5 * "One Ring to rule them all, One Ring to find them..."
8 /* This file contains functions for executing a regular expression. See
9 * also regcomp.c which funnily enough, contains functions for compiling
10 * a regular expression.
12 * This file is also copied at build time to ext/re/re_exec.c, where
13 * it's built with -DPERL_EXT_RE_BUILD -DPERL_EXT_RE_DEBUG -DPERL_EXT.
14 * This causes the main functions to be compiled under new names and with
15 * debugging support added, which makes "use re 'debug'" work.
19 /* NOTE: this is derived from Henry Spencer's regexp code, and should not
20 * confused with the original package (see point 3 below). Thanks, Henry!
23 /* Additional note: this code is very heavily munged from Henry's version
24 * in places. In some spots I've traded clarity for efficiency, so don't
25 * blame Henry for some of the lack of readability.
28 /* The names of the functions have been changed from regcomp and
29 * regexec to pregcomp and pregexec in order to avoid conflicts
30 * with the POSIX routines of the same names.
33 #ifdef PERL_EXT_RE_BUILD
34 /* need to replace pregcomp et al, so enable that */
35 # ifndef PERL_IN_XSUB_RE
36 # define PERL_IN_XSUB_RE
38 /* need access to debugger hooks */
39 # if defined(PERL_EXT_RE_DEBUG) && !defined(DEBUGGING)
44 #ifdef PERL_IN_XSUB_RE
45 /* We *really* need to overwrite these symbols: */
46 # define Perl_regexec_flags my_regexec
47 # define Perl_regdump my_regdump
48 # define Perl_regprop my_regprop
49 # define Perl_re_intuit_start my_re_intuit_start
50 /* *These* symbols are masked to allow static link. */
51 # define Perl_pregexec my_pregexec
52 # define Perl_reginitcolors my_reginitcolors
53 # define Perl_regclass_swash my_regclass_swash
55 # define PERL_NO_GET_CONTEXT
59 * pregcomp and pregexec -- regsub and regerror are not used in perl
61 * Copyright (c) 1986 by University of Toronto.
62 * Written by Henry Spencer. Not derived from licensed software.
64 * Permission is granted to anyone to use this software for any
65 * purpose on any computer system, and to redistribute it freely,
66 * subject to the following restrictions:
68 * 1. The author is not responsible for the consequences of use of
69 * this software, no matter how awful, even if they arise
72 * 2. The origin of this software must not be misrepresented, either
73 * by explicit claim or by omission.
75 * 3. Altered versions must be plainly marked as such, and must not
76 * be misrepresented as being the original software.
78 **** Alterations to Henry's code are...
80 **** Copyright (C) 1991, 1992, 1993, 1994, 1995, 1996, 1997, 1998, 1999,
81 **** 2000, 2001, 2002, 2003, 2004, 2005, 2006, by Larry Wall and others
83 **** You may distribute under the terms of either the GNU General Public
84 **** License or the Artistic License, as specified in the README file.
86 * Beware that some of this code is subtly aware of the way operator
87 * precedence is structured in regular expressions. Serious changes in
88 * regular-expression syntax might require a total rethink.
91 #define PERL_IN_REGEXEC_C
96 #define RF_tainted 1 /* tainted information used? */
97 #define RF_warned 2 /* warned about big count? */
98 #define RF_evaled 4 /* Did an EVAL with setting? */
99 #define RF_utf8 8 /* String contains multibyte chars? */
101 #define UTF ((PL_reg_flags & RF_utf8) != 0)
103 #define RS_init 1 /* eval environment created */
104 #define RS_set 2 /* replsv value is set */
107 #define STATIC static
110 #define REGINCLASS(p,c) (ANYOF_FLAGS(p) ? reginclass(p,c,0,0) : ANYOF_BITMAP_TEST(p,*(c)))
116 #define CHR_SVLEN(sv) (do_utf8 ? sv_len_utf8(sv) : SvCUR(sv))
117 #define CHR_DIST(a,b) (PL_reg_match_utf8 ? utf8_distance(a,b) : a - b)
119 #define HOPc(pos,off) ((char *)(PL_reg_match_utf8 \
120 ? reghop3((U8*)pos, off, (U8*)(off >= 0 ? PL_regeol : PL_bostr)) \
122 #define HOPBACKc(pos, off) ((char*) \
123 ((PL_reg_match_utf8) \
124 ? reghopmaybe3((U8*)pos, -off, ((U8*)(off < 0 ? PL_regeol : PL_bostr))) \
125 : (pos - off >= PL_bostr) \
130 #define reghopmaybe3_c(pos,off,lim) ((char*)reghopmaybe3((U8*)pos, off, (U8*)lim))
131 #define HOP3(pos,off,lim) (PL_reg_match_utf8 ? reghop3((U8*)pos, off, (U8*)lim) : (U8*)(pos + off))
132 #define HOP3c(pos,off,lim) ((char*)HOP3(pos,off,lim))
134 #define LOAD_UTF8_CHARCLASS(class,str) STMT_START { \
135 if (!CAT2(PL_utf8_,class)) { bool ok; ENTER; save_re_context(); ok=CAT2(is_utf8_,class)((const U8*)str); assert(ok); LEAVE; } } STMT_END
136 #define LOAD_UTF8_CHARCLASS_ALNUM() LOAD_UTF8_CHARCLASS(alnum,"a")
137 #define LOAD_UTF8_CHARCLASS_DIGIT() LOAD_UTF8_CHARCLASS(digit,"0")
138 #define LOAD_UTF8_CHARCLASS_SPACE() LOAD_UTF8_CHARCLASS(space," ")
139 #define LOAD_UTF8_CHARCLASS_MARK() LOAD_UTF8_CHARCLASS(mark, "\xcd\x86")
141 /* for use after a quantifier and before an EXACT-like node -- japhy */
142 #define JUMPABLE(rn) ( \
143 OP(rn) == OPEN || OP(rn) == CLOSE || OP(rn) == EVAL || \
144 OP(rn) == SUSPEND || OP(rn) == IFMATCH || \
145 OP(rn) == PLUS || OP(rn) == MINMOD || \
146 (PL_regkind[(U8)OP(rn)] == CURLY && ARG1(rn) > 0) \
149 #define HAS_TEXT(rn) ( \
150 PL_regkind[(U8)OP(rn)] == EXACT || PL_regkind[(U8)OP(rn)] == REF \
154 Search for mandatory following text node; for lookahead, the text must
155 follow but for lookbehind (rn->flags != 0) we skip to the next step.
157 #define FIND_NEXT_IMPT(rn) STMT_START { \
158 while (JUMPABLE(rn)) \
159 if (OP(rn) == SUSPEND || PL_regkind[(U8)OP(rn)] == CURLY) \
160 rn = NEXTOPER(NEXTOPER(rn)); \
161 else if (OP(rn) == PLUS) \
163 else if (OP(rn) == IFMATCH) \
164 rn = (rn->flags == 0) ? NEXTOPER(NEXTOPER(rn)) : rn + ARG(rn); \
165 else rn += NEXT_OFF(rn); \
168 static void restore_pos(pTHX_ void *arg);
171 S_regcppush(pTHX_ I32 parenfloor)
174 const int retval = PL_savestack_ix;
175 #define REGCP_PAREN_ELEMS 4
176 const int paren_elems_to_push = (PL_regsize - parenfloor) * REGCP_PAREN_ELEMS;
179 if (paren_elems_to_push < 0)
180 Perl_croak(aTHX_ "panic: paren_elems_to_push < 0");
182 #define REGCP_OTHER_ELEMS 6
183 SSGROW(paren_elems_to_push + REGCP_OTHER_ELEMS);
184 for (p = PL_regsize; p > parenfloor; p--) {
185 /* REGCP_PARENS_ELEMS are pushed per pairs of parentheses. */
186 SSPUSHINT(PL_regendp[p]);
187 SSPUSHINT(PL_regstartp[p]);
188 SSPUSHPTR(PL_reg_start_tmp[p]);
191 /* REGCP_OTHER_ELEMS are pushed in any case, parentheses or no. */
192 SSPUSHINT(PL_regsize);
193 SSPUSHINT(*PL_reglastparen);
194 SSPUSHINT(*PL_reglastcloseparen);
195 SSPUSHPTR(PL_reginput);
196 #define REGCP_FRAME_ELEMS 2
197 /* REGCP_FRAME_ELEMS are part of the REGCP_OTHER_ELEMS and
198 * are needed for the regexp context stack bookkeeping. */
199 SSPUSHINT(paren_elems_to_push + REGCP_OTHER_ELEMS - REGCP_FRAME_ELEMS);
200 SSPUSHINT(SAVEt_REGCONTEXT); /* Magic cookie. */
205 /* These are needed since we do not localize EVAL nodes: */
206 # define REGCP_SET(cp) DEBUG_EXECUTE_r(PerlIO_printf(Perl_debug_log, \
207 " Setting an EVAL scope, savestack=%"IVdf"\n", \
208 (IV)PL_savestack_ix)); cp = PL_savestack_ix
210 # define REGCP_UNWIND(cp) DEBUG_EXECUTE_r(cp != PL_savestack_ix ? \
211 PerlIO_printf(Perl_debug_log, \
212 " Clearing an EVAL scope, savestack=%"IVdf"..%"IVdf"\n", \
213 (IV)(cp), (IV)PL_savestack_ix) : 0); regcpblow(cp)
216 S_regcppop(pTHX_ const regexp *rex)
222 GET_RE_DEBUG_FLAGS_DECL;
224 /* Pop REGCP_OTHER_ELEMS before the parentheses loop starts. */
226 assert(i == SAVEt_REGCONTEXT); /* Check that the magic cookie is there. */
227 i = SSPOPINT; /* Parentheses elements to pop. */
228 input = (char *) SSPOPPTR;
229 *PL_reglastcloseparen = SSPOPINT;
230 *PL_reglastparen = SSPOPINT;
231 PL_regsize = SSPOPINT;
233 /* Now restore the parentheses context. */
234 for (i -= (REGCP_OTHER_ELEMS - REGCP_FRAME_ELEMS);
235 i > 0; i -= REGCP_PAREN_ELEMS) {
237 U32 paren = (U32)SSPOPINT;
238 PL_reg_start_tmp[paren] = (char *) SSPOPPTR;
239 PL_regstartp[paren] = SSPOPINT;
241 if (paren <= *PL_reglastparen)
242 PL_regendp[paren] = tmps;
244 PerlIO_printf(Perl_debug_log,
245 " restoring \\%"UVuf" to %"IVdf"(%"IVdf")..%"IVdf"%s\n",
246 (UV)paren, (IV)PL_regstartp[paren],
247 (IV)(PL_reg_start_tmp[paren] - PL_bostr),
248 (IV)PL_regendp[paren],
249 (paren > *PL_reglastparen ? "(no)" : ""));
253 if ((I32)(*PL_reglastparen + 1) <= rex->nparens) {
254 PerlIO_printf(Perl_debug_log,
255 " restoring \\%"IVdf"..\\%"IVdf" to undef\n",
256 (IV)(*PL_reglastparen + 1), (IV)rex->nparens);
260 /* It would seem that the similar code in regtry()
261 * already takes care of this, and in fact it is in
262 * a better location to since this code can #if 0-ed out
263 * but the code in regtry() is needed or otherwise tests
264 * requiring null fields (pat.t#187 and split.t#{13,14}
265 * (as of patchlevel 7877) will fail. Then again,
266 * this code seems to be necessary or otherwise
267 * building DynaLoader will fail:
268 * "Error: '*' not in typemap in DynaLoader.xs, line 164"
270 for (i = *PL_reglastparen + 1; i <= rex->nparens; i++) {
272 PL_regstartp[i] = -1;
279 #define regcpblow(cp) LEAVE_SCOPE(cp) /* Ignores regcppush()ed data. */
281 #define TRYPAREN(paren, n, input, where) { \
284 PL_regstartp[paren] = HOPc(input, -1) - PL_bostr; \
285 PL_regendp[paren] = input - PL_bostr; \
288 PL_regendp[paren] = -1; \
290 REGMATCH(next, where); \
294 PL_regendp[paren] = -1; \
299 * pregexec and friends
303 - pregexec - match a regexp against a string
306 Perl_pregexec(pTHX_ register regexp *prog, char *stringarg, register char *strend,
307 char *strbeg, I32 minend, SV *screamer, U32 nosave)
308 /* strend: pointer to null at end of string */
309 /* strbeg: real beginning of string */
310 /* minend: end of match must be >=minend after stringarg. */
311 /* nosave: For optimizations. */
314 regexec_flags(prog, stringarg, strend, strbeg, minend, screamer, NULL,
315 nosave ? 0 : REXEC_COPY_STR);
320 * Need to implement the following flags for reg_anch:
322 * USE_INTUIT_NOML - Useful to call re_intuit_start() first
324 * INTUIT_AUTORITATIVE_NOML - Can trust a positive answer
325 * INTUIT_AUTORITATIVE_ML
326 * INTUIT_ONCE_NOML - Intuit can match in one location only.
329 * Another flag for this function: SECOND_TIME (so that float substrs
330 * with giant delta may be not rechecked).
333 /* Assumptions: if ANCH_GPOS, then strpos is anchored. XXXX Check GPOS logic */
335 /* If SCREAM, then SvPVX_const(sv) should be compatible with strpos and strend.
336 Otherwise, only SvCUR(sv) is used to get strbeg. */
338 /* XXXX We assume that strpos is strbeg unless sv. */
340 /* XXXX Some places assume that there is a fixed substring.
341 An update may be needed if optimizer marks as "INTUITable"
342 RExen without fixed substrings. Similarly, it is assumed that
343 lengths of all the strings are no more than minlen, thus they
344 cannot come from lookahead.
345 (Or minlen should take into account lookahead.) */
347 /* A failure to find a constant substring means that there is no need to make
348 an expensive call to REx engine, thus we celebrate a failure. Similarly,
349 finding a substring too deep into the string means that less calls to
350 regtry() should be needed.
352 REx compiler's optimizer found 4 possible hints:
353 a) Anchored substring;
355 c) Whether we are anchored (beginning-of-line or \G);
356 d) First node (of those at offset 0) which may distingush positions;
357 We use a)b)d) and multiline-part of c), and try to find a position in the
358 string which does not contradict any of them.
361 /* Most of decisions we do here should have been done at compile time.
362 The nodes of the REx which we used for the search should have been
363 deleted from the finite automaton. */
366 Perl_re_intuit_start(pTHX_ regexp *prog, SV *sv, char *strpos,
367 char *strend, U32 flags, re_scream_pos_data *data)
370 register I32 start_shift = 0;
371 /* Should be nonnegative! */
372 register I32 end_shift = 0;
377 const int do_utf8 = sv ? SvUTF8(sv) : 0; /* if no sv we have to assume bytes */
379 register char *other_last = NULL; /* other substr checked before this */
380 char *check_at = NULL; /* check substr found at this pos */
381 const I32 multiline = prog->reganch & PMf_MULTILINE;
383 const char * const i_strpos = strpos;
384 SV * const dsv = PERL_DEBUG_PAD_ZERO(0);
387 GET_RE_DEBUG_FLAGS_DECL;
389 RX_MATCH_UTF8_set(prog,do_utf8);
391 if (prog->reganch & ROPT_UTF8) {
392 DEBUG_EXECUTE_r(PerlIO_printf(Perl_debug_log,
393 "UTF-8 regex...\n"));
394 PL_reg_flags |= RF_utf8;
398 const char *s = PL_reg_match_utf8 ?
399 sv_uni_display(dsv, sv, 60, UNI_DISPLAY_REGEX) :
401 const int len = PL_reg_match_utf8 ?
402 strlen(s) : strend - strpos;
405 if (PL_reg_match_utf8)
406 DEBUG_EXECUTE_r(PerlIO_printf(Perl_debug_log,
407 "UTF-8 target...\n"));
408 PerlIO_printf(Perl_debug_log,
409 "%sGuessing start of match, REx%s \"%s%.60s%s%s\" against \"%s%.*s%s%s\"...\n",
410 PL_colors[4], PL_colors[5], PL_colors[0],
413 (strlen(prog->precomp) > 60 ? "..." : ""),
415 (int)(len > 60 ? 60 : len),
417 (len > 60 ? "..." : "")
421 /* CHR_DIST() would be more correct here but it makes things slow. */
422 if (prog->minlen > strend - strpos) {
423 DEBUG_EXECUTE_r(PerlIO_printf(Perl_debug_log,
424 "String too short... [re_intuit_start]\n"));
427 strbeg = (sv && SvPOK(sv)) ? strend - SvCUR(sv) : strpos;
430 if (!prog->check_utf8 && prog->check_substr)
431 to_utf8_substr(prog);
432 check = prog->check_utf8;
434 if (!prog->check_substr && prog->check_utf8)
435 to_byte_substr(prog);
436 check = prog->check_substr;
438 if (check == &PL_sv_undef) {
439 DEBUG_EXECUTE_r(PerlIO_printf(Perl_debug_log,
440 "Non-utf string cannot match utf check string\n"));
443 if (prog->reganch & ROPT_ANCH) { /* Match at beg-of-str or after \n */
444 ml_anch = !( (prog->reganch & ROPT_ANCH_SINGLE)
445 || ( (prog->reganch & ROPT_ANCH_BOL)
446 && !multiline ) ); /* Check after \n? */
449 if ( !(prog->reganch & (ROPT_ANCH_GPOS /* Checked by the caller */
450 | ROPT_IMPLICIT)) /* not a real BOL */
451 /* SvCUR is not set on references: SvRV and SvPVX_const overlap */
453 && (strpos != strbeg)) {
454 DEBUG_EXECUTE_r(PerlIO_printf(Perl_debug_log, "Not at start...\n"));
457 if (prog->check_offset_min == prog->check_offset_max &&
458 !(prog->reganch & ROPT_CANY_SEEN)) {
459 /* Substring at constant offset from beg-of-str... */
462 s = HOP3c(strpos, prog->check_offset_min, strend);
464 slen = SvCUR(check); /* >= 1 */
466 if ( strend - s > slen || strend - s < slen - 1
467 || (strend - s == slen && strend[-1] != '\n')) {
468 DEBUG_EXECUTE_r(PerlIO_printf(Perl_debug_log, "String too long...\n"));
471 /* Now should match s[0..slen-2] */
473 if (slen && (*SvPVX_const(check) != *s
475 && memNE(SvPVX_const(check), s, slen)))) {
477 DEBUG_EXECUTE_r(PerlIO_printf(Perl_debug_log, "String not equal...\n"));
481 else if (*SvPVX_const(check) != *s
482 || ((slen = SvCUR(check)) > 1
483 && memNE(SvPVX_const(check), s, slen)))
486 goto success_at_start;
489 /* Match is anchored, but substr is not anchored wrt beg-of-str. */
491 start_shift = prog->check_offset_min; /* okay to underestimate on CC */
492 end_shift = prog->minlen - start_shift -
493 CHR_SVLEN(check) + (SvTAIL(check) != 0);
495 const I32 end = prog->check_offset_max + CHR_SVLEN(check)
496 - (SvTAIL(check) != 0);
497 const I32 eshift = CHR_DIST((U8*)strend, (U8*)s) - end;
499 if (end_shift < eshift)
503 else { /* Can match at random position */
506 start_shift = prog->check_offset_min; /* okay to underestimate on CC */
507 /* Should be nonnegative! */
508 end_shift = prog->minlen - start_shift -
509 CHR_SVLEN(check) + (SvTAIL(check) != 0);
512 #ifdef DEBUGGING /* 7/99: reports of failure (with the older version) */
514 Perl_croak(aTHX_ "panic: end_shift");
518 /* Find a possible match in the region s..strend by looking for
519 the "check" substring in the region corrected by start/end_shift. */
520 if (flags & REXEC_SCREAM) {
521 I32 p = -1; /* Internal iterator of scream. */
522 I32 * const pp = data ? data->scream_pos : &p;
524 if (PL_screamfirst[BmRARE(check)] >= 0
525 || ( BmRARE(check) == '\n'
526 && (BmPREVIOUS(check) == SvCUR(check) - 1)
528 s = screaminstr(sv, check,
529 start_shift + (s - strbeg), end_shift, pp, 0);
532 /* we may be pointing at the wrong string */
533 if (s && RX_MATCH_COPIED(prog))
534 s = strbeg + (s - SvPVX_const(sv));
536 *data->scream_olds = s;
538 else if (prog->reganch & ROPT_CANY_SEEN)
539 s = fbm_instr((U8*)(s + start_shift),
540 (U8*)(strend - end_shift),
541 check, multiline ? FBMrf_MULTILINE : 0);
543 s = fbm_instr(HOP3(s, start_shift, strend),
544 HOP3(strend, -end_shift, strbeg),
545 check, multiline ? FBMrf_MULTILINE : 0);
547 /* Update the count-of-usability, remove useless subpatterns,
550 DEBUG_EXECUTE_r(PerlIO_printf(Perl_debug_log, "%s %s substr \"%s%.*s%s\"%s%s",
551 (s ? "Found" : "Did not find"),
552 (check == (do_utf8 ? prog->anchored_utf8 : prog->anchored_substr) ? "anchored" : "floating"),
554 (int)(SvCUR(check) - (SvTAIL(check)!=0)),
556 PL_colors[1], (SvTAIL(check) ? "$" : ""),
557 (s ? " at offset " : "...\n") ) );
564 /* Finish the diagnostic message */
565 DEBUG_EXECUTE_r(PerlIO_printf(Perl_debug_log, "%ld...\n", (long)(s - i_strpos)) );
567 /* Got a candidate. Check MBOL anchoring, and the *other* substr.
