projects / wxWidgets.git / blob
? search:
summary | shortlog | log | commit | commitdiff | tree
blame | history | raw | HEAD
Fix for [ 877037 ] It is no longer possible to build using wxWindows.dsw
[wxWidgets.git] / src / regex / regcomp.c
1 /*
2 * re_*comp and friends - compile REs
3 * This file #includes several others (see the bottom).
4 *
5 * Copyright (c) 1998, 1999 Henry Spencer. All rights reserved.
6 *
7 * Development of this software was funded, in part, by Cray Research Inc.,
8 * UUNET Communications Services Inc., Sun Microsystems Inc., and Scriptics
9 * Corporation, none of whom are responsible for the results. The author
10 * thanks all of them.
11 *
12 * Redistribution and use in source and binary forms -- with or without
13 * modification -- are permitted for any purpose, provided that
14 * redistributions in source form retain this entire copyright notice and
15 * indicate the origin and nature of any modifications.
16 *
17 * I'd appreciate being given credit for this package in the documentation
18 * of software which uses it, but that is not a requirement.
19 *
20 * THIS SOFTWARE IS PROVIDED ``AS IS'' AND ANY EXPRESS OR IMPLIED WARRANTIES,
21 * INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY
22 * AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL
23 * HENRY SPENCER BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL,
24 * EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO,
25 * PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS;
26 * OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY,
27 * WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR
28 * OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF
29 * ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
30 *
31 * $Header: /projects/cvsroot/pgsql-server/src/backend/regex/regcomp.c,v 1.38 2003/08/08 21:41:56 momjian Exp $
32 *
33 */
34
35 #include "regguts.h"
36
37 /*
38 * forward declarations, up here so forward datatypes etc. are defined early
39 */
40 /* === regcomp.c === */
41 static void moresubs(struct vars *, int);
42 static int freev(struct vars *, int);
43 static void makesearch(struct vars *, struct nfa *);
44 static struct subre *parse(struct vars *, int, int, struct state *, struct state *);
45 static struct subre *parsebranch(struct vars *, int, int, struct state *, struct state *, int);
46 static void parseqatom(struct vars *, int, int, struct state *, struct state *, struct subre *);
47 static void nonword(struct vars *, int, struct state *, struct state *);
48 static void word(struct vars *, int, struct state *, struct state *);
49 static int scannum(struct vars *);
50 static void repeat(struct vars *, struct state *, struct state *, int, int);
51 static void bracket(struct vars *, struct state *, struct state *);
52 static void cbracket(struct vars *, struct state *, struct state *);
53 static void brackpart(struct vars *, struct state *, struct state *);
54 static chr *scanplain(struct vars *);
55 static void leaders(struct vars *, struct cvec *);
56 static void onechr(struct vars *, chr, struct state *, struct state *);
57 static void dovec(struct vars *, struct cvec *, struct state *, struct state *);
58 static celt nextleader(struct vars *, chr, chr);
59 static void wordchrs(struct vars *);
60 static struct subre *subre(struct vars *, int, int, struct state *, struct state *);
61 static void freesubre(struct vars *, struct subre *);
62 static void freesrnode(struct vars *, struct subre *);
63 static void optst(struct vars *, struct subre *);
64 static int numst(struct subre *, int);
65 static void markst(struct subre *);
66 static void cleanst(struct vars *);
67 static long nfatree(struct vars *, struct subre *, FILE *);
68 static long nfanode(struct vars *, struct subre *, FILE *);
69 static int newlacon(struct vars *, struct state *, struct state *, int);
70 static void freelacons(struct subre *, int);
71 static void rfree(regex_t *);
72
73 #ifdef REG_DEBUG
74 static void dump(regex_t *, FILE *);
75 static void dumpst(struct subre *, FILE *, int);
76 static void stdump(struct subre *, FILE *, int);
77 static char *stid(struct subre *, char *, size_t);
78 #endif
79 /* === regc_lex.c === */
80 static void lexstart(struct vars *);
81 static void prefixes(struct vars *);
82 static void lexnest(struct vars *, chr *, chr *);
83 static void lexword(struct vars *);
84 static int next(struct vars *);
85 static int lexescape(struct vars *);
86 static chr lexdigits(struct vars *, int, int, int);
87 static int brenext(struct vars *, chr);
88 static void skip(struct vars *);
89 static chr newline(void);
90 static chr chrnamed(struct vars *, chr *, chr *, chr);
91
92 /* === regc_color.c === */
93 static void initcm(struct vars *, struct colormap *);
94 static void freecm(struct colormap *);
95 static void cmtreefree(struct colormap *, union tree *, int);
96 static color setcolor(struct colormap *, chr, pcolor);
97 static color maxcolor(struct colormap *);
98 static color newcolor(struct colormap *);
99 static void freecolor(struct colormap *, pcolor);
100 static color pseudocolor(struct colormap *);
101 static color subcolor(struct colormap *, chr c);
102 static color newsub(struct colormap *, pcolor);
103 static void subrange(struct vars *, chr, chr, struct state *, struct state *);
104 static void subblock(struct vars *, chr, struct state *, struct state *);
105 static void okcolors(struct nfa *, struct colormap *);
106 static void colorchain(struct colormap *, struct arc *);
107 static void uncolorchain(struct colormap *, struct arc *);
108 static int singleton(struct colormap *, chr c);
109 static void rainbow(struct nfa *, struct colormap *, int, pcolor, struct state *, struct state *);
110 static void colorcomplement(struct nfa *, struct colormap *, int, struct state *, struct state *, struct state *);
111
112 #ifdef REG_DEBUG
113 static void dumpcolors(struct colormap *, FILE *);
114 static void fillcheck(struct colormap *, union tree *, int, FILE *);
115 static void dumpchr(chr, FILE *);
116 #endif
117 /* === regc_nfa.c === */
118 static struct nfa *newnfa(struct vars *, struct colormap *, struct nfa *);
119 static void freenfa(struct nfa *);
120 static struct state *newstate(struct nfa *);
121 static struct state *newfstate(struct nfa *, int flag);
122 static void dropstate(struct nfa *, struct state *);
123 static void freestate(struct nfa *, struct state *);
124 static void destroystate(struct nfa *, struct state *);
125 static void newarc(struct nfa *, int, pcolor, struct state *, struct state *);
126 static struct arc *allocarc(struct nfa *, struct state *);
127 static void freearc(struct nfa *, struct arc *);
128 static struct arc *findarc(struct state *, int, pcolor);
129 static void cparc(struct nfa *, struct arc *, struct state *, struct state *);
130 static void moveins(struct nfa *, struct state *, struct state *);
131 static void copyins(struct nfa *, struct state *, struct state *);
132 static void moveouts(struct nfa *, struct state *, struct state *);
133 static void copyouts(struct nfa *, struct state *, struct state *);
134 static void cloneouts(struct nfa *, struct state *, struct state *, struct state *, int);
135 static void delsub(struct nfa *, struct state *, struct state *);
136 static void deltraverse(struct nfa *, struct state *, struct state *);
137 static void dupnfa(struct nfa *, struct state *, struct state *, struct state *, struct state *);
138 static void duptraverse(struct nfa *, struct state *, struct state *);
139 static void cleartraverse(struct nfa *, struct state *);
140 static void specialcolors(struct nfa *);
141 static long optimize(struct nfa *, FILE *);
142 static void pullback(struct nfa *, FILE *);
143 static int pull(struct nfa *, struct arc *);
144 static void pushfwd(struct nfa *, FILE *);
145 static int push(struct nfa *, struct arc *);
146
147 #define INCOMPATIBLE 1 /* destroys arc */
148 #define SATISFIED 2 /* constraint satisfied */
149 #define COMPATIBLE 3 /* compatible but not satisfied yet */
150 static int combine(struct arc *, struct arc *);
151 static void fixempties(struct nfa *, FILE *);
152 static int unempty(struct nfa *, struct arc *);
153 static void cleanup(struct nfa *);
154 static void markreachable(struct nfa *, struct state *, struct state *, struct state *);
155 static void markcanreach(struct nfa *, struct state *, struct state *, struct state *);
156 static long analyze(struct nfa *);
157 static void compact(struct nfa *, struct cnfa *);
158 static void carcsort(struct carc *, struct carc *);
159 static void freecnfa(struct cnfa *);
160 static void dumpnfa(struct nfa *, FILE *);
161
162 #ifdef REG_DEBUG
163 static void dumpstate(struct state *, FILE *);
164 static void dumparcs(struct state *, FILE *);
165 static int dumprarcs(struct arc *, struct state *, FILE *, int);
166 static void dumparc(struct arc *, struct state *, FILE *);
167 static void dumpcnfa(struct cnfa *, FILE *);
168 static void dumpcstate(int, struct carc *, struct cnfa *, FILE *);
169 #endif
170 /* === regc_cvec.c === */
171 static struct cvec *newcvec(int, int, int);
172 static struct cvec *clearcvec(struct cvec *);
173 static void addchr(struct cvec *, chr);
174 static void addrange(struct cvec *, chr, chr);
175 static void addmcce(struct cvec *, chr *, chr *);
176 static int haschr(struct cvec *, chr);
177 static struct cvec *getcvec(struct vars *, int, int, int);
178 static void freecvec(struct cvec *);
179
180 /* === regc_locale.c === */
181 extern int wx_isdigit(wx_wchar c);
182 extern int wx_isalpha(wx_wchar c);
183 extern int wx_isalnum(wx_wchar c);
184 extern int wx_isupper(wx_wchar c);
185 extern int wx_islower(wx_wchar c);
186 extern int wx_isgraph(wx_wchar c);
187 extern int wx_ispunct(wx_wchar c);
188 extern int wx_isspace(wx_wchar c);
189 extern wx_wchar wx_toupper(wx_wchar c);
190 extern wx_wchar wx_tolower(wx_wchar c);
191 static int nmcces(struct vars *);
192 static int nleaders(struct vars *);
193 static struct cvec *allmcces(struct vars *, struct cvec *);
