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.
18 /* NOTE: this is derived from Henry Spencer's regexp code, and should not
19 * confused with the original package (see point 3 below). Thanks, Henry!
22 /* Additional note: this code is very heavily munged from Henry's version
23 * in places. In some spots I've traded clarity for efficiency, so don't
24 * blame Henry for some of the lack of readability.
27 /* The names of the functions have been changed from regcomp and
28 * regexec to pregcomp and pregexec in order to avoid conflicts
29 * with the POSIX routines of the same names.
32 #ifdef PERL_EXT_RE_BUILD
37 * pregcomp and pregexec -- regsub and regerror are not used in perl
39 * Copyright (c) 1986 by University of Toronto.
40 * Written by Henry Spencer. Not derived from licensed software.
42 * Permission is granted to anyone to use this software for any
43 * purpose on any computer system, and to redistribute it freely,
44 * subject to the following restrictions:
46 * 1. The author is not responsible for the consequences of use of
47 * this software, no matter how awful, even if they arise
50 * 2. The origin of this software must not be misrepresented, either
51 * by explicit claim or by omission.
53 * 3. Altered versions must be plainly marked as such, and must not
54 * be misrepresented as being the original software.
56 **** Alterations to Henry's code are...
58 **** Copyright (C) 1991, 1992, 1993, 1994, 1995, 1996, 1997, 1998, 1999,
59 **** 2000, 2001, 2002, 2003, 2004, 2005, 2006, by Larry Wall and others
61 **** You may distribute under the terms of either the GNU General Public
62 **** License or the Artistic License, as specified in the README file.
64 * Beware that some of this code is subtly aware of the way operator
65 * precedence is structured in regular expressions. Serious changes in
66 * regular-expression syntax might require a total rethink.
69 #define PERL_IN_REGEXEC_C
72 #ifdef PERL_IN_XSUB_RE
78 #define RF_tainted 1 /* tainted information used? */
79 #define RF_warned 2 /* warned about big count? */
81 #define RF_utf8 8 /* Pattern contains multibyte chars? */
83 #define UTF ((PL_reg_flags & RF_utf8) != 0)
85 #define RS_init 1 /* eval environment created */
86 #define RS_set 2 /* replsv value is set */
92 #define REGINCLASS(prog,p,c) (ANYOF_FLAGS(p) ? reginclass(prog,p,c,0,0) : ANYOF_BITMAP_TEST(p,*(c)))
98 #define CHR_SVLEN(sv) (do_utf8 ? sv_len_utf8(sv) : SvCUR(sv))
99 #define CHR_DIST(a,b) (PL_reg_match_utf8 ? utf8_distance(a,b) : a - b)
101 #define HOPc(pos,off) \
102 (char *)(PL_reg_match_utf8 \
103 ? reghop3((U8*)pos, off, (U8*)(off >= 0 ? PL_regeol : PL_bostr)) \
105 #define HOPBACKc(pos, off) \
106 (char*)(PL_reg_match_utf8\
107 ? reghopmaybe3((U8*)pos, -off, (U8*)PL_bostr) \
108 : (pos - off >= PL_bostr) \
112 #define HOP3(pos,off,lim) (PL_reg_match_utf8 ? reghop3((U8*)(pos), off, (U8*)(lim)) : (U8*)(pos + off))
113 #define HOP3c(pos,off,lim) ((char*)HOP3(pos,off,lim))
115 #define LOAD_UTF8_CHARCLASS(class,str) STMT_START { \
116 if (!CAT2(PL_utf8_,class)) { bool ok; ENTER; save_re_context(); ok=CAT2(is_utf8_,class)((const U8*)str); assert(ok); LEAVE; } } STMT_END
117 #define LOAD_UTF8_CHARCLASS_ALNUM() LOAD_UTF8_CHARCLASS(alnum,"a")
118 #define LOAD_UTF8_CHARCLASS_DIGIT() LOAD_UTF8_CHARCLASS(digit,"0")
119 #define LOAD_UTF8_CHARCLASS_SPACE() LOAD_UTF8_CHARCLASS(space," ")
120 #define LOAD_UTF8_CHARCLASS_MARK() LOAD_UTF8_CHARCLASS(mark, "\xcd\x86")
122 /* TODO: Combine JUMPABLE and HAS_TEXT to cache OP(rn) */
124 /* for use after a quantifier and before an EXACT-like node -- japhy */
125 #define JUMPABLE(rn) ( \
126 OP(rn) == OPEN || OP(rn) == CLOSE || OP(rn) == EVAL || \
127 OP(rn) == SUSPEND || OP(rn) == IFMATCH || \
128 OP(rn) == PLUS || OP(rn) == MINMOD || \
129 (PL_regkind[OP(rn)] == CURLY && ARG1(rn) > 0) \
132 #define HAS_TEXT(rn) ( \
133 PL_regkind[OP(rn)] == EXACT || PL_regkind[OP(rn)] == REF \
137 Search for mandatory following text node; for lookahead, the text must
138 follow but for lookbehind (rn->flags != 0) we skip to the next step.
140 #define FIND_NEXT_IMPT(rn) STMT_START { \
141 while (JUMPABLE(rn)) { \
142 const OPCODE type = OP(rn); \
143 if (type == SUSPEND || PL_regkind[type] == CURLY) \
144 rn = NEXTOPER(NEXTOPER(rn)); \
145 else if (type == PLUS) \
147 else if (type == IFMATCH) \
148 rn = (rn->flags == 0) ? NEXTOPER(NEXTOPER(rn)) : rn + ARG(rn); \
149 else rn += NEXT_OFF(rn); \
154 static void restore_pos(pTHX_ void *arg);
157 S_regcppush(pTHX_ I32 parenfloor)
160 const int retval = PL_savestack_ix;
161 #define REGCP_PAREN_ELEMS 4
162 const int paren_elems_to_push = (PL_regsize - parenfloor) * REGCP_PAREN_ELEMS;
164 GET_RE_DEBUG_FLAGS_DECL;
166 if (paren_elems_to_push < 0)
167 Perl_croak(aTHX_ "panic: paren_elems_to_push < 0");
169 #define REGCP_OTHER_ELEMS 8
170 SSGROW(paren_elems_to_push + REGCP_OTHER_ELEMS);
172 for (p = PL_regsize; p > parenfloor; p--) {
173 /* REGCP_PARENS_ELEMS are pushed per pairs of parentheses. */
174 SSPUSHINT(PL_regendp[p]);
175 SSPUSHINT(PL_regstartp[p]);
176 SSPUSHPTR(PL_reg_start_tmp[p]);
178 DEBUG_EXECUTE_r(PerlIO_printf(Perl_debug_log,
179 " saving \\%"UVuf" %"IVdf"(%"IVdf")..%"IVdf"\n",
180 (UV)p, (IV)PL_regstartp[p],
181 (IV)(PL_reg_start_tmp[p] - PL_bostr),