568 Start with the other substr.
569 XXXX no SCREAM optimization yet - and a very coarse implementation
570 XXXX /ttx+/ results in anchored="ttx", floating="x". floating will
571 *always* match. Probably should be marked during compile...
572 Probably it is right to do no SCREAM here...
575 if (do_utf8 ? (prog->float_utf8 && prog->anchored_utf8) : (prog->float_substr && prog->anchored_substr)) {
576 /* Take into account the "other" substring. */
577 /* XXXX May be hopelessly wrong for UTF... */
580 if (check == (do_utf8 ? prog->float_utf8 : prog->float_substr)) {
583 char * const last = HOP3c(s, -start_shift, strbeg);
588 t = s - prog->check_offset_max;
589 if (s - strpos > prog->check_offset_max /* signed-corrected t > strpos */
591 || ((t = reghopmaybe3_c(s, -(prog->check_offset_max), strpos))
596 t = HOP3c(t, prog->anchored_offset, strend);
597 if (t < other_last) /* These positions already checked */
599 last2 = last1 = HOP3c(strend, -prog->minlen, strbeg);
602 /* XXXX It is not documented what units *_offsets are in. Assume bytes. */
603 /* On end-of-str: see comment below. */
604 must = do_utf8 ? prog->anchored_utf8 : prog->anchored_substr;
605 if (must == &PL_sv_undef) {
607 DEBUG_EXECUTE_r(must = prog->anchored_utf8); /* for debug */
612 HOP3(HOP3(last1, prog->anchored_offset, strend)
613 + SvCUR(must), -(SvTAIL(must)!=0), strbeg),
615 multiline ? FBMrf_MULTILINE : 0
617 DEBUG_EXECUTE_r(PerlIO_printf(Perl_debug_log,
618 "%s anchored substr \"%s%.*s%s\"%s",
619 (s ? "Found" : "Contradicts"),
622 - (SvTAIL(must)!=0)),
624 PL_colors[1], (SvTAIL(must) ? "$" : "")));
626 if (last1 >= last2) {
627 DEBUG_EXECUTE_r(PerlIO_printf(Perl_debug_log,
628 ", giving up...\n"));
631 DEBUG_EXECUTE_r(PerlIO_printf(Perl_debug_log,
632 ", trying floating at offset %ld...\n",
633 (long)(HOP3c(s1, 1, strend) - i_strpos)));
634 other_last = HOP3c(last1, prog->anchored_offset+1, strend);
635 s = HOP3c(last, 1, strend);
639 DEBUG_EXECUTE_r(PerlIO_printf(Perl_debug_log, " at offset %ld...\n",
640 (long)(s - i_strpos)));
641 t = HOP3c(s, -prog->anchored_offset, strbeg);
642 other_last = HOP3c(s, 1, strend);
650 else { /* Take into account the floating substring. */
655 t = HOP3c(s, -start_shift, strbeg);
657 HOP3c(strend, -prog->minlen + prog->float_min_offset, strbeg);
658 if (CHR_DIST((U8*)last, (U8*)t) > prog->float_max_offset)
659 last = HOP3c(t, prog->float_max_offset, strend);
660 s = HOP3c(t, prog->float_min_offset, strend);
663 /* XXXX It is not documented what units *_offsets are in. Assume bytes. */
664 must = do_utf8 ? prog->float_utf8 : prog->float_substr;
665 /* fbm_instr() takes into account exact value of end-of-str
666 if the check is SvTAIL(ed). Since false positives are OK,
667 and end-of-str is not later than strend we are OK. */
668 if (must == &PL_sv_undef) {
670 DEBUG_EXECUTE_r(must = prog->float_utf8); /* for debug message */
673 s = fbm_instr((unsigned char*)s,
674 (unsigned char*)last + SvCUR(must)
676 must, multiline ? FBMrf_MULTILINE : 0);
677 DEBUG_EXECUTE_r(PerlIO_printf(Perl_debug_log, "%s floating substr \"%s%.*s%s\"%s",
678 (s ? "Found" : "Contradicts"),
680 (int)(SvCUR(must) - (SvTAIL(must)!=0)),
682 PL_colors[1], (SvTAIL(must) ? "$" : "")));
685 DEBUG_EXECUTE_r(PerlIO_printf(Perl_debug_log,
686 ", giving up...\n"));
689 DEBUG_EXECUTE_r(PerlIO_printf(Perl_debug_log,
690 ", trying anchored starting at offset %ld...\n",
691 (long)(s1 + 1 - i_strpos)));
693 s = HOP3c(t, 1, strend);
697 DEBUG_EXECUTE_r(PerlIO_printf(Perl_debug_log, " at offset %ld...\n",
698 (long)(s - i_strpos)));
699 other_last = s; /* Fix this later. --Hugo */
708 t = s - prog->check_offset_max;
709 if (s - strpos > prog->check_offset_max /* signed-corrected t > strpos */
711 || ((t = reghopmaybe3_c(s, -prog->check_offset_max, strpos))
713 /* Fixed substring is found far enough so that the match
714 cannot start at strpos. */
716 if (ml_anch && t[-1] != '\n') {
717 /* Eventually fbm_*() should handle this, but often
718 anchored_offset is not 0, so this check will not be wasted. */
719 /* XXXX In the code below we prefer to look for "^" even in
720 presence of anchored substrings. And we search even
721 beyond the found float position. These pessimizations
722 are historical artefacts only. */
724 while (t < strend - prog->minlen) {
726 if (t < check_at - prog->check_offset_min) {
727 if (do_utf8 ? prog->anchored_utf8 : prog->anchored_substr) {
728 /* Since we moved from the found position,
729 we definitely contradict the found anchored
730 substr. Due to the above check we do not
731 contradict "check" substr.
732 Thus we can arrive here only if check substr
733 is float. Redo checking for "other"=="fixed".
736 DEBUG_EXECUTE_r(PerlIO_printf(Perl_debug_log, "Found /%s^%s/m at offset %ld, rescanning for anchored from offset %ld...\n",
737 PL_colors[0], PL_colors[1], (long)(strpos - i_strpos), (long)(strpos - i_strpos + prog->anchored_offset)));
738 goto do_other_anchored;
740 /* We don't contradict the found floating substring. */
741 /* XXXX Why not check for STCLASS? */
743 DEBUG_EXECUTE_r(PerlIO_printf(Perl_debug_log, "Found /%s^%s/m at offset %ld...\n",
744 PL_colors[0], PL_colors[1], (long)(s - i_strpos)));
747 /* Position contradicts check-string */
748 /* XXXX probably better to look for check-string
749 than for "\n", so one should lower the limit for t? */
750 DEBUG_EXECUTE_r(PerlIO_printf(Perl_debug_log, "Found /%s^%s/m, restarting lookup for check-string at offset %ld...\n",
751 PL_colors[0], PL_colors[1], (long)(t + 1 - i_strpos)));
752 other_last = strpos = s = t + 1;
757 DEBUG_EXECUTE_r(PerlIO_printf(Perl_debug_log, "Did not find /%s^%s/m...\n",
758 PL_colors[0], PL_colors[1]));
762 DEBUG_EXECUTE_r(PerlIO_printf(Perl_debug_log, "Starting position does not contradict /%s^%s/m...\n",
763 PL_colors[0], PL_colors[1]));
767 ++BmUSEFUL(do_utf8 ? prog->check_utf8 : prog->check_substr); /* hooray/5 */
770 /* The found string does not prohibit matching at strpos,
771 - no optimization of calling REx engine can be performed,
772 unless it was an MBOL and we are not after MBOL,
773 or a future STCLASS check will fail this. */
775 /* Even in this situation we may use MBOL flag if strpos is offset
776 wrt the start of the string. */
777 if (ml_anch && sv && !SvROK(sv) /* See prev comment on SvROK */
778 && (strpos != strbeg) && strpos[-1] != '\n'
779 /* May be due to an implicit anchor of m{.*foo} */
780 && !(prog->reganch & ROPT_IMPLICIT))
785 DEBUG_EXECUTE_r( if (ml_anch)
786 PerlIO_printf(Perl_debug_log, "Position at offset %ld does not contradict /%s^%s/m...\n",
787 (long)(strpos - i_strpos), PL_colors[0], PL_colors[1]);
790 if (!(prog->reganch & ROPT_NAUGHTY) /* XXXX If strpos moved? */
792 prog->check_utf8 /* Could be deleted already */
793 && --BmUSEFUL(prog->check_utf8) < 0
794 && (prog->check_utf8 == prog->float_utf8)
796 prog->check_substr /* Could be deleted already */
797 && --BmUSEFUL(prog->check_substr) < 0
798 && (prog->check_substr == prog->float_substr)
801 /* If flags & SOMETHING - do not do it many times on the same match */
802 DEBUG_EXECUTE_r(PerlIO_printf(Perl_debug_log, "... Disabling check substring...\n"));
803 SvREFCNT_dec(do_utf8 ? prog->check_utf8 : prog->check_substr);
804 if (do_utf8 ? prog->check_substr : prog->check_utf8)
805 SvREFCNT_dec(do_utf8 ? prog->check_substr : prog->check_utf8);
806 prog->check_substr = prog->check_utf8 = NULL; /* disable */
807 prog->float_substr = prog->float_utf8 = NULL; /* clear */
808 check = NULL; /* abort */
810 /* XXXX This is a remnant of the old implementation. It
811 looks wasteful, since now INTUIT can use many
813 prog->reganch &= ~RE_USE_INTUIT;
820 /* XXXX BmUSEFUL already changed, maybe multiple change is meaningful... */
821 if (prog->regstclass) {
822 /* minlen == 0 is possible if regstclass is \b or \B,
823 and the fixed substr is ''$.
824 Since minlen is already taken into account, s+1 is before strend;
825 accidentally, minlen >= 1 guaranties no false positives at s + 1
826 even for \b or \B. But (minlen? 1 : 0) below assumes that
827 regstclass does not come from lookahead... */
828 /* If regstclass takes bytelength more than 1: If charlength==1, OK.
829 This leaves EXACTF only, which is dealt with in find_byclass(). */
830 const U8* const str = (U8*)STRING(prog->regstclass);
831 const int cl_l = (PL_regkind[(U8)OP(prog->regstclass)] == EXACT
832 ? CHR_DIST(str+STR_LEN(prog->regstclass), str)
834 const char * const endpos = (prog->anchored_substr || prog->anchored_utf8 || ml_anch)
835 ? HOP3c(s, (prog->minlen ? cl_l : 0), strend)
836 : (prog->float_substr || prog->float_utf8
837 ? HOP3c(HOP3c(check_at, -start_shift, strbeg),
843 s = find_byclass(prog, prog->regstclass, s, endpos, 1);
846 const char *what = NULL;
848 if (endpos == strend) {
849 DEBUG_EXECUTE_r( PerlIO_printf(Perl_debug_log,
850 "Could not match STCLASS...\n") );
853 DEBUG_EXECUTE_r( PerlIO_printf(Perl_debug_log,
854 "This position contradicts STCLASS...\n") );
855 if ((prog->reganch & ROPT_ANCH) && !ml_anch)
857 /* Contradict one of substrings */
858 if (prog->anchored_substr || prog->anchored_utf8) {
859 if ((do_utf8 ? prog->anchored_utf8 : prog->anchored_substr) == check) {
860 DEBUG_EXECUTE_r( what = "anchored" );
862 s = HOP3c(t, 1, strend);
863 if (s + start_shift + end_shift > strend) {
864 /* XXXX Should be taken into account earlier? */
865 DEBUG_EXECUTE_r( PerlIO_printf(Perl_debug_log,
866 "Could not match STCLASS...\n") );
871 DEBUG_EXECUTE_r( PerlIO_printf(Perl_debug_log,
872 "Looking for %s substr starting at offset %ld...\n",
873 what, (long)(s + start_shift - i_strpos)) );
876 /* Have both, check_string is floating */
877 if (t + start_shift >= check_at) /* Contradicts floating=check */
878 goto retry_floating_check;
879 /* Recheck anchored substring, but not floating... */
883 DEBUG_EXECUTE_r( PerlIO_printf(Perl_debug_log,
884 "Looking for anchored substr starting at offset %ld...\n",
885 (long)(other_last - i_strpos)) );
886 goto do_other_anchored;
888 /* Another way we could have checked stclass at the
889 current position only: */
894 DEBUG_EXECUTE_r( PerlIO_printf(Perl_debug_log,
895 "Looking for /%s^%s/m starting at offset %ld...\n",
896 PL_colors[0], PL_colors[1], (long)(t - i_strpos)) );
899 if (!(do_utf8 ? prog->float_utf8 : prog->float_substr)) /* Could have been deleted */
901 /* Check is floating subtring. */
902 retry_floating_check:
903 t = check_at - start_shift;
904 DEBUG_EXECUTE_r( what = "floating" );
905 goto hop_and_restart;
908 DEBUG_EXECUTE_r(PerlIO_printf(Perl_debug_log,
909 "By STCLASS: moving %ld --> %ld\n",
910 (long)(t - i_strpos), (long)(s - i_strpos))
914 DEBUG_EXECUTE_r(PerlIO_printf(Perl_debug_log,
915 "Does not contradict STCLASS...\n");
920 DEBUG_EXECUTE_r(PerlIO_printf(Perl_debug_log, "%s%s:%s match at offset %ld\n",
921 PL_colors[4], (check ? "Guessed" : "Giving up"),
922 PL_colors[5], (long)(s - i_strpos)) );
925 fail_finish: /* Substring not found */
926 if (prog->check_substr || prog->check_utf8) /* could be removed already */
927 BmUSEFUL(do_utf8 ? prog->check_utf8 : prog->check_substr) += 5; /* hooray */
929 DEBUG_EXECUTE_r(PerlIO_printf(Perl_debug_log, "%sMatch rejected by optimizer%s\n",
930 PL_colors[4], PL_colors[5]));
934 /* We know what class REx starts with. Try to find this position... */
936 S_find_byclass(pTHX_ regexp * prog, const regnode *c, char *s, const char *strend, I32 norun)
939 const I32 doevery = (prog->reganch & ROPT_SKIP) == 0;
943 register STRLEN uskip;
947 register I32 tmp = 1; /* Scratch variable? */
948 register const bool do_utf8 = PL_reg_match_utf8;
950 /* We know what class it must start with. */
954 while (s + (uskip = UTF8SKIP(s)) <= strend) {
955 if ((ANYOF_FLAGS(c) & ANYOF_UNICODE) ||
956 !UTF8_IS_INVARIANT((U8)s[0]) ?
957 reginclass(c, (U8*)s, 0, do_utf8) :
958 REGINCLASS(c, (U8*)s)) {
959 if (tmp && (norun || regtry(prog, s)))
973 if (REGINCLASS(c, (U8*)s) ||
974 (ANYOF_FOLD_SHARP_S(c, s, strend) &&
975 /* The assignment of 2 is intentional:
976 * for the folded sharp s, the skip is 2. */
977 (skip = SHARP_S_SKIP))) {
978 if (tmp && (norun || regtry(prog, s)))
991 if (tmp && (norun || regtry(prog, s)))
1000 ln = STR_LEN(c); /* length to match in octets/bytes */
1001 lnc = (I32) ln; /* length to match in characters */
1003 STRLEN ulen1, ulen2;
1005 U8 tmpbuf1[UTF8_MAXBYTES_CASE+1];
1006 U8 tmpbuf2[UTF8_MAXBYTES_CASE+1];
1007 const U32 uniflags = UTF8_ALLOW_DEFAULT;
1009 to_utf8_lower((U8*)m, tmpbuf1, &ulen1);
1010 to_utf8_upper((U8*)m, tmpbuf2, &ulen2);
1012 c1 = utf8n_to_uvchr(tmpbuf1, UTF8_MAXBYTES_CASE,
1014 c2 = utf8n_to_uvchr(tmpbuf2, UTF8_MAXBYTES_CASE,
1017 while (sm < ((U8 *) m + ln)) {
1032 c2 = PL_fold_locale[c1];
1034 e = HOP3c(strend, -((I32)lnc), s);
1037 e = s; /* Due to minlen logic of intuit() */
1039 /* The idea in the EXACTF* cases is to first find the
1040 * first character of the EXACTF* node and then, if
1041 * necessary, case-insensitively compare the full
1042 * text of the node. The c1 and c2 are the first
1043 * characters (though in Unicode it gets a bit
1044 * more complicated because there are more cases
1045 * than just upper and lower: one needs to use
1046 * the so-called folding case for case-insensitive
1047 * matching (called "loose matching" in Unicode).
1048 * ibcmp_utf8() will do just that. */
1052 U8 tmpbuf [UTF8_MAXBYTES+1];
1053 STRLEN len, foldlen;
1054 const U32 uniflags = UTF8_ALLOW_DEFAULT;
1056 /* Upper and lower of 1st char are equal -
1057 * probably not a "letter". */
1059 c = utf8n_to_uvchr((U8*)s, UTF8_MAXBYTES, &len,
1063 ibcmp_utf8(s, (char **)0, 0, do_utf8,
1064 m, (char **)0, ln, (bool)UTF))
1065 && (norun || regtry(prog, s)) )
1068 U8 foldbuf[UTF8_MAXBYTES_CASE+1];
1069 uvchr_to_utf8(tmpbuf, c);
1070 f = to_utf8_fold(tmpbuf, foldbuf, &foldlen);
1072 && (f == c1 || f == c2)
1073 && (ln == foldlen ||
1074 !ibcmp_utf8((char *) foldbuf,
1075 (char **)0, foldlen, do_utf8,
1077 (char **)0, ln, (bool)UTF))
1078 && (norun || regtry(prog, s)) )
1086 c = utf8n_to_uvchr((U8*)s, UTF8_MAXBYTES, &len,
1089 /* Handle some of the three Greek sigmas cases.