194 static celt element(struct vars *, chr *, chr *);
195 static struct cvec *range(struct vars *, celt, celt, int);
196 static int before(celt, celt);
197 static struct cvec *eclass(struct vars *, celt, int);
198 static struct cvec *cclass(struct vars *, chr *, chr *, int);
199 static struct cvec *allcases(struct vars *, chr);
200 static int cmp(const chr *, const chr *, size_t);
201 static int casecmp(const chr *, const chr *, size_t);
202
203
204 /* internal variables, bundled for easy passing around */
205 struct vars
206 {
207 regex_t *re;
208 chr *now; /* scan pointer into string */
209 chr *stop; /* end of string */
210 chr *savenow; /* saved now and stop for "subroutine
211 * call" */
212 chr *savestop;
213 int err; /* error code (0 if none) */
214 int cflags; /* copy of compile flags */
215 int lasttype; /* type of previous token */
216 int nexttype; /* type of next token */
217 chr nextvalue; /* value (if any) of next token */
218 int lexcon; /* lexical context type (see lex.c) */
219 int nsubexp; /* subexpression count */
220 struct subre **subs; /* subRE pointer vector */
221 size_t nsubs; /* length of vector */
222 struct subre *sub10[10]; /* initial vector, enough for most */
223 struct nfa *nfa; /* the NFA */
224 struct colormap *cm; /* character color map */
225 color nlcolor; /* color of newline */
226 struct state *wordchrs; /* state in nfa holding word-char outarcs */
227 struct subre *tree; /* subexpression tree */
228 struct subre *treechain; /* all tree nodes allocated */
229 struct subre *treefree; /* any free tree nodes */
230 int ntree; /* number of tree nodes */
231 struct cvec *cv; /* interface cvec */
232 struct cvec *cv2; /* utility cvec */
233 struct cvec *mcces; /* collating-element information */
234 #define ISCELEADER(v,c) (v->mcces != NULL && haschr(v->mcces, (c)))
235 struct state *mccepbegin; /* in nfa, start of MCCE prototypes */
236 struct state *mccepend; /* in nfa, end of MCCE prototypes */
237 struct subre *lacons; /* lookahead-constraint vector */
238 int nlacons; /* size of lacons */
239 };
240
241 /* parsing macros; most know that `v' is the struct vars pointer */
242 #define NEXT() (next(v)) /* advance by one token */
243 #define SEE(t) (v->nexttype == (t)) /* is next token this? */
244 #define EAT(t) (SEE(t) && next(v)) /* if next is this, swallow it */
245 #define VISERR(vv) ((vv)->err != 0) /* have we seen an error yet? */
246 #define ISERR() VISERR(v)
247 #define VERR(vv,e) ((vv)->nexttype = EOS, ((vv)->err) ? (vv)->err :\
248 ((vv)->err = (e)))
249 #define ERR(e) VERR(v, e) /* record an error */
250 #define NOERR() {if (ISERR()) return;} /* if error seen, return */
251 #define NOERRN() {if (ISERR()) return NULL;} /* NOERR with retval */
252 #define NOERRZ() {if (ISERR()) return 0;} /* NOERR with retval */
253 #define INSIST(c, e) ((c) ? 0 : ERR(e)) /* if condition false,
254 * error */
255 #define NOTE(b) (v->re->re_info |= (b)) /* note visible condition */
256 #define EMPTYARC(x, y) newarc(v->nfa, EMPTY, 0, x, y)
257
258 /* token type codes, some also used as NFA arc types */
259 #define EMPTY 'n' /* no token present */
260 #define EOS 'e' /* end of string */
261 #define PLAIN 'p' /* ordinary character */
262 #define DIGIT 'd' /* digit (in bound) */
263 #define BACKREF 'b' /* back reference */
264 #define COLLEL 'I' /* start of [. */
265 #define ECLASS 'E' /* start of [= */
266 #define CCLASS 'C' /* start of [: */
267 #define END 'X' /* end of [. [= [: */
268 #define RANGE 'R' /* - within [] which might be range delim. */
269 #define LACON 'L' /* lookahead constraint subRE */
270 #define AHEAD 'a' /* color-lookahead arc */
271 #define BEHIND 'r' /* color-lookbehind arc */
272 #define WBDRY 'w' /* word boundary constraint */
273 #define NWBDRY 'W' /* non-word-boundary constraint */
274 #define SBEGIN 'A' /* beginning of string (even if not BOL) */
275 #define SEND 'Z' /* end of string (even if not EOL) */
276 #define PREFER 'P' /* length preference */
277
278 /* is an arc colored, and hence on a color chain? */
279 #define COLORED(a) ((a)->type == PLAIN || (a)->type == AHEAD || \
280 (a)->type == BEHIND)
281
282
283
284 /* static function list */
285 static struct fns functions = {
286 rfree, /* regfree insides */
287 };
288
289
290
291 /*
292 * regcomp - compile regular expression
293 */
294 int
295 regcomp(regex_t *re,
296 const chr *string,
297 int flags)
298 {
299 return wx_regcomp(re, string, wx_strlen(string), flags);
300 }
301 int
302 wx_regcomp(regex_t *re,
303 const chr *string,
304 size_t len,
305 int flags)
306 {
307 struct vars var;
308 struct vars *v = &var;
309 struct guts *g;
310 int i;
311 size_t j;
312
313 #ifdef REG_DEBUG
314 FILE *debug = (flags & REG_PROGRESS) ? stdout : (FILE *) NULL;
315
316 #else
317 FILE *debug = (FILE *) NULL;
318 #endif
319
320 #define CNOERR() { if (ISERR()) return freev(v, v->err); }
321
322 /* sanity checks */
323
324 if (re == NULL || string == NULL)
325 return REG_INVARG;
326 if ((flags & REG_QUOTE) &&
327 (flags & (REG_ADVANCED | REG_EXPANDED | REG_NEWLINE)))
328 return REG_INVARG;
329 if (!(flags & REG_EXTENDED) && (flags & REG_ADVF))
330 return REG_INVARG;
331
332 /* initial setup (after which freev() is callable) */
333 v->re = re;
334 v->now = (chr *) string;
335 v->stop = v->now + len;
336 v->savenow = v->savestop = NULL;
337 v->err = 0;
338 v->cflags = flags;
339 v->nsubexp = 0;
340 v->subs = v->sub10;
341 v->nsubs = 10;
342 for (j = 0; j < v->nsubs; j++)
343 v->subs[j] = NULL;
344 v->nfa = NULL;
345 v->cm = NULL;
346 v->nlcolor = COLORLESS;
347 v->wordchrs = NULL;
348 v->tree = NULL;
349 v->treechain = NULL;
350 v->treefree = NULL;
351 v->cv = NULL;
352 v->cv2 = NULL;
353 v->mcces = NULL;
354 v->lacons = NULL;
355 v->nlacons = 0;
356 re->re_magic = REMAGIC;
357 re->re_info = 0; /* bits get set during parse */
358 re->re_csize = sizeof(chr);
359 re->re_guts = NULL;
360 re->re_fns = VS(&functions);
361
362 /* more complex setup, malloced things */
363 re->re_guts = VS(MALLOC(sizeof(struct guts)));
364 if (re->re_guts == NULL)
365 return freev(v, REG_ESPACE);
366 g = (struct guts *) re->re_guts;
367 g->tree = NULL;
368 initcm(v, &g->cmap);
369 v->cm = &g->cmap;
370 g->lacons = NULL;
371 g->nlacons = 0;
372 ZAPCNFA(g->search);
373 v->nfa = newnfa(v, v->cm, (struct nfa *) NULL);
374 CNOERR();
375 v->cv = newcvec(100, 20, 10);
376 if (v->cv == NULL)
377 return freev(v, REG_ESPACE);
378 i = nmcces(v);
379 if (i > 0)
380 {
381 v->mcces = newcvec(nleaders(v), 0, i);
382 CNOERR();
383 v->mcces = allmcces(v, v->mcces);
384 leaders(v, v->mcces);
385 addmcce(v->mcces, (chr *) NULL, (chr *) NULL); /* dummy */
386 }
387 CNOERR();
388
389 /* parsing */
390 lexstart(v); /* also handles prefixes */
391 if ((v->cflags & REG_NLSTOP) || (v->cflags & REG_NLANCH))
392 {
393 /* assign newline a unique color */
394 v->nlcolor = subcolor(v->cm, newline());
395 okcolors(v->nfa, v->cm);
396 }
397 CNOERR();
398 v->tree = parse(v, EOS, PLAIN, v->nfa->init, v->nfa->final);
399 assert(SEE(EOS)); /* even if error; ISERR() => SEE(EOS) */
400 CNOERR();
401 assert(v->tree != NULL);
402
403 /* finish setup of nfa and its subre tree */
404 specialcolors(v->nfa);
405 CNOERR();
406 #ifdef REG_DEBUG
407 if (debug != NULL)
408 {
409 fprintf(debug, "\n\n\n========= RAW ==========\n");
410 dumpnfa(v->nfa, debug);
411 dumpst(v->tree, debug, 1);
412 }
413 #endif
414 optst(v, v->tree);
415 v->ntree = numst(v->tree, 1);
416 markst(v->tree);
417 cleanst(v);
418 #ifdef REG_DEBUG
419 if (debug != NULL)
420 {
421 fprintf(debug, "\n\n\n========= TREE FIXED ==========\n");
422 dumpst(v->tree, debug, 1);
423 }
424 #endif
425
426 /* build compacted NFAs for tree and lacons */
427 re->re_info |= nfatree(v, v->tree, debug);
428 CNOERR();
429 assert(v->nlacons == 0 || v->lacons != NULL);
430 for (i = 1; i < v->nlacons; i++)
431 {
432 #ifdef REG_DEBUG
433 if (debug != NULL)
434 fprintf(debug, "\n\n\n========= LA%d ==========\n", i);
435 #endif
436 nfanode(v, &v->lacons[i], debug);
437 }
438 CNOERR();
439 if (v->tree->flags & SHORTER)
440 NOTE(REG_USHORTEST);
441
442 /* build compacted NFAs for tree, lacons, fast search */
443 #ifdef REG_DEBUG
444 if (debug != NULL)
445 fprintf(debug, "\n\n\n========= SEARCH ==========\n");
446 #endif
447 /* can sacrifice main NFA now, so use it as work area */
448 (DISCARD) optimize(v->nfa, debug);
449 CNOERR();
450 makesearch(v, v->nfa);
451 CNOERR();
452 compact(v->nfa, &g->search);
453 CNOERR();
454
455 /* looks okay, package it up */
456 re->re_nsub = v->nsubexp;
457 v->re = NULL; /* freev no longer frees re */
458 g->magic = GUTSMAGIC;
459 g->cflags = v->cflags;
460 g->info = re->re_info;
461 g->nsub = re->re_nsub;
462 g->tree = v->tree;
463 v->tree = NULL;
464 g->ntree = v->ntree;
465 g->compare = (v->cflags & REG_ICASE) ? casecmp : cmp;
466 g->lacons = v->lacons;
467 v->lacons = NULL;
468 g->nlacons = v->nlacons;
469
470 #ifdef REG_DEBUG
471 if (flags & REG_DUMP)
472 dump(re, stdout);
473 #endif
474
475 assert(v->err == 0);
476 return freev(v, 0);
477 }
478
479 /*
480 * moresubs - enlarge subRE vector
481 */
482 static void
483 moresubs(struct vars * v,
484 int wanted) /* want enough room for this one */
485 {
486 struct subre **p;
487 size_t n;
488
489 assert(wanted > 0 && (size_t) wanted >= v->nsubs);
490 n = (size_t) wanted *3 / 2 + 1;
491
492 if (v->subs == v->sub10)
493 {
494 p = (struct subre **) MALLOC(n * sizeof(struct subre *));
495 if (p != NULL)
496 memcpy(VS(p), VS(v->subs),
497 v->nsubs * sizeof(struct subre *));
498 }
499 else
500 p = (struct subre **) REALLOC(v->subs, n * sizeof(struct subre *));
501 if (p == NULL)
502 {
503 ERR(REG_ESPACE);
504 return;
505 }
506 v->subs = p;
507 for (p = &v->subs[v->nsubs]; v->nsubs < n; p++, v->nsubs++)
508 *p = NULL;
509 assert(v->nsubs == n);
510 assert((size_t) wanted < v->nsubs);
511 }
512
513 /*
514 * freev - free vars struct's substructures where necessary
515 *
516 * Optionally does error-number setting, and always returns error code
517 * (if any), to make error-handling code terser.