185 /* REGCP_OTHER_ELEMS are pushed in any case, parentheses or no. */
186 SSPUSHPTR(PL_regstartp);
187 SSPUSHPTR(PL_regendp);
188 SSPUSHINT(PL_regsize);
189 SSPUSHINT(*PL_reglastparen);
190 SSPUSHINT(*PL_reglastcloseparen);
191 SSPUSHPTR(PL_reginput);
192 #define REGCP_FRAME_ELEMS 2
193 /* REGCP_FRAME_ELEMS are part of the REGCP_OTHER_ELEMS and
194 * are needed for the regexp context stack bookkeeping. */
195 SSPUSHINT(paren_elems_to_push + REGCP_OTHER_ELEMS - REGCP_FRAME_ELEMS);
196 SSPUSHINT(SAVEt_REGCONTEXT); /* Magic cookie. */
201 /* These are needed since we do not localize EVAL nodes: */
202 #define REGCP_SET(cp) \
204 PerlIO_printf(Perl_debug_log, \
205 " Setting an EVAL scope, savestack=%"IVdf"\n", \
206 (IV)PL_savestack_ix)); \
209 #define REGCP_UNWIND(cp) \
211 if (cp != PL_savestack_ix) \
212 PerlIO_printf(Perl_debug_log, \
213 " Clearing an EVAL scope, savestack=%"IVdf"..%"IVdf"\n", \
214 (IV)(cp), (IV)PL_savestack_ix)); \
218 S_regcppop(pTHX_ const regexp *rex)
224 GET_RE_DEBUG_FLAGS_DECL;
226 /* Pop REGCP_OTHER_ELEMS before the parentheses loop starts. */
228 assert(i == SAVEt_REGCONTEXT); /* Check that the magic cookie is there. */
229 i = SSPOPINT; /* Parentheses elements to pop. */
230 input = (char *) SSPOPPTR;
231 *PL_reglastcloseparen = SSPOPINT;
232 *PL_reglastparen = SSPOPINT;
233 PL_regsize = SSPOPINT;
234 PL_regendp=(I32 *) SSPOPPTR;
235 PL_regstartp=(I32 *) SSPOPPTR;
238 /* Now restore the parentheses context. */
239 for (i -= (REGCP_OTHER_ELEMS - REGCP_FRAME_ELEMS);
240 i > 0; i -= REGCP_PAREN_ELEMS) {
242 U32 paren = (U32)SSPOPINT;
243 PL_reg_start_tmp[paren] = (char *) SSPOPPTR;
244 PL_regstartp[paren] = SSPOPINT;
246 if (paren <= *PL_reglastparen)
247 PL_regendp[paren] = tmps;
249 PerlIO_printf(Perl_debug_log,
250 " restoring \\%"UVuf" to %"IVdf"(%"IVdf")..%"IVdf"%s\n",
251 (UV)paren, (IV)PL_regstartp[paren],
252 (IV)(PL_reg_start_tmp[paren] - PL_bostr),
253 (IV)PL_regendp[paren],
254 (paren > *PL_reglastparen ? "(no)" : ""));
258 if (*PL_reglastparen + 1 <= rex->nparens) {
259 PerlIO_printf(Perl_debug_log,
260 " restoring \\%"IVdf"..\\%"IVdf" to undef\n",
261 (IV)(*PL_reglastparen + 1), (IV)rex->nparens);
265 /* It would seem that the similar code in regtry()
266 * already takes care of this, and in fact it is in
267 * a better location to since this code can #if 0-ed out
268 * but the code in regtry() is needed or otherwise tests
269 * requiring null fields (pat.t#187 and split.t#{13,14}
270 * (as of patchlevel 7877) will fail. Then again,
271 * this code seems to be necessary or otherwise
272 * building DynaLoader will fail:
273 * "Error: '*' not in typemap in DynaLoader.xs, line 164"
275 for (i = *PL_reglastparen + 1; i <= rex->nparens; i++) {
277 PL_regstartp[i] = -1;
284 #define regcpblow(cp) LEAVE_SCOPE(cp) /* Ignores regcppush()ed data. */
287 * pregexec and friends
290 #ifndef PERL_IN_XSUB_RE
292 - pregexec - match a regexp against a string
295 Perl_pregexec(pTHX_ register regexp *prog, char *stringarg, register char *strend,
296 char *strbeg, I32 minend, SV *screamer, U32 nosave)
297 /* strend: pointer to null at end of string */
298 /* strbeg: real beginning of string */
299 /* minend: end of match must be >=minend after stringarg. */
300 /* nosave: For optimizations. */
303 regexec_flags(prog, stringarg, strend, strbeg, minend, screamer, NULL,
304 nosave ? 0 : REXEC_COPY_STR);
309 * Need to implement the following flags for reg_anch:
311 * USE_INTUIT_NOML - Useful to call re_intuit_start() first
313 * INTUIT_AUTORITATIVE_NOML - Can trust a positive answer
314 * INTUIT_AUTORITATIVE_ML
315 * INTUIT_ONCE_NOML - Intuit can match in one location only.
318 * Another flag for this function: SECOND_TIME (so that float substrs
319 * with giant delta may be not rechecked).
322 /* Assumptions: if ANCH_GPOS, then strpos is anchored. XXXX Check GPOS logic */
324 /* If SCREAM, then SvPVX_const(sv) should be compatible with strpos and strend.
325 Otherwise, only SvCUR(sv) is used to get strbeg. */
327 /* XXXX We assume that strpos is strbeg unless sv. */
329 /* XXXX Some places assume that there is a fixed substring.
330 An update may be needed if optimizer marks as "INTUITable"
331 RExen without fixed substrings. Similarly, it is assumed that
332 lengths of all the strings are no more than minlen, thus they
333 cannot come from lookahead.
334 (Or minlen should take into account lookahead.)
335 NOTE: Some of this comment is not correct. minlen does now take account
336 of lookahead/behind. Further research is required. -- demerphq
340 /* A failure to find a constant substring means that there is no need to make
341 an expensive call to REx engine, thus we celebrate a failure. Similarly,
342 finding a substring too deep into the string means that less calls to
343 regtry() should be needed.
345 REx compiler's optimizer found 4 possible hints:
346 a) Anchored substring;
348 c) Whether we are anchored (beginning-of-line or \G);
349 d) First node (of those at offset 0) which may distingush positions;
350 We use a)b)d) and multiline-part of c), and try to find a position in the
351 string which does not contradict any of them.
354 /* Most of decisions we do here should have been done at compile time.