1090 * Note that not all the possible combinations
1091 * are handled here: some of them are handled
1092 * by the standard folding rules, and some of
1093 * them (the character class or ANYOF cases)
1094 * are handled during compiletime in
1095 * regexec.c:S_regclass(). */
1096 if (c == (UV)UNICODE_GREEK_CAPITAL_LETTER_SIGMA ||
1097 c == (UV)UNICODE_GREEK_SMALL_LETTER_FINAL_SIGMA)
1098 c = (UV)UNICODE_GREEK_SMALL_LETTER_SIGMA;
1100 if ( (c == c1 || c == c2)
1102 ibcmp_utf8(s, (char **)0, 0, do_utf8,
1103 m, (char **)0, ln, (bool)UTF))
1104 && (norun || regtry(prog, s)) )
1107 U8 foldbuf[UTF8_MAXBYTES_CASE+1];
1108 uvchr_to_utf8(tmpbuf, c);
1109 f = to_utf8_fold(tmpbuf, foldbuf, &foldlen);
1111 && (f == c1 || f == c2)
1112 && (ln == foldlen ||
1113 !ibcmp_utf8((char *) foldbuf,
1114 (char **)0, foldlen, do_utf8,
1116 (char **)0, ln, (bool)UTF))
1117 && (norun || regtry(prog, s)) )
1128 && (ln == 1 || !(OP(c) == EXACTF
1130 : ibcmp_locale(s, m, ln)))
1131 && (norun || regtry(prog, s)) )
1137 if ( (*(U8*)s == c1 || *(U8*)s == c2)
1138 && (ln == 1 || !(OP(c) == EXACTF
1140 : ibcmp_locale(s, m, ln)))
1141 && (norun || regtry(prog, s)) )
1148 PL_reg_flags |= RF_tainted;
1155 U8 * const r = reghop3((U8*)s, -1, (U8*)PL_bostr);
1156 tmp = utf8n_to_uvchr(r, UTF8SKIP(r), 0, UTF8_ALLOW_DEFAULT);
1158 tmp = ((OP(c) == BOUND ?
1159 isALNUM_uni(tmp) : isALNUM_LC_uvchr(UNI_TO_NATIVE(tmp))) != 0);
1160 LOAD_UTF8_CHARCLASS_ALNUM();
1161 while (s + (uskip = UTF8SKIP(s)) <= strend) {
1162 if (tmp == !(OP(c) == BOUND ?
1163 swash_fetch(PL_utf8_alnum, (U8*)s, do_utf8) :
1164 isALNUM_LC_utf8((U8*)s)))
1167 if ((norun || regtry(prog, s)))
1174 tmp = (s != PL_bostr) ? UCHARAT(s - 1) : '\n';
1175 tmp = ((OP(c) == BOUND ? isALNUM(tmp) : isALNUM_LC(tmp)) != 0);
1176 while (s < strend) {
1178 !(OP(c) == BOUND ? isALNUM(*s) : isALNUM_LC(*s))) {
1180 if ((norun || regtry(prog, s)))
1186 if ((!prog->minlen && tmp) && (norun || regtry(prog, s)))
1190 PL_reg_flags |= RF_tainted;
1197 U8 * const r = reghop3((U8*)s, -1, (U8*)PL_bostr);
1198 tmp = utf8n_to_uvchr(r, UTF8SKIP(r), 0, UTF8_ALLOW_DEFAULT);
1200 tmp = ((OP(c) == NBOUND ?
1201 isALNUM_uni(tmp) : isALNUM_LC_uvchr(UNI_TO_NATIVE(tmp))) != 0);
1202 LOAD_UTF8_CHARCLASS_ALNUM();
1203 while (s + (uskip = UTF8SKIP(s)) <= strend) {
1204 if (tmp == !(OP(c) == NBOUND ?
1205 swash_fetch(PL_utf8_alnum, (U8*)s, do_utf8) :
1206 isALNUM_LC_utf8((U8*)s)))
1208 else if ((norun || regtry(prog, s)))
1214 tmp = (s != PL_bostr) ? UCHARAT(s - 1) : '\n';
1215 tmp = ((OP(c) == NBOUND ?
1216 isALNUM(tmp) : isALNUM_LC(tmp)) != 0);
1217 while (s < strend) {
1219 !(OP(c) == NBOUND ? isALNUM(*s) : isALNUM_LC(*s)))
1221 else if ((norun || regtry(prog, s)))
1226 if ((!prog->minlen && !tmp) && (norun || regtry(prog, s)))
1231 LOAD_UTF8_CHARCLASS_ALNUM();
1232 while (s + (uskip = UTF8SKIP(s)) <= strend) {
1233 if (swash_fetch(PL_utf8_alnum, (U8*)s, do_utf8)) {
1234 if (tmp && (norun || regtry(prog, s)))
1245 while (s < strend) {
1247 if (tmp && (norun || regtry(prog, s)))
1259 PL_reg_flags |= RF_tainted;
1261 while (s + (uskip = UTF8SKIP(s)) <= strend) {
1262 if (isALNUM_LC_utf8((U8*)s)) {
1263 if (tmp && (norun || regtry(prog, s)))
1274 while (s < strend) {
1275 if (isALNUM_LC(*s)) {
1276 if (tmp && (norun || regtry(prog, s)))
1289 LOAD_UTF8_CHARCLASS_ALNUM();
1290 while (s + (uskip = UTF8SKIP(s)) <= strend) {
1291 if (!swash_fetch(PL_utf8_alnum, (U8*)s, do_utf8)) {
1292 if (tmp && (norun || regtry(prog, s)))
1303 while (s < strend) {
1305 if (tmp && (norun || regtry(prog, s)))
1317 PL_reg_flags |= RF_tainted;
1319 while (s + (uskip = UTF8SKIP(s)) <= strend) {
1320 if (!isALNUM_LC_utf8((U8*)s)) {
1321 if (tmp && (norun || regtry(prog, s)))
1332 while (s < strend) {
1333 if (!isALNUM_LC(*s)) {
1334 if (tmp && (norun || regtry(prog, s)))
1347 LOAD_UTF8_CHARCLASS_SPACE();
1348 while (s + (uskip = UTF8SKIP(s)) <= strend) {
1349 if (*s == ' ' || swash_fetch(PL_utf8_space,(U8*)s, do_utf8)) {
1350 if (tmp && (norun || regtry(prog, s)))
1361 while (s < strend) {
1363 if (tmp && (norun || regtry(prog, s)))
1375 PL_reg_flags |= RF_tainted;
1377 while (s + (uskip = UTF8SKIP(s)) <= strend) {
1378 if (*s == ' ' || isSPACE_LC_utf8((U8*)s)) {
1379 if (tmp && (norun || regtry(prog, s)))
1390 while (s < strend) {
1391 if (isSPACE_LC(*s)) {
1392 if (tmp && (norun || regtry(prog, s)))
1405 LOAD_UTF8_CHARCLASS_SPACE();
1406 while (s + (uskip = UTF8SKIP(s)) <= strend) {
1407 if (!(*s == ' ' || swash_fetch(PL_utf8_space,(U8*)s, do_utf8))) {
1408 if (tmp && (norun || regtry(prog, s)))
1419 while (s < strend) {
1421 if (tmp && (norun || regtry(prog, s)))
1433 PL_reg_flags |= RF_tainted;
1435 while (s + (uskip = UTF8SKIP(s)) <= strend) {
1436 if (!(*s == ' ' || isSPACE_LC_utf8((U8*)s))) {
1437 if (tmp && (norun || regtry(prog, s)))
1448 while (s < strend) {
1449 if (!isSPACE_LC(*s)) {
1450 if (tmp && (norun || regtry(prog, s)))
1463 LOAD_UTF8_CHARCLASS_DIGIT();
1464 while (s + (uskip = UTF8SKIP(s)) <= strend) {
1465 if (swash_fetch(PL_utf8_digit,(U8*)s, do_utf8)) {
1466 if (tmp && (norun || regtry(prog, s)))
1477 while (s < strend) {
1479 if (tmp && (norun || regtry(prog, s)))
1491 PL_reg_flags |= RF_tainted;
1493 while (s + (uskip = UTF8SKIP(s)) <= strend) {
1494 if (isDIGIT_LC_utf8((U8*)s)) {
1495 if (tmp && (norun || regtry(prog, s)))
1506 while (s < strend) {
1507 if (isDIGIT_LC(*s)) {
1508 if (tmp && (norun || regtry(prog, s)))
1521 LOAD_UTF8_CHARCLASS_DIGIT();
1522 while (s + (uskip = UTF8SKIP(s)) <= strend) {
1523 if (!swash_fetch(PL_utf8_digit,(U8*)s, do_utf8)) {
1524 if (tmp && (norun || regtry(prog, s)))
1535 while (s < strend) {
1537 if (tmp && (norun || regtry(prog, s)))
1549 PL_reg_flags |= RF_tainted;
1551 while (s + (uskip = UTF8SKIP(s)) <= strend) {
1552 if (!isDIGIT_LC_utf8((U8*)s)) {
1553 if (tmp && (norun || regtry(prog, s)))
1564 while (s < strend) {
1565 if (!isDIGIT_LC(*s)) {
1566 if (tmp && (norun || regtry(prog, s)))
1578 Perl_croak(aTHX_ "panic: unknown regstclass %d", (int)OP(c));
1587 - regexec_flags - match a regexp against a string
1590 Perl_regexec_flags(pTHX_ register regexp *prog, char *stringarg, register char *strend,
1591 char *strbeg, I32 minend, SV *sv, void *data, U32 flags)
1592 /* strend: pointer to null at end of string */
1593 /* strbeg: real beginning of string */
1594 /* minend: end of match must be >=minend after stringarg. */
1595 /* data: May be used for some additional optimizations. */
1596 /* nosave: For optimizations. */
1600 register regnode *c;
1601 register char *startpos = stringarg;
1602 I32 minlen; /* must match at least this many chars */
1603 I32 dontbother = 0; /* how many characters not to try at end */
1604 I32 end_shift = 0; /* Same for the end. */ /* CC */
1605 I32 scream_pos = -1; /* Internal iterator of scream. */
1606 char *scream_olds = NULL;
1607 SV* oreplsv = GvSV(PL_replgv);
1608 const bool do_utf8 = DO_UTF8(sv);
1609 const I32 multiline = prog->reganch & PMf_MULTILINE;
1611 SV * const dsv0 = PERL_DEBUG_PAD_ZERO(0);
1612 SV * const dsv1 = PERL_DEBUG_PAD_ZERO(1);
1615 GET_RE_DEBUG_FLAGS_DECL;
1617 PERL_UNUSED_ARG(data);
1619 /* Be paranoid... */
1620 if (prog == NULL || startpos == NULL) {
1621 Perl_croak(aTHX_ "NULL regexp parameter");
1627 PL_regnarrate = DEBUG_r_TEST;
1630 RX_MATCH_UTF8_set(prog, do_utf8);
1632 minlen = prog->minlen;
1633 if (strend - startpos < minlen) {
1634 DEBUG_EXECUTE_r(PerlIO_printf(Perl_debug_log,
1635 "String too short [regexec_flags]...\n"));
1639 /* Check validity of program. */
1640 if (UCHARAT(prog->program) != REG_MAGIC) {
1641 Perl_croak(aTHX_ "corrupted regexp program");
1645 PL_reg_eval_set = 0;
1648 if (prog->reganch & ROPT_UTF8)
1649 PL_reg_flags |= RF_utf8;
1651 /* Mark beginning of line for ^ and lookbehind. */
1652 PL_regbol = startpos;
1656 /* Mark end of line for $ (and such) */
1659 /* see how far we have to get to not match where we matched before */
1660 PL_regtill = startpos+minend;
1662 /* If there is a "must appear" string, look for it. */
1665 if (prog->reganch & ROPT_GPOS_SEEN) { /* Need to have PL_reg_ganch */
1668 if (flags & REXEC_IGNOREPOS) /* Means: check only at start */
1669 PL_reg_ganch = startpos;
1670 else if (sv && SvTYPE(sv) >= SVt_PVMG
1672 && (mg = mg_find(sv, PERL_MAGIC_regex_global))
1673 && mg->mg_len >= 0) {
1674 PL_reg_ganch = strbeg + mg->mg_len; /* Defined pos() */
1675 if (prog->reganch & ROPT_ANCH_GPOS) {
1676 if (s > PL_reg_ganch)
1681 else /* pos() not defined */
1682 PL_reg_ganch = strbeg;
1685 if (!(flags & REXEC_CHECKED) && (prog->check_substr != NULL || prog->check_utf8 != NULL)) {
1686 re_scream_pos_data d;
1688 d.scream_olds = &scream_olds;
1689 d.scream_pos = &scream_pos;
1690 s = re_intuit_start(prog, sv, s, strend, flags, &d);
1692 DEBUG_EXECUTE_r(PerlIO_printf(Perl_debug_log, "Not present...\n"));
1693 goto phooey; /* not present */
1698 const char * const s0 = UTF
1699 ? pv_uni_display(dsv0, (U8*)prog->precomp, prog->prelen, 60,
1702 const int len0 = UTF ? SvCUR(dsv0) : prog->prelen;
1703 const char * const s1 = do_utf8 ? sv_uni_display(dsv1, sv, 60,
1704 UNI_DISPLAY_REGEX) : startpos;
1705 const int len1 = do_utf8 ? SvCUR(dsv1) : strend - startpos;
1708 PerlIO_printf(Perl_debug_log,
1709 "%sMatching REx%s \"%s%*.*s%s%s\" against \"%s%.*s%s%s\"\n",
1710 PL_colors[4], PL_colors[5], PL_colors[0],
1713 len0 > 60 ? "..." : "",
1715 (int)(len1 > 60 ? 60 : len1),
1717 (len1 > 60 ? "..." : "")
1721 /* Simplest case: anchored match need be tried only once. */
1722 /* [unless only anchor is BOL and multiline is set] */
1723 if (prog->reganch & (ROPT_ANCH & ~ROPT_ANCH_GPOS)) {
1724 if (s == startpos && regtry(prog, startpos))
1726 else if (multiline || (prog->reganch & ROPT_IMPLICIT)
1727 || (prog->reganch & ROPT_ANCH_MBOL)) /* XXXX SBOL? */
1732 dontbother = minlen - 1;
1733 end = HOP3c(strend, -dontbother, strbeg) - 1;
1734 /* for multiline we only have to try after newlines */
1735 if (prog->check_substr || prog->check_utf8) {
1739 if (regtry(prog, s))
1744 if (prog->reganch & RE_USE_INTUIT) {
1745 s = re_intuit_start(prog, sv, s + 1, strend, flags, NULL);
1756 if (*s++ == '\n') { /* don't need PL_utf8skip here */
1757 if (regtry(prog, s))
1764 } else if (prog->reganch & ROPT_ANCH_GPOS) {
1765 if (regtry(prog, PL_reg_ganch))
1770 /* Messy cases: unanchored match. */
1771 if ((prog->anchored_substr || prog->anchored_utf8) && prog->reganch & ROPT_SKIP) {
1772 /* we have /x+whatever/ */
1773 /* it must be a one character string (XXXX Except UTF?) */
1778 if (!(do_utf8 ? prog->anchored_utf8 : prog->anchored_substr))
1779 do_utf8 ? to_utf8_substr(prog) : to_byte_substr(prog);
1780 ch = SvPVX_const(do_utf8 ? prog->anchored_utf8 : prog->anchored_substr)[0];
1783 while (s < strend) {
1785 DEBUG_EXECUTE_r( did_match = 1 );
1786 if (regtry(prog, s)) goto got_it;
1788 while (s < strend && *s == ch)
1795 while (s < strend) {
1797 DEBUG_EXECUTE_r( did_match = 1 );
1798 if (regtry(prog, s)) goto got_it;
1800 while (s < strend && *s == ch)
1806 DEBUG_EXECUTE_r(if (!did_match)
1807 PerlIO_printf(Perl_debug_log,
1808 "Did not find anchored character...\n")
1811 else if (prog->anchored_substr != NULL
1812 || prog->anchored_utf8 != NULL
1813 || ((prog->float_substr != NULL || prog->float_utf8 != NULL)
1814 && prog->float_max_offset < strend - s)) {
1819 char *last1; /* Last position checked before */
1823 if (prog->anchored_substr || prog->anchored_utf8) {
1824 if (!(do_utf8 ? prog->anchored_utf8 : prog->anchored_substr))
1825 do_utf8 ? to_utf8_substr(prog) : to_byte_substr(prog);
1826 must = do_utf8 ? prog->anchored_utf8 : prog->anchored_substr;
1827 back_max = back_min = prog->anchored_offset;
1829 if (!(do_utf8 ? prog->float_utf8 : prog->float_substr))
1830 do_utf8 ? to_utf8_substr(prog) : to_byte_substr(prog);
1831 must = do_utf8 ? prog->float_utf8 : prog->float_substr;
1832 back_max = prog->float_max_offset;
1833 back_min = prog->float_min_offset;
1835 if (must == &PL_sv_undef)
1836 /* could not downgrade utf8 check substring, so must fail */
1839 last = HOP3c(strend, /* Cannot start after this */
1840 -(I32)(CHR_SVLEN(must)
1841 - (SvTAIL(must) != 0) + back_min), strbeg);
1844 last1 = HOPc(s, -1);
1846 last1 = s - 1; /* bogus */
1848 /* XXXX check_substr already used to find "s", can optimize if
1849 check_substr==must. */
1851 dontbother = end_shift;
1852 strend = HOPc(strend, -dontbother);
1853 while ( (s <= last) &&
1854 ((flags & REXEC_SCREAM)
1855 ? (s = screaminstr(sv, must, HOP3c(s, back_min, strend) - strbeg,
1856 end_shift, &scream_pos, 0))
1857 : (s = fbm_instr((unsigned char*)HOP3(s, back_min, strend),
1858 (unsigned char*)strend, must,
1859 multiline ? FBMrf_MULTILINE : 0))) ) {
1860 /* we may be pointing at the wrong string */
1861 if ((flags & REXEC_SCREAM) && RX_MATCH_COPIED(prog))
1862 s = strbeg + (s - SvPVX_const(sv));
1863 DEBUG_EXECUTE_r( did_match = 1 );
1864 if (HOPc(s, -back_max) > last1) {
1865 last1 = HOPc(s, -back_min);
1866 s = HOPc(s, -back_max);
1869 char * const t = (last1 >= PL_bostr) ? HOPc(last1, 1) : last1 + 1;
1871 last1 = HOPc(s, -back_min);
1875 while (s <= last1) {
1876 if (regtry(prog, s))
1882 while (s <= last1) {
1883 if (regtry(prog, s))
1889 DEBUG_EXECUTE_r(if (!did_match)
1890 PerlIO_printf(Perl_debug_log,
1891 "Did not find %s substr \"%s%.*s%s\"%s...\n",
1892 ((must == prog->anchored_substr || must == prog->anchored_utf8)
1893 ? "anchored" : "floating"),
1895 (int)(SvCUR(must) - (SvTAIL(must)!=0)),
1897 PL_colors[1], (SvTAIL(must) ? "$" : ""))
1901 else if ((c = prog->regstclass)) {
1903 I32 op = (U8)OP(prog->regstclass);
1904 /* don't bother with what can't match */
1905 if (PL_regkind[op] != EXACT && op != CANY)
1906 strend = HOPc(strend, -(minlen - 1));
1909 SV *prop = sv_newmortal();
1917 pv_uni_display(dsv0, (U8*)SvPVX_const(prop), SvCUR(prop), 60,
1918 UNI_DISPLAY_REGEX) :
1920 len0 = UTF ? SvCUR(dsv0) : SvCUR(prop);
1922 sv_uni_display(dsv1, sv, 60, UNI_DISPLAY_REGEX) : s;
1923 len1 = UTF ? SvCUR(dsv1) : strend - s;
1924 PerlIO_printf(Perl_debug_log,
1925 "Matching stclass \"%*.*s\" against \"%*.*s\"\n",
1929 if (find_byclass(prog, c, s, strend, 0))
1931 DEBUG_EXECUTE_r(PerlIO_printf(Perl_debug_log, "Contradicts stclass...\n"));
1935 if (prog->float_substr != NULL || prog->float_utf8 != NULL) {
1940 if (!(do_utf8 ? prog->float_utf8 : prog->float_substr))
1941 do_utf8 ? to_utf8_substr(prog) : to_byte_substr(prog);
1942 float_real = do_utf8 ? prog->float_utf8 : prog->float_substr;
1944 if (flags & REXEC_SCREAM) {
1945 last = screaminstr(sv, float_real, s - strbeg,
1946 end_shift, &scream_pos, 1); /* last one */
1948 last = scream_olds; /* Only one occurrence. */
1949 /* we may be pointing at the wrong string */
1950 else if (RX_MATCH_COPIED(prog))
1951 s = strbeg + (s - SvPVX_const(sv));
1955 const char * const little = SvPV_const(float_real, len);
1957 if (SvTAIL(float_real)) {
1958 if (memEQ(strend - len + 1, little, len - 1))
1959 last = strend - len + 1;
1960 else if (!multiline)