518 */
519 static int
520 freev(struct vars * v,
521 int err)
522 {
523 if (v->re != NULL)
524 rfree(v->re);
525 if (v->subs != v->sub10)
526 FREE(v->subs);
527 if (v->nfa != NULL)
528 freenfa(v->nfa);
529 if (v->tree != NULL)
530 freesubre(v, v->tree);
531 if (v->treechain != NULL)
532 cleanst(v);
533 if (v->cv != NULL)
534 freecvec(v->cv);
535 if (v->cv2 != NULL)
536 freecvec(v->cv2);
537 if (v->mcces != NULL)
538 freecvec(v->mcces);
539 if (v->lacons != NULL)
540 freelacons(v->lacons, v->nlacons);
541 ERR(err); /* nop if err==0 */
542
543 return v->err;
544 }
545
546 /*
547 * makesearch - turn an NFA into a search NFA (implicit prepend of .*?)
548 * NFA must have been optimize()d already.
549 */
550 static void
551 makesearch(struct vars * v,
552 struct nfa * nfa)
553 {
554 struct arc *a;
555 struct arc *b;
556 struct state *pre = nfa->pre;
557 struct state *s;
558 struct state *s2;
559 struct state *slist;
560
561 /* no loops are needed if it's anchored */
562 for (a = pre->outs; a != NULL; a = a->outchain)
563 {
564 assert(a->type == PLAIN);
565 if (a->co != nfa->bos[0] && a->co != nfa->bos[1])
566 break;
567 }
568 if (a != NULL)
569 {
570 /* add implicit .* in front */
571 rainbow(nfa, v->cm, PLAIN, COLORLESS, pre, pre);
572
573 /* and ^* and \A* too -- not always necessary, but harmless */
574 newarc(nfa, PLAIN, nfa->bos[0], pre, pre);
575 newarc(nfa, PLAIN, nfa->bos[1], pre, pre);
576 }
577
578 /*
579 * Now here's the subtle part. Because many REs have no lookback
580 * constraints, often knowing when you were in the pre state tells you
581 * little; it's the next state(s) that are informative. But some of
582 * them may have other inarcs, i.e. it may be possible to make actual
583 * progress and then return to one of them. We must de-optimize such
584 * cases, splitting each such state into progress and no-progress
585 * states.
586 */
587
588 /* first, make a list of the states */
589 slist = NULL;
590 for (a = pre->outs; a != NULL; a = a->outchain)
591 {
592 s = a->to;
593 for (b = s->ins; b != NULL; b = b->inchain)
594 if (b->from != pre)
595 break;
596 if (b != NULL)
597 { /* must be split */
598 s->tmp = slist;
599 slist = s;
600 }
601 }
602
603 /* do the splits */
604 for (s = slist; s != NULL; s = s2)
605 {
606 s2 = newstate(nfa);
607 copyouts(nfa, s, s2);
608 for (a = s->ins; a != NULL; a = b)
609 {
610 b = a->inchain;
611 if (a->from != pre)
612 {
613 cparc(nfa, a, a->from, s2);
614 freearc(nfa, a);
615 }
616 }
617 s2 = s->tmp;
618 s->tmp = NULL; /* clean up while we're at it */
619 }
620 }
621
622 /*
623 * parse - parse an RE
624 *
625 * This is actually just the top level, which parses a bunch of branches
626 * tied together with '|'. They appear in the tree as the left children
627 * of a chain of '|' subres.
628 */
629 static struct subre *
630 parse(struct vars * v,
631 int stopper, /* EOS or ')' */
632 int type, /* LACON (lookahead subRE) or PLAIN */
633 struct state * init, /* initial state */
634 struct state * final) /* final state */
635 {
636 struct state *left; /* scaffolding for branch */
637 struct state *right;
638 struct subre *branches; /* top level */
639 struct subre *branch; /* current branch */
640 struct subre *t; /* temporary */
641 int firstbranch; /* is this the first branch? */
642
643 assert(stopper == ')' || stopper == EOS);
644
645 branches = subre(v, '|', LONGER, init, final);
646 NOERRN();
647 branch = branches;
648 firstbranch = 1;
649 do
650 { /* a branch */
651 if (!firstbranch)
652 {
653 /* need a place to hang it */
654 branch->right = subre(v, '|', LONGER, init, final);
655 NOERRN();
656 branch = branch->right;
657 }
658 firstbranch = 0;
659 left = newstate(v->nfa);
660 right = newstate(v->nfa);
661 NOERRN();
662 EMPTYARC(init, left);
663 EMPTYARC(right, final);
664 NOERRN();
665 branch->left = parsebranch(v, stopper, type, left, right, 0);
666 NOERRN();
667 branch->flags |= UP(branch->flags | branch->left->flags);
668 if ((branch->flags & ~branches->flags) != 0) /* new flags */
669 for (t = branches; t != branch; t = t->right)
670 t->flags |= branch->flags;
671 } while (EAT('|'));
672 assert(SEE(stopper) || SEE(EOS));
673
674 if (!SEE(stopper))
675 {
676 assert(stopper == ')' && SEE(EOS));
677 ERR(REG_EPAREN);
678 }
679
680 /* optimize out simple cases */
681 if (branch == branches)
682 { /* only one branch */
683 assert(branch->right == NULL);
684 t = branch->left;
685 branch->left = NULL;
686 freesubre(v, branches);
687 branches = t;
688 }
689 else if (!MESSY(branches->flags))
690 { /* no interesting innards */
691 freesubre(v, branches->left);
692 branches->left = NULL;
693 freesubre(v, branches->right);
694 branches->right = NULL;
695 branches->op = '=';
696 }
697
698 return branches;
699 }
700
701 /*
702 * parsebranch - parse one branch of an RE
703 *
704 * This mostly manages concatenation, working closely with parseqatom().
705 * Concatenated things are bundled up as much as possible, with separate
706 * ',' nodes introduced only when necessary due to substructure.
707 */
708 static struct subre *
709 parsebranch(struct vars * v,
710 int stopper, /* EOS or ')' */
711 int type, /* LACON (lookahead subRE) or PLAIN */
712 struct state * left, /* leftmost state */
713 struct state * right, /* rightmost state */
714 int partial) /* is this only part of a branch? */
715 {
716 struct state *lp; /* left end of current construct */
717 int seencontent; /* is there anything in this branch yet? */
718 struct subre *t;
719
720 lp = left;
721 seencontent = 0;
722 t = subre(v, '=', 0, left, right); /* op '=' is tentative */
723 NOERRN();
724 while (!SEE('|') && !SEE(stopper) && !SEE(EOS))
725 {
726 if (seencontent)
727 { /* implicit concat operator */
728 lp = newstate(v->nfa);
729 NOERRN();
730 moveins(v->nfa, right, lp);
731 }
732 seencontent = 1;
733
734 /* NB, recursion in parseqatom() may swallow rest of branch */
735 parseqatom(v, stopper, type, lp, right, t);
736 }
737
738 if (!seencontent)
739 { /* empty branch */
740 if (!partial)
741 NOTE(REG_UUNSPEC);
742 assert(lp == left);
743 EMPTYARC(left, right);
744 }
745
746 return t;
747 }
748
749 /*
750 * parseqatom - parse one quantified atom or constraint of an RE
751 *
752 * The bookkeeping near the end cooperates very closely with parsebranch();
753 * in particular, it contains a recursion that can involve parsing the rest
754 * of the branch, making this function's name somewhat inaccurate.