355 The nodes of the REx which we used for the search should have been
356 deleted from the finite automaton. */
359 Perl_re_intuit_start(pTHX_ regexp *prog, SV *sv, char *strpos,
360 char *strend, U32 flags, re_scream_pos_data *data)
363 register I32 start_shift = 0;
364 /* Should be nonnegative! */
365 register I32 end_shift = 0;
370 const bool do_utf8 = (sv && SvUTF8(sv)) ? 1 : 0; /* if no sv we have to assume bytes */
372 register char *other_last = NULL; /* other substr checked before this */
373 char *check_at = NULL; /* check substr found at this pos */
374 const I32 multiline = prog->extflags & RXf_PMf_MULTILINE;
375 RXi_GET_DECL(prog,progi);
377 const char * const i_strpos = strpos;
380 GET_RE_DEBUG_FLAGS_DECL;
382 RX_MATCH_UTF8_set(prog,do_utf8);
384 if (prog->extflags & RXf_UTF8) {
385 PL_reg_flags |= RF_utf8;
388 debug_start_match(prog, do_utf8, strpos, strend,
389 sv ? "Guessing start of match in sv for"
390 : "Guessing start of match in string for");
393 /* CHR_DIST() would be more correct here but it makes things slow. */
394 if (prog->minlen > strend - strpos) {
395 DEBUG_EXECUTE_r(PerlIO_printf(Perl_debug_log,
396 "String too short... [re_intuit_start]\n"));
400 strbeg = (sv && SvPOK(sv)) ? strend - SvCUR(sv) : strpos;
403 if (!prog->check_utf8 && prog->check_substr)
404 to_utf8_substr(prog);
405 check = prog->check_utf8;
407 if (!prog->check_substr && prog->check_utf8)
408 to_byte_substr(prog);
409 check = prog->check_substr;
411 if (check == &PL_sv_undef) {
412 DEBUG_EXECUTE_r(PerlIO_printf(Perl_debug_log,
413 "Non-utf8 string cannot match utf8 check string\n"));
416 if (prog->extflags & RXf_ANCH) { /* Match at beg-of-str or after \n */
417 ml_anch = !( (prog->extflags & RXf_ANCH_SINGLE)
418 || ( (prog->extflags & RXf_ANCH_BOL)
419 && !multiline ) ); /* Check after \n? */
422 if ( !(prog->extflags & RXf_ANCH_GPOS) /* Checked by the caller */
423 && !(prog->intflags & PREGf_IMPLICIT) /* not a real BOL */
424 /* SvCUR is not set on references: SvRV and SvPVX_const overlap */
426 && (strpos != strbeg)) {
427 DEBUG_EXECUTE_r(PerlIO_printf(Perl_debug_log, "Not at start...\n"));
430 if (prog->check_offset_min == prog->check_offset_max &&
431 !(prog->extflags & RXf_CANY_SEEN)) {
432 /* Substring at constant offset from beg-of-str... */
435 s = HOP3c(strpos, prog->check_offset_min, strend);
438 slen = SvCUR(check); /* >= 1 */
440 if ( strend - s > slen || strend - s < slen - 1
441 || (strend - s == slen && strend[-1] != '\n')) {
442 DEBUG_EXECUTE_r(PerlIO_printf(Perl_debug_log, "String too long...\n"));
445 /* Now should match s[0..slen-2] */
447 if (slen && (*SvPVX_const(check) != *s
449 && memNE(SvPVX_const(check), s, slen)))) {
451 DEBUG_EXECUTE_r(PerlIO_printf(Perl_debug_log, "String not equal...\n"));
455 else if (*SvPVX_const(check) != *s
456 || ((slen = SvCUR(check)) > 1
457 && memNE(SvPVX_const(check), s, slen)))
460 goto success_at_start;
463 /* Match is anchored, but substr is not anchored wrt beg-of-str. */
465 start_shift = prog->check_offset_min; /* okay to underestimate on CC */
466 end_shift = prog->check_end_shift;
469 const I32 end = prog->check_offset_max + CHR_SVLEN(check)
470 - (SvTAIL(check) != 0);
471 const I32 eshift = CHR_DIST((U8*)strend, (U8*)s) - end;
473 if (end_shift < eshift)
477 else { /* Can match at random position */
480 start_shift = prog->check_offset_min; /* okay to underestimate on CC */
481 end_shift = prog->check_end_shift;
483 /* end shift should be non negative here */
486 #ifdef DEBUGGING /* 7/99: reports of failure (with the older version) */
488 Perl_croak(aTHX_ "panic: end_shift: %"IVdf" pattern:\n%s\n ",
489 (IV)end_shift, prog->precomp);
493 /* Find a possible match in the region s..strend by looking for
494 the "check" substring in the region corrected by start/end_shift. */
497 I32 srch_start_shift = start_shift;
498 I32 srch_end_shift = end_shift;
499 if (srch_start_shift < 0 && strbeg - s > srch_start_shift) {
500 srch_end_shift -= ((strbeg - s) - srch_start_shift);
501 srch_start_shift = strbeg - s;
503 DEBUG_OPTIMISE_MORE_r({
504 PerlIO_printf(Perl_debug_log, "Check offset min: %"IVdf" Start shift: %"IVdf" End shift %"IVdf" Real End Shift: %"IVdf"\n",
505 (IV)prog->check_offset_min,
506 (IV)srch_start_shift,
508 (IV)prog->check_end_shift);
511 if (flags & REXEC_SCREAM) {
512 I32 p = -1; /* Internal iterator of scream. */
513 I32 * const pp = data ? data->scream_pos : &p;
515 if (PL_screamfirst[BmRARE(check)] >= 0
516 || ( BmRARE(check) == '\n'
517 && (BmPREVIOUS(check) == SvCUR(check) - 1)
519 s = screaminstr(sv, check,
520 srch_start_shift + (s - strbeg), srch_end_shift, pp, 0);
523 /* we may be pointing at the wrong string */
524 if (s && RX_MATCH_COPIED(prog))
525 s = strbeg + (s - SvPVX_const(sv));
527 *data->scream_olds = s;
532 if (prog->extflags & RXf_CANY_SEEN) {
533 start_point= (U8*)(s + srch_start_shift);
534 end_point= (U8*)(strend - srch_end_shift);
536 start_point= HOP3(s, srch_start_shift, srch_start_shift < 0 ? strbeg : strend);
537 end_point= HOP3(strend, -srch_end_shift, strbeg);
539 DEBUG_OPTIMISE_MORE_r({
540 PerlIO_printf(Perl_debug_log, "fbm_instr len=%d str=<%.*s>\n",
541 (int)(end_point - start_point),
542 (int)(end_point - start_point) > 20 ? 20 : (int)(end_point - start_point),
546 s = fbm_instr( start_point, end_point,
547 check, multiline ? FBMrf_MULTILINE : 0);
550 /* Update the count-of-usability, remove useless subpatterns,
554 RE_PV_QUOTED_DECL(quoted, do_utf8, PERL_DEBUG_PAD_ZERO(0),
555 SvPVX_const(check), RE_SV_DUMPLEN(check), 30);
556 PerlIO_printf(Perl_debug_log, "%s %s substr %s%s%s",
557 (s ? "Found" : "Did not find"),
558 (check == (do_utf8 ? prog->anchored_utf8 : prog->anchored_substr)
559 ? "anchored" : "floating"),
562 (s ? " at offset " : "...\n") );
567 /* Finish the diagnostic message */
568 DEBUG_EXECUTE_r(PerlIO_printf(Perl_debug_log, "%ld...\n", (long)(s - i_strpos)) );
570 /* XXX dmq: first branch is for positive lookbehind...
571 Our check string is offset from the beginning of the pattern.
572 So we need to do any stclass tests offset forward from that
581 /* Got a candidate. Check MBOL anchoring, and the *other* substr.
582 Start with the other substr.
583 XXXX no SCREAM optimization yet - and a very coarse implementation
584 XXXX /ttx+/ results in anchored="ttx", floating="x". floating will
585 *always* match. Probably should be marked during compile...
586 Probably it is right to do no SCREAM here...
589 if (do_utf8 ? (prog->float_utf8 && prog->anchored_utf8)
590 : (prog->float_substr && prog->anchored_substr))
592 /* Take into account the "other" substring. */
593 /* XXXX May be hopelessly wrong for UTF... */
596 if (check == (do_utf8 ? prog->float_utf8 : prog->float_substr)) {
599 char * const last = HOP3c(s, -start_shift, strbeg);
601 char * const saved_s = s;
604 t = s - prog->check_offset_max;
605 if (s - strpos > prog->check_offset_max /* signed-corrected t > strpos */
607 || ((t = (char*)reghopmaybe3((U8*)s, -(prog->check_offset_max), (U8*)strpos))
612 t = HOP3c(t, prog->anchored_offset, strend);
613 if (t < other_last) /* These positions already checked */
615 last2 = last1 = HOP3c(strend, -prog->minlen, strbeg);
618 /* XXXX It is not documented what units *_offsets are in.