1961 last = memEQ(strend - len, little, len)
1962 ? strend - len : NULL;
1968 last = rninstr(s, strend, little, little + len);
1970 last = strend; /* matching "$" */
1974 DEBUG_EXECUTE_r(PerlIO_printf(Perl_debug_log,
1975 "%sCan't trim the tail, match fails (should not happen)%s\n",
1976 PL_colors[4], PL_colors[5]));
1977 goto phooey; /* Should not happen! */
1979 dontbother = strend - last + prog->float_min_offset;
1981 if (minlen && (dontbother < minlen))
1982 dontbother = minlen - 1;
1983 strend -= dontbother; /* this one's always in bytes! */
1984 /* We don't know much -- general case. */
1987 if (regtry(prog, s))
1996 if (regtry(prog, s))
1998 } while (s++ < strend);
2006 RX_MATCH_TAINTED_set(prog, PL_reg_flags & RF_tainted);
2008 if (PL_reg_eval_set) {
2009 /* Preserve the current value of $^R */
2010 if (oreplsv != GvSV(PL_replgv))
2011 sv_setsv(oreplsv, GvSV(PL_replgv));/* So that when GvSV(replgv) is
2012 restored, the value remains
2014 restore_pos(aTHX_ prog);
2017 /* make sure $`, $&, $', and $digit will work later */
2018 if ( !(flags & REXEC_NOT_FIRST) ) {
2019 RX_MATCH_COPY_FREE(prog);
2020 if (flags & REXEC_COPY_STR) {
2021 I32 i = PL_regeol - startpos + (stringarg - strbeg);
2022 #ifdef PERL_OLD_COPY_ON_WRITE
2024 || (SvFLAGS(sv) & CAN_COW_MASK) == CAN_COW_FLAGS)) {
2026 PerlIO_printf(Perl_debug_log,
2027 "Copy on write: regexp capture, type %d\n",
2030 prog->saved_copy = sv_setsv_cow(prog->saved_copy, sv);
2031 prog->subbeg = (char *)SvPVX_const(prog->saved_copy);
2032 assert (SvPOKp(prog->saved_copy));
2036 RX_MATCH_COPIED_on(prog);
2037 s = savepvn(strbeg, i);
2043 prog->subbeg = strbeg;
2044 prog->sublen = PL_regeol - strbeg; /* strend may have been modified */
2051 DEBUG_EXECUTE_r(PerlIO_printf(Perl_debug_log, "%sMatch failed%s\n",
2052 PL_colors[4], PL_colors[5]));
2053 if (PL_reg_eval_set)
2054 restore_pos(aTHX_ prog);
2059 - regtry - try match at specific point
2061 STATIC I32 /* 0 failure, 1 success */
2062 S_regtry(pTHX_ regexp *prog, char *startpos)
2068 GET_RE_DEBUG_FLAGS_DECL;
2071 PL_regindent = 0; /* XXXX Not good when matches are reenterable... */
2073 if ((prog->reganch & ROPT_EVAL_SEEN) && !PL_reg_eval_set) {
2076 PL_reg_eval_set = RS_init;
2077 DEBUG_EXECUTE_r(DEBUG_s(
2078 PerlIO_printf(Perl_debug_log, " setting stack tmpbase at %"IVdf"\n",
2079 (IV)(PL_stack_sp - PL_stack_base));
2081 SAVEI32(cxstack[cxstack_ix].blk_oldsp);
2082 cxstack[cxstack_ix].blk_oldsp = PL_stack_sp - PL_stack_base;
2083 /* Otherwise OP_NEXTSTATE will free whatever on stack now. */
2085 /* Apparently this is not needed, judging by wantarray. */
2086 /* SAVEI8(cxstack[cxstack_ix].blk_gimme);
2087 cxstack[cxstack_ix].blk_gimme = G_SCALAR; */
2090 /* Make $_ available to executed code. */
2091 if (PL_reg_sv != DEFSV) {
2096 if (!(SvTYPE(PL_reg_sv) >= SVt_PVMG && SvMAGIC(PL_reg_sv)
2097 && (mg = mg_find(PL_reg_sv, PERL_MAGIC_regex_global)))) {
2098 /* prepare for quick setting of pos */
2099 #ifdef PERL_OLD_COPY_ON_WRITE
2101 sv_force_normal_flags(sv, 0);
2103 mg = sv_magicext(PL_reg_sv, (SV*)0, PERL_MAGIC_regex_global,
2104 &PL_vtbl_mglob, NULL, 0);
2108 PL_reg_oldpos = mg->mg_len;
2109 SAVEDESTRUCTOR_X(restore_pos, prog);
2111 if (!PL_reg_curpm) {
2112 Newxz(PL_reg_curpm, 1, PMOP);
2115 SV* repointer = newSViv(0);
2116 /* so we know which PL_regex_padav element is PL_reg_curpm */
2117 SvFLAGS(repointer) |= SVf_BREAK;
2118 av_push(PL_regex_padav,repointer);
2119 PL_reg_curpm->op_pmoffset = av_len(PL_regex_padav);
2120 PL_regex_pad = AvARRAY(PL_regex_padav);
2124 PM_SETRE(PL_reg_curpm, prog);
2125 PL_reg_oldcurpm = PL_curpm;
2126 PL_curpm = PL_reg_curpm;
2127 if (RX_MATCH_COPIED(prog)) {
2128 /* Here is a serious problem: we cannot rewrite subbeg,
2129 since it may be needed if this match fails. Thus
2130 $` inside (?{}) could fail... */
2131 PL_reg_oldsaved = prog->subbeg;
2132 PL_reg_oldsavedlen = prog->sublen;
2133 #ifdef PERL_OLD_COPY_ON_WRITE
2134 PL_nrs = prog->saved_copy;
2136 RX_MATCH_COPIED_off(prog);
2139 PL_reg_oldsaved = NULL;
2140 prog->subbeg = PL_bostr;
2141 prog->sublen = PL_regeol - PL_bostr; /* strend may have been modified */
2143 prog->startp[0] = startpos - PL_bostr;
2144 PL_reginput = startpos;
2145 PL_regstartp = prog->startp;
2146 PL_regendp = prog->endp;
2147 PL_reglastparen = &prog->lastparen;
2148 PL_reglastcloseparen = &prog->lastcloseparen;
2149 prog->lastparen = 0;
2150 prog->lastcloseparen = 0;
2152 DEBUG_EXECUTE_r(PL_reg_starttry = startpos);
2153 if (PL_reg_start_tmpl <= prog->nparens) {
2154 PL_reg_start_tmpl = prog->nparens*3/2 + 3;
2155 if(PL_reg_start_tmp)
2156 Renew(PL_reg_start_tmp, PL_reg_start_tmpl, char*);
2158 Newx(PL_reg_start_tmp, PL_reg_start_tmpl, char*);
2161 /* XXXX What this code is doing here?!!! There should be no need
2162 to do this again and again, PL_reglastparen should take care of
2165 /* Tests pat.t#187 and split.t#{13,14} seem to depend on this code.
2166 * Actually, the code in regcppop() (which Ilya may be meaning by
2167 * PL_reglastparen), is not needed at all by the test suite
2168 * (op/regexp, op/pat, op/split), but that code is needed, oddly
2169 * enough, for building DynaLoader, or otherwise this
2170 * "Error: '*' not in typemap in DynaLoader.xs, line 164"
2171 * will happen. Meanwhile, this code *is* needed for the
2172 * above-mentioned test suite tests to succeed. The common theme
2173 * on those tests seems to be returning null fields from matches.
2178 if (prog->nparens) {
2180 for (i = prog->nparens; i > (I32)*PL_reglastparen; i--) {
2187 if (regmatch(prog, prog->program + 1)) {
2188 prog->endp[0] = PL_reginput - PL_bostr;
2191 REGCP_UNWIND(lastcp);
2195 #define RE_UNWIND_BRANCH 1
2196 #define RE_UNWIND_BRANCHJ 2
2200 typedef struct { /* XX: makes sense to enlarge it... */
2204 } re_unwind_generic_t;
2218 } re_unwind_branch_t;
2220 typedef union re_unwind_t {
2222 re_unwind_generic_t generic;
2223 re_unwind_branch_t branch;
2226 #define sayYES goto yes
2227 #define sayNO goto no
2228 #define sayNO_ANYOF goto no_anyof
2229 #define sayYES_FINAL goto yes_final
2230 #define sayYES_LOUD goto yes_loud
2231 #define sayNO_FINAL goto no_final
2232 #define sayNO_SILENT goto do_no
2233 #define saySAME(x) if (x) goto yes; else goto no
2235 #define POSCACHE_SUCCESS 0 /* caching success rather than failure */
2236 #define POSCACHE_SEEN 1 /* we know what we're caching */
2237 #define POSCACHE_START 2 /* the real cache: this bit maps to pos 0 */
2238 #define CACHEsayYES STMT_START { \
2239 if (st->u.whilem.cache_offset | st->u.whilem.cache_bit) { \
2240 if (!(PL_reg_poscache[0] & (1<<POSCACHE_SEEN))) \
2241 PL_reg_poscache[0] |= (1<<POSCACHE_SUCCESS) || (1<<POSCACHE_SEEN); \
2242 else if (!(PL_reg_poscache[0] & (1<<POSCACHE_SUCCESS))) { \
2243 /* cache records failure, but this is success */ \
2245 PerlIO_printf(Perl_debug_log, \
2246 "%*s (remove success from failure cache)\n", \
2247 REPORT_CODE_OFF+PL_regindent*2, "") \
2249 PL_reg_poscache[st->u.whilem.cache_offset] &= ~(1<<st->u.whilem.cache_bit); \
2254 #define CACHEsayNO STMT_START { \
2255 if (st->u.whilem.cache_offset | st->u.whilem.cache_bit) { \
2256 if (!(PL_reg_poscache[0] & (1<<POSCACHE_SEEN))) \
2257 PL_reg_poscache[0] |= (1<<POSCACHE_SEEN); \
2258 else if ((PL_reg_poscache[0] & (1<<POSCACHE_SUCCESS))) { \
2259 /* cache records success, but this is failure */ \
2261 PerlIO_printf(Perl_debug_log, \
2262 "%*s (remove failure from success cache)\n", \
2263 REPORT_CODE_OFF+PL_regindent*2, "") \
2265 PL_reg_poscache[st->u.whilem.cache_offset] &= ~(1<<st->u.whilem.cache_bit); \
2271 /* this is used to determine how far from the left messages like
2272 'failed...' are printed. Currently 29 makes these messages line
2273 up with the opcode they refer to. Earlier perls used 25 which
2274 left these messages outdented making reviewing a debug output
2277 #define REPORT_CODE_OFF 29
2280 /* Make sure there is a test for this +1 options in re_tests */
2281 #define TRIE_INITAL_ACCEPT_BUFFLEN 4;
2283 /* grab a new slab and return the first slot in it */
2285 STATIC regmatch_state *
2288 regmatch_slab *s = PL_regmatch_slab->next;
2290 Newx(s, 1, regmatch_slab);
2291 s->prev = PL_regmatch_slab;
2293 PL_regmatch_slab->next = s;
2295 PL_regmatch_slab = s;
2296 return &s->states[0];
2299 /* simulate a recursive call to regmatch */
2301 #define REGMATCH(ns, where) \
2304 st->resume_state = resume_##where; \
2305 goto start_recurse; \
2306 resume_point_##where:
2309 /* push a new regex state. Set newst to point to it */
2311 #define PUSH_STATE(newst, resume) \
2313 DEBUG_EXECUTE_r(PerlIO_printf(Perl_debug_log, "PUSH STATE(%d)\n", depth)); \
2317 st->locinput = locinput; \
2318 st->resume_state = resume; \
2320 if (newst > &(PL_regmatch_slab->states[PERL_REGMATCH_SLAB_SLOTS-1])) \
2321 newst = S_push_slab(aTHX); \
2322 PL_regmatch_state = newst; \
2324 newst->minmod = 0; \
2326 newst->logical = 0; \
2327 newst->unwind = 0; \
2328 locinput = PL_reginput; \
2329 nextchr = UCHARAT(locinput);
2332 DEBUG_EXECUTE_r(PerlIO_printf(Perl_debug_log, "POP STATE(%d)\n", depth)); \
2335 if (st < &PL_regmatch_slab->states[0]) { \
2336 PL_regmatch_slab = PL_regmatch_slab->prev; \
2337 st = &PL_regmatch_slab->states[PERL_REGMATCH_SLAB_SLOTS-1]; \
2339 PL_regmatch_state = st; \
2343 locinput = st->locinput; \
2344 nextchr = UCHARAT(locinput);
2347 - regmatch - main matching routine
2349 * Conceptually the strategy is simple: check to see whether the current
2350 * node matches, call self recursively to see whether the rest matches,
2351 * and then act accordingly. In practice we make some effort to avoid
2352 * recursion, in particular by going through "ordinary" nodes (that don't
2353 * need to know whether the rest of the match failed) by a loop instead of
2356 /* [lwall] I've hoisted the register declarations to the outer block in order to
2357 * maybe save a little bit of pushing and popping on the stack. It also takes
2358 * advantage of machines that use a register save mask on subroutine entry.
2360 * This function used to be heavily recursive, but since this had the
2361 * effect of blowing the CPU stack on complex regexes, it has been
2362 * restructured to be iterative, and to save state onto the heap rather
2363 * than the stack. Essentially whereever regmatch() used to be called, it
2364 * pushes the current state, notes where to return, then jumps back into
2367 * Originally the structure of this function used to look something like
2372 while (scan != NULL) {
2373 a++; // do stuff with a and b
2379 if (regmatch(...)) // recurse
2389 * Now it looks something like this:
2397 regmatch_state *st = new();
2399 st->a++; // do stuff with a and b
2401 while (scan != NULL) {
2409 st->resume_state = resume_FOO;
2410 goto start_recurse; // recurse
2419 st = new(); push a new state
2420 st->a = 1; st->b = 2;
2427 switch (resume_state) {
2429 goto resume_point_FOO;
2436 * WARNING: this means that any line in this function that contains a
2437 * REGMATCH() or TRYPAREN() is actually simulating a recursive call to
2438 * regmatch() using gotos instead. Thus the values of any local variables
2439 * not saved in the regmatch_state structure will have been lost when
2440 * execution resumes on the next line .
2442 * States (ie the st pointer) are allocated in slabs of about 4K in size.
2443 * PL_regmatch_state always points to the currently active state, and
2444 * PL_regmatch_slab points to the slab currently containing PL_regmatch_state.
2445 * The first time regmatch is called, the first slab is allocated, and is
2446 * never freed until interpreter desctruction. When the slab is full,
2447 * a new one is allocated chained to the end. At exit from regmatch, slabs
2448 * allocated since entry are freed.
2452 STATIC I32 /* 0 failure, 1 success */
2453 S_regmatch(pTHX_ regexp *rex, regnode *prog)
2456 register const bool do_utf8 = PL_reg_match_utf8;
2457 const U32 uniflags = UTF8_ALLOW_DEFAULT;
2459 regmatch_slab *orig_slab;
2460 regmatch_state *orig_state;
2462 /* the current state. This is a cached copy of PL_regmatch_state */
2463 register regmatch_state *st;
2465 /* cache heavy used fields of st in registers */
2466 register regnode *scan;
2467 register regnode *next;
2468 register I32 n = 0; /* initialize to shut up compiler warning */
2469 register char *locinput = PL_reginput;
2471 /* these variables are NOT saved during a recusive RFEGMATCH: */
2472 register I32 nextchr; /* is always set to UCHARAT(locinput) */
2473 bool result; /* return value of S_regmatch */
2474 regnode *inner; /* Next node in internal branch. */
2475 int depth = 0; /* depth of recursion */
2476 regmatch_state *newst; /* when pushing a state, this is the new one */
2477 regmatch_state *cur_eval = NULL; /* most recent (??{}) state */
2480 SV *re_debug_flags = NULL;
2485 /* on first ever call to regmatch, allocate first slab */
2486 if (!PL_regmatch_slab) {
2487 Newx(PL_regmatch_slab, 1, regmatch_slab);
2488 PL_regmatch_slab->prev = NULL;
2489 PL_regmatch_slab->next = NULL;
2490 PL_regmatch_state = &PL_regmatch_slab->states[0] - 1;
2493 /* remember current high-water mark for exit */
2494 /* XXX this should be done with SAVE* instead */
2495 orig_slab = PL_regmatch_slab;
2496 orig_state = PL_regmatch_state;
2498 /* grab next free state slot */
2499 st = ++PL_regmatch_state;
2500 if (st > &(PL_regmatch_slab->states[PERL_REGMATCH_SLAB_SLOTS-1]))
2501 st = PL_regmatch_state = S_push_slab(aTHX);
2508 /* Note that nextchr is a byte even in UTF */
2509 nextchr = UCHARAT(locinput);
2511 while (scan != NULL) {
2514 SV * const prop = sv_newmortal();
2515 const int docolor = *PL_colors[0];
2516 const int taill = (docolor ? 10 : 7); /* 3 chars for "> <" */
2517 int l = (PL_regeol - locinput) > taill ? taill : (PL_regeol - locinput);
2518 /* The part of the string before starttry has one color
2519 (pref0_len chars), between starttry and current
2520 position another one (pref_len - pref0_len chars),
2521 after the current position the third one.
2522 We assume that pref0_len <= pref_len, otherwise we
2523 decrease pref0_len. */
2524 int pref_len = (locinput - PL_bostr) > (5 + taill) - l
2525 ? (5 + taill) - l : locinput - PL_bostr;
2528 while (do_utf8 && UTF8_IS_CONTINUATION(*(U8*)(locinput - pref_len)))
2530 pref0_len = pref_len - (locinput - PL_reg_starttry);
2531 if (l + pref_len < (5 + taill) && l < PL_regeol - locinput)
2532 l = ( PL_regeol - locinput > (5 + taill) - pref_len
2533 ? (5 + taill) - pref_len : PL_regeol - locinput);
2534 while (do_utf8 && UTF8_IS_CONTINUATION(*(U8*)(locinput + l)))
2538 if (pref0_len > pref_len)
2539 pref0_len = pref_len;
2540 regprop(prop, scan);
2542 const char * const s0 =
2543 do_utf8 && OP(scan) != CANY ?
2544 pv_uni_display(PERL_DEBUG_PAD(0), (U8*)(locinput - pref_len),
2545 pref0_len, 60, UNI_DISPLAY_REGEX) :
2546 locinput - pref_len;
2547 const int len0 = do_utf8 ? strlen(s0) : pref0_len;
2548 const char * const s1 = do_utf8 && OP(scan) != CANY ?
