '\n' && | |
556 | (m->g->cflags®_NEWLINE)) ) | |
557 | { /* yes */ } | |
558 | else | |
559 | return(NULL); | |
560 | break; | |
561 | case OEOL: | |
562 | if ( (sp == m->endp && !(m->eflags®_NOTEOL)) || | |
563 | (sp < m->endp && *sp == '\n' && | |
564 | (m->g->cflags®_NEWLINE)) ) | |
565 | { /* yes */ } | |
566 | else | |
567 | return(NULL); | |
568 | break; | |
569 | case OBOW: | |
570 | if (( (sp == m->beginp && !(m->eflags®_NOTBOL)) || | |
571 | (sp < m->endp && *(sp-1) == '\n' && | |
572 | (m->g->cflags®_NEWLINE)) || | |
573 | (sp > m->beginp && | |
574 | !ISWORD(*(sp-1))) ) && | |
575 | (sp < m->endp && ISWORD(*sp)) ) | |
576 | { /* yes */ } | |
577 | else | |
578 | return(NULL); | |
579 | break; | |
580 | case OEOW: | |
581 | if (( (sp == m->endp && !(m->eflags®_NOTEOL)) || | |
582 | (sp < m->endp && *sp == '\n' && | |
583 | (m->g->cflags®_NEWLINE)) || | |
584 | (sp < m->endp && !ISWORD(*sp)) ) && | |
585 | (sp > m->beginp && ISWORD(*(sp-1))) ) | |
586 | { /* yes */ } | |
587 | else | |
588 | return(NULL); | |
589 | break; | |
590 | case O_QUEST: | |
591 | break; | |
592 | case OOR1: /* matches null but needs to skip */ | |
593 | ss++; | |
594 | s = m->g->strip[ss]; | |
595 | do { | |
596 | assert(OP(s) == OOR2); | |
597 | ss += OPND(s); | |
598 | } while (OP(s = m->g->strip[ss]) != O_CH); | |
599 | /* note that the ss++ gets us past the O_CH */ | |
600 | break; | |
601 | default: /* have to make a choice */ | |
602 | hard = 1; | |
603 | break; | |
604 | } | |
605 | if (!hard) { /* that was it! */ | |
606 | if (sp != stop) | |
607 | return(NULL); | |
608 | return(sp); | |
609 | } | |
610 | ss--; /* adjust for the for's final increment */ | |
611 | ||
612 | /* the hard stuff */ | |
613 | AT("hard", sp, stop, ss, stopst); | |
614 | s = m->g->strip[ss]; | |
615 | switch (OP(s)) { | |
616 | case OBACK_: /* the vilest depths */ | |
617 | i = OPND(s); | |
618 | assert(0 < i && i <= m->g->nsub); | |
619 | if (m->pmatch[i].rm_eo == -1) | |
620 | return(NULL); | |
621 | assert(m->pmatch[i].rm_so != -1); | |
622 | len = m->pmatch[i].rm_eo - m->pmatch[i].rm_so; | |
623 | assert(stop - m->beginp >= len); | |
624 | if (sp > stop - len) | |
625 | return(NULL); /* not enough left to match */ | |
626 | ssp = m->offp + m->pmatch[i].rm_so; | |
627 | if (memcmp(sp, ssp, len) != 0) | |
628 | return(NULL); | |
629 | while (m->g->strip[ss] != SOP(O_BACK, i)) | |
630 | ss++; | |
631 | return(backref(m, sp+len, stop, ss+1, stopst, lev)); | |
632 | break; | |
633 | case OQUEST_: /* to null or not */ | |
634 | dp = backref(m, sp, stop, ss+1, stopst, lev); | |
635 | if (dp != NULL) | |
636 | return(dp); /* not */ | |
637 | return(backref(m, sp, stop, ss+OPND(s)+1, stopst, lev)); | |
638 | break; | |
639 | case OPLUS_: | |
640 | assert(m->lastpos != NULL); | |
641 | assert(lev+1 <= m->g->nplus); | |
642 | m->lastpos[lev+1] = sp; | |
643 | return(backref(m, sp, stop, ss+1, stopst, lev+1)); | |
644 | break; | |
645 | case O_PLUS: | |