619 We assume bytes, but this is clearly wrong.
620 Meaning this code needs to be carefully reviewed for errors.
624 /* On end-of-str: see comment below. */
625 must = do_utf8 ? prog->anchored_utf8 : prog->anchored_substr;
626 if (must == &PL_sv_undef) {
628 DEBUG_r(must = prog->anchored_utf8); /* for debug */
633 HOP3(HOP3(last1, prog->anchored_offset, strend)
634 + SvCUR(must), -(SvTAIL(must)!=0), strbeg),
636 multiline ? FBMrf_MULTILINE : 0
639 RE_PV_QUOTED_DECL(quoted, do_utf8, PERL_DEBUG_PAD_ZERO(0),
640 SvPVX_const(must), RE_SV_DUMPLEN(must), 30);
641 PerlIO_printf(Perl_debug_log, "%s anchored substr %s%s",
642 (s ? "Found" : "Contradicts"),
643 quoted, RE_SV_TAIL(must));
648 if (last1 >= last2) {
649 DEBUG_EXECUTE_r(PerlIO_printf(Perl_debug_log,
650 ", giving up...\n"));
653 DEBUG_EXECUTE_r(PerlIO_printf(Perl_debug_log,
654 ", trying floating at offset %ld...\n",
655 (long)(HOP3c(saved_s, 1, strend) - i_strpos)));
656 other_last = HOP3c(last1, prog->anchored_offset+1, strend);
657 s = HOP3c(last, 1, strend);
661 DEBUG_EXECUTE_r(PerlIO_printf(Perl_debug_log, " at offset %ld...\n",
662 (long)(s - i_strpos)));
663 t = HOP3c(s, -prog->anchored_offset, strbeg);
664 other_last = HOP3c(s, 1, strend);
672 else { /* Take into account the floating substring. */
674 char * const saved_s = s;
677 t = HOP3c(s, -start_shift, strbeg);
679 HOP3c(strend, -prog->minlen + prog->float_min_offset, strbeg);
680 if (CHR_DIST((U8*)last, (U8*)t) > prog->float_max_offset)
681 last = HOP3c(t, prog->float_max_offset, strend);
682 s = HOP3c(t, prog->float_min_offset, strend);
685 /* XXXX It is not documented what units *_offsets are in. Assume bytes. */
686 must = do_utf8 ? prog->float_utf8 : prog->float_substr;
687 /* fbm_instr() takes into account exact value of end-of-str
688 if the check is SvTAIL(ed). Since false positives are OK,
689 and end-of-str is not later than strend we are OK. */
690 if (must == &PL_sv_undef) {
692 DEBUG_r(must = prog->float_utf8); /* for debug message */
695 s = fbm_instr((unsigned char*)s,
696 (unsigned char*)last + SvCUR(must)
698 must, multiline ? FBMrf_MULTILINE : 0);
700 RE_PV_QUOTED_DECL(quoted, do_utf8, PERL_DEBUG_PAD_ZERO(0),
701 SvPVX_const(must), RE_SV_DUMPLEN(must), 30);
702 PerlIO_printf(Perl_debug_log, "%s floating substr %s%s",
703 (s ? "Found" : "Contradicts"),
704 quoted, RE_SV_TAIL(must));
708 DEBUG_EXECUTE_r(PerlIO_printf(Perl_debug_log,
709 ", giving up...\n"));
712 DEBUG_EXECUTE_r(PerlIO_printf(Perl_debug_log,
713 ", trying anchored starting at offset %ld...\n",
714 (long)(saved_s + 1 - i_strpos)));
716 s = HOP3c(t, 1, strend);
720 DEBUG_EXECUTE_r(PerlIO_printf(Perl_debug_log, " at offset %ld...\n",
721 (long)(s - i_strpos)));
722 other_last = s; /* Fix this later. --Hugo */
732 t= (char*)HOP3( s, -prog->check_offset_max, (prog->check_offset_max<0) ? strend : strpos);
734 DEBUG_OPTIMISE_MORE_r(
735 PerlIO_printf(Perl_debug_log,
736 "Check offset min:%"IVdf" max:%"IVdf" S:%"IVdf" t:%"IVdf" D:%"IVdf" end:%"IVdf"\n",
737 (IV)prog->check_offset_min,
738 (IV)prog->check_offset_max,
746 if (s - strpos > prog->check_offset_max /* signed-corrected t > strpos */
748 || ((t = (char*)reghopmaybe3((U8*)s, -prog->check_offset_max, (U8*) ((prog->check_offset_max<0) ? strend : strpos)))
751 /* Fixed substring is found far enough so that the match
752 cannot start at strpos. */
754 if (ml_anch && t[-1] != '\n') {
755 /* Eventually fbm_*() should handle this, but often
756 anchored_offset is not 0, so this check will not be wasted. */
757 /* XXXX In the code below we prefer to look for "^" even in
758 presence of anchored substrings. And we search even
759 beyond the found float position. These pessimizations
760 are historical artefacts only. */
762 while (t < strend - prog->minlen) {
764 if (t < check_at - prog->check_offset_min) {
765 if (do_utf8 ? prog->anchored_utf8 : prog->anchored_substr) {
766 /* Since we moved from the found position,
767 we definitely contradict the found anchored
768 substr. Due to the above check we do not
769 contradict "check" substr.
770 Thus we can arrive here only if check substr
771 is float. Redo checking for "other"=="fixed".
774 DEBUG_EXECUTE_r(PerlIO_printf(Perl_debug_log, "Found /%s^%s/m at offset %ld, rescanning for anchored from offset %ld...\n",
775 PL_colors[0], PL_colors[1], (long)(strpos - i_strpos), (long)(strpos - i_strpos + prog->anchored_offset)));
776 goto do_other_anchored;
778 /* We don't contradict the found floating substring. */
779 /* XXXX Why not check for STCLASS? */
781 DEBUG_EXECUTE_r(PerlIO_printf(Perl_debug_log, "Found /%s^%s/m at offset %ld...\n",
782 PL_colors[0], PL_colors[1], (long)(s - i_strpos)));
785 /* Position contradicts check-string */
786 /* XXXX probably better to look for check-string
787 than for "\n", so one should lower the limit for t? */
788 DEBUG_EXECUTE_r(PerlIO_printf(Perl_debug_log, "Found /%s^%s/m, restarting lookup for check-string at offset %ld...\n",
789 PL_colors[0], PL_colors[1], (long)(t + 1 - i_strpos)));
790 other_last = strpos = s = t + 1;
795 DEBUG_EXECUTE_r(PerlIO_printf(Perl_debug_log, "Did not find /%s^%s/m...\n",
796 PL_colors[0], PL_colors[1]));
800 DEBUG_EXECUTE_r(PerlIO_printf(Perl_debug_log, "Starting position does not contradict /%s^%s/m...\n",
801 PL_colors[0], PL_colors[1]));
805 ++BmUSEFUL(do_utf8 ? prog->check_utf8 : prog->check_substr); /* hooray/5 */
808 /* The found string does not prohibit matching at strpos,
809 - no optimization of calling REx engine can be performed,
810 unless it was an MBOL and we are not after MBOL,
811 or a future STCLASS check will fail this. */
813 /* Even in this situation we may use MBOL flag if strpos is offset
814 wrt the start of the string. */
815 if (ml_anch && sv && !SvROK(sv) /* See prev comment on SvROK */
816 && (strpos != strbeg) && strpos[-1] != '\n'
817 /* May be due to an implicit anchor of m{.*foo} */
818 && !(prog->intflags & PREGf_IMPLICIT))
823 DEBUG_EXECUTE_r( if (ml_anch)
824 PerlIO_printf(Perl_debug_log, "Position at offset %ld does not contradict /%s^%s/m...\n",
825 (long)(strpos - i_strpos), PL_colors[0], PL_colors[1]);
828 if (!(prog->intflags & PREGf_NAUGHTY) /* XXXX If strpos moved? */
830 prog->check_utf8 /* Could be deleted already */
831 && --BmUSEFUL(prog->check_utf8) < 0
832 && (prog->check_utf8 == prog->float_utf8)
834 prog->check_substr /* Could be deleted already */
835 && --BmUSEFUL(prog->check_substr) < 0
836 && (prog->check_substr == prog->float_substr)
839 /* If flags & SOMETHING - do not do it many times on the same match */
840 DEBUG_EXECUTE_r(PerlIO_printf(Perl_debug_log, "... Disabling check substring...\n"));
841 SvREFCNT_dec(do_utf8 ? prog->check_utf8 : prog->check_substr);
842 if (do_utf8 ? prog->check_substr : prog->check_utf8)
843 SvREFCNT_dec(do_utf8 ? prog->check_substr : prog->check_utf8);
844 prog->check_substr = prog->check_utf8 = NULL; /* disable */
845 prog->float_substr = prog->float_utf8 = NULL; /* clear */
846 check = NULL; /* abort */
848 /* XXXX This is a remnant of the old implementation. It
849 looks wasteful, since now INTUIT can use many
851 prog->extflags &= ~RXf_USE_INTUIT;
858 /* XXXX BmUSEFUL already changed, maybe multiple change is meaningful... */
859 /* trie stclasses are too expensive to use here, we are better off to
860 leave it to regmatch itself */
861 if (progi->regstclass && PL_regkind[OP(progi->regstclass)]!=TRIE) {
862 /* minlen == 0 is possible if regstclass is \b or \B,
863 and the fixed substr is ''$.