755 */
756 static void
757 parseqatom(struct vars * v,
758 int stopper, /* EOS or ')' */
759 int type, /* LACON (lookahead subRE) or PLAIN */
760 struct state * lp, /* left state to hang it on */
761 struct state * rp, /* right state to hang it on */
762 struct subre * top) /* subtree top */
763 {
764 struct state *s; /* temporaries for new states */
765 struct state *s2;
766
767 #define ARCV(t, val) newarc(v->nfa, t, val, lp, rp)
768 int m,
769 n;
770 struct subre *atom; /* atom's subtree */
771 struct subre *t;
772 int cap; /* capturing parens? */
773 int pos; /* positive lookahead? */
774 int subno; /* capturing-parens or backref number */
775 int atomtype;
776 int qprefer; /* quantifier short/long preference */
777 int f;
778 struct subre **atomp; /* where the pointer to atom is */
779
780 /* initial bookkeeping */
781 atom = NULL;
782 assert(lp->nouts == 0); /* must string new code */
783 assert(rp->nins == 0); /* between lp and rp */
784 subno = 0; /* just to shut lint up */
785
786 /* an atom or constraint... */
787 atomtype = v->nexttype;
788 switch (atomtype)
789 {
790 /* first, constraints, which end by returning */
791 case '^':
792 ARCV('^', 1);
793 if (v->cflags & REG_NLANCH)
794 ARCV(BEHIND, v->nlcolor);
795 NEXT();
796 return;
797 break;
798 case '$':
799 ARCV('$', 1);
800 if (v->cflags & REG_NLANCH)
801 ARCV(AHEAD, v->nlcolor);
802 NEXT();
803 return;
804 break;
805 case SBEGIN:
806 ARCV('^', 1); /* BOL */
807 ARCV('^', 0); /* or BOS */
808 NEXT();
809 return;
810 break;
811 case SEND:
812 ARCV('$', 1); /* EOL */
813 ARCV('$', 0); /* or EOS */
814 NEXT();
815 return;
816 break;
817 case '<':
818 wordchrs(v); /* does NEXT() */
819 s = newstate(v->nfa);
820 NOERR();
821 nonword(v, BEHIND, lp, s);
822 word(v, AHEAD, s, rp);
823 return;
824 break;
825 case '>':
826 wordchrs(v); /* does NEXT() */
827 s = newstate(v->nfa);
828 NOERR();
829 word(v, BEHIND, lp, s);
830 nonword(v, AHEAD, s, rp);
831 return;
832 break;
833 case WBDRY:
834 wordchrs(v); /* does NEXT() */
835 s = newstate(v->nfa);
836 NOERR();
837 nonword(v, BEHIND, lp, s);
838 word(v, AHEAD, s, rp);
839 s = newstate(v->nfa);
840 NOERR();
841 word(v, BEHIND, lp, s);
842 nonword(v, AHEAD, s, rp);
843 return;
844 break;
845 case NWBDRY:
846 wordchrs(v); /* does NEXT() */
847 s = newstate(v->nfa);
848 NOERR();
849 word(v, BEHIND, lp, s);
850 word(v, AHEAD, s, rp);
851 s = newstate(v->nfa);
852 NOERR();
853 nonword(v, BEHIND, lp, s);
854 nonword(v, AHEAD, s, rp);
855 return;
856 break;
857 case LACON: /* lookahead constraint */
858 pos = v->nextvalue;
859 NEXT();
860 s = newstate(v->nfa);
861 s2 = newstate(v->nfa);
862 NOERR();
863 t = parse(v, ')', LACON, s, s2);
864 freesubre(v, t); /* internal structure irrelevant */
865 assert(SEE(')') || ISERR());
866 NEXT();
867 n = newlacon(v, s, s2, pos);
868 NOERR();
869 ARCV(LACON, n);
870 return;
871 break;
872 /* then errors, to get them out of the way */
873 case '*':
874 case '+':
875 case '?':
876 case '{':
877 ERR(REG_BADRPT);
878 return;
879 break;
880 default:
881 ERR(REG_ASSERT);
882 return;
883 break;
884 /* then plain characters, and minor variants on that theme */
885 case ')': /* unbalanced paren */
886 if ((v->cflags & REG_ADVANCED) != REG_EXTENDED)
887 {
888 ERR(REG_EPAREN);
889 return;
890 }
891 /* legal in EREs due to specification botch */
892 NOTE(REG_UPBOTCH);
893 /* fallthrough into case PLAIN */
894 case PLAIN:
895 onechr(v, v->nextvalue, lp, rp);
896 okcolors(v->nfa, v->cm);
897 NOERR();
898 NEXT();
899 break;
900 case '[':
901 if (v->nextvalue == 1)
902 bracket(v, lp, rp);
903 else
904 cbracket(v, lp, rp);
905 assert(SEE(']') || ISERR());
906 NEXT();
907 break;
908 case '.':
909 rainbow(v->nfa, v->cm, PLAIN,
910 (v->cflags & REG_NLSTOP) ? v->nlcolor : COLORLESS,
911 lp, rp);
912 NEXT();
913 break;
914 /* and finally the ugly stuff */
915 case '(': /* value flags as capturing or non */
916 cap = (type == LACON) ? 0 : v->nextvalue;
917 if (cap)
918 {
919 v->nsubexp++;
920 subno = v->nsubexp;
921 if ((size_t) subno >= v->nsubs)
922 moresubs(v, subno);
923 assert((size_t) subno < v->nsubs);
924 }
925 else
926 atomtype = PLAIN; /* something that's not '(' */
927 NEXT();
928 /* need new endpoints because tree will contain pointers */
929 s = newstate(v->nfa);
930 s2 = newstate(v->nfa);
931 NOERR();
932 EMPTYARC(lp, s);
933 EMPTYARC(s2, rp);
934 NOERR();
935 atom = parse(v, ')', PLAIN, s, s2);
936 assert(SEE(')') || ISERR());
937 NEXT();
938 NOERR();
939 if (cap)
940 {
941 v->subs[subno] = atom;
942 t = subre(v, '(', atom->flags | CAP, lp, rp);
943 NOERR();
944 t->subno = subno;
945 t->left = atom;
946 atom = t;
947 }
948 /* postpone everything else pending possible {0} */
949 break;
950 case BACKREF: /* the Feature From The Black Lagoon */
951 INSIST(type != LACON, REG_ESUBREG);
952 INSIST(v->nextvalue < v->nsubs, REG_ESUBREG);
953 INSIST(v->subs[v->nextvalue] != NULL, REG_ESUBREG);
954 NOERR();
955 assert(v->nextvalue > 0);
956 atom = subre(v, 'b', BACKR, lp, rp);
957 subno = v->nextvalue;
958 atom->subno = subno;
959 EMPTYARC(lp, rp); /* temporarily, so there's something */
960 NEXT();
961 break;
962 }
963
964 /* ...and an atom may be followed by a quantifier */
965 switch (v->nexttype)
966 {
967 case '*':
968 m = 0;
969 n = INFINITY;
970 qprefer = (v->nextvalue) ? LONGER : SHORTER;
971 NEXT();
972 break;
973 case '+':
974 m = 1;
975 n = INFINITY;
976 qprefer = (v->nextvalue) ? LONGER : SHORTER;
977 NEXT();
978 break;
979 case '?':
980 m = 0;
981 n = 1;
982 qprefer = (v->nextvalue) ? LONGER : SHORTER;
983 NEXT();
984 break;
985 case '{':
986 NEXT();
987 m = scannum(v);
988 if (EAT(','))
989 {
990 if (SEE(DIGIT))
991 n = scannum(v);
992 else
993 n = INFINITY;
994 if (m > n)
995 {
996 ERR(REG_BADBR);
997 return;
998 }
999 /* {m,n} exercises preference, even if it's {m,m} */
1000 qprefer = (v->nextvalue) ? LONGER : SHORTER;
1001 }
1002 else
1003 {
1004 n = m;
1005 /* {m} passes operand's preference through */
1006 qprefer = 0;
1007 }
1008 if (!SEE('}'))
1009 { /* catches errors too */
1010 ERR(REG_BADBR);
1011 return;
1012 }
1013 NEXT();
1014 break;
1015 default: /* no quantifier */
1016 m = n = 1;
1017 qprefer = 0;
1018 break;
1019 }
1020
1021 /* annoying special case: {0} or {0,0} cancels everything */
1022 if (m == 0 && n == 0)
1023 {
1024 if (atom != NULL)
1025 freesubre(v, atom);
1026 if (atomtype == '(')
1027 v->subs[subno] = NULL;
1028 delsub(v->nfa, lp, rp);
1029 EMPTYARC(lp, rp);
1030 return;
1031 }
1032
1033 /* if not a messy case, avoid hard part */
1034 assert(!MESSY(top->flags));
1035 f = top->flags | qprefer | ((atom != NULL) ? atom->flags : 0);
1036 if (atomtype != '(' && atomtype != BACKREF && !MESSY(UP(f)))
1037 {
1038 if (!(m == 1 && n == 1))
1039 repeat(v, lp, rp, m, n);
1040 if (atom != NULL)
1041 freesubre(v, atom);
1042 top->flags = f;
1043 return;
1044 }
1045
1046 /*
1047 * hard part: something messy That is, capturing parens, back
1048 * reference, short/long clash, or an atom with substructure
1049 * containing one of those.