646 | if (sp == m->lastpos[lev]) /* last pass matched null */ | |
647 | return(backref(m, sp, stop, ss+1, stopst, lev-1)); | |
648 | /* try another pass */ | |
649 | m->lastpos[lev] = sp; | |
650 | dp = backref(m, sp, stop, ss-OPND(s)+1, stopst, lev); | |
651 | if (dp == NULL) | |
652 | return(backref(m, sp, stop, ss+1, stopst, lev-1)); | |
653 | else | |
654 | return(dp); | |
655 | break; | |
656 | case OCH_: /* find the right one, if any */ | |
657 | ssub = ss + 1; | |
658 | esub = ss + OPND(s) - 1; | |
659 | assert(OP(m->g->strip[esub]) == OOR1); | |
660 | for (;;) { /* find first matching branch */ | |
661 | dp = backref(m, sp, stop, ssub, esub, lev); | |
662 | if (dp != NULL) | |
663 | return(dp); | |
664 | /* that one missed, try next one */ | |
665 | if (OP(m->g->strip[esub]) == O_CH) | |
666 | return(NULL); /* there is none */ | |
667 | esub++; | |
668 | assert(OP(m->g->strip[esub]) == OOR2); | |
669 | ssub = esub + 1; | |
670 | esub += OPND(m->g->strip[esub]); | |
671 | if (OP(m->g->strip[esub]) == OOR2) | |
672 | esub--; | |
673 | else | |
674 | assert(OP(m->g->strip[esub]) == O_CH); | |
675 | } | |
676 | break; | |
677 | case OLPAREN: /* must undo assignment if rest fails */ | |
678 | i = OPND(s); | |
679 | assert(0 < i && i <= m->g->nsub); | |
680 | offsave = m->pmatch[i].rm_so; | |
681 | m->pmatch[i].rm_so = sp - m->offp; | |
682 | dp = backref(m, sp, stop, ss+1, stopst, lev); | |
683 | if (dp != NULL) | |
684 | return(dp); | |
685 | m->pmatch[i].rm_so = offsave; | |
686 | return(NULL); | |
687 | break; | |
688 | case ORPAREN: /* must undo assignment if rest fails */ | |
689 | i = OPND(s); | |
690 | assert(0 < i && i <= m->g->nsub); | |
691 | offsave = m->pmatch[i].rm_eo; | |
692 | m->pmatch[i].rm_eo = sp - m->offp; | |
693 | dp = backref(m, sp, stop, ss+1, stopst, lev); | |
694 | if (dp != NULL) | |
695 | return(dp); | |
696 | m->pmatch[i].rm_eo = offsave; | |
697 | return(NULL); | |
698 | break; | |
699 | default: /* uh oh */ | |
700 | assert(nope); | |
701 | break; | |
702 | } | |
703 | ||
704 | /* "can't happen" */ | |
705 | assert(nope); | |
706 | /* NOTREACHED */ | |
707 | } | |
708 | ||
709 | /* | |
710 | - fast - step through the string at top speed | |
711 | == static char *fast(register struct match *m, char *start, \ | |
712 | == char *stop, sopno startst, sopno stopst); | |
713 | */ | |
714 | static char * /* where tentative match ended, or NULL */ | |
715 | fast(m, start, stop, startst, stopst) | |
716 | register struct match *m; | |
717 | char *start; | |
718 | char *stop; | |
719 | sopno startst; | |
720 | sopno stopst; | |
721 | { | |
722 | register states st = m->st; | |
723 | register states fresh = m->fresh; | |
724 | register states tmp = m->tmp; | |
725 | register char *p = start; | |
726 | register int c = (start == m->beginp) ? OUT : *(start-1); | |
727 | register int lastc; /* previous c */ | |
728 | register int flagch; | |
729 | register int i; | |
730 | register char *coldp; /* last p after which no match was underway */ | |
731 | ||