864 Since minlen is already taken into account, s+1 is before strend;
865 accidentally, minlen >= 1 guaranties no false positives at s + 1
866 even for \b or \B. But (minlen? 1 : 0) below assumes that
867 regstclass does not come from lookahead... */
868 /* If regstclass takes bytelength more than 1: If charlength==1, OK.
869 This leaves EXACTF only, which is dealt with in find_byclass(). */
870 const U8* const str = (U8*)STRING(progi->regstclass);
871 const int cl_l = (PL_regkind[OP(progi->regstclass)] == EXACT
872 ? CHR_DIST(str+STR_LEN(progi->regstclass), str)
875 if (prog->anchored_substr || prog->anchored_utf8 || ml_anch)
876 endpos= HOP3c(s, (prog->minlen ? cl_l : 0), strend);
877 else if (prog->float_substr || prog->float_utf8)
878 endpos= HOP3c(HOP3c(check_at, -start_shift, strbeg), cl_l, strend);
882 DEBUG_EXECUTE_r(PerlIO_printf(Perl_debug_log, "start_shift: %"IVdf" check_at: %"IVdf" s: %"IVdf" endpos: %"IVdf"\n",
883 (IV)start_shift, (IV)(check_at - strbeg), (IV)(s - strbeg), (IV)(endpos - strbeg)));
886 s = find_byclass(prog, progi->regstclass, s, endpos, NULL);
889 const char *what = NULL;
891 if (endpos == strend) {
892 DEBUG_EXECUTE_r( PerlIO_printf(Perl_debug_log,
893 "Could not match STCLASS...\n") );
896 DEBUG_EXECUTE_r( PerlIO_printf(Perl_debug_log,
897 "This position contradicts STCLASS...\n") );
898 if ((prog->extflags & RXf_ANCH) && !ml_anch)
900 /* Contradict one of substrings */
901 if (prog->anchored_substr || prog->anchored_utf8) {
902 if ((do_utf8 ? prog->anchored_utf8 : prog->anchored_substr) == check) {
903 DEBUG_EXECUTE_r( what = "anchored" );
905 s = HOP3c(t, 1, strend);
906 if (s + start_shift + end_shift > strend) {
907 /* XXXX Should be taken into account earlier? */
908 DEBUG_EXECUTE_r( PerlIO_printf(Perl_debug_log,
909 "Could not match STCLASS...\n") );
914 DEBUG_EXECUTE_r( PerlIO_printf(Perl_debug_log,
915 "Looking for %s substr starting at offset %ld...\n",
916 what, (long)(s + start_shift - i_strpos)) );
919 /* Have both, check_string is floating */
920 if (t + start_shift >= check_at) /* Contradicts floating=check */
921 goto retry_floating_check;
922 /* Recheck anchored substring, but not floating... */
926 DEBUG_EXECUTE_r( PerlIO_printf(Perl_debug_log,
927 "Looking for anchored substr starting at offset %ld...\n",
928 (long)(other_last - i_strpos)) );
929 goto do_other_anchored;
931 /* Another way we could have checked stclass at the
932 current position only: */
937 DEBUG_EXECUTE_r( PerlIO_printf(Perl_debug_log,
938 "Looking for /%s^%s/m starting at offset %ld...\n",
939 PL_colors[0], PL_colors[1], (long)(t - i_strpos)) );
942 if (!(do_utf8 ? prog->float_utf8 : prog->float_substr)) /* Could have been deleted */
944 /* Check is floating subtring. */
945 retry_floating_check:
946 t = check_at - start_shift;
947 DEBUG_EXECUTE_r( what = "floating" );
948 goto hop_and_restart;
951 DEBUG_EXECUTE_r(PerlIO_printf(Perl_debug_log,
952 "By STCLASS: moving %ld --> %ld\n",
953 (long)(t - i_strpos), (long)(s - i_strpos))
957 DEBUG_EXECUTE_r(PerlIO_printf(Perl_debug_log,
958 "Does not contradict STCLASS...\n");
963 DEBUG_EXECUTE_r(PerlIO_printf(Perl_debug_log, "%s%s:%s match at offset %ld\n",
964 PL_colors[4], (check ? "Guessed" : "Giving up"),
965 PL_colors[5], (long)(s - i_strpos)) );
968 fail_finish: /* Substring not found */
969 if (prog->check_substr || prog->check_utf8) /* could be removed already */
970 BmUSEFUL(do_utf8 ? prog->check_utf8 : prog->check_substr) += 5; /* hooray */
972 DEBUG_EXECUTE_r(PerlIO_printf(Perl_debug_log, "%sMatch rejected by optimizer%s\n",
973 PL_colors[4], PL_colors[5]));
979 #define REXEC_TRIE_READ_CHAR(trie_type, trie, uc, uscan, len, uvc, charid, \
980 foldlen, foldbuf, uniflags) STMT_START { \
981 switch (trie_type) { \
982 case trie_utf8_fold: \
984 uvc = utf8n_to_uvuni( uscan, UTF8_MAXLEN, &len, uniflags ); \
989 uvc = utf8n_to_uvuni( (U8*)uc, UTF8_MAXLEN, &len, uniflags ); \
990 uvc = to_uni_fold( uvc, foldbuf, &foldlen ); \
991 foldlen -= UNISKIP( uvc ); \
992 uscan = foldbuf + UNISKIP( uvc ); \
996 uvc = utf8n_to_uvuni( (U8*)uc, UTF8_MAXLEN, &len, uniflags ); \
1004 charid = trie->charmap[ uvc ]; \
1008 if (trie->widecharmap) { \
1009 SV** const svpp = hv_fetch(trie->widecharmap, \
1010 (char*)&uvc, sizeof(UV), 0); \
1012 charid = (U16)SvIV(*svpp); \
1017 #define REXEC_FBC_EXACTISH_CHECK(CoNd) \
1020 ibcmp_utf8(s, NULL, 0, do_utf8, \
1021 m, NULL, ln, (bool)UTF)) \
1022 && (!reginfo || regtry(reginfo, &s)) ) \
1025 U8 foldbuf[UTF8_MAXBYTES_CASE+1]; \
1026 uvchr_to_utf8(tmpbuf, c); \
1027 f = to_utf8_fold(tmpbuf, foldbuf, &foldlen); \
1029 && (f == c1 || f == c2) \
1030 && (ln == foldlen || \
1031 !ibcmp_utf8((char *) foldbuf, \
1032 NULL, foldlen, do_utf8, \
1034 NULL, ln, (bool)UTF)) \
1035 && (!reginfo || regtry(reginfo, &s)) ) \
1040 #define REXEC_FBC_EXACTISH_SCAN(CoNd) \
1044 && (ln == 1 || !(OP(c) == EXACTF \
1046 : ibcmp_locale(s, m, ln))) \
1047 && (!reginfo || regtry(reginfo, &s)) ) \
1053 #define REXEC_FBC_UTF8_SCAN(CoDe) \