2549 pv_uni_display(PERL_DEBUG_PAD(1),
2550 (U8*)(locinput - pref_len + pref0_len),
2551 pref_len - pref0_len, 60, UNI_DISPLAY_REGEX) :
2552 locinput - pref_len + pref0_len;
2553 const int len1 = do_utf8 ? strlen(s1) : pref_len - pref0_len;
2554 const char * const s2 = do_utf8 && OP(scan) != CANY ?
2555 pv_uni_display(PERL_DEBUG_PAD(2), (U8*)locinput,
2556 PL_regeol - locinput, 60, UNI_DISPLAY_REGEX) :
2558 const int len2 = do_utf8 ? strlen(s2) : l;
2559 PerlIO_printf(Perl_debug_log,
2560 "%4"IVdf" <%s%.*s%s%s%.*s%s%s%s%.*s%s>%*s|%3"IVdf":%*s%s\n",
2561 (IV)(locinput - PL_bostr),
2568 (docolor ? "" : "> <"),
2572 15 - l - pref_len + 1,
2574 (IV)(scan - rex->program), PL_regindent*2, "",
2579 next = scan + NEXT_OFF(scan);
2585 if (locinput == PL_bostr)
2587 /* regtill = regbol; */
2592 if (locinput == PL_bostr ||
2593 ((nextchr || locinput < PL_regeol) && locinput[-1] == '\n'))
2599 if (locinput == PL_bostr)
2603 if (locinput == PL_reg_ganch)
2609 if ((nextchr || locinput < PL_regeol) && nextchr != '\n')
2614 if ((nextchr || locinput < PL_regeol) && nextchr != '\n')
2616 if (PL_regeol - locinput > 1)
2620 if (PL_regeol != locinput)
2624 if (!nextchr && locinput >= PL_regeol)
2627 locinput += PL_utf8skip[nextchr];
2628 if (locinput > PL_regeol)
2630 nextchr = UCHARAT(locinput);
2633 nextchr = UCHARAT(++locinput);
2636 if (!nextchr && locinput >= PL_regeol)
2638 nextchr = UCHARAT(++locinput);
2641 if ((!nextchr && locinput >= PL_regeol) || nextchr == '\n')
2644 locinput += PL_utf8skip[nextchr];
2645 if (locinput > PL_regeol)
2647 nextchr = UCHARAT(locinput);
2650 nextchr = UCHARAT(++locinput);
2656 traverse the TRIE keeping track of all accepting states
2657 we transition through until we get to a failing node.
2665 U8 *uc = ( U8* )locinput;
2672 U8 *uscan = (U8*)NULL;
2674 SV *sv_accept_buff = NULL;
2675 const enum { trie_plain, trie_utf8, trie_uft8_fold }
2676 trie_type = do_utf8 ?
2677 (OP(scan) == TRIE ? trie_utf8 : trie_uft8_fold)
2680 /* what trie are we using right now */
2682 = (reg_trie_data*)PL_reg_re->data->data[ ARG( scan ) ];
2683 st->u.trie.accepted = 0; /* how many accepting states we have seen */
2686 while ( state && uc <= (U8*)PL_regeol ) {
2688 if (trie->states[ state ].wordnum) {
2689 if (!st->u.trie.accepted ) {
2692 bufflen = TRIE_INITAL_ACCEPT_BUFFLEN;
2693 sv_accept_buff=newSV(bufflen *
2694 sizeof(reg_trie_accepted) - 1);
2695 SvCUR_set(sv_accept_buff,
2696 sizeof(reg_trie_accepted));
2697 SvPOK_on(sv_accept_buff);
2698 sv_2mortal(sv_accept_buff);
2699 st->u.trie.accept_buff =
2700 (reg_trie_accepted*)SvPV_nolen(sv_accept_buff );
2703 if (st->u.trie.accepted >= bufflen) {
2705 st->u.trie.accept_buff =(reg_trie_accepted*)
2706 SvGROW(sv_accept_buff,
2707 bufflen * sizeof(reg_trie_accepted));
2709 SvCUR_set(sv_accept_buff,SvCUR(sv_accept_buff)
2710 + sizeof(reg_trie_accepted));
2712 st->u.trie.accept_buff[st->u.trie.accepted].wordnum = trie->states[state].wordnum;
2713 st->u.trie.accept_buff[st->u.trie.accepted].endpos = uc;
2714 ++st->u.trie.accepted;
2717 base = trie->states[ state ].trans.base;
2719 DEBUG_TRIE_EXECUTE_r(
2720 PerlIO_printf( Perl_debug_log,
2721 "%*s %sState: %4"UVxf", Base: %4"UVxf", Accepted: %4"UVxf" ",
2722 REPORT_CODE_OFF + PL_regindent * 2, "", PL_colors[4],
2723 (UV)state, (UV)base, (UV)st->u.trie.accepted );
2727 switch (trie_type) {
2728 case trie_uft8_fold:
2730 uvc = utf8n_to_uvuni( uscan, UTF8_MAXLEN, &len, uniflags );
2735 U8 foldbuf[ UTF8_MAXBYTES_CASE + 1 ];
2736 uvc = utf8n_to_uvuni( (U8*)uc, UTF8_MAXLEN, &len, uniflags );
2737 uvc = to_uni_fold( uvc, foldbuf, &foldlen );
2738 foldlen -= UNISKIP( uvc );
2739 uscan = foldbuf + UNISKIP( uvc );
2743 uvc = utf8n_to_uvuni( (U8*)uc, UTF8_MAXLEN,
2752 charid = trie->charmap[ uvc ];
2756 if (trie->widecharmap) {
2757 SV** svpp = (SV**)NULL;
2758 svpp = hv_fetch(trie->widecharmap,
2759 (char*)&uvc, sizeof(UV), 0);
2761 charid = (U16)SvIV(*svpp);
2766 (base + charid > trie->uniquecharcount )
2767 && (base + charid - 1 - trie->uniquecharcount
2769 && trie->trans[base + charid - 1 -
2770 trie->uniquecharcount].check == state)
2772 state = trie->trans[base + charid - 1 -
2773 trie->uniquecharcount ].next;
2784 DEBUG_TRIE_EXECUTE_r(
2785 PerlIO_printf( Perl_debug_log,
2786 "Charid:%3x CV:%4"UVxf" After State: %4"UVxf"%s\n",
2787 charid, uvc, (UV)state, PL_colors[5] );
2790 if (!st->u.trie.accepted )
2794 There was at least one accepting state that we
2795 transitioned through. Presumably the number of accepting
2796 states is going to be low, typically one or two. So we
2797 simply scan through to find the one with lowest wordnum.
2798 Once we find it, we swap the last state into its place
2799 and decrement the size. We then try to match the rest of
2800 the pattern at the point where the word ends, if we
2801 succeed then we end the loop, otherwise the loop
2802 eventually terminates once all of the accepting states
2806 if ( st->u.trie.accepted == 1 ) {
2808 SV ** const tmp = av_fetch( trie->words, st->u.trie.accept_buff[ 0 ].wordnum-1, 0 );
2809 PerlIO_printf( Perl_debug_log,
2810 "%*s %sonly one match : #%d <%s>%s\n",
2811 REPORT_CODE_OFF+PL_regindent*2, "", PL_colors[4],
2812 st->u.trie.accept_buff[ 0 ].wordnum,
2813 tmp ? SvPV_nolen_const( *tmp ) : "not compiled under -Dr",
2816 PL_reginput = (char *)st->u.trie.accept_buff[ 0 ].endpos;
2817 /* in this case we free tmps/leave before we call regmatch
2818 as we wont be using accept_buff again. */
2821 REGMATCH(scan + NEXT_OFF(scan), TRIE1);
2822 /*** all unsaved local vars undefined at this point */
2825 PerlIO_printf( Perl_debug_log,"%*s %sgot %"IVdf" possible matches%s\n",
2826 REPORT_CODE_OFF + PL_regindent * 2, "", PL_colors[4], (IV)st->u.trie.accepted,
2829 while ( !result && st->u.trie.accepted-- ) {
2832 for( cur = 1 ; cur <= st->u.trie.accepted ; cur++ ) {
2833 DEBUG_TRIE_EXECUTE_r(
2834 PerlIO_printf( Perl_debug_log,
2835 "%*s %sgot %"IVdf" (%d) as best, looking at %"IVdf" (%d)%s\n",
2836 REPORT_CODE_OFF + PL_regindent * 2, "", PL_colors[4],
2837 (IV)best, st->u.trie.accept_buff[ best ].wordnum, (IV)cur,
2838 st->u.trie.accept_buff[ cur ].wordnum, PL_colors[5] );
2841 if (st->u.trie.accept_buff[cur].wordnum <
2842 st->u.trie.accept_buff[best].wordnum)
2846 reg_trie_data * const trie = (reg_trie_data*)
2847 PL_reg_re->data->data[ARG(scan)];
2848 SV ** const tmp = av_fetch( trie->words, st->u.trie.accept_buff[ best ].wordnum - 1, 0 );
2849 PerlIO_printf( Perl_debug_log, "%*s %strying alternation #%d <%s> at 0x%p%s\n",
2850 REPORT_CODE_OFF+PL_regindent*2, "", PL_colors[4],
2851 st->u.trie.accept_buff[best].wordnum,
2852 tmp ? SvPV_nolen_const( *tmp ) : "not compiled under -Dr",scan,
2855 if ( best<st->u.trie.accepted ) {
2856 reg_trie_accepted tmp = st->u.trie.accept_buff[ best ];
2857 st->u.trie.accept_buff[ best ] = st->u.trie.accept_buff[ st->u.trie.accepted ];
2858 st->u.trie.accept_buff[ st->u.trie.accepted ] = tmp;
2859 best = st->u.trie.accepted;
2861 PL_reginput = (char *)st->u.trie.accept_buff[ best ].endpos;
2864 as far as I can tell we only need the SAVETMPS/FREETMPS
2865 for re's with EVAL in them but I'm leaving them in for
2866 all until I can be sure.
2869 REGMATCH(scan + NEXT_OFF(scan), TRIE2);
2870 /*** all unsaved local vars undefined at this point */
2883 /* unreached codepoint */
2885 char *s = STRING(scan);
2886 st->ln = STR_LEN(scan);
2887 if (do_utf8 != UTF) {
2888 /* The target and the pattern have differing utf8ness. */
2890 const char *e = s + st->ln;
2893 /* The target is utf8, the pattern is not utf8. */
2898 if (NATIVE_TO_UNI(*(U8*)s) !=
2899 utf8n_to_uvuni((U8*)l, UTF8_MAXBYTES, &ulen,
2907 /* The target is not utf8, the pattern is utf8. */
2912 if (NATIVE_TO_UNI(*((U8*)l)) !=
2913 utf8n_to_uvuni((U8*)s, UTF8_MAXBYTES, &ulen,
2921 nextchr = UCHARAT(locinput);
2924 /* The target and the pattern have the same utf8ness. */
2925 /* Inline the first character, for speed. */
2926 if (UCHARAT(s) != nextchr)
2928 if (PL_regeol - locinput < st->ln)
2930 if (st->ln > 1 && memNE(s, locinput, st->ln))
2933 nextchr = UCHARAT(locinput);
2937 PL_reg_flags |= RF_tainted;
2940 char *s = STRING(scan);
2941 st->ln = STR_LEN(scan);
2943 if (do_utf8 || UTF) {
2944 /* Either target or the pattern are utf8. */
2946 char *e = PL_regeol;
2948 if (ibcmp_utf8(s, 0, st->ln, (bool)UTF,
2949 l, &e, 0, do_utf8)) {
2950 /* One more case for the sharp s:
2951 * pack("U0U*", 0xDF) =~ /ss/i,
2952 * the 0xC3 0x9F are the UTF-8
2953 * byte sequence for the U+00DF. */
2955 toLOWER(s[0]) == 's' &&
2957 toLOWER(s[1]) == 's' &&
2964 nextchr = UCHARAT(locinput);
2968 /* Neither the target and the pattern are utf8. */
2970 /* Inline the first character, for speed. */
2971 if (UCHARAT(s) != nextchr &&
2972 UCHARAT(s) != ((OP(scan) == EXACTF)
2973 ? PL_fold : PL_fold_locale)[nextchr])
2975 if (PL_regeol - locinput < st->ln)
2977 if (st->ln > 1 && (OP(scan) == EXACTF
2978 ? ibcmp(s, locinput, st->ln)
2979 : ibcmp_locale(s, locinput, st->ln)))
2982 nextchr = UCHARAT(locinput);
2987 STRLEN inclasslen = PL_regeol - locinput;
2989 if (!reginclass(scan, (U8*)locinput, &inclasslen, do_utf8))
2991 if (locinput >= PL_regeol)
2993 locinput += inclasslen ? inclasslen : UTF8SKIP(locinput);
2994 nextchr = UCHARAT(locinput);
2999 nextchr = UCHARAT(locinput);
3000 if (!REGINCLASS(scan, (U8*)locinput))
3002 if (!nextchr && locinput >= PL_regeol)
3004 nextchr = UCHARAT(++locinput);
3008 /* If we might have the case of the German sharp s
3009 * in a casefolding Unicode character class. */
3011 if (ANYOF_FOLD_SHARP_S(scan, locinput, PL_regeol)) {
3012 locinput += SHARP_S_SKIP;
3013 nextchr = UCHARAT(locinput);
3019 PL_reg_flags |= RF_tainted;
3025 LOAD_UTF8_CHARCLASS_ALNUM();
3026 if (!(OP(scan) == ALNUM
3027 ? swash_fetch(PL_utf8_alnum, (U8*)locinput, do_utf8)
3028 : isALNUM_LC_utf8((U8*)locinput)))
3032 locinput += PL_utf8skip[nextchr];
3033 nextchr = UCHARAT(locinput);
3036 if (!(OP(scan) == ALNUM
3037 ? isALNUM(nextchr) : isALNUM_LC(nextchr)))
3039 nextchr = UCHARAT(++locinput);
3042 PL_reg_flags |= RF_tainted;
3045 if (!nextchr && locinput >= PL_regeol)
3048 LOAD_UTF8_CHARCLASS_ALNUM();
3049 if (OP(scan) == NALNUM
3050 ? swash_fetch(PL_utf8_alnum, (U8*)locinput, do_utf8)
3051 : isALNUM_LC_utf8((U8*)locinput))
3055 locinput += PL_utf8skip[nextchr];
3056 nextchr = UCHARAT(locinput);
3059 if (OP(scan) == NALNUM
3060 ? isALNUM(nextchr) : isALNUM_LC(nextchr))
3062 nextchr = UCHARAT(++locinput);
3066 PL_reg_flags |= RF_tainted;
3070 /* was last char in word? */
3072 if (locinput == PL_bostr)
3075 const U8 * const r = reghop3((U8*)locinput, -1, (U8*)PL_bostr);
3077 st->ln = utf8n_to_uvchr(r, UTF8SKIP(r), 0, uniflags);
3079 if (OP(scan) == BOUND || OP(scan) == NBOUND) {
3080 st->ln = isALNUM_uni(st->ln);
3081 LOAD_UTF8_CHARCLASS_ALNUM();
3082 n = swash_fetch(PL_utf8_alnum, (U8*)locinput, do_utf8);
3085 st->ln = isALNUM_LC_uvchr(UNI_TO_NATIVE(st->ln));
3086 n = isALNUM_LC_utf8((U8*)locinput);
3090 st->ln = (locinput != PL_bostr) ?
3091 UCHARAT(locinput - 1) : '\n';
3092 if (OP(scan) == BOUND || OP(scan) == NBOUND) {
3093 st->ln = isALNUM(st->ln);
3094 n = isALNUM(nextchr);
3097 st->ln = isALNUM_LC(st->ln);
3098 n = isALNUM_LC(nextchr);
3101 if (((!st->ln) == (!n)) == (OP(scan) == BOUND ||
3102 OP(scan) == BOUNDL))
3106 PL_reg_flags |= RF_tainted;
3112 if (UTF8_IS_CONTINUED(nextchr)) {
3113 LOAD_UTF8_CHARCLASS_SPACE();
3114 if (!(OP(scan) == SPACE
3115 ? swash_fetch(PL_utf8_space, (U8*)locinput, do_utf8)
3116 : isSPACE_LC_utf8((U8*)locinput)))
3120 locinput += PL_utf8skip[nextchr];
3121 nextchr = UCHARAT(locinput);
3124 if (!(OP(scan) == SPACE
3125 ? isSPACE(nextchr) : isSPACE_LC(nextchr)))
3127 nextchr = UCHARAT(++locinput);
3130 if (!(OP(scan) == SPACE
3131 ? isSPACE(nextchr) : isSPACE_LC(nextchr)))
3133 nextchr = UCHARAT(++locinput);
3137 PL_reg_flags |= RF_tainted;
3140 if (!nextchr && locinput >= PL_regeol)
3143 LOAD_UTF8_CHARCLASS_SPACE();
3144 if (OP(scan) == NSPACE
3145 ? swash_fetch(PL_utf8_space, (U8*)locinput, do_utf8)
3146 : isSPACE_LC_utf8((U8*)locinput))
3150 locinput += PL_utf8skip[nextchr];
3151 nextchr = UCHARAT(locinput);
3154 if (OP(scan) == NSPACE
3155 ? isSPACE(nextchr) : isSPACE_LC(nextchr))
3157 nextchr = UCHARAT(++locinput);
3160 PL_reg_flags |= RF_tainted;
3166 LOAD_UTF8_CHARCLASS_DIGIT();
3167 if (!(OP(scan) == DIGIT
3168 ? swash_fetch(PL_utf8_digit, (U8*)locinput, do_utf8)
3169 : isDIGIT_LC_utf8((U8*)locinput)))
3173 locinput += PL_utf8skip[nextchr];
3174 nextchr = UCHARAT(locinput);
3177 if (!(OP(scan) == DIGIT
3178 ? isDIGIT(nextchr) : isDIGIT_LC(nextchr)))
3180 nextchr = UCHARAT(++locinput);
3183 PL_reg_flags |= RF_tainted;
3186 if (!nextchr && locinput >= PL_regeol)
3189 LOAD_UTF8_CHARCLASS_DIGIT();
3190 if (OP(scan) == NDIGIT
3191 ? swash_fetch(PL_utf8_digit, (U8*)locinput, do_utf8)
3192 : isDIGIT_LC_utf8((U8*)locinput))
3196 locinput += PL_utf8skip[nextchr];
3197 nextchr = UCHARAT(locinput);
3200 if (OP(scan) == NDIGIT
3201 ? isDIGIT(nextchr) : isDIGIT_LC(nextchr))
3203 nextchr = UCHARAT(++locinput);
3206 if (locinput >= PL_regeol)
3209 LOAD_UTF8_CHARCLASS_MARK();
3210 if (swash_fetch(PL_utf8_mark,(U8*)locinput, do_utf8))
3212 locinput += PL_utf8skip[nextchr];
3213 while (locinput < PL_regeol &&
3214 swash_fetch(PL_utf8_mark,(U8*)locinput, do_utf8))
3215 locinput += UTF8SKIP(locinput);
3216 if (locinput > PL_regeol)
3221 nextchr = UCHARAT(locinput);
3224 PL_reg_flags |= RF_tainted;
3229 n = ARG(scan); /* which paren pair */
3230 st->ln = PL_regstartp[n];
3231 PL_reg_leftiter = PL_reg_maxiter; /* Void cache */
3232 if ((I32)*PL_reglastparen < n || st->ln == -1)
3233 sayNO; /* Do not match unless seen CLOSEn. */
3234 if (st->ln == PL_regendp[n])
3237 s = PL_bostr + st->ln;
3238 if (do_utf8 && OP(scan) != REF) { /* REF can do byte comparison */
3240 const char *e = PL_bostr + PL_regendp[n];
3242 * Note that we can't do the "other character" lookup trick as
3243 * in the 8-bit case (no pun intended) because in Unicode we
3244 * have to map both upper and title case to lower case.