1050 */
1051
1052 /* now we'll need a subre for the contents even if they're boring */
1053 if (atom == NULL)
1054 {
1055 atom = subre(v, '=', 0, lp, rp);
1056 NOERR();
1057 }
1058
1059 /*
1060 * prepare a general-purpose state skeleton
1061 *
1062 * ---> [s] ---prefix---> [begin] ---atom---> [end] ----rest---> [rp] / /
1063 * [lp] ----> [s2] ----bypass---------------------
1064 *
1065 * where bypass is an empty, and prefix is some repetitions of atom
1066 */
1067 s = newstate(v->nfa); /* first, new endpoints for the atom */
1068 s2 = newstate(v->nfa);
1069 NOERR();
1070 moveouts(v->nfa, lp, s);
1071 moveins(v->nfa, rp, s2);
1072 NOERR();
1073 atom->begin = s;
1074 atom->end = s2;
1075 s = newstate(v->nfa); /* and spots for prefix and bypass */
1076 s2 = newstate(v->nfa);
1077 NOERR();
1078 EMPTYARC(lp, s);
1079 EMPTYARC(lp, s2);
1080 NOERR();
1081
1082 /* break remaining subRE into x{...} and what follows */
1083 t = subre(v, '.', COMBINE(qprefer, atom->flags), lp, rp);
1084 t->left = atom;
1085 atomp = &t->left;
1086 /* here we should recurse... but we must postpone that to the end */
1087
1088 /* split top into prefix and remaining */
1089 assert(top->op == '=' && top->left == NULL && top->right == NULL);
1090 top->left = subre(v, '=', top->flags, top->begin, lp);
1091 top->op = '.';
1092 top->right = t;
1093
1094 /* if it's a backref, now is the time to replicate the subNFA */
1095 if (atomtype == BACKREF)
1096 {
1097 assert(atom->begin->nouts == 1); /* just the EMPTY */
1098 delsub(v->nfa, atom->begin, atom->end);
1099 assert(v->subs[subno] != NULL);
1100 /* and here's why the recursion got postponed: it must */
1101 /* wait until the skeleton is filled in, because it may */
1102 /* hit a backref that wants to copy the filled-in skeleton */
1103 dupnfa(v->nfa, v->subs[subno]->begin, v->subs[subno]->end,
1104 atom->begin, atom->end);
1105 NOERR();
1106 }
1107
1108 /* it's quantifier time; first, turn x{0,...} into x{1,...}|empty */
1109 if (m == 0)
1110 {
1111 EMPTYARC(s2, atom->end); /* the bypass */
1112 assert(PREF(qprefer) != 0);
1113 f = COMBINE(qprefer, atom->flags);
1114 t = subre(v, '|', f, lp, atom->end);
1115 NOERR();
1116 t->left = atom;
1117 t->right = subre(v, '|', PREF(f), s2, atom->end);
1118 NOERR();
1119 t->right->left = subre(v, '=', 0, s2, atom->end);
1120 NOERR();
1121 *atomp = t;
1122 atomp = &t->left;
1123 m = 1;
1124 }
1125
1126 /* deal with the rest of the quantifier */
1127 if (atomtype == BACKREF)
1128 {
1129 /* special case: backrefs have internal quantifiers */
1130 EMPTYARC(s, atom->begin); /* empty prefix */
1131 /* just stuff everything into atom */
1132 repeat(v, atom->begin, atom->end, m, n);
1133 atom->min = (short) m;
1134 atom->max = (short) n;
1135 atom->flags |= COMBINE(qprefer, atom->flags);
1136 }
1137 else if (m == 1 && n == 1)
1138 {
1139 /* no/vacuous quantifier: done */
1140 EMPTYARC(s, atom->begin); /* empty prefix */
1141 }
1142 else
1143 {
1144 /* turn x{m,n} into x{m-1,n-1}x, with capturing */
1145 /* parens in only second x */
1146 dupnfa(v->nfa, atom->begin, atom->end, s, atom->begin);
1147 assert(m >= 1 && m != INFINITY && n >= 1);
1148 repeat(v, s, atom->begin, m - 1, (n == INFINITY) ? n : n - 1);
1149 f = COMBINE(qprefer, atom->flags);
1150 t = subre(v, '.', f, s, atom->end); /* prefix and atom */
1151 NOERR();
1152 t->left = subre(v, '=', PREF(f), s, atom->begin);
1153 NOERR();
1154 t->right = atom;
1155 *atomp = t;
1156 }
1157
1158 /* and finally, look after that postponed recursion */
1159 t = top->right;
1160 if (!(SEE('|') || SEE(stopper) || SEE(EOS)))
1161 t->right = parsebranch(v, stopper, type, atom->end, rp, 1);
1162 else
1163 {
1164 EMPTYARC(atom->end, rp);
1165 t->right = subre(v, '=', 0, atom->end, rp);
1166 }
1167 assert(SEE('|') || SEE(stopper) || SEE(EOS));
1168 t->flags |= COMBINE(t->flags, t->right->flags);
1169 top->flags |= COMBINE(top->flags, t->flags);
1170 }
1171
1172 /*
1173 * nonword - generate arcs for non-word-character ahead or behind
1174 */
1175 static void
1176 nonword(struct vars * v,
1177 int dir, /* AHEAD or BEHIND */
1178 struct state * lp,
1179 struct state * rp)
1180 {
1181 int anchor = (dir == AHEAD) ? '$' : '^';
1182
1183 assert(dir == AHEAD || dir == BEHIND);
1184 newarc(v->nfa, anchor, 1, lp, rp);
1185 newarc(v->nfa, anchor, 0, lp, rp);
1186 colorcomplement(v->nfa, v->cm, dir, v->wordchrs, lp, rp);
1187 /* (no need for special attention to \n) */
1188 }
1189
1190 /*
1191 * word - generate arcs for word character ahead or behind
1192 */
1193 static void
1194 word(struct vars * v,
1195 int dir, /* AHEAD or BEHIND */
1196 struct state * lp,
1197 struct state * rp)
1198 {
1199 assert(dir == AHEAD || dir == BEHIND);
1200 cloneouts(v->nfa, v->wordchrs, lp, rp, dir);
1201 /* (no need for special attention to \n) */
1202 }
1203
1204 /*
1205 * scannum - scan a number
1206 */
1207 static int /* value, <= DUPMAX */
1208 scannum(struct vars * v)
1209 {
1210 int n = 0;
1211
1212 while (SEE(DIGIT) && n < DUPMAX)
1213 {
1214 n = n * 10 + v->nextvalue;
1215 NEXT();
1216 }
1217 if (SEE(DIGIT) || n > DUPMAX)
1218 {
1219 ERR(REG_BADBR);
1220 return 0;
1221 }
1222 return n;
1223 }
1224
1225 /*
1226 * repeat - replicate subNFA for quantifiers
1227 *
1228 * The duplication sequences used here are chosen carefully so that any
1229 * pointers starting out pointing into the subexpression end up pointing into
1230 * the last occurrence. (Note that it may not be strung between the same
1231 * left and right end states, however!) This used to be important for the
1232 * subRE tree, although the important bits are now handled by the in-line
1233 * code in parse(), and when this is called, it doesn't matter any more.
1234 */
1235 static void
1236 repeat(struct vars * v,
1237 struct state * lp,
1238 struct state * rp,
1239 int m,
1240 int n)
1241 {
1242 #define SOME 2
1243 #define INF 3
1244 #define PAIR(x, y) ((x)*4 + (y))
1245 #define REDUCE(x) ( ((x) == INFINITY) ? INF : (((x) > 1) ? SOME : (x)) )
1246 const int rm = REDUCE(m);
1247 const int rn = REDUCE(n);
1248 struct state *s;
1249 struct state *s2;
1250
1251 switch (PAIR(rm, rn))
1252 {
1253 case PAIR(0, 0): /* empty string */
1254 delsub(v->nfa, lp, rp);
1255 EMPTYARC(lp, rp);
1256 break;
1257 case PAIR(0, 1): /* do as x| */
1258 EMPTYARC(lp, rp);
1259 break;
1260 case PAIR(0, SOME): /* do as x{1,n}| */
1261 repeat(v, lp, rp, 1, n);
1262 NOERR();
1263 EMPTYARC(lp, rp);
1264 break;
1265 case PAIR(0, INF): /* loop x around */
1266 s = newstate(v->nfa);
1267 NOERR();
1268 moveouts(v->nfa, lp, s);
1269 moveins(v->nfa, rp, s);
1270 EMPTYARC(lp, s);
1271 EMPTYARC(s, rp);
1272 break;
1273 case PAIR(1, 1): /* no action required */
1274 break;
1275 case PAIR(1, SOME): /* do as x{0,n-1}x = (x{1,n-1}|)x */
1276 s = newstate(v->nfa);
1277 NOERR();
1278 moveouts(v->nfa, lp, s);
1279 dupnfa(v->nfa, s, rp, lp, s);
1280 NOERR();
1281 repeat(v, lp, s, 1, n - 1);
1282 NOERR();
1283 EMPTYARC(lp, s);
1284 break;
1285 case PAIR(1, INF): /* add loopback arc */
1286 s = newstate(v->nfa);
1287 s2 = newstate(v->nfa);
1288 NOERR();
1289 moveouts(v->nfa, lp, s);
1290 moveins(v->nfa, rp, s2);
1291 EMPTYARC(lp, s);
1292 EMPTYARC(s2, rp);
1293 EMPTYARC(s2, s);
1294 break;
1295 case PAIR(SOME, SOME): /* do as x{m-1,n-1}x */
1296 s = newstate(v->nfa);
1297 NOERR();
1298 moveouts(v->nfa, lp, s);
1299 dupnfa(v->nfa, s, rp, lp, s);
1300 NOERR();
1301 repeat(v, lp, s, m - 1, n - 1);
1302 break;
1303 case PAIR(SOME, INF): /* do as x{m-1,}x */
1304 s = newstate(v->nfa);
1305 NOERR();
1306 moveouts(v->nfa, lp, s);
1307 dupnfa(v->nfa, s, rp, lp, s);
1308 NOERR();
1309 repeat(v, lp, s, m - 1, n);
1310 break;
1311 default:
1312 ERR(REG_ASSERT);
1313 break;
1314 }
1315 }
1316
1317 /*
1318 * bracket - handle non-complemented bracket expression
1319 * Also called from cbracket for complemented bracket expressions.