1055 while (s + (uskip = UTF8SKIP(s)) <= strend) { \
1061 #define REXEC_FBC_SCAN(CoDe) \
1063 while (s < strend) { \
1069 #define REXEC_FBC_UTF8_CLASS_SCAN(CoNd) \
1070 REXEC_FBC_UTF8_SCAN( \
1072 if (tmp && (!reginfo || regtry(reginfo, &s))) \
1081 #define REXEC_FBC_CLASS_SCAN(CoNd) \
1084 if (tmp && (!reginfo || regtry(reginfo, &s))) \
1093 #define REXEC_FBC_TRYIT \
1094 if ((!reginfo || regtry(reginfo, &s))) \
1097 #define REXEC_FBC_CSCAN_PRELOAD(UtFpReLoAd,CoNdUtF8,CoNd) \
1100 REXEC_FBC_UTF8_CLASS_SCAN(CoNdUtF8); \
1103 REXEC_FBC_CLASS_SCAN(CoNd); \
1107 #define REXEC_FBC_CSCAN_TAINT(CoNdUtF8,CoNd) \
1108 PL_reg_flags |= RF_tainted; \
1110 REXEC_FBC_UTF8_CLASS_SCAN(CoNdUtF8); \
1113 REXEC_FBC_CLASS_SCAN(CoNd); \
1117 #define DUMP_EXEC_POS(li,s,doutf8) \
1118 dump_exec_pos(li,s,(PL_regeol),(PL_bostr),(PL_reg_starttry),doutf8)
1120 /* We know what class REx starts with. Try to find this position... */
1121 /* if reginfo is NULL, its a dryrun */
1122 /* annoyingly all the vars in this routine have different names from their counterparts
1123 in regmatch. /grrr */
1126 S_find_byclass(pTHX_ regexp * prog, const regnode *c, char *s,
1127 const char *strend, regmatch_info *reginfo)
1130 const I32 doevery = (prog->intflags & PREGf_SKIP) == 0;
1134 register STRLEN uskip;
1138 register I32 tmp = 1; /* Scratch variable? */
1139 register const bool do_utf8 = PL_reg_match_utf8;
1140 RXi_GET_DECL(prog,progi);
1142 /* We know what class it must start with. */
1146 REXEC_FBC_UTF8_CLASS_SCAN((ANYOF_FLAGS(c) & ANYOF_UNICODE) ||
1147 !UTF8_IS_INVARIANT((U8)s[0]) ?
1148 reginclass(prog, c, (U8*)s, 0, do_utf8) :
1149 REGINCLASS(prog, c, (U8*)s));
1152 while (s < strend) {
1155 if (REGINCLASS(prog, c, (U8*)s) ||
1156 (ANYOF_FOLD_SHARP_S(c, s, strend) &&
1157 /* The assignment of 2 is intentional:
1158 * for the folded sharp s, the skip is 2. */
1159 (skip = SHARP_S_SKIP))) {
1160 if (tmp && (!reginfo || regtry(reginfo, &s)))
1173 if (tmp && (!reginfo || regtry(reginfo, &s)))
1181 ln = STR_LEN(c); /* length to match in octets/bytes */
1182 lnc = (I32) ln; /* length to match in characters */
1184 STRLEN ulen1, ulen2;
1186 U8 tmpbuf1[UTF8_MAXBYTES_CASE+1];
1187 U8 tmpbuf2[UTF8_MAXBYTES_CASE+1];
1188 const U32 uniflags = UTF8_ALLOW_DEFAULT;
1190 to_utf8_lower((U8*)m, tmpbuf1, &ulen1);
1191 to_utf8_upper((U8*)m, tmpbuf2, &ulen2);
1193 c1 = utf8n_to_uvchr(tmpbuf1, UTF8_MAXBYTES_CASE,
1195 c2 = utf8n_to_uvchr(tmpbuf2, UTF8_MAXBYTES_CASE,
1198 while (sm < ((U8 *) m + ln)) {
1213 c2 = PL_fold_locale[c1];
1215 e = HOP3c(strend, -((I32)lnc), s);
1217 if (!reginfo && e < s)
1218 e = s; /* Due to minlen logic of intuit() */
1220 /* The idea in the EXACTF* cases is to first find the
1221 * first character of the EXACTF* node and then, if
1222 * necessary, case-insensitively compare the full
1223 * text of the node. The c1 and c2 are the first
1224 * characters (though in Unicode it gets a bit
1225 * more complicated because there are more cases
1226 * than just upper and lower: one needs to use
1227 * the so-called folding case for case-insensitive
1228 * matching (called "loose matching" in Unicode).
1229 * ibcmp_utf8() will do just that. */
1233 U8 tmpbuf [UTF8_MAXBYTES+1];
1234 STRLEN len, foldlen;
1235 const U32 uniflags = UTF8_ALLOW_DEFAULT;
1237 /* Upper and lower of 1st char are equal -
1238 * probably not a "letter". */
1240 c = utf8n_to_uvchr((U8*)s, UTF8_MAXBYTES, &len,
1242 REXEC_FBC_EXACTISH_CHECK(c == c1);
1247 c = utf8n_to_uvchr((U8*)s, UTF8_MAXBYTES, &len,
1250 /* Handle some of the three Greek sigmas cases.
1251 * Note that not all the possible combinations
1252 * are handled here: some of them are handled
1253 * by the standard folding rules, and some of
1254 * them (the character class or ANYOF cases)
1255 * are handled during compiletime in
1256 * regexec.c:S_regclass(). */
1257 if (c == (UV)UNICODE_GREEK_CAPITAL_LETTER_SIGMA ||
1258 c == (UV)UNICODE_GREEK_SMALL_LETTER_FINAL_SIGMA)
1259 c = (UV)UNICODE_GREEK_SMALL_LETTER_SIGMA;
1261 REXEC_FBC_EXACTISH_CHECK(c == c1 || c == c2);
1267 REXEC_FBC_EXACTISH_SCAN(*(U8*)s == c1);
1269 REXEC_FBC_EXACTISH_SCAN(*(U8*)s == c1 || *(U8*)s == c2);
1273 PL_reg_flags |= RF_tainted;
1280 U8 * const r = reghop3((U8*)s, -1, (U8*)PL_bostr);
1281 tmp = utf8n_to_uvchr(r, UTF8SKIP(r), 0, UTF8_ALLOW_DEFAULT);
1283 tmp = ((OP(c) == BOUND ?
1284 isALNUM_uni(tmp) : isALNUM_LC_uvchr(UNI_TO_NATIVE(tmp))) != 0);
1285 LOAD_UTF8_CHARCLASS_ALNUM();
1286 REXEC_FBC_UTF8_SCAN(
1287 if (tmp == !(OP(c) == BOUND ?
1288 (bool)swash_fetch(PL_utf8_alnum, (U8*)s, do_utf8) :
1289 isALNUM_LC_utf8((U8*)s)))
1297 tmp = (s != PL_bostr) ? UCHARAT(s - 1) : '\n';
1298 tmp = ((OP(c) == BOUND ? isALNUM(tmp) : isALNUM_LC(tmp)) != 0);
1301 !(OP(c) == BOUND ? isALNUM(*s) : isALNUM_LC(*s))) {
1307 if ((!prog->minlen && tmp) && (!reginfo || regtry(reginfo, &s)))
1311 PL_reg_flags |= RF_tainted;
1318 U8 * const r = reghop3((U8*)s, -1, (U8*)PL_bostr);
1319 tmp = utf8n_to_uvchr(r, UTF8SKIP(r), 0, UTF8_ALLOW_DEFAULT);
1321 tmp = ((OP(c) == NBOUND ?