732 | CLEAR(st); | |
733 | SET1(st, startst); | |
734 | st = step(m->g, startst, stopst, st, NOTHING, st); | |
735 | ASSIGN(fresh, st); | |
736 | SP("start", st, *p); | |
737 | coldp = NULL; | |
738 | for (;;) { | |
739 | /* next character */ | |
740 | lastc = c; | |
741 | c = (p == m->endp) ? OUT : *p; | |
742 | if (EQ(st, fresh)) | |
743 | coldp = p; | |
744 | ||
745 | /* is there an EOL and/or BOL between lastc and c? */ | |
746 | flagch = '\0'; | |
747 | i = 0; | |
748 | if ( (lastc == '\n' && m->g->cflags®_NEWLINE) || | |
749 | (lastc == OUT && !(m->eflags®_NOTBOL)) ) { | |
750 | flagch = BOL; | |
751 | i = m->g->nbol; | |
752 | } | |
753 | if ( (c == '\n' && m->g->cflags®_NEWLINE) || | |
754 | (c == OUT && !(m->eflags®_NOTEOL)) ) { | |
755 | flagch = (flagch == BOL) ? BOLEOL : EOL; | |
756 | i += m->g->neol; | |
757 | } | |
758 | if (i != 0) { | |
759 | for (; i > 0; i--) | |
760 | st = step(m->g, startst, stopst, st, flagch, st); | |
761 | SP("boleol", st, c); | |
762 | } | |
763 | ||
764 | /* how about a word boundary? */ | |
765 | if ( (flagch == BOL || (lastc != OUT && !ISWORD(lastc))) && | |
766 | (c != OUT && ISWORD(c)) ) { | |
767 | flagch = BOW; | |
768 | } | |
769 | if ( (lastc != OUT && ISWORD(lastc)) && | |
770 | (flagch == EOL || (c != OUT && !ISWORD(c))) ) { | |
771 | flagch = EOW; | |
772 | } | |
773 | if (flagch == BOW || flagch == EOW) { | |
774 | st = step(m->g, startst, stopst, st, flagch, st); | |
775 | SP("boweow", st, c); | |
776 | } | |
777 | ||
778 | /* are we done? */ | |
779 | if (ISSET(st, stopst) || p == stop) | |
780 | break; /* NOTE BREAK OUT */ | |
781 | ||
782 | /* no, we must deal with this character */ | |
783 | ASSIGN(tmp, st); | |
784 | ASSIGN(st, fresh); | |
785 | assert(c != OUT); | |
786 | st = step(m->g, startst, stopst, tmp, c, st); | |
787 | SP("aft", st, c); | |
788 | assert(EQ(step(m->g, startst, stopst, st, NOTHING, st), st)); | |
789 | p++; | |
790 | } | |
791 | ||
792 | assert(coldp != NULL); | |
793 | m->coldp = coldp; | |
794 | if (ISSET(st, stopst)) | |
795 | return(p+1); | |
796 | else | |
797 | return(NULL); | |
798 | } | |
799 | ||
800 | /* | |
801 | - slow - step through the string more deliberately | |
802 | == static char *slow(register struct match *m, char *start, \ | |
803 | == char *stop, sopno startst, sopno stopst); | |
804 | */ | |
805 | static char * /* where it ended */ | |
806 | slow(m, start, stop, startst, stopst) | |
807 | register struct match *m; | |
808 | char *start; | |
809 | char *stop; | |
810 | sopno startst; | |
811 | sopno stopst; | |
812 | { | |
813 | register states st = m->st; | |
814 | register states empty = m->empty; | |
815 | register states tmp = m->tmp; | |
816 | register char *p = start; | |
817 | register int c = (start == m->beginp) ? OUT : *(start-1); | |
818 | register int lastc; /* previous c */ | |
819 | register int flagch; | |
820 | register int i; | |
821 | register char *matchp; /* last p at which a match ended */ | |
822 | ||