3246 if (OP(scan) == REFF) {
3248 STRLEN ulen1, ulen2;
3249 U8 tmpbuf1[UTF8_MAXBYTES_CASE+1];
3250 U8 tmpbuf2[UTF8_MAXBYTES_CASE+1];
3254 toLOWER_utf8((U8*)s, tmpbuf1, &ulen1);
3255 toLOWER_utf8((U8*)l, tmpbuf2, &ulen2);
3256 if (ulen1 != ulen2 || memNE((char *)tmpbuf1, (char *)tmpbuf2, ulen1))
3263 nextchr = UCHARAT(locinput);
3267 /* Inline the first character, for speed. */
3268 if (UCHARAT(s) != nextchr &&
3270 (UCHARAT(s) != ((OP(scan) == REFF
3271 ? PL_fold : PL_fold_locale)[nextchr]))))
3273 st->ln = PL_regendp[n] - st->ln;
3274 if (locinput + st->ln > PL_regeol)
3276 if (st->ln > 1 && (OP(scan) == REF
3277 ? memNE(s, locinput, st->ln)
3279 ? ibcmp(s, locinput, st->ln)
3280 : ibcmp_locale(s, locinput, st->ln))))
3283 nextchr = UCHARAT(locinput);
3296 /* execute the code in the {...} */
3298 SV ** const before = SP;
3299 OP_4tree * const oop = PL_op;
3300 COP * const ocurcop = PL_curcop;
3302 struct regexp * const oreg = PL_reg_re;
3305 PL_op = (OP_4tree*)PL_reg_re->data->data[n];
3306 DEBUG_EXECUTE_r( PerlIO_printf(Perl_debug_log, " re_eval 0x%"UVxf"\n", PTR2UV(PL_op)) );
3307 PAD_SAVE_LOCAL(old_comppad, (PAD*)PL_reg_re->data->data[n + 2]);
3308 PL_regendp[0] = PL_reg_magic->mg_len = locinput - PL_bostr;
3310 CALLRUNOPS(aTHX); /* Scalar context. */
3313 ret = &PL_sv_undef; /* protect against empty (?{}) blocks. */
3320 PAD_RESTORE_LOCAL(old_comppad);
3321 PL_curcop = ocurcop;
3325 sv_setsv(save_scalar(PL_replgv), ret);
3329 if (st->logical == 2) { /* Postponed subexpression: /(??{...})/ */
3332 /* extract RE object from returned value; compiling if
3337 if(SvROK(ret) && SvSMAGICAL(sv = SvRV(ret)))
3338 mg = mg_find(sv, PERL_MAGIC_qr);
3339 else if (SvSMAGICAL(ret)) {
3340 if (SvGMAGICAL(ret))
3341 sv_unmagic(ret, PERL_MAGIC_qr);
3343 mg = mg_find(ret, PERL_MAGIC_qr);
3347 re = (regexp *)mg->mg_obj;
3348 (void)ReREFCNT_inc(re);
3352 const char * const t = SvPV_const(ret, len);
3354 const I32 osize = PL_regsize;
3357 if (DO_UTF8(ret)) pm.op_pmdynflags |= PMdf_DYN_UTF8;
3358 re = CALLREGCOMP(aTHX_ (char*)t, (char*)t + len, &pm);
3360 & (SVs_TEMP | SVs_PADTMP | SVf_READONLY
3362 sv_magic(ret,(SV*)ReREFCNT_inc(re),
3368 /* run the pattern returned from (??{...}) */
3371 PerlIO_printf(Perl_debug_log,
3372 "Entering embedded \"%s%.60s%s%s\"\n",
3376 (strlen(re->precomp) > 60 ? "..." : ""))
3379 st->u.eval.cp = regcppush(0); /* Save *all* the positions. */
3380 REGCP_SET(st->u.eval.lastcp);
3381 *PL_reglastparen = 0;
3382 *PL_reglastcloseparen = 0;
3383 PL_reginput = locinput;
3385 /* XXXX This is too dramatic a measure... */
3389 st->u.eval.toggleutf = ((PL_reg_flags & RF_utf8) != 0) ^
3390 ((re->reganch & ROPT_UTF8) != 0);
3391 if (st->u.eval.toggleutf) PL_reg_flags ^= RF_utf8;
3392 st->u.eval.prev_rex = rex;
3393 assert(rex == PL_reg_re); /* XXX */
3395 PL_reg_re = rex; /* XXX */
3397 st->u.eval.prev_eval = cur_eval;
3398 st->u.eval.prev_slab = PL_regmatch_slab;
3399 st->u.eval.depth = depth;
3401 PUSH_STATE(newst, resume_EVAL);
3404 /* now continue from first node in postoned RE */
3405 next = re->program + 1;
3409 /* /(?(?{...})X|Y)/ */
3410 st->sw = SvTRUE(ret);
3415 n = ARG(scan); /* which paren pair */
3416 PL_reg_start_tmp[n] = locinput;
3421 n = ARG(scan); /* which paren pair */
3422 PL_regstartp[n] = PL_reg_start_tmp[n] - PL_bostr;
3423 PL_regendp[n] = locinput - PL_bostr;
3424 if (n > (I32)*PL_reglastparen)
3425 *PL_reglastparen = n;
3426 *PL_reglastcloseparen = n;
3429 n = ARG(scan); /* which paren pair */
3430 st->sw = ((I32)*PL_reglastparen >= n && PL_regendp[n] != -1);
3433 PL_reg_leftiter = PL_reg_maxiter; /* Void cache */
3435 next = NEXTOPER(NEXTOPER(scan));
3437 next = scan + ARG(scan);
3438 if (OP(next) == IFTHEN) /* Fake one. */
3439 next = NEXTOPER(NEXTOPER(next));
3443 st->logical = scan->flags;
3445 /*******************************************************************
3446 cc points to the regmatch_state associated with the most recent CURLYX.
3447 This struct contains info about the innermost (...)* loop (an
3448 "infoblock"), and a pointer to the next outer cc.
3450 Here is how Y(A)*Z is processed (if it is compiled into CURLYX/WHILEM):
3452 1) After matching Y, regnode for CURLYX is processed;
3454 2) This regnode populates cc, and calls regmatch() recursively
3455 with the starting point at WHILEM node;
3457 3) Each hit of WHILEM node tries to match A and Z (in the order
3458 depending on the current iteration, min/max of {min,max} and
3459 greediness). The information about where are nodes for "A"
3460 and "Z" is read from cc, as is info on how many times "A"
3461 was already matched, and greediness.
3463 4) After A matches, the same WHILEM node is hit again.
3465 5) Each time WHILEM is hit, cc is the infoblock created by CURLYX
3466 of the same pair. Thus when WHILEM tries to match Z, it temporarily
3467 resets cc, since this Y(A)*Z can be a part of some other loop:
3468 as in (Y(A)*Z)*. If Z matches, the automaton will hit the WHILEM node
3469 of the external loop.
3471 Currently present infoblocks form a tree with a stem formed by st->cc
3472 and whatever it mentions via ->next, and additional attached trees
3473 corresponding to temporarily unset infoblocks as in "5" above.
3475 In the following picture, infoblocks for outer loop of
3476 (Y(A)*?Z)*?T are denoted O, for inner I. NULL starting block
3477 is denoted by x. The matched string is YAAZYAZT. Temporarily postponed
3478 infoblocks are drawn below the "reset" infoblock.
3480 In fact in the picture below we do not show failed matches for Z and T
3481 by WHILEM blocks. [We illustrate minimal matches, since for them it is
3482 more obvious *why* one needs to *temporary* unset infoblocks.]
3484 Matched REx position InfoBlocks Comment
3488 Y A)*?Z)*?T x <- O <- I
3489 YA )*?Z)*?T x <- O <- I
3490 YA A)*?Z)*?T x <- O <- I
3491 YAA )*?Z)*?T x <- O <- I
3492 YAA Z)*?T x <- O # Temporary unset I
3495 YAAZ Y(A)*?Z)*?T x <- O
3498 YAAZY (A)*?Z)*?T x <- O
3501 YAAZY A)*?Z)*?T x <- O <- I
3504 YAAZYA )*?Z)*?T x <- O <- I
3507 YAAZYA Z)*?T x <- O # Temporary unset I
3513 YAAZYAZ T x # Temporary unset O
3520 *******************************************************************/
3523 /* No need to save/restore up to this paren */
3524 I32 parenfloor = scan->flags;
3526 if (OP(PREVOPER(next)) == NOTHING) /* LONGJMP */
3528 /* XXXX Probably it is better to teach regpush to support
3529 parenfloor > PL_regsize... */
3530 if (parenfloor > (I32)*PL_reglastparen)
3531 parenfloor = *PL_reglastparen; /* Pessimization... */
3533 st->u.curlyx.cp = PL_savestack_ix;
3534 st->u.curlyx.outercc = st->cc;
3536 /* these fields contain the state of the current curly.
3537 * they are accessed by subsequent WHILEMs;
3538 * cur and lastloc are also updated by WHILEM */
3539 st->u.curlyx.parenfloor = parenfloor;
3540 st->u.curlyx.cur = -1; /* this will be updated by WHILEM */
3541 st->u.curlyx.min = ARG1(scan);
3542 st->u.curlyx.max = ARG2(scan);
3543 st->u.curlyx.scan = NEXTOPER(scan) + EXTRA_STEP_2ARGS;
3544 st->u.curlyx.lastloc = 0;
3545 /* st->next and st->minmod are also read by WHILEM */
3547 PL_reginput = locinput;
3548 REGMATCH(PREVOPER(next), CURLYX); /* start on the WHILEM */
3549 /*** all unsaved local vars undefined at this point */
3550 regcpblow(st->u.curlyx.cp);
3551 st->cc = st->u.curlyx.outercc;
3557 * This is really hard to understand, because after we match
3558 * what we're trying to match, we must make sure the rest of
3559 * the REx is going to match for sure, and to do that we have
3560 * to go back UP the parse tree by recursing ever deeper. And
3561 * if it fails, we have to reset our parent's current state
3562 * that we can try again after backing off.
3565 st->u.whilem.lastloc = st->cc->u.curlyx.lastloc; /* Detection of 0-len. */
3566 st->u.whilem.cache_offset = 0;
3567 st->u.whilem.cache_bit = 0;
3569 n = st->cc->u.curlyx.cur + 1; /* how many we know we matched */
3570 PL_reginput = locinput;
3573 PerlIO_printf(Perl_debug_log,
3574 "%*s %ld out of %ld..%ld cc=%"UVxf"\n",
3575 REPORT_CODE_OFF+PL_regindent*2, "",
3576 (long)n, (long)st->cc->u.curlyx.min,
3577 (long)st->cc->u.curlyx.max, PTR2UV(st->cc))
3580 /* If degenerate scan matches "", assume scan done. */
3582 if (locinput == st->cc->u.curlyx.lastloc && n >= st->cc->u.curlyx.min) {
3583 st->u.whilem.savecc = st->cc;
3584 st->cc = st->cc->u.curlyx.outercc;
3586 st->ln = st->cc->u.curlyx.cur;
3588 PerlIO_printf(Perl_debug_log,
3589 "%*s empty match detected, try continuation...\n",
3590 REPORT_CODE_OFF+PL_regindent*2, "")
3592 REGMATCH(st->u.whilem.savecc->next, WHILEM1);
3593 /*** all unsaved local vars undefined at this point */
3594 st->cc = st->u.whilem.savecc;
3597 if (st->cc->u.curlyx.outercc)
3598 st->cc->u.curlyx.outercc->u.curlyx.cur = st->ln;
3602 /* First just match a string of min scans. */
3604 if (n < st->cc->u.curlyx.min) {
3605 st->cc->u.curlyx.cur = n;
3606 st->cc->u.curlyx.lastloc = locinput;
3607 REGMATCH(st->cc->u.curlyx.scan, WHILEM2);
3608 /*** all unsaved local vars undefined at this point */
3611 st->cc->u.curlyx.cur = n - 1;
3612 st->cc->u.curlyx.lastloc = st->u.whilem.lastloc;
3617 /* Check whether we already were at this position.
3618 Postpone detection until we know the match is not
3619 *that* much linear. */
3620 if (!PL_reg_maxiter) {
3621 PL_reg_maxiter = (PL_regeol - PL_bostr + 1) * (scan->flags>>4);
3622 PL_reg_leftiter = PL_reg_maxiter;
3624 if (PL_reg_leftiter-- == 0) {
3625 const I32 size = (PL_reg_maxiter + 7 + POSCACHE_START)/8;
3626 if (PL_reg_poscache) {
3627 if ((I32)PL_reg_poscache_size < size) {
3628 Renew(PL_reg_poscache, size, char);
3629 PL_reg_poscache_size = size;
3631 Zero(PL_reg_poscache, size, char);
3634 PL_reg_poscache_size = size;
3635 Newxz(PL_reg_poscache, size, char);
3638 PerlIO_printf(Perl_debug_log,
3639 "%sDetected a super-linear match, switching on caching%s...\n",
3640 PL_colors[4], PL_colors[5])
3643 if (PL_reg_leftiter < 0) {
3644 st->u.whilem.cache_offset = locinput - PL_bostr;
3646 st->u.whilem.cache_offset = (scan->flags & 0xf) - 1 + POSCACHE_START
3647 + st->u.whilem.cache_offset * (scan->flags>>4);
3648 st->u.whilem.cache_bit = st->u.whilem.cache_offset % 8;
3649 st->u.whilem.cache_offset /= 8;
3650 if (PL_reg_poscache[st->u.whilem.cache_offset] & (1<<st->u.whilem.cache_bit)) {
3652 PerlIO_printf(Perl_debug_log,
3653 "%*s already tried at this position...\n",
3654 REPORT_CODE_OFF+PL_regindent*2, "")
3656 if (PL_reg_poscache[0] & (1<<POSCACHE_SUCCESS))
3657 /* cache records success */
3660 /* cache records failure */
3663 PL_reg_poscache[st->u.whilem.cache_offset] |= (1<<st->u.whilem.cache_bit);
3667 /* Prefer next over scan for minimal matching. */
3669 if (st->cc->minmod) {
3670 st->u.whilem.savecc = st->cc;
3671 st->cc = st->cc->u.curlyx.outercc;
3673 st->ln = st->cc->u.curlyx.cur;
3674 st->u.whilem.cp = regcppush(st->u.whilem.savecc->u.curlyx.parenfloor);
3675 REGCP_SET(st->u.whilem.lastcp);
3676 REGMATCH(st->u.whilem.savecc->next, WHILEM3);
3677 /*** all unsaved local vars undefined at this point */
3678 st->cc = st->u.whilem.savecc;
3680 regcpblow(st->u.whilem.cp);
3681 CACHEsayYES; /* All done. */
3683 REGCP_UNWIND(st->u.whilem.lastcp);
3685 if (st->cc->u.curlyx.outercc)
3686 st->cc->u.curlyx.outercc->u.curlyx.cur = st->ln;
3688 if (n >= st->cc->u.curlyx.max) { /* Maximum greed exceeded? */
3689 if (ckWARN(WARN_REGEXP) && n >= REG_INFTY
3690 && !(PL_reg_flags & RF_warned)) {
3691 PL_reg_flags |= RF_warned;
3692 Perl_warner(aTHX_ packWARN(WARN_REGEXP), "%s limit (%d) exceeded",
3693 "Complex regular subexpression recursion",
3700 PerlIO_printf(Perl_debug_log,
3701 "%*s trying longer...\n",
3702 REPORT_CODE_OFF+PL_regindent*2, "")
3704 /* Try scanning more and see if it helps. */
3705 PL_reginput = locinput;
3706 st->cc->u.curlyx.cur = n;
3707 st->cc->u.curlyx.lastloc = locinput;
3708 st->u.whilem.cp = regcppush(st->cc->u.curlyx.parenfloor);
3709 REGCP_SET(st->u.whilem.lastcp);
3710 REGMATCH(st->cc->u.curlyx.scan, WHILEM4);
3711 /*** all unsaved local vars undefined at this point */
3713 regcpblow(st->u.whilem.cp);
3716 REGCP_UNWIND(st->u.whilem.lastcp);
3718 st->cc->u.curlyx.cur = n - 1;
3719 st->cc->u.curlyx.lastloc = st->u.whilem.lastloc;
3723 /* Prefer scan over next for maximal matching. */
3725 if (n < st->cc->u.curlyx.max) { /* More greed allowed? */
3726 st->u.whilem.cp = regcppush(st->cc->u.curlyx.parenfloor);
3727 st->cc->u.curlyx.cur = n;
3728 st->cc->u.curlyx.lastloc = locinput;
3729 REGCP_SET(st->u.whilem.lastcp);
3730 REGMATCH(st->cc->u.curlyx.scan, WHILEM5);
3731 /*** all unsaved local vars undefined at this point */
3733 regcpblow(st->u.whilem.cp);
3736 REGCP_UNWIND(st->u.whilem.lastcp);
3737 regcppop(rex); /* Restore some previous $<digit>s? */
3738 PL_reginput = locinput;
3740 PerlIO_printf(Perl_debug_log,
3741 "%*s failed, try continuation...\n",
3742 REPORT_CODE_OFF+PL_regindent*2, "")
3745 if (ckWARN(WARN_REGEXP) && n >= REG_INFTY
3746 && !(PL_reg_flags & RF_warned)) {
3747 PL_reg_flags |= RF_warned;
3748 Perl_warner(aTHX_ packWARN(WARN_REGEXP), "%s limit (%d) exceeded",
3749 "Complex regular subexpression recursion",
3753 /* Failed deeper matches of scan, so see if this one works. */
3754 st->u.whilem.savecc = st->cc;
3755 st->cc = st->cc->u.curlyx.outercc;
3757 st->ln = st->cc->u.curlyx.cur;
3758 REGMATCH(st->u.whilem.savecc->next, WHILEM6);
3759 /*** all unsaved local vars undefined at this point */
3760 st->cc = st->u.whilem.savecc;
3763 if (st->cc->u.curlyx.outercc)
3764 st->cc->u.curlyx.outercc->u.curlyx.cur = st->ln;
3765 st->cc->u.curlyx.cur = n - 1;
3766 st->cc->u.curlyx.lastloc = st->u.whilem.lastloc;
3771 next = scan + ARG(scan);
3774 inner = NEXTOPER(NEXTOPER(scan));
3777 inner = NEXTOPER(scan);
3782 if (OP(next) != type) /* No choice. */
3783 next = inner; /* Avoid recursion. */
3785 const I32 lastparen = *PL_reglastparen;
3786 /* Put unwinding data on stack */
3787 const I32 unwind1 = SSNEWt(1,re_unwind_branch_t);
3788 re_unwind_branch_t * const uw = SSPTRt(unwind1,re_unwind_branch_t);