1322 isALNUM_uni(tmp) : isALNUM_LC_uvchr(UNI_TO_NATIVE(tmp))) != 0);
1323 LOAD_UTF8_CHARCLASS_ALNUM();
1324 REXEC_FBC_UTF8_SCAN(
1325 if (tmp == !(OP(c) == NBOUND ?
1326 (bool)swash_fetch(PL_utf8_alnum, (U8*)s, do_utf8) :
1327 isALNUM_LC_utf8((U8*)s)))
1329 else REXEC_FBC_TRYIT;
1333 tmp = (s != PL_bostr) ? UCHARAT(s - 1) : '\n';
1334 tmp = ((OP(c) == NBOUND ?
1335 isALNUM(tmp) : isALNUM_LC(tmp)) != 0);
1338 !(OP(c) == NBOUND ? isALNUM(*s) : isALNUM_LC(*s)))
1340 else REXEC_FBC_TRYIT;
1343 if ((!prog->minlen && !tmp) && (!reginfo || regtry(reginfo, &s)))
1347 REXEC_FBC_CSCAN_PRELOAD(
1348 LOAD_UTF8_CHARCLASS_ALNUM(),
1349 swash_fetch(PL_utf8_alnum, (U8*)s, do_utf8),
1353 REXEC_FBC_CSCAN_TAINT(
1354 isALNUM_LC_utf8((U8*)s),
1358 REXEC_FBC_CSCAN_PRELOAD(
1359 LOAD_UTF8_CHARCLASS_ALNUM(),
1360 !swash_fetch(PL_utf8_alnum, (U8*)s, do_utf8),
1364 REXEC_FBC_CSCAN_TAINT(
1365 !isALNUM_LC_utf8((U8*)s),
1369 REXEC_FBC_CSCAN_PRELOAD(
1370 LOAD_UTF8_CHARCLASS_SPACE(),
1371 *s == ' ' || swash_fetch(PL_utf8_space,(U8*)s, do_utf8),
1375 REXEC_FBC_CSCAN_TAINT(
1376 *s == ' ' || isSPACE_LC_utf8((U8*)s),
1380 REXEC_FBC_CSCAN_PRELOAD(
1381 LOAD_UTF8_CHARCLASS_SPACE(),
1382 !(*s == ' ' || swash_fetch(PL_utf8_space,(U8*)s, do_utf8)),
1386 REXEC_FBC_CSCAN_TAINT(
1387 !(*s == ' ' || isSPACE_LC_utf8((U8*)s)),
1391 REXEC_FBC_CSCAN_PRELOAD(
1392 LOAD_UTF8_CHARCLASS_DIGIT(),
1393 swash_fetch(PL_utf8_digit,(U8*)s, do_utf8),
1397 REXEC_FBC_CSCAN_TAINT(
1398 isDIGIT_LC_utf8((U8*)s),
1402 REXEC_FBC_CSCAN_PRELOAD(
1403 LOAD_UTF8_CHARCLASS_DIGIT(),
1404 !swash_fetch(PL_utf8_digit,(U8*)s, do_utf8),
1408 REXEC_FBC_CSCAN_TAINT(
1409 !isDIGIT_LC_utf8((U8*)s),
1415 const enum { trie_plain, trie_utf8, trie_utf8_fold }
1416 trie_type = do_utf8 ?
1417 (c->flags == EXACT ? trie_utf8 : trie_utf8_fold)
1419 /* what trie are we using right now */
1421 = (reg_ac_data*)progi->data->data[ ARG( c ) ];
1422 reg_trie_data *trie=aho->trie;
1424 const char *last_start = strend - trie->minlen;
1426 const char *real_start = s;
1428 STRLEN maxlen = trie->maxlen;
1430 U8 **points; /* map of where we were in the input string
1431 when reading a given char. For ASCII this
1432 is unnecessary overhead as the relationship
1433 is always 1:1, but for unicode, especially
1434 case folded unicode this is not true. */
1435 U8 foldbuf[ UTF8_MAXBYTES_CASE + 1 ];
1439 GET_RE_DEBUG_FLAGS_DECL;
1441 /* We can't just allocate points here. We need to wrap it in
1442 * an SV so it gets freed properly if there is a croak while
1443 * running the match */
1446 sv_points=newSV(maxlen * sizeof(U8 *));
1447 SvCUR_set(sv_points,
1448 maxlen * sizeof(U8 *));
1449 SvPOK_on(sv_points);
1450 sv_2mortal(sv_points);
1451 points=(U8**)SvPV_nolen(sv_points );
1452 if ( trie_type != trie_utf8_fold
1453 && (trie->bitmap || OP(c)==AHOCORASICKC) )
1456 bitmap=(U8*)trie->bitmap;
1458 bitmap=(U8*)ANYOF_BITMAP(c);
1460 /* this is the Aho-Corasick algorithm modified a touch
1461 to include special handling for long "unknown char"
1462 sequences. The basic idea being that we use AC as long
1463 as we are dealing with a possible matching char, when
1464 we encounter an unknown char (and we have not encountered
1465 an accepting state) we scan forward until we find a legal
1467 AC matching is basically that of trie matching, except
1468 that when we encounter a failing transition, we fall back
1469 to the current states "fail state", and try the current char
1470 again, a process we repeat until we reach the root state,
1471 state 1, or a legal transition. If we fail on the root state
1472 then we can either terminate if we have reached an accepting
1473 state previously, or restart the entire process from the beginning
1477 while (s <= last_start) {
1478 const U32 uniflags = UTF8_ALLOW_DEFAULT;
1486 U8 *uscan = (U8*)NULL;
1487 U8 *leftmost = NULL;
1489 U32 accepted_word= 0;
1493 while ( state && uc <= (U8*)strend ) {
1495 U32 word = aho->states[ state ].wordnum;
1499 DEBUG_TRIE_EXECUTE_r(
1500 if ( uc <= (U8*)last_start && !BITMAP_TEST(bitmap,*uc) ) {
1501 dump_exec_pos( (char *)uc, c, strend, real_start,
1502 (char *)uc, do_utf8 );
1503 PerlIO_printf( Perl_debug_log,
1504 " Scanning for legal start char...\n");
1507 while ( uc <= (U8*)last_start && !BITMAP_TEST(bitmap,*uc) ) {
1512 if (uc >(U8*)last_start) break;
1516 U8 *lpos= points[ (pointpos - trie->wordlen[word-1] ) % maxlen ];
1517 if (!leftmost || lpos < leftmost) {
1518 DEBUG_r(accepted_word=word);
1524 points[pointpos++ % maxlen]= uc;
1525 REXEC_TRIE_READ_CHAR(trie_type, trie, uc, uscan, len,
1526 uvc, charid, foldlen, foldbuf, uniflags);
1527 DEBUG_TRIE_EXECUTE_r({
1528 dump_exec_pos( (char *)uc, c, strend, real_start,
1530 PerlIO_printf(Perl_debug_log,
1531 " Charid:%3u CP:%4"UVxf" ",
1537 word = aho->states[ state ].wordnum;
1539 base = aho->states[ state ].trans.base;
1541 DEBUG_TRIE_EXECUTE_r({
1543 dump_exec_pos( (char *)uc, c, strend, real_start,
1545 PerlIO_printf( Perl_debug_log,
1546 "%sState: %4"UVxf", word=%"UVxf,
1547 failed ? " Fail transition to " : "",
1548 (UV)state, (UV)word);
1553 (base + charid > trie->uniquecharcount )
1554 && (base + charid - 1 - trie->uniquecharcount
1556 && trie->trans[base + charid - 1 -
1557 trie->uniquecharcount].check == state
1558 && (tmp=trie->trans[base + charid - 1 -
1559 trie->uniquecharcount ].next))
1561 DEBUG_TRIE_EXECUTE_r(
1562 PerlIO_printf( Perl_debug_log," - legal\n"));
1567 DEBUG_TRIE_EXECUTE_r(
1568 PerlIO_printf( Perl_debug_log," - fail\n"));