1320 */
1321 static void
1322 bracket(struct vars * v,
1323 struct state * lp,
1324 struct state * rp)
1325 {
1326 assert(SEE('['));
1327 NEXT();
1328 while (!SEE(']') && !SEE(EOS))
1329 brackpart(v, lp, rp);
1330 assert(SEE(']') || ISERR());
1331 okcolors(v->nfa, v->cm);
1332 }
1333
1334 /*
1335 * cbracket - handle complemented bracket expression
1336 * We do it by calling bracket() with dummy endpoints, and then complementing
1337 * the result. The alternative would be to invoke rainbow(), and then delete
1338 * arcs as the b.e. is seen... but that gets messy.
1339 */
1340 static void
1341 cbracket(struct vars * v,
1342 struct state * lp,
1343 struct state * rp)
1344 {
1345 struct state *left = newstate(v->nfa);
1346 struct state *right = newstate(v->nfa);
1347 struct state *s;
1348 struct arc *a; /* arc from lp */
1349 struct arc *ba; /* arc from left, from bracket() */
1350 struct arc *pa; /* MCCE-prototype arc */
1351 color co;
1352 chr *p;
1353 int i;
1354
1355 NOERR();
1356 bracket(v, left, right);
1357 if (v->cflags & REG_NLSTOP)
1358 newarc(v->nfa, PLAIN, v->nlcolor, left, right);
1359 NOERR();
1360
1361 assert(lp->nouts == 0); /* all outarcs will be ours */
1362
1363 /* easy part of complementing */
1364 colorcomplement(v->nfa, v->cm, PLAIN, left, lp, rp);
1365 NOERR();
1366 if (v->mcces == NULL)
1367 { /* no MCCEs -- we're done */
1368 dropstate(v->nfa, left);
1369 assert(right->nins == 0);
1370 freestate(v->nfa, right);
1371 return;
1372 }
1373
1374 /* but complementing gets messy in the presence of MCCEs... */
1375 NOTE(REG_ULOCALE);
1376 for (p = v->mcces->chrs, i = v->mcces->nchrs; i > 0; p++, i--)
1377 {
1378 co = GETCOLOR(v->cm, *p);
1379 a = findarc(lp, PLAIN, co);
1380 ba = findarc(left, PLAIN, co);
1381 if (ba == NULL)
1382 {
1383 assert(a != NULL);
1384 freearc(v->nfa, a);
1385 }
1386 else
1387 assert(a == NULL);
1388 s = newstate(v->nfa);
1389 NOERR();
1390 newarc(v->nfa, PLAIN, co, lp, s);
1391 NOERR();
1392 pa = findarc(v->mccepbegin, PLAIN, co);
1393 assert(pa != NULL);
1394 if (ba == NULL)
1395 { /* easy case, need all of them */
1396 cloneouts(v->nfa, pa->to, s, rp, PLAIN);
1397 newarc(v->nfa, '$', 1, s, rp);
1398 newarc(v->nfa, '$', 0, s, rp);
1399 colorcomplement(v->nfa, v->cm, AHEAD, pa->to, s, rp);
1400 }
1401 else
1402 { /* must be selective */
1403 if (findarc(ba->to, '$', 1) == NULL)
1404 {
1405 newarc(v->nfa, '$', 1, s, rp);
1406 newarc(v->nfa, '$', 0, s, rp);
1407 colorcomplement(v->nfa, v->cm, AHEAD, pa->to,
1408 s, rp);
1409 }
1410 for (pa = pa->to->outs; pa != NULL; pa = pa->outchain)
1411 if (findarc(ba->to, PLAIN, pa->co) == NULL)
1412 newarc(v->nfa, PLAIN, pa->co, s, rp);
1413 if (s->nouts == 0) /* limit of selectivity: none */
1414 dropstate(v->nfa, s); /* frees arc too */
1415 }
1416 NOERR();
1417 }
1418
1419 delsub(v->nfa, left, right);
1420 assert(left->nouts == 0);
1421 freestate(v->nfa, left);
1422 assert(right->nins == 0);
1423 freestate(v->nfa, right);
1424 }
1425
1426 /*
1427 * brackpart - handle one item (or range) within a bracket expression
1428 */
1429 static void
1430 brackpart(struct vars * v,
1431 struct state * lp,
1432 struct state * rp)
1433 {
1434 celt startc;
1435 celt endc;
1436 struct cvec *cv;
1437 chr *startp;
1438 chr *endp;
1439 chr c[1];
1440
1441 /* parse something, get rid of special cases, take shortcuts */
1442 switch (v->nexttype)
1443 {
1444 case RANGE: /* a-b-c or other botch */
1445 ERR(REG_ERANGE);
1446 return;
1447 break;
1448 case PLAIN:
1449 c[0] = v->nextvalue;
1450 NEXT();
1451 /* shortcut for ordinary chr (not range, not MCCE leader) */
1452 if (!SEE(RANGE) && !ISCELEADER(v, c[0]))
1453 {
1454 onechr(v, c[0], lp, rp);
1455 return;
1456 }
1457 startc = element(v, c, c + 1);
1458 NOERR();
1459 break;
1460 case COLLEL:
1461 startp = v->now;
1462 endp = scanplain(v);
1463 INSIST(startp < endp, REG_ECOLLATE);
1464 NOERR();
1465 startc = element(v, startp, endp);
1466 NOERR();
1467 break;
1468 case ECLASS:
1469 startp = v->now;
1470 endp = scanplain(v);
1471 INSIST(startp < endp, REG_ECOLLATE);
1472 NOERR();
1473 startc = element(v, startp, endp);
1474 NOERR();
1475 cv = eclass(v, startc, (v->cflags & REG_ICASE));
1476 NOERR();
1477 dovec(v, cv, lp, rp);
1478 return;
1479 break;
1480 case CCLASS:
1481 startp = v->now;
1482 endp = scanplain(v);
1483 INSIST(startp < endp, REG_ECTYPE);
1484 NOERR();
1485 cv = cclass(v, startp, endp, (v->cflags & REG_ICASE));
1486 NOERR();
1487 dovec(v, cv, lp, rp);
1488 return;
1489 break;
1490 default:
1491 ERR(REG_ASSERT);
1492 return;
1493 break;
1494 }
1495
1496 if (SEE(RANGE))
1497 {
1498 NEXT();
1499 switch (v->nexttype)
1500 {
1501 case PLAIN:
1502 case RANGE:
1503 c[0] = v->nextvalue;
1504 NEXT();
1505 endc = element(v, c, c + 1);
1506 NOERR();
1507 break;
1508 case COLLEL:
1509 startp = v->now;
1510 endp = scanplain(v);
1511 INSIST(startp < endp, REG_ECOLLATE);
1512 NOERR();
1513 endc = element(v, startp, endp);
1514 NOERR();
1515 break;
1516 default:
1517 ERR(REG_ERANGE);
1518 return;
1519 break;
1520 }
1521 }
1522 else
1523 endc = startc;
1524
1525 /*
1526 * Ranges are unportable. Actually, standard C does guarantee that
1527 * digits are contiguous, but making that an exception is just too
1528 * complicated.
1529 */
1530 if (startc != endc)
1531 NOTE(REG_UUNPORT);
1532 cv = range(v, startc, endc, (v->cflags & REG_ICASE));
1533 NOERR();
1534 dovec(v, cv, lp, rp);
1535 }
1536
1537 /*
1538 * scanplain - scan PLAIN contents of [. etc.
1539 *
1540 * Certain bits of trickery in lex.c know that this code does not try
1541 * to look past the final bracket of the [. etc.
1542 */
1543 static chr * /* just after end of sequence */
1544 scanplain(struct vars * v)
1545 {
1546 chr *endp;
1547
1548 assert(SEE(COLLEL) || SEE(ECLASS) || SEE(CCLASS));
1549 NEXT();
1550
1551 endp = v->now;
1552 while (SEE(PLAIN))
1553 {
1554 endp = v->now;
1555 NEXT();
1556 }
1557
1558 assert(SEE(END) || ISERR());
1559 NEXT();
1560
1561 return endp;
1562 }
1563
1564 /*
1565 * leaders - process a cvec of collating elements to also include leaders
1566 * Also gives all characters involved their own colors, which is almost
1567 * certainly necessary, and sets up little disconnected subNFA.
1568 */
1569 static void
1570 leaders(struct vars * v,
1571 struct cvec * cv)
1572 {
1573 int mcce;
1574 chr *p;
1575 chr leader;
1576 struct state *s;
1577 struct arc *a;
1578
1579 v->mccepbegin = newstate(v->nfa);
1580 v->mccepend = newstate(v->nfa);
1581 NOERR();
1582
1583 for (mcce = 0; mcce < cv->nmcces; mcce++)
1584 {
1585 p = cv->mcces[mcce];
1586 leader = *p;
1587 if (!haschr(cv, leader))
1588 {
1589 addchr(cv, leader);
1590 s = newstate(v->nfa);
1591 newarc(v->nfa, PLAIN, subcolor(v->cm, leader),
1592 v->mccepbegin, s);
1593 okcolors(v->nfa, v->cm);
1594 }
1595 else
1596 {
1597 a = findarc(v->mccepbegin, PLAIN,
1598 GETCOLOR(v->cm, leader));
1599 assert(a != NULL);
1600 s = a->to;
1601 assert(s != v->mccepend);
1602 }
1603 p++;
1604 assert(*p != 0 && *(p + 1) == 0); /* only 2-char MCCEs for
1605 * now */
1606 newarc(v->nfa, PLAIN, subcolor(v->cm, *p), s, v->mccepend);
1607 okcolors(v->nfa, v->cm);
1608 }
1609 }
1610
1611 /*
1612 * onechr - fill in arcs for a plain character, and possible case complements
1613 * This is mostly a shortcut for efficient handling of the common case.