823 | AT("slow", start, stop, startst, stopst); | |
824 | CLEAR(st); | |
825 | SET1(st, startst); | |
826 | SP("sstart", st, *p); | |
827 | st = step(m->g, startst, stopst, st, NOTHING, st); | |
828 | matchp = NULL; | |
829 | for (;;) { | |
830 | /* next character */ | |
831 | lastc = c; | |
832 | c = (p == m->endp) ? OUT : *p; | |
833 | ||
834 | /* is there an EOL and/or BOL between lastc and c? */ | |
835 | flagch = '\0'; | |
836 | i = 0; | |
837 | if ( (lastc == '\n' && m->g->cflags®_NEWLINE) || | |
838 | (lastc == OUT && !(m->eflags®_NOTBOL)) ) { | |
839 | flagch = BOL; | |
840 | i = m->g->nbol; | |
841 | } | |
842 | if ( (c == '\n' && m->g->cflags®_NEWLINE) || | |
843 | (c == OUT && !(m->eflags®_NOTEOL)) ) { | |
844 | flagch = (flagch == BOL) ? BOLEOL : EOL; | |
845 | i += m->g->neol; | |
846 | } | |
847 | if (i != 0) { | |
848 | for (; i > 0; i--) | |
849 | st = step(m->g, startst, stopst, st, flagch, st); | |
850 | SP("sboleol", st, c); | |
851 | } | |
852 | ||
853 | /* how about a word boundary? */ | |
854 | if ( (flagch == BOL || (lastc != OUT && !ISWORD(lastc))) && | |
855 | (c != OUT && ISWORD(c)) ) { | |
856 | flagch = BOW; | |
857 | } | |
858 | if ( (lastc != OUT && ISWORD(lastc)) && | |
859 | (flagch == EOL || (c != OUT && !ISWORD(c))) ) { | |
860 | flagch = EOW; | |
861 | } | |
862 | if (flagch == BOW || flagch == EOW) { | |
863 | st = step(m->g, startst, stopst, st, flagch, st); | |
864 | SP("sboweow", st, c); | |
865 | } | |
866 | ||
867 | /* are we done? */ | |
868 | if (ISSET(st, stopst)) | |
869 | matchp = p; | |
870 | if (EQ(st, empty) || p == stop) | |
871 | break; /* NOTE BREAK OUT */ | |
872 | ||
873 | /* no, we must deal with this character */ | |
874 | ASSIGN(tmp, st); | |
875 | ASSIGN(st, empty); | |
876 | assert(c != OUT); | |
877 | st = step(m->g, startst, stopst, tmp, c, st); | |
878 | SP("saft", st, c); | |
879 | assert(EQ(step(m->g, startst, stopst, st, NOTHING, st), st)); | |
880 | p++; | |
881 | } | |
882 | ||
883 | return(matchp); | |
884 | } | |
885 | ||
886 | ||
887 | /* | |
888 | - step - map set of states reachable before char to set reachable after | |
889 | == static states step(register struct re_guts *g, sopno start, sopno stop, \ | |
890 | == register states bef, int ch, register states aft); | |
891 | == #define BOL (OUT+1) | |
892 | == #define EOL (BOL+1) | |
893 | == #define BOLEOL (BOL+2) | |
894 | == #define NOTHING (BOL+3) | |
895 | == #define BOW (BOL+4) | |
896 | == #define EOW (BOL+5) | |
897 | == #define CODEMAX (BOL+5) // highest code used | |
898 | == #define NONCHAR(c) ((c) > CHAR_MAX) | |
899 | == #define NNONCHAR (CODEMAX-CHAR_MAX) | |
900 | */ | |
901 | static states | |
902 | step(g, start, stop, bef, ch, aft) | |
903 | register struct re_guts *g; | |
904 | sopno start; /* start state within strip */ | |
905 | sopno stop; /* state after stop state within strip */ | |
906 | register states bef; /* states reachable before */ | |
907 | int ch; /* character or NONCHAR code */ | |