1570 state = aho->fail[state];
1574 /* we must be accepting here */
1575 DEBUG_TRIE_EXECUTE_r(
1576 PerlIO_printf( Perl_debug_log," - accepting\n"));
1585 if (!state) state = 1;
1588 if ( aho->states[ state ].wordnum ) {
1589 U8 *lpos = points[ (pointpos - trie->wordlen[aho->states[ state ].wordnum-1]) % maxlen ];
1590 if (!leftmost || lpos < leftmost) {
1591 DEBUG_r(accepted_word=aho->states[ state ].wordnum);
1596 s = (char*)leftmost;
1597 DEBUG_TRIE_EXECUTE_r({
1599 Perl_debug_log,"Matches word #%"UVxf" at position %"IVdf". Trying full pattern...\n",
1600 (UV)accepted_word, (IV)(s - real_start)
1603 if (!reginfo || regtry(reginfo, &s)) {
1609 DEBUG_TRIE_EXECUTE_r({
1610 PerlIO_printf( Perl_debug_log,"Pattern failed. Looking for new start point...\n");
1613 DEBUG_TRIE_EXECUTE_r(
1614 PerlIO_printf( Perl_debug_log,"No match.\n"));
1623 Perl_croak(aTHX_ "panic: unknown regstclass %d", (int)OP(c));
1632 - regexec_flags - match a regexp against a string
1635 Perl_regexec_flags(pTHX_ register regexp *prog, char *stringarg, register char *strend,
1636 char *strbeg, I32 minend, SV *sv, void *data, U32 flags)
1637 /* strend: pointer to null at end of string */
1638 /* strbeg: real beginning of string */
1639 /* minend: end of match must be >=minend after stringarg. */
1640 /* data: May be used for some additional optimizations.
1641 Currently its only used, with a U32 cast, for transmitting
1642 the ganch offset when doing a /g match. This will change */
1643 /* nosave: For optimizations. */
1646 /*register*/ char *s;
1647 register regnode *c;
1648 /*register*/ char *startpos = stringarg;
1649 I32 minlen; /* must match at least this many chars */
1650 I32 dontbother = 0; /* how many characters not to try at end */
1651 I32 end_shift = 0; /* Same for the end. */ /* CC */
1652 I32 scream_pos = -1; /* Internal iterator of scream. */
1653 char *scream_olds = NULL;
1654 SV* const oreplsv = GvSV(PL_replgv);
1655 const bool do_utf8 = (bool)DO_UTF8(sv);
1657 RXi_GET_DECL(prog,progi);
1658 regmatch_info reginfo; /* create some info to pass to regtry etc */
1660 GET_RE_DEBUG_FLAGS_DECL;
1662 PERL_UNUSED_ARG(data);
1664 /* Be paranoid... */
1665 if (prog == NULL || startpos == NULL) {
1666 Perl_croak(aTHX_ "NULL regexp parameter");
1670 multiline = prog->extflags & RXf_PMf_MULTILINE;
1671 reginfo.prog = prog;
1673 RX_MATCH_UTF8_set(prog, do_utf8);
1675 debug_start_match(prog, do_utf8, startpos, strend,
1679 minlen = prog->minlen;
1681 if (strend - startpos < (minlen+(prog->check_offset_min<0?prog->check_offset_min:0))) {
1682 DEBUG_EXECUTE_r(PerlIO_printf(Perl_debug_log,
1683 "String too short [regexec_flags]...\n"));
1688 /* Check validity of program. */
1689 if (UCHARAT(progi->program) != REG_MAGIC) {
1690 Perl_croak(aTHX_ "corrupted regexp program");
1694 PL_reg_eval_set = 0;
1697 if (prog->extflags & RXf_UTF8)
1698 PL_reg_flags |= RF_utf8;
1700 /* Mark beginning of line for ^ and lookbehind. */
1701 reginfo.bol = startpos; /* XXX not used ??? */
1705 /* Mark end of line for $ (and such) */
1708 /* see how far we have to get to not match where we matched before */
1709 reginfo.till = startpos+minend;
1711 /* If there is a "must appear" string, look for it. */
1714 if (prog->extflags & RXf_GPOS_SEEN) { /* Need to set reginfo->ganch */
1717 if (flags & REXEC_IGNOREPOS) /* Means: check only at start */
1718 reginfo.ganch = startpos + prog->gofs;
1719 else if (sv && SvTYPE(sv) >= SVt_PVMG
1721 && (mg = mg_find(sv, PERL_MAGIC_regex_global))
1722 && mg->mg_len >= 0) {
1723 reginfo.ganch = strbeg + mg->mg_len; /* Defined pos() */
1724 if (prog->extflags & RXf_ANCH_GPOS) {
1725 if (s > reginfo.ganch)
1727 s = reginfo.ganch - prog->gofs;
1731 reginfo.ganch = strbeg + PTR2UV(data);
1732 } else /* pos() not defined */
1733 reginfo.ganch = strbeg;
1735 if (PL_curpm && (PM_GETRE(PL_curpm) == prog)) {
1738 /* We have to be careful. If the previous successful match
1739 was from this regex we don't want a subsequent paritally
1740 successful match to clobber the old results.
1741 So when we detect this possibility we add a swap buffer
1742 to the re, and switch the buffer each match. If we fail
1743 we switch it back, otherwise we leave it swapped.
1745 Newxz(progi->swap, 1, regexp_paren_ofs);
1746 /* no need to copy these */
1747 Newxz(progi->swap->startp, prog->nparens + 1, I32);
1748 Newxz(progi->swap->endp, prog->nparens + 1, I32);
1750 t = progi->swap->startp;
1751 progi->swap->startp = prog->startp;
1753 t = progi->swap->endp;
1754 progi->swap->endp = prog->endp;
1757 if (!(flags & REXEC_CHECKED) && (prog->check_substr != NULL || prog->check_utf8 != NULL)) {
1758 re_scream_pos_data d;
1760 d.scream_olds = &scream_olds;
1761 d.scream_pos = &scream_pos;
1762 s = re_intuit_start(prog, sv, s, strend, flags, &d);
1764 DEBUG_EXECUTE_r(PerlIO_printf(Perl_debug_log, "Not present...\n"));
1765 goto phooey; /* not present */
1771 /* Simplest case: anchored match need be tried only once. */
1772 /* [unless only anchor is BOL and multiline is set] */
1773 if (prog->extflags & (RXf_ANCH & ~RXf_ANCH_GPOS)) {
1774 if (s == startpos && regtry(®info, &startpos))
1776 else if (multiline || (prog->intflags & PREGf_IMPLICIT)
1777 || (prog->extflags & RXf_ANCH_MBOL)) /* XXXX SBOL? */
1782 dontbother = minlen - 1;
1783 end = HOP3c(strend, -dontbother, strbeg) - 1;
1784 /* for multiline we only have to try after newlines */
1785 if (prog->check_substr || prog->check_utf8) {
1789 if (regtry(®info, &s))
1794 if (prog->extflags