1614 */
1615 static void
1616 onechr(struct vars * v,
1617 chr c,
1618 struct state * lp,
1619 struct state * rp)
1620 {
1621 if (!(v->cflags & REG_ICASE))
1622 {
1623 newarc(v->nfa, PLAIN, subcolor(v->cm, c), lp, rp);
1624 return;
1625 }
1626
1627 /* rats, need general case anyway... */
1628 dovec(v, allcases(v, c), lp, rp);
1629 }
1630
1631 /*
1632 * dovec - fill in arcs for each element of a cvec
1633 * This one has to handle the messy cases, like MCCEs and MCCE leaders.
1634 */
1635 static void
1636 dovec(struct vars * v,
1637 struct cvec * cv,
1638 struct state * lp,
1639 struct state * rp)
1640 {
1641 chr ch,
1642 from,
1643 to;
1644 celt ce;
1645 chr *p;
1646 int i;
1647 color co;
1648 struct cvec *leads;
1649 struct arc *a;
1650 struct arc *pa; /* arc in prototype */
1651 struct state *s;
1652 struct state *ps; /* state in prototype */
1653
1654 /* need a place to store leaders, if any */
1655 if (nmcces(v) > 0)
1656 {
1657 assert(v->mcces != NULL);
1658 if (v->cv2 == NULL || v->cv2->nchrs < v->mcces->nchrs)
1659 {
1660 if (v->cv2 != NULL)
1661 free(v->cv2);
1662 v->cv2 = newcvec(v->mcces->nchrs, 0, v->mcces->nmcces);
1663 NOERR();
1664 leads = v->cv2;
1665 }
1666 else
1667 leads = clearcvec(v->cv2);
1668 }
1669 else
1670 leads = NULL;
1671
1672 /* first, get the ordinary characters out of the way */
1673 for (p = cv->chrs, i = cv->nchrs; i > 0; p++, i--)
1674 {
1675 ch = *p;
1676 if (!ISCELEADER(v, ch))
1677 newarc(v->nfa, PLAIN, subcolor(v->cm, ch), lp, rp);
1678 else
1679 {
1680 assert(singleton(v->cm, ch));
1681 assert(leads != NULL);
1682 if (!haschr(leads, ch))
1683 addchr(leads, ch);
1684 }
1685 }
1686
1687 /* and the ranges */
1688 for (p = cv->ranges, i = cv->nranges; i > 0; p += 2, i--)
1689 {
1690 from = *p;
1691 to = *(p + 1);
1692 while (from <= to && (ce = nextleader(v, from, to)) != NOCELT)
1693 {
1694 if (from < ce)
1695 subrange(v, from, ce - 1, lp, rp);
1696 assert(singleton(v->cm, ce));
1697 assert(leads != NULL);
1698 if (!haschr(leads, ce))
1699 addchr(leads, ce);
1700 from = ce + 1;
1701 }
1702 if (from <= to)
1703 subrange(v, from, to, lp, rp);
1704 }
1705
1706 if ((leads == NULL || leads->nchrs == 0) && cv->nmcces == 0)
1707 return;
1708
1709 /* deal with the MCCE leaders */
1710 NOTE(REG_ULOCALE);
1711 for (p = leads->chrs, i = leads->nchrs; i > 0; p++, i--)
1712 {
1713 co = GETCOLOR(v->cm, *p);
1714 a = findarc(lp, PLAIN, co);
1715 if (a != NULL)
1716 s = a->to;
1717 else
1718 {
1719 s = newstate(v->nfa);
1720 NOERR();
1721 newarc(v->nfa, PLAIN, co, lp, s);
1722 NOERR();
1723 }
1724 pa = findarc(v->mccepbegin, PLAIN, co);
1725 assert(pa != NULL);
1726 ps = pa->to;
1727 newarc(v->nfa, '$', 1, s, rp);
1728 newarc(v->nfa, '$', 0, s, rp);
1729 colorcomplement(v->nfa, v->cm, AHEAD, ps, s, rp);
1730 NOERR();
1731 }
1732
1733 /* and the MCCEs */
1734 for (i = 0; i < cv->nmcces; i++)
1735 {
1736 p = cv->mcces[i];
1737 assert(singleton(v->cm, *p));
1738 if (!singleton(v->cm, *p))
1739 {
1740 ERR(REG_ASSERT);
1741 return;
1742 }
1743 ch = *p++;
1744 co = GETCOLOR(v->cm, ch);
1745 a = findarc(lp, PLAIN, co);
1746 if (a != NULL)
1747 s = a->to;
1748 else
1749 {
1750 s = newstate(v->nfa);
1751 NOERR();
1752 newarc(v->nfa, PLAIN, co, lp, s);
1753 NOERR();
1754 }
1755 assert(*p != 0); /* at least two chars */
1756 assert(singleton(v->cm, *p));
1757 ch = *p++;
1758 co = GETCOLOR(v->cm, ch);
1759 assert(*p == 0); /* and only two, for now */
1760 newarc(v->nfa, PLAIN, co, s, rp);
1761 NOERR();
1762 }
1763 }
1764
1765 /*
1766 * nextleader - find next MCCE leader within range
1767 */
1768 static celt /* NOCELT means none */
1769 nextleader(struct vars * v,
1770 chr from,
1771 chr to)
1772 {
1773 int i;
1774 chr *p;
1775 chr ch;
1776 celt it = NOCELT;
1777
1778 if (v->mcces == NULL)
1779 return it;
1780
1781 for (i = v->mcces->nchrs, p = v->mcces->chrs; i > 0; i--, p++)
1782 {
1783 ch = *p;
1784 if (from <= ch && ch <= to)
1785 if (it == NOCELT || ch < it)
1786 it = ch;
1787 }
1788 return it;
1789 }
1790
1791 /*
1792 * wordchrs - set up word-chr list for word-boundary stuff, if needed
1793 *
1794 * The list is kept as a bunch of arcs between two dummy states; it's
1795 * disposed of by the unreachable-states sweep in NFA optimization.
1796 * Does NEXT(). Must not be called from any unusual lexical context.
1797 * This should be reconciled with the \w etc. handling in lex.c, and
1798 * should be cleaned up to reduce dependencies on input scanning.
1799 */
1800 static void
1801 wordchrs(struct vars * v)
1802 {
1803 struct state *left;
1804 struct state *right;
1805
1806 if (v->wordchrs != NULL)
1807 {
1808 NEXT(); /* for consistency */
1809 return;
1810 }
1811
1812 left = newstate(v->nfa);
1813 right = newstate(v->nfa);
1814 NOERR();
1815 /* fine point: implemented with [::], and lexer will set REG_ULOCALE */
1816 lexword(v);
1817 NEXT();
1818 assert(v->savenow != NULL && SEE('['));
1819 bracket(v, left, right);
1820 assert((v->savenow != NULL && SEE(']')) || ISERR());
1821 NEXT();
1822 NOERR();
1823 v->wordchrs = left;
1824 }
1825
1826 /*
1827 * subre - allocate a subre
1828 */
1829 static struct subre *
1830 subre(struct vars * v,
1831 int op,
1832 int flags,
1833 struct state * begin,
1834 struct state * end)
1835 {
1836 struct subre *ret;
1837
1838 ret = v->treefree;
1839 if (ret != NULL)
1840 v->treefree = ret->left;
1841 else
1842 {
1843 ret = (struct subre *) MALLOC(sizeof(struct subre));
1844 if (ret == NULL)
1845 {
1846 ERR(REG_ESPACE);
1847 return NULL;
1848 }
1849 ret->chain = v->treechain;
1850 v->treechain = ret;
1851 }
1852
1853 assert(strchr("|.b(=", op) != NULL);
1854
1855 ret->op = op;
1856 ret->flags = flags;
1857 ret->retry = 0;
1858 ret->subno = 0;
1859 ret->min = ret->max = 1;
1860 ret->left = NULL;
1861 ret->right = NULL;
1862 ret->begin = begin;
1863 ret->end = end;
1864 ZAPCNFA(ret->cnfa);
1865
1866 return ret;
1867 }
1868
1869 /*
1870 * freesubre - free a subRE subtree
1871 */
1872 static void
1873 freesubre(struct vars * v, /* might be NULL */
1874 struct subre * sr)
1875 {
1876 if (sr == NULL)
1877 return;
1878
1879 if (sr->left != NULL)
1880 freesubre(v, sr->left);
1881 if (sr->right != NULL)
1882 freesubre(v, sr->right);
1883
1884 freesrnode(v, sr);
1885 }
1886
1887 /*
1888 * freesrnode - free one node in a subRE subtree
1889 */
1890 static void
1891 freesrnode(struct vars * v, /* might be NULL */
1892 struct subre * sr)
1893 {
1894 if (sr == NULL)
1895 return;
1896
1897 if (!NULLCNFA(sr->cnfa))
1898 freecnfa(&sr->cnfa);
1899 sr->flags = 0;
1900
1901 if (v != NULL)
1902 {
1903 sr->left = v->treefree;
1904 v->treefree = sr;
1905 }
1906 else
1907 FREE(sr);
1908 }
1909
1910 /*
1911 * optst - optimize a subRE subtree
1912 */
1913 static void
1914 optst(struct vars * v,
1915 struct subre * t)
1916 {
1917 if (t == NULL)
1918 return;
1919
1920 /* recurse through children */
1921 if (t->left != NULL)
1922 optst(v, t->left);
1923 if (t->right != NULL)
1924 optst(v, t->right);
1925 }
1926
1927 /*
1928 * numst - number tree nodes (assigning retry indexes)
1929 */
1930 static int /* next number */
1931 numst(struct subre * t,
1932 int start) /* starting point for subtree numbers */
1933 {
1934 int i;
1935
1936 assert(t != NULL);
1937
1938 i = start;
1939 t->retry = (short) i++;
1940 if (t->left != NULL)