908 | register states aft; /* states already known reachable after */ | |
909 | { | |
910 | register cset *cs; | |
911 | register sop s; | |
912 | register sopno pc; | |
913 | register onestate here; /* note, macros know this name */ | |
914 | register sopno look; | |
915 | register int i; | |
916 | ||
917 | for (pc = start, INIT(here, pc); pc != stop; pc++, INC(here)) { | |
918 | s = g->strip[pc]; | |
919 | switch (OP(s)) { | |
920 | case OEND: | |
921 | assert(pc == stop-1); | |
922 | break; | |
923 | case OCHAR: | |
924 | /* only characters can match */ | |
925 | assert(!NONCHAR(ch) || ch != (char)OPND(s)); | |
926 | if (ch == (char)OPND(s)) | |
927 | FWD(aft, bef, 1); | |
928 | break; | |
929 | case OBOL: | |
930 | if (ch == BOL || ch == BOLEOL) | |
931 | FWD(aft, bef, 1); | |
932 | break; | |
933 | case OEOL: | |
934 | if (ch == EOL || ch == BOLEOL) | |
935 | FWD(aft, bef, 1); | |
936 | break; | |
937 | case OBOW: | |
938 | if (ch == BOW) | |
939 | FWD(aft, bef, 1); | |
940 | break; | |
941 | case OEOW: | |
942 | if (ch == EOW) | |
943 | FWD(aft, bef, 1); | |
944 | break; | |
945 | case OANY: | |
946 | if (!NONCHAR(ch)) | |
947 | FWD(aft, bef, 1); | |
948 | break; | |
949 | case OANYOF: | |
950 | cs = &g->sets[OPND(s)]; | |
951 | if (!NONCHAR(ch) && CHIN(cs, ch)) | |
952 | FWD(aft, bef, 1); | |
953 | break; | |
954 | case OBACK_: /* ignored here */ | |
955 | case O_BACK: | |
956 | FWD(aft, aft, 1); | |
957 | break; | |
958 | case OPLUS_: /* forward, this is just an empty */ | |
959 | FWD(aft, aft, 1); | |
960 | break; | |
961 | case O_PLUS: /* both forward and back */ | |
962 | FWD(aft, aft, 1); | |
963 | i = ISSETBACK(aft, OPND(s)); | |
964 | BACK(aft, aft, OPND(s)); | |
965 | if (!i && ISSETBACK(aft, OPND(s))) { | |
966 | /* oho, must reconsider loop body */ | |
967 | pc -= OPND(s) + 1; | |
968 | INIT(here, pc); | |
969 | } | |
970 | break; | |
971 | case OQUEST_: /* two branches, both forward */ | |
972 | FWD(aft, aft, 1); | |
973 | FWD(aft, aft, OPND(s)); | |
974 | break; | |
975 | case O_QUEST: /* just an empty */ | |
976 | FWD(aft, aft, 1); | |
977 | break; | |
978 | case OLPAREN: /* not significant here */ | |
979 | case ORPAREN: | |
980 | FWD(aft, aft, 1); | |
981 | break; | |
982 | case OCH_: /* mark the first two branches */ | |
983 | FWD(aft, aft, 1); | |
984 | assert(OP(g->strip[pc+OPND(s)]) == OOR2); | |
985 | FWD(aft, aft, OPND(s)); | |
986 | break; | |
987 | case OOR1: /* done a branch, find the O_CH */ | |
988 | if (ISSTATEIN(aft, here)) { | |
989 | for (look = 1; | |
990 | OP(s = g->strip[pc+look]) != O_CH; | |
991 | look += OPND(s)) | |
992 | assert(OP(s) == OOR2); | |
993 | FWD(aft, aft, look); | |
994 | } | |
995 | break; | |
996 | case OOR2: /* propagate OCH_'s marking */ | |
997 | FWD(aft, aft, 1); | |
998 | if (OP(g->strip[pc+OPND(s)]) != O_CH) { | |
999 | assert(OP(g->strip[pc+OPND(s)]) == OOR2); | |
1000 | FWD(aft, aft, OPND(s)); | |
1001 | } | |
1002 | break; | |
1003 | case O_CH: /* just empty */ | |
1004 | FWD(aft, aft, 1); | |