3790 uw->prev = st->unwind;
3791 st->unwind = unwind1;
3792 uw->type = ((type == BRANCH)
3794 : RE_UNWIND_BRANCHJ);
3795 uw->lastparen = lastparen;
3797 uw->locinput = locinput;
3798 uw->nextchr = nextchr;
3799 uw->minmod = st->minmod;
3801 uw->regindent = ++PL_regindent;
3804 REGCP_SET(uw->lastcp);
3806 /* Now go into the first branch */
3816 st->u.curlym.l = st->u.curlym.matches = 0;
3818 /* We suppose that the next guy does not need
3819 backtracking: in particular, it is of constant non-zero length,
3820 and has no parenths to influence future backrefs. */
3821 st->ln = ARG1(scan); /* min to match */
3822 n = ARG2(scan); /* max to match */
3823 st->u.curlym.paren = scan->flags;
3824 if (st->u.curlym.paren) {
3825 if (st->u.curlym.paren > PL_regsize)
3826 PL_regsize = st->u.curlym.paren;
3827 if (st->u.curlym.paren > (I32)*PL_reglastparen)
3828 *PL_reglastparen = st->u.curlym.paren;
3830 scan = NEXTOPER(scan) + NODE_STEP_REGNODE;
3831 if (st->u.curlym.paren)
3832 scan += NEXT_OFF(scan); /* Skip former OPEN. */
3833 PL_reginput = locinput;
3834 st->u.curlym.maxwanted = st->minmod ? st->ln : n;
3835 if (st->u.curlym.maxwanted) {
3836 while (PL_reginput < PL_regeol && st->u.curlym.matches < st->u.curlym.maxwanted) {
3837 REGMATCH(scan, CURLYM1);
3838 /*** all unsaved local vars undefined at this point */
3841 /* on first match, determine length, u.curlym.l */
3842 if (!st->u.curlym.matches++) {
3843 if (PL_reg_match_utf8) {
3845 while (s < PL_reginput) {
3851 st->u.curlym.l = PL_reginput - locinput;
3853 if (st->u.curlym.l == 0) {
3854 st->u.curlym.matches = st->u.curlym.maxwanted;
3858 locinput = PL_reginput;
3862 PL_reginput = locinput;
3866 if (st->ln && st->u.curlym.matches < st->ln)
3868 if (HAS_TEXT(next) || JUMPABLE(next)) {
3869 regnode *text_node = next;
3871 if (! HAS_TEXT(text_node)) FIND_NEXT_IMPT(text_node);
3873 if (! HAS_TEXT(text_node)) st->u.curlym.c1 = st->u.curlym.c2 = -1000;
3875 if (PL_regkind[(U8)OP(text_node)] == REF) {
3876 st->u.curlym.c1 = st->u.curlym.c2 = -1000;
3879 else { st->u.curlym.c1 = (U8)*STRING(text_node); }
3880 if (OP(text_node) == EXACTF || OP(text_node) == REFF)
3881 st->u.curlym.c2 = PL_fold[st->u.curlym.c1];
3882 else if (OP(text_node) == EXACTFL || OP(text_node) == REFFL)
3883 st->u.curlym.c2 = PL_fold_locale[st->u.curlym.c1];
3885 st->u.curlym.c2 = st->u.curlym.c1;
3889 st->u.curlym.c1 = st->u.curlym.c2 = -1000;
3891 REGCP_SET(st->u.curlym.lastcp);
3892 while (n >= st->ln || (n == REG_INFTY && st->ln > 0)) { /* ln overflow ? */
3893 /* If it could work, try it. */
3894 if (st->u.curlym.c1 == -1000 ||
3895 UCHARAT(PL_reginput) == st->u.curlym.c1 ||
3896 UCHARAT(PL_reginput) == st->u.curlym.c2)
3898 if (st->u.curlym.paren) {
3900 PL_regstartp[st->u.curlym.paren] =
3901 HOPc(PL_reginput, -st->u.curlym.l) - PL_bostr;
3902 PL_regendp[st->u.curlym.paren] = PL_reginput - PL_bostr;
3905 PL_regendp[st->u.curlym.paren] = -1;
3907 REGMATCH(next, CURLYM2);
3908 /*** all unsaved local vars undefined at this point */
3911 REGCP_UNWIND(st->u.curlym.lastcp);
3913 /* Couldn't or didn't -- move forward. */
3914 PL_reginput = locinput;
3915 REGMATCH(scan, CURLYM3);
3916 /*** all unsaved local vars undefined at this point */
3919 locinput = PL_reginput;
3927 PerlIO_printf(Perl_debug_log,
3928 "%*s matched %"IVdf" times, len=%"IVdf"...\n",
3929 (int)(REPORT_CODE_OFF+PL_regindent*2), "",
3930 (IV) st->u.curlym.matches, (IV)st->u.curlym.l)
3932 if (st->u.curlym.matches >= st->ln) {
3933 if (HAS_TEXT(next) || JUMPABLE(next)) {
3934 regnode *text_node = next;
3936 if (! HAS_TEXT(text_node)) FIND_NEXT_IMPT(text_node);
3938 if (! HAS_TEXT(text_node)) st->u.curlym.c1 = st->u.curlym.c2 = -1000;
3940 if (PL_regkind[(U8)OP(text_node)] == REF) {
3941 st->u.curlym.c1 = st->u.curlym.c2 = -1000;
3944 else { st->u.curlym.c1 = (U8)*STRING(text_node); }
3946 if (OP(text_node) == EXACTF || OP(text_node) == REFF)
3947 st->u.curlym.c2 = PL_fold[st->u.curlym.c1];
3948 else if (OP(text_node) == EXACTFL || OP(text_node) == REFFL)
3949 st->u.curlym.c2 = PL_fold_locale[st->u.curlym.c1];
3951 st->u.curlym.c2 = st->u.curlym.c1;
3955 st->u.curlym.c1 = st->u.curlym.c2 = -1000;
3958 REGCP_SET(st->u.curlym.lastcp);
3959 while (st->u.curlym.matches >= st->ln) {
3960 /* If it could work, try it. */
3961 if (st->u.curlym.c1 == -1000 ||
3962 UCHARAT(PL_reginput) == st->u.curlym.c1 ||
3963 UCHARAT(PL_reginput) == st->u.curlym.c2)
3966 PerlIO_printf(Perl_debug_log,
3967 "%*s trying tail with matches=%"IVdf"...\n",
3968 (int)(REPORT_CODE_OFF+PL_regindent*2),
3969 "", (IV)st->u.curlym.matches)
3971 if (st->u.curlym.paren) {
3972 if (st->u.curlym.matches) {
3973 PL_regstartp[st->u.curlym.paren]
3974 = HOPc(PL_reginput, -st->u.curlym.l) - PL_bostr;
3975 PL_regendp[st->u.curlym.paren] = PL_reginput - PL_bostr;
3978 PL_regendp[st->u.curlym.paren] = -1;
3980 REGMATCH(next, CURLYM4);
3981 /*** all unsaved local vars undefined at this point */
3984 REGCP_UNWIND(st->u.curlym.lastcp);
3986 /* Couldn't or didn't -- back up. */
3987 st->u.curlym.matches--;
3988 locinput = HOPc(locinput, -st->u.curlym.l);
3989 PL_reginput = locinput;
3997 st->u.plus.paren = scan->flags; /* Which paren to set */
3998 if (st->u.plus.paren > PL_regsize)
3999 PL_regsize = st->u.plus.paren;
4000 if (st->u.plus.paren > (I32)*PL_reglastparen)
4001 *PL_reglastparen = st->u.plus.paren;
4002 st->ln = ARG1(scan); /* min to match */
4003 n = ARG2(scan); /* max to match */
4004 scan = regnext(NEXTOPER(scan) + NODE_STEP_REGNODE);
4007 st->u.plus.paren = 0;
4008 st->ln = ARG1(scan); /* min to match */
4009 n = ARG2(scan); /* max to match */
4010 scan = NEXTOPER(scan) + NODE_STEP_REGNODE;
4015 scan = NEXTOPER(scan);
4016 st->u.plus.paren = 0;
4021 scan = NEXTOPER(scan);
4022 st->u.plus.paren = 0;
4025 * Lookahead to avoid useless match attempts
4026 * when we know what character comes next.
4030 * Used to only do .*x and .*?x, but now it allows
4031 * for )'s, ('s and (?{ ... })'s to be in the way
4032 * of the quantifier and the EXACT-like node. -- japhy
4035 if (HAS_TEXT(next) || JUMPABLE(next)) {
4037 regnode *text_node = next;
4039 if (! HAS_TEXT(text_node)) FIND_NEXT_IMPT(text_node);
4041 if (! HAS_TEXT(text_node)) st->u.plus.c1 = st->u.plus.c2 = -1000;
4043 if (PL_regkind[(U8)OP(text_node)] == REF) {
4044 st->u.plus.c1 = st->u.plus.c2 = -1000;
4045 goto assume_ok_easy;
4047 else { s = (U8*)STRING(text_node); }
4050 st->u.plus.c2 = st->u.plus.c1 = *s;
4051 if (OP(text_node) == EXACTF || OP(text_node) == REFF)
4052 st->u.plus.c2 = PL_fold[st->u.plus.c1];
4053 else if (OP(text_node) == EXACTFL || OP(text_node) == REFFL)
4054 st->u.plus.c2 = PL_fold_locale[st->u.plus.c1];
4057 if (OP(text_node) == EXACTF || OP(text_node) == REFF) {
4058 STRLEN ulen1, ulen2;
4059 U8 tmpbuf1[UTF8_MAXBYTES_CASE+1];
4060 U8 tmpbuf2[UTF8_MAXBYTES_CASE+1];
4062 to_utf8_lower((U8*)s, tmpbuf1, &ulen1);
4063 to_utf8_upper((U8*)s, tmpbuf2, &ulen2);
4065 st->u.plus.c1 = utf8n_to_uvuni(tmpbuf1, UTF8_MAXBYTES, 0,
4067 st->u.plus.c2 = utf8n_to_uvuni(tmpbuf2, UTF8_MAXBYTES, 0,
4071 st->u.plus.c2 = st->u.plus.c1 = utf8n_to_uvchr(s, UTF8_MAXBYTES, 0,
4078 st->u.plus.c1 = st->u.plus.c2 = -1000;
4080 PL_reginput = locinput;
4083 if (st->ln && regrepeat(scan, st->ln) < st->ln)
4085 locinput = PL_reginput;
4086 REGCP_SET(st->u.plus.lastcp);
4087 if (st->u.plus.c1 != -1000) {
4088 st->u.plus.old = locinput;
4089 st->u.plus.count = 0;
4091 if (n == REG_INFTY) {
4092 st->u.plus.e = PL_regeol - 1;
4094 while (UTF8_IS_CONTINUATION(*(U8*)st->u.plus.e))
4099 for (st->u.plus.e = locinput;
4100 m >0 && st->u.plus.e + UTF8SKIP(st->u.plus.e) <= PL_regeol; m--)
4101 st->u.plus.e += UTF8SKIP(st->u.plus.e);
4104 st->u.plus.e = locinput + n - st->ln;
4105 if (st->u.plus.e >= PL_regeol)
4106 st->u.plus.e = PL_regeol - 1;
4109 /* Find place 'next' could work */
4111 if (st->u.plus.c1 == st->u.plus.c2) {
4112 while (locinput <= st->u.plus.e &&
4113 UCHARAT(locinput) != st->u.plus.c1)
4116 while (locinput <= st->u.plus.e
4117 && UCHARAT(locinput) != st->u.plus.c1
4118 && UCHARAT(locinput) != st->u.plus.c2)
4121 st->u.plus.count = locinput - st->u.plus.old;
4124 if (st->u.plus.c1 == st->u.plus.c2) {
4126 /* count initialised to
4127 * utf8_distance(old, locinput) */
4128 while (locinput <= st->u.plus.e &&
4129 utf8n_to_uvchr((U8*)locinput,
4130 UTF8_MAXBYTES, &len,
4131 uniflags) != (UV)st->u.plus.c1) {
4137 /* count initialised to
4138 * utf8_distance(old, locinput) */
4139 while (locinput <= st->u.plus.e) {
4140 UV c = utf8n_to_uvchr((U8*)locinput,
4141 UTF8_MAXBYTES, &len,
4143 if (c == (UV)st->u.plus.c1 || c == (UV)st->u.plus.c2)
4150 if (locinput > st->u.plus.e)
4152 /* PL_reginput == old now */
4153 if (locinput != st->u.plus.old) {
4154 st->ln = 1; /* Did some */
4155 if (regrepeat(scan, st->u.plus.count) < st->u.plus.count)
4158 /* PL_reginput == locinput now */
4159 TRYPAREN(st->u.plus.paren, st->ln, locinput, PLUS1);
4160 /*** all unsaved local vars undefined at this point */
4161 PL_reginput = locinput; /* Could be reset... */
4162 REGCP_UNWIND(st->u.plus.lastcp);
4163 /* Couldn't or didn't -- move forward. */
4164 st->u.plus.old = locinput;
4166 locinput += UTF8SKIP(locinput);
4169 st->u.plus.count = 1;
4173 while (n >= st->ln || (n == REG_INFTY && st->ln > 0)) { /* ln overflow ? */
4175 if (st->u.plus.c1 != -1000) {
4177 c = utf8n_to_uvchr((U8*)PL_reginput,
4181 c = UCHARAT(PL_reginput);
4182 /* If it could work, try it. */
4183 if (c == (UV)st->u.plus.c1 || c == (UV)st->u.plus.c2)
4185 TRYPAREN(st->u.plus.paren, st->ln, PL_reginput, PLUS2);
4186 /*** all unsaved local vars undefined at this point */
4187 REGCP_UNWIND(st->u.plus.lastcp);
4190 /* If it could work, try it. */
4191 else if (st->u.plus.c1 == -1000)
4193 TRYPAREN(st->u.plus.paren, st->ln, PL_reginput, PLUS3);
4194 /*** all unsaved local vars undefined at this point */
4195 REGCP_UNWIND(st->u.plus.lastcp);
4197 /* Couldn't or didn't -- move forward. */
4198 PL_reginput = locinput;
4199 if (regrepeat(scan, 1)) {
4201 locinput = PL_reginput;
4208 n = regrepeat(scan, n);
4209 locinput = PL_reginput;
4210 if (st->ln < n && PL_regkind[(U8)OP(next)] == EOL &&
4211 (OP(next) != MEOL ||
4212 OP(next) == SEOL || OP(next) == EOS))
4214 st->ln = n; /* why back off? */
4215 /* ...because $ and \Z can match before *and* after
4216 newline at the end. Consider "\n\n" =~ /\n+\Z\n/.
4217 We should back off by one in this case. */
4218 if (UCHARAT(PL_reginput - 1) == '\n' && OP(next) != EOS)
4221 REGCP_SET(st->u.plus.lastcp);
4224 while (n >= st->ln) {
4225 if (st->u.plus.c1 != -1000) {
4227 c = utf8n_to_uvchr((U8*)PL_reginput,
4231 c = UCHARAT(PL_reginput);
4233 /* If it could work, try it. */
4234 if (st->u.plus.c1 == -1000 || c == (UV)st->u.plus.c1 || c == (UV)st->u.plus.c2)
4236 TRYPAREN(st->u.plus.paren, n, PL_reginput, PLUS4);
4237 /*** all unsaved local vars undefined at this point */
4238 REGCP_UNWIND(st->u.plus.lastcp);
4240 /* Couldn't or didn't -- back up. */
4242 PL_reginput = locinput = HOPc(locinput, -1);
4250 /* we have successfully completed the execution of a
4251 * postponed re. Pop all states back to the last EVAL
4252 * then continue with the node following the (??{...})
4255 /* this simulates a POP_STATE, except that it pops several
4256 * levels, and doesn't restore locinput */
4259 PL_regmatch_slab = st->u.eval.prev_slab;
4260 cur_eval = st->u.eval.prev_eval;
4261 depth = st->u.eval.depth;
4263 PL_regmatch_state = st;
4268 if (st->u.eval.toggleutf)
4269 PL_reg_flags ^= RF_utf8;
4271 rex = st->u.eval.prev_rex;
4272 PL_reg_re = rex; /* XXX */
4273 /* XXXX This is too dramatic a measure... */
4276 /* Restore parens of the caller without popping the
4279 I32 tmp = PL_savestack_ix;
4280 PL_savestack_ix = st->u.eval.lastcp;
4282 PL_savestack_ix = tmp;
4286 PL_reginput = locinput;
4287 /* resume at node following the (??{...}) */
4292 if (locinput < PL_regtill) {
4293 DEBUG_EXECUTE_r(PerlIO_printf(Perl_debug_log,
4294 "%sMatch possible, but length=%ld is smaller than requested=%ld, failing!%s\n",
4296 (long)(locinput - PL_reg_starttry),
4297 (long)(PL_regtill - PL_reg_starttry),
4299 sayNO_FINAL; /* Cannot match: too short. */
4301 PL_reginput = locinput; /* put where regtry can find it */
4302 sayYES_FINAL; /* Success! */
4304 PL_reginput = locinput; /* put where regtry can find it */
4305 sayYES_LOUD; /* Success! */
4308 PL_reginput = locinput;
4313 char * const s = HOPBACKc(locinput, scan->flags);
4319 PL_reginput = locinput;
4324 char * const s = HOPBACKc(locinput, scan->flags);
4330 PL_reginput = locinput;
4333 REGMATCH(NEXTOPER(NEXTOPER(scan)), IFMATCH);
4334 /*** all unsaved local vars undefined at this point */
4350 if (OP(scan) == SUSPEND) {
4351 locinput = PL_reginput;
4352 nextchr = UCHARAT(locinput);
4357 next = scan + ARG(scan);
4362 PerlIO_printf(Perl_error_log, "%"UVxf" %d\n",
4363 PTR2UV(scan), OP(scan));
4364 Perl_croak(aTHX_ "regexp memory corruption");
4372 /* simulate recursively calling regmatch(), but without actually
4373 * recursing - ie save the current state on the heap rather than on
4374 * the stack, then re-enter the loop. This avoids complex regexes
4375 * blowing the processor stack */
4379 /* push new state */
4380 regmatch_state *oldst = st;
4384 /* grab the next free state slot */
4386 if (st > &(PL_regmatch_slab->states[PERL_REGMATCH_SLAB_SLOTS-1]))
4387 st = S_push_slab(aTHX);
4388 PL_regmatch_state = st;
4392 oldst->locinput = locinput;
4395 locinput = PL_reginput;
4396 nextchr = UCHARAT(locinput);
4410 * We get here only if there's trouble -- normally "case END" is
4411 * the terminating point.
4413 Perl_croak(aTHX_ "corrupted regexp pointers");
4419 PerlIO_printf(Perl_debug_log,
4420 "%*s %scould match...%s\n",
4421 REPORT_CODE_OFF+PL_regindent*2, "", PL_colors[4], PL_colors[5])
4425 DEBUG_EXECUTE_r(PerlIO_printf(Perl_debug_log, "%sMatch successful!%s\n",
4426 PL_colors[4], PL_colors[5]));
4433 /* XXX this is duplicate(ish) code to that in the do_no section.