1941 i = numst(t->left, i);
1942 if (t->right != NULL)
1943 i = numst(t->right, i);
1944 return i;
1945 }
1946
1947 /*
1948 * markst - mark tree nodes as INUSE
1949 */
1950 static void
1951 markst(struct subre * t)
1952 {
1953 assert(t != NULL);
1954
1955 t->flags |= INUSE;
1956 if (t->left != NULL)
1957 markst(t->left);
1958 if (t->right != NULL)
1959 markst(t->right);
1960 }
1961
1962 /*
1963 * cleanst - free any tree nodes not marked INUSE
1964 */
1965 static void
1966 cleanst(struct vars * v)
1967 {
1968 struct subre *t;
1969 struct subre *next;
1970
1971 for (t = v->treechain; t != NULL; t = next)
1972 {
1973 next = t->chain;
1974 if (!(t->flags & INUSE))
1975 FREE(t);
1976 }
1977 v->treechain = NULL;
1978 v->treefree = NULL; /* just on general principles */
1979 }
1980
1981 /*
1982 * nfatree - turn a subRE subtree into a tree of compacted NFAs
1983 */
1984 static long /* optimize results from top node */
1985 nfatree(struct vars * v,
1986 struct subre * t,
1987 FILE *f) /* for debug output */
1988 {
1989 assert(t != NULL && t->begin != NULL);
1990
1991 if (t->left != NULL)
1992 (DISCARD) nfatree(v, t->left, f);
1993 if (t->right != NULL)
1994 (DISCARD) nfatree(v, t->right, f);
1995
1996 return nfanode(v, t, f);
1997 }
1998
1999 /*
2000 * nfanode - do one NFA for nfatree
2001 */
2002 static long /* optimize results */
2003 nfanode(struct vars * v,
2004 struct subre * t,
2005 FILE *f) /* for debug output */
2006 {
2007 struct nfa *nfa;
2008 long ret = 0;
2009
2010 assert(t->begin != NULL);
2011
2012 #ifdef REG_DEBUG
2013 if (f != NULL)
2014 {
2015 char idbuf[50];
2016
2017 fprintf(f, "\n\n\n========= TREE NODE %s ==========\n",
2018 stid(t, idbuf, sizeof(idbuf)));
2019 }
2020 #endif
2021 nfa = newnfa(v, v->cm, v->nfa);
2022 NOERRZ();
2023 dupnfa(nfa, t->begin, t->end, nfa->init, nfa->final);
2024 if (!ISERR())
2025 {
2026 specialcolors(nfa);
2027 ret = optimize(nfa, f);
2028 }
2029 if (!ISERR())
2030 compact(nfa, &t->cnfa);
2031
2032 freenfa(nfa);
2033 return ret;
2034 }
2035
2036 /*
2037 * newlacon - allocate a lookahead-constraint subRE
2038 */
2039 static int /* lacon number */
2040 newlacon(struct vars * v,
2041 struct state * begin,
2042 struct state * end,
2043 int pos)
2044 {
2045 int n;
2046 struct subre *sub;
2047
2048 if (v->nlacons == 0)
2049 {
2050 v->lacons = (struct subre *) MALLOC(2 * sizeof(struct subre));
2051 n = 1; /* skip 0th */
2052 v->nlacons = 2;
2053 }
2054 else
2055 {
2056 v->lacons = (struct subre *) REALLOC(v->lacons,
2057 (v->nlacons + 1) * sizeof(struct subre));
2058 n = v->nlacons++;
2059 }
2060 if (v->lacons == NULL)
2061 {
2062 ERR(REG_ESPACE);
2063 return 0;
2064 }
2065 sub = &v->lacons[n];
2066 sub->begin = begin;
2067 sub->end = end;
2068 sub->subno = pos;
2069 ZAPCNFA(sub->cnfa);
2070 return n;
2071 }
2072
2073 /*
2074 * freelacons - free lookahead-constraint subRE vector
2075 */
2076 static void
2077 freelacons(struct subre * subs,
2078 int n)
2079 {
2080 struct subre *sub;
2081 int i;
2082
2083 assert(n > 0);
2084 for (sub = subs + 1, i = n - 1; i > 0; sub++, i--) /* no 0th */
2085 if (!NULLCNFA(sub->cnfa))
2086 freecnfa(&sub->cnfa);
2087 FREE(subs);
2088 }
2089
2090 /*
2091 * rfree - free a whole RE (insides of regfree)
2092 */
2093 static void
2094 rfree(regex_t *re)
2095 {
2096 struct guts *g;
2097
2098 if (re == NULL || re->re_magic != REMAGIC)
2099 return;
2100
2101 re->re_magic = 0; /* invalidate RE */
2102 g = (struct guts *) re->re_guts;
2103 re->re_guts = NULL;
2104 re->re_fns = NULL;
2105 g->magic = 0;
2106 freecm(&g->cmap);
2107 if (g->tree != NULL)
2108 freesubre((struct vars *) NULL, g->tree);
2109 if (g->lacons != NULL)
2110 freelacons(g->lacons, g->nlacons);
2111 if (!NULLCNFA(g->search))
2112 freecnfa(&g->search);
2113 FREE(g);
2114 }
2115
2116 #ifdef REG_DEBUG
2117
2118 /*
2119 * dump - dump an RE in human-readable form
2120 */
2121 static void
2122 dump(regex_t *re,
2123 FILE *f)
2124 {
2125 struct guts *g;
2126 int i;
2127
2128 if (re->re_magic != REMAGIC)
2129 fprintf(f, "bad magic number (0x%x not 0x%x)\n", re->re_magic,
2130 REMAGIC);
2131 if (re->re_guts == NULL)
2132 {
2133 fprintf(f, "NULL guts!!!\n");
2134 return;
2135 }
2136 g = (struct guts *) re->re_guts;
2137 if (g->magic != GUTSMAGIC)
2138 fprintf(f, "bad guts magic number (0x%x not 0x%x)\n", g->magic,
2139 GUTSMAGIC);
2140
2141 fprintf(f, "\n\n\n========= DUMP ==========\n");
2142 fprintf(f, "nsub %d, info 0%lo, csize %d, ntree %d\n",
2143 re->re_nsub, re->re_info, re->re_csize, g->ntree);
2144
2145 dumpcolors(&g->cmap, f);
2146 if (!NULLCNFA(g->search))
2147 {
2148 printf("\nsearch:\n");
2149 dumpcnfa(&g->search, f);
2150 }
2151 for (i = 1; i < g->nlacons; i++)
2152 {
2153 fprintf(f, "\nla%d (%s):\n", i,
2154 (g->lacons[i].subno) ? "positive" : "negative");
2155 dumpcnfa(&g->lacons[i].cnfa, f);
2156 }
2157 fprintf(f, "\n");
2158 dumpst(g->tree, f, 0);
2159 }
2160
2161 /*
2162 * dumpst - dump a subRE tree
2163 */
2164 static void
2165 dumpst(struct subre * t,
2166 FILE *f,
2167 int nfapresent) /* is the original NFA still around? */
2168 {
2169 if (t == NULL)
2170 fprintf(f, "null tree\n");
2171 else
2172 stdump(t, f, nfapresent);
2173 fflush(f);
2174 }
2175
2176 /*
2177 * stdump - recursive guts of dumpst
2178 */
2179 static void
2180 stdump(struct subre * t,
2181 FILE *f,
2182 int nfapresent) /* is the original NFA still around? */
2183 {
2184 char idbuf[50];
2185
2186 fprintf(f, "%s. `%c'", stid(t, idbuf, sizeof(idbuf)), t->op);
2187 if (t->flags & LONGER)
2188 fprintf(f, " longest");
2189 if (t->flags & SHORTER)
2190 fprintf(f, " shortest");
2191 if (t->flags & MIXED)
2192 fprintf(f, " hasmixed");
2193 if (t->flags & CAP)
2194 fprintf(f, " hascapture");
2195 if (t->flags & BACKR)
2196 fprintf(f, " hasbackref");
2197 if (!(t->flags & INUSE))
2198 fprintf(f, " UNUSED");
2199 if (t->subno != 0)
2200 fprintf(f, " (#%d)", t->subno);
2201 if (t->min != 1 || t->max != 1)
2202 {
2203 fprintf(f, " {%d,", t->min);
2204 if (t->max != INFINITY)
2205 fprintf(f, "%d", t->max);
2206 fprintf(f, "}");
2207 }
2208 if (nfapresent)
2209 fprintf(f, " %ld-%ld", (long) t->begin->no, (long) t->end->no);
2210 if (t->left != NULL)
2211 fprintf(f, " L:%s", stid(t->left, idbuf, sizeof(idbuf)));
2212 if (t->right != NULL)
2213 fprintf(f, " R:%s", stid(t->right, idbuf, sizeof(idbuf)));
2214 if (!NULLCNFA(t->cnfa))
2215 {
2216 fprintf(f, "\n");
2217 dumpcnfa(&t->cnfa, f);
2218 fprintf(f, "\n");
2219 }
2220 if (t->left != NULL)
2221 stdump(t->left, f, nfapresent);
2222 if (t->right != NULL)
2223 stdump(t->right, f, nfapresent);
2224 }
2225
2226 /*
2227 * stid - identify a subtree node for dumping
2228 */
2229 static char * /* points to buf or constant string */
2230 stid(struct subre * t,
2231 char *buf,
2232 size_t bufsize)
2233 {
2234 /* big enough for hex int or decimal t->retry? */
2235 if (bufsize < sizeof(int) * 2 + 3 || bufsize < sizeof(t->retry) * 3 + 1)
2236 return "unable";
2237 if (t->retry != 0)
2238 sprintf(buf, "%d", t->retry);
2239 else
2240 sprintf(buf, "0x%x", (int) t); /* may lose bits, that's okay */
2241 return buf;
2242 }
2243 #endif /* REG_DEBUG */
2244
2245
2246 #include "regc_lex.c"
2247 #include "regc_color.c"
2248 #include "regc_nfa.c"
2249 #include "regc_cvec.c"
2250 #include "regc_locale.c"
wxWidgets (Impactor)