1005 | break; | |
1006 | default: /* ooooops... */ | |
1007 | assert(nope); | |
1008 | break; | |
1009 | } | |
1010 | } | |
1011 | ||
1012 | return(aft); | |
1013 | } | |
1014 | ||
1015 | #ifdef REDEBUG | |
1016 | /* | |
1017 | - print - print a set of states | |
1018 | == #ifdef REDEBUG | |
1019 | == static void print(struct match *m, char *caption, states st, \ | |
1020 | == int ch, FILE *d); | |
1021 | == #endif | |
1022 | */ | |
1023 | static void | |
1024 | print(m, caption, st, ch, d) | |
1025 | struct match *m; | |
1026 | char *caption; | |
1027 | states st; | |
1028 | int ch; | |
1029 | FILE *d; | |
1030 | { | |
1031 | register struct re_guts *g = m->g; | |
1032 | register int i; | |
1033 | register int first = 1; | |
1034 | ||
1035 | if (!(m->eflags®_TRACE)) | |
1036 | return; | |
1037 | ||
1038 | fprintf(d, "%s", caption); | |
1039 | if (ch != '\0') | |
1040 | fprintf(d, " %s", pchar(ch)); | |
1041 | for (i = 0; i < g->nstates; i++) | |
1042 | if (ISSET(st, i)) { | |
1043 | fprintf(d, "%s%d", (first) ? "\t" : ", ", i); | |
1044 | first = 0; | |
1045 | } | |
1046 | fprintf(d, "\n"); | |
1047 | } | |
1048 | ||
1049 | /* | |
1050 | - at - print current situation | |
1051 | == #ifdef REDEBUG | |
1052 | == static void at(struct match *m, char *title, char *start, char *stop, \ | |
1053 | == sopno startst, sopno stopst); | |
1054 | == #endif | |
1055 | */ | |
1056 | static void | |
1057 | at(m, title, start, stop, startst, stopst) | |
1058 | struct match *m; | |
1059 | char *title; | |
1060 | char *start; | |
1061 | char *stop; | |
1062 | sopno startst; | |
1063 | sopno stopst; | |
1064 | { | |
1065 | if (!(m->eflags®_TRACE)) | |
1066 | return; | |
1067 | ||
1068 | printf("%s %s-", title, pchar(*start)); | |
1069 | printf("%s ", pchar(*stop)); | |
1070 | printf("%ld-%ld\n", (long)startst, (long)stopst); | |
1071 | } | |
1072 | ||
1073 | #ifndef PCHARDONE | |
1074 | #define PCHARDONE /* never again */ | |
1075 | /* | |
1076 | - pchar - make a character printable | |
1077 | == #ifdef REDEBUG | |
1078 | == static char *pchar(int ch); | |
1079 | == #endif | |
1080 | * | |
1081 | * Is this identical to regchar() over in debug.c? Well, yes. But a | |
1082 | * duplicate here avoids having a debugging-capable regexec.o tied to | |
1083 | * a matching debug.o, and this is convenient. It all disappears in | |
1084 | * the non-debug compilation anyway, so it doesn't matter much. | |
1085 | */ | |
1086 | static char * /* -> representation */ | |
1087 | pchar(ch) | |
1088 | int ch; | |
1089 | { | |
1090 | static char pbuf[10]; | |
1091 | ||
1092 | if (isprint(ch) || ch == ' ') | |
1093 | sprintf(pbuf, "%c", ch); | |
1094 | else | |
1095 | sprintf(pbuf, "\\%o", ch); | |
1096 | return(pbuf); | |
1097 | } | |
1098 | #endif | |
1099 | #endif | |
1100 | ||
1101 | #undef matcher | |
1102 | #undef fast | |
1103 | #undef slow | |
1104 | #undef dissect | |
1105 | #undef backref | |
1106 | #undef step | |
1107 | #undef print | |
1108 | #undef at | |
1109 | #undef match |