4434 * eventually a yes should just pop the whole stack */
4436 /* restore previous state and re-enter */
4439 switch (st->resume_state) {
4441 goto resume_point_TRIE1;
4443 goto resume_point_TRIE2;
4447 goto resume_point_CURLYX;
4448 case resume_WHILEM1:
4449 goto resume_point_WHILEM1;
4450 case resume_WHILEM2:
4451 goto resume_point_WHILEM2;
4452 case resume_WHILEM3:
4453 goto resume_point_WHILEM3;
4454 case resume_WHILEM4:
4455 goto resume_point_WHILEM4;
4456 case resume_WHILEM5:
4457 goto resume_point_WHILEM5;
4458 case resume_WHILEM6:
4459 goto resume_point_WHILEM6;
4460 case resume_CURLYM1:
4461 goto resume_point_CURLYM1;
4462 case resume_CURLYM2:
4463 goto resume_point_CURLYM2;
4464 case resume_CURLYM3:
4465 goto resume_point_CURLYM3;
4466 case resume_CURLYM4:
4467 goto resume_point_CURLYM4;
4468 case resume_IFMATCH:
4469 goto resume_point_IFMATCH;
4471 goto resume_point_PLUS1;
4473 goto resume_point_PLUS2;
4475 goto resume_point_PLUS3;
4477 goto resume_point_PLUS4;
4479 Perl_croak(aTHX_ "regexp resume memory corruption");
4486 PerlIO_printf(Perl_debug_log,
4487 "%*s %sfailed...%s\n",
4488 REPORT_CODE_OFF+PL_regindent*2, "", PL_colors[4], PL_colors[5])
4494 re_unwind_t * const uw = SSPTRt(st->unwind,re_unwind_t);
4497 case RE_UNWIND_BRANCH:
4498 case RE_UNWIND_BRANCHJ:
4500 re_unwind_branch_t * const uwb = &(uw->branch);
4501 const I32 lastparen = uwb->lastparen;
4503 REGCP_UNWIND(uwb->lastcp);
4504 for (n = *PL_reglastparen; n > lastparen; n--)
4506 *PL_reglastparen = n;
4507 scan = next = uwb->next;
4508 st->minmod = uwb->minmod;
4510 OP(scan) != (uwb->type == RE_UNWIND_BRANCH
4511 ? BRANCH : BRANCHJ) ) { /* Failure */
4512 st->unwind = uwb->prev;
4518 /* Have more choice yet. Reuse the same uwb. */
4519 if ((n = (uwb->type == RE_UNWIND_BRANCH
4520 ? NEXT_OFF(next) : ARG(next))))
4523 next = NULL; /* XXXX Needn't unwinding in this case... */
4525 next = NEXTOPER(scan);
4526 if (uwb->type == RE_UNWIND_BRANCHJ)
4527 next = NEXTOPER(next);
4528 locinput = uwb->locinput;
4529 nextchr = uwb->nextchr;
4531 PL_regindent = uwb->regindent;
4538 Perl_croak(aTHX_ "regexp unwind memory corruption");
4549 /* there's a previous state to backtrack to */
4551 switch (st->resume_state) {
4553 goto resume_point_TRIE1;
4555 goto resume_point_TRIE2;
4557 /* we have failed an (??{...}). Restore state to the outer re
4558 * then re-throw the failure */
4559 if (st->u.eval.toggleutf)
4560 PL_reg_flags ^= RF_utf8;
4562 rex = st->u.eval.prev_rex;
4563 PL_reg_re = rex; /* XXX */
4564 cur_eval = st->u.eval.prev_eval;
4566 /* XXXX This is too dramatic a measure... */
4569 PL_reginput = locinput;
4570 REGCP_UNWIND(st->u.eval.lastcp);
4575 goto resume_point_CURLYX;
4576 case resume_WHILEM1:
4577 goto resume_point_WHILEM1;
4578 case resume_WHILEM2:
4579 goto resume_point_WHILEM2;
4580 case resume_WHILEM3:
4581 goto resume_point_WHILEM3;
4582 case resume_WHILEM4:
4583 goto resume_point_WHILEM4;
4584 case resume_WHILEM5:
4585 goto resume_point_WHILEM5;
4586 case resume_WHILEM6:
4587 goto resume_point_WHILEM6;
4588 case resume_CURLYM1:
4589 goto resume_point_CURLYM1;
4590 case resume_CURLYM2:
4591 goto resume_point_CURLYM2;
4592 case resume_CURLYM3:
4593 goto resume_point_CURLYM3;
4594 case resume_CURLYM4:
4595 goto resume_point_CURLYM4;
4596 case resume_IFMATCH:
4597 goto resume_point_IFMATCH;
4599 goto resume_point_PLUS1;
4601 goto resume_point_PLUS2;
4603 goto resume_point_PLUS3;
4605 goto resume_point_PLUS4;
4607 Perl_croak(aTHX_ "regexp resume memory corruption");
4613 /* restore original high-water mark */
4614 PL_regmatch_slab = orig_slab;
4615 PL_regmatch_state = orig_state;
4617 /* free all slabs above current one */
4618 if (orig_slab->next) {
4619 regmatch_slab *osl, *sl = orig_slab->next;
4620 orig_slab->next = NULL;
4633 - regrepeat - repeatedly match something simple, report how many
4636 * [This routine now assumes that it will only match on things of length 1.
4637 * That was true before, but now we assume scan - reginput is the count,
4638 * rather than incrementing count on every character. [Er, except utf8.]]
4641 S_regrepeat(pTHX_ const regnode *p, I32 max)
4644 register char *scan;
4646 register char *loceol = PL_regeol;
4647 register I32 hardcount = 0;
4648 register bool do_utf8 = PL_reg_match_utf8;
4651 if (max == REG_INFTY)
4653 else if (max < loceol - scan)
4654 loceol = scan + max;
4659 while (scan < loceol && hardcount < max && *scan != '\n') {
4660 scan += UTF8SKIP(scan);
4664 while (scan < loceol && *scan != '\n')
4671 while (scan < loceol && hardcount < max) {
4672 scan += UTF8SKIP(scan);
4682 case EXACT: /* length of string is 1 */
4684 while (scan < loceol && UCHARAT(scan) == c)
4687 case EXACTF: /* length of string is 1 */
4689 while (scan < loceol &&
4690 (UCHARAT(scan) == c || UCHARAT(scan) == PL_fold[c]))
4693 case EXACTFL: /* length of string is 1 */
4694 PL_reg_flags |= RF_tainted;
4696 while (scan < loceol &&
4697 (UCHARAT(scan) == c || UCHARAT(scan) == PL_fold_locale[c]))
4703 while (hardcount < max && scan < loceol &&
4704 reginclass(p, (U8*)scan, 0, do_utf8)) {
4705 scan += UTF8SKIP(scan);
4709 while (scan < loceol && REGINCLASS(p, (U8*)scan))
4716 LOAD_UTF8_CHARCLASS_ALNUM();
4717 while (hardcount < max && scan < loceol &&
4718 swash_fetch(PL_utf8_alnum, (U8*)scan, do_utf8)) {
4719 scan += UTF8SKIP(scan);
4723 while (scan < loceol && isALNUM(*scan))
4728 PL_reg_flags |= RF_tainted;
4731 while (hardcount < max && scan < loceol &&
4732 isALNUM_LC_utf8((U8*)scan)) {
4733 scan += UTF8SKIP(scan);
4737 while (scan < loceol && isALNUM_LC(*scan))
4744 LOAD_UTF8_CHARCLASS_ALNUM();
4745 while (hardcount < max && scan < loceol &&
4746 !swash_fetch(PL_utf8_alnum, (U8*)scan, do_utf8)) {
4747 scan += UTF8SKIP(scan);
4751 while (scan < loceol && !isALNUM(*scan))
4756 PL_reg_flags |= RF_tainted;
4759 while (hardcount < max && scan < loceol &&
4760 !isALNUM_LC_utf8((U8*)scan)) {
4761 scan += UTF8SKIP(scan);
4765 while (scan < loceol && !isALNUM_LC(*scan))
4772 LOAD_UTF8_CHARCLASS_SPACE();
4773 while (hardcount < max && scan < loceol &&
4775 swash_fetch(PL_utf8_space,(U8*)scan, do_utf8))) {
4776 scan += UTF8SKIP(scan);
4780 while (scan < loceol && isSPACE(*scan))
4785 PL_reg_flags |= RF_tainted;
4788 while (hardcount < max && scan < loceol &&
4789 (*scan == ' ' || isSPACE_LC_utf8((U8*)scan))) {
4790 scan += UTF8SKIP(scan);
4794 while (scan < loceol && isSPACE_LC(*scan))
4801 LOAD_UTF8_CHARCLASS_SPACE();
4802 while (hardcount < max && scan < loceol &&
4804 swash_fetch(PL_utf8_space,(U8*)scan, do_utf8))) {
4805 scan += UTF8SKIP(scan);
4809 while (scan < loceol && !isSPACE(*scan))
4814 PL_reg_flags |= RF_tainted;
4817 while (hardcount < max && scan < loceol &&
4818 !(*scan == ' ' || isSPACE_LC_utf8((U8*)scan))) {
4819 scan += UTF8SKIP(scan);
4823 while (scan < loceol && !isSPACE_LC(*scan))
4830 LOAD_UTF8_CHARCLASS_DIGIT();
4831 while (hardcount < max && scan < loceol &&
4832 swash_fetch(PL_utf8_digit, (U8*)scan, do_utf8)) {
4833 scan += UTF8SKIP(scan);
4837 while (scan < loceol && isDIGIT(*scan))
4844 LOAD_UTF8_CHARCLASS_DIGIT();
4845 while (hardcount < max && scan < loceol &&
4846 !swash_fetch(PL_utf8_digit, (U8*)scan, do_utf8)) {
4847 scan += UTF8SKIP(scan);
4851 while (scan < loceol && !isDIGIT(*scan))
4855 default: /* Called on something of 0 width. */
4856 break; /* So match right here or not at all. */
4862 c = scan - PL_reginput;
4866 SV *re_debug_flags = NULL;
4867 SV * const prop = sv_newmortal();
4871 PerlIO_printf(Perl_debug_log,
4872 "%*s %s can match %"IVdf" times out of %"IVdf"...\n",
4873 REPORT_CODE_OFF+1, "", SvPVX_const(prop),(IV)c,(IV)max);
4882 - regclass_swash - prepare the utf8 swash
4886 Perl_regclass_swash(pTHX_ register const regnode* node, bool doinit, SV** listsvp, SV **altsvp)
4892 const struct reg_data *data = PL_reg_re ? PL_reg_re->data : NULL;
4894 if (data && data->count) {
4895 const U32 n = ARG(node);
4897 if (data->what[n] == 's') {
4898 SV * const rv = (SV*)data->data[n];
4899 AV * const av = (AV*)SvRV((SV*)rv);
4900 SV **const ary = AvARRAY(av);
4903 /* See the end of regcomp.c:S_regclass() for
4904 * documentation of these array elements. */
4907 a = SvROK(ary[1]) ? &ary[1] : 0;
4908 b = SvTYPE(ary[2]) == SVt_PVAV ? &ary[2] : 0;
4912 else if (si && doinit) {
4913 sw = swash_init("utf8", "", si, 1, 0);
4914 (void)av_store(av, 1, sw);
4930 - reginclass - determine if a character falls into a character class
4932 The n is the ANYOF regnode, the p is the target string, lenp
4933 is pointer to the maximum length of how far to go in the p
4934 (if the lenp is zero, UTF8SKIP(p) is used),
4935 do_utf8 tells whether the target string is in UTF-8.
4940 S_reginclass(pTHX_ register const regnode *n, register const U8* p, STRLEN* lenp, register bool do_utf8)
4943 const char flags = ANYOF_FLAGS(n);
4949 if (do_utf8 && !UTF8_IS_INVARIANT(c)) {
4950 c = utf8n_to_uvchr(p, UTF8_MAXBYTES, &len,
4951 (UTF8_ALLOW_DEFAULT & UTF8_ALLOW_ANYUV) | UTF8_CHECK_ONLY);
4952 /* see [perl #37836] for UTF8_ALLOW_ANYUV */
4953 if (len == (STRLEN)-1)
4954 Perl_croak(aTHX_ "Malformed UTF-8 character (fatal)");
4957 plen = lenp ? *lenp : UNISKIP(NATIVE_TO_UNI(c));
4958 if (do_utf8 || (flags & ANYOF_UNICODE)) {
4961 if (do_utf8 && !ANYOF_RUNTIME(n)) {
4962 if (len != (STRLEN)-1 && c < 256 && ANYOF_BITMAP_TEST(n, c))
4965 if (!match && do_utf8 && (flags & ANYOF_UNICODE_ALL) && c >= 256)
4969 SV * const sw = regclass_swash(n, TRUE, 0, (SV**)&av);
4972 if (swash_fetch(sw, p, do_utf8))
4974 else if (flags & ANYOF_FOLD) {
4975 if (!match && lenp && av) {
4977 for (i = 0; i <= av_len(av); i++) {
4978 SV* const sv = *av_fetch(av, i, FALSE);
4980 const char * const s = SvPV_const(sv, len);
4982 if (len <= plen && memEQ(s, (char*)p, len)) {
4990 U8 tmpbuf[UTF8_MAXBYTES_CASE+1];
4993 to_utf8_fold(p, tmpbuf, &tmplen);
4994 if (swash_fetch(sw, tmpbuf, do_utf8))
5000 if (match && lenp && *lenp == 0)
5001 *lenp = UNISKIP(NATIVE_TO_UNI(c));
5003 if (!match && c < 256) {
5004 if (ANYOF_BITMAP_TEST(n, c))
5006 else if (flags & ANYOF_FOLD) {
5009 if (flags & ANYOF_LOCALE) {
5010 PL_reg_flags |= RF_tainted;
5011 f = PL_fold_locale[c];
5015 if (f != c && ANYOF_BITMAP_TEST(n, f))
5019 if (!match && (flags & ANYOF_CLASS)) {
5020 PL_reg_flags |= RF_tainted;
5022 (ANYOF_CLASS_TEST(n, ANYOF_ALNUM) && isALNUM_LC(c)) ||
5023 (ANYOF_CLASS_TEST(n, ANYOF_NALNUM) && !isALNUM_LC(c)) ||
5024 (ANYOF_CLASS_TEST(n, ANYOF_SPACE) && isSPACE_LC(c)) ||
5025 (ANYOF_CLASS_TEST(n, ANYOF_NSPACE) && !isSPACE_LC(c)) ||
5026 (ANYOF_CLASS_TEST(n, ANYOF_DIGIT) && isDIGIT_LC(c)) ||
5027 (ANYOF_CLASS_TEST(n, ANYOF_NDIGIT) && !isDIGIT_LC(c)) ||
5028 (ANYOF_CLASS_TEST(n, ANYOF_ALNUMC) && isALNUMC_LC(c)) ||
5029 (ANYOF_CLASS_TEST(n, ANYOF_NALNUMC) && !isALNUMC_LC(c)) ||
5030 (ANYOF_CLASS_TEST(n, ANYOF_ALPHA) && isALPHA_LC(c)) ||
5031 (ANYOF_CLASS_TEST(n, ANYOF_NALPHA) && !isALPHA_LC(c)) ||
5032 (ANYOF_CLASS_TEST(n, ANYOF_ASCII) && isASCII(c)) ||
5033 (ANYOF_CLASS_TEST(n, ANYOF_NASCII) && !isASCII(c)) ||
5034 (ANYOF_CLASS_TEST(n, ANYOF_CNTRL) && isCNTRL_LC(c)) ||
5035 (ANYOF_CLASS_TEST(n, ANYOF_NCNTRL) && !isCNTRL_LC(c)) ||
5036 (ANYOF_CLASS_TEST(n, ANYOF_GRAPH) && isGRAPH_LC(c)) ||
5037 (ANYOF_CLASS_TEST(n, ANYOF_NGRAPH) && !isGRAPH_LC(c)) ||
5038 (ANYOF_CLASS_TEST(n, ANYOF_LOWER) && isLOWER_LC(c)) ||
5039 (ANYOF_CLASS_TEST(n, ANYOF_NLOWER) && !isLOWER_LC(c)) ||
5040 (ANYOF_CLASS_TEST(n, ANYOF_PRINT) && isPRINT_LC(c)) ||
5041 (ANYOF_CLASS_TEST(n, ANYOF_NPRINT) && !isPRINT_LC(c)) ||
5042 (ANYOF_CLASS_TEST(n, ANYOF_PUNCT) && isPUNCT_LC(c)) ||
5043 (ANYOF_CLASS_TEST(n, ANYOF_NPUNCT) && !isPUNCT_LC(c)) ||
5044 (ANYOF_CLASS_TEST(n, ANYOF_UPPER) && isUPPER_LC(c)) ||
5045 (ANYOF_CLASS_TEST(n, ANYOF_NUPPER) && !isUPPER_LC(c)) ||
5046 (ANYOF_CLASS_TEST(n, ANYOF_XDIGIT) && isXDIGIT(c)) ||
5047 (ANYOF_CLASS_TEST(n, ANYOF_NXDIGIT) && !isXDIGIT(c)) ||
5048 (ANYOF_CLASS_TEST(n, ANYOF_PSXSPC) && isPSXSPC(c)) ||
5049 (ANYOF_CLASS_TEST(n, ANYOF_NPSXSPC) && !isPSXSPC(c)) ||
5050 (ANYOF_CLASS_TEST(n, ANYOF_BLANK) && isBLANK(c)) ||
5051 (ANYOF_CLASS_TEST(n, ANYOF_NBLANK) && !isBLANK(c))
5052 ) /* How's that for a conditional? */
5059 return (flags & ANYOF_INVERT) ? !match : match;
5063 S_reghop3(U8 *s, I32 off, U8* lim)
5067 while (off-- && s < lim) {
5068 /* XXX could check well-formedness here */
5076 if (UTF8_IS_CONTINUED(*s)) {
5077 while (s > (U8*)lim && UTF8_IS_CONTINUATION(*s))
5080 /* XXX could check well-formedness here */
5088 S_reghopmaybe3(U8* s, I32 off, U8* lim)
5092 while (off-- && s < lim) {
5093 /* XXX could check well-formedness here */
5103 if (UTF8_IS_CONTINUED(*s)) {
5104 while (s > (U8*)lim && UTF8_IS_CONTINUATION(*s))
5107 /* XXX could check well-formedness here */
5119 restore_pos(pTHX_ void *arg)
5122 regexp * const rex = (regexp *)arg;
5123 if (PL_reg_eval_set) {
5124 if (PL_reg_oldsaved) {
5125 rex->subbeg = PL_reg_oldsaved;
5126 rex->sublen = PL_reg_oldsavedlen;
5127 #ifdef PERL_OLD_COPY_ON_WRITE
5128 rex->saved_copy = PL_nrs;
5130 RX_MATCH_COPIED_on(rex);
5132 PL_reg_magic->mg_len = PL_reg_oldpos;
5133 PL_reg_eval_set = 0;
5134 PL_curpm = PL_reg_oldcurpm;
5139 S_to_utf8_substr(pTHX_ register regexp *prog)
5141 if (prog->float_substr && !prog->float_utf8) {
5142 SV* const sv = newSVsv(prog->float_substr);
5143 prog->float_utf8 = sv;
5144 sv_utf8_upgrade(sv);
5145 if (SvTAIL(prog->float_substr))
5147 if (prog->float_substr == prog->check_substr)
5148 prog->check_utf8 = sv;
5150 if (prog->anchored_substr && !prog->anchored_utf8) {
5151 SV* const sv = newSVsv(prog->anchored_substr);
5152 prog->anchored_utf8 = sv;
5153 sv_utf8_upgrade(sv);
5154 if (SvTAIL(prog->anchored_substr))
5156 if (prog->anchored_substr == prog->check_substr)
5157 prog->check_utf8 = sv;
5162 S_to_byte_substr(pTHX_ register regexp *prog)
5165 if (prog->float_utf8 && !prog->float_substr) {
5166 SV* sv = newSVsv(prog->float_utf8);
5167 prog->float_substr = sv;
5168 if (sv_utf8_downgrade(sv, TRUE)) {
5169 if (SvTAIL(prog->float_utf8))
5173 prog->float_substr = sv = &PL_sv_undef;
5175 if (prog->float_utf8 == prog->check_utf8)
5176 prog->check_substr = sv;
5178 if (prog->anchored_utf8 && !prog->anchored_substr) {
5179 SV* sv = newSVsv(prog->anchored_utf8);
5180 prog->anchored_substr = sv;
5181 if (sv_utf8_downgrade(sv, TRUE)) {
5182 if (SvTAIL(prog->anchored_utf8))
5186 prog->anchored_substr = sv = &PL_sv_undef;
5188 if (prog->anchored_utf8 == prog->check_utf8)
5189 prog->check_substr = sv;
5195 * c-indentation-style: bsd
5197 * indent-tabs-mode: t
5200 * ex: set ts=8 sts=4 sw=4 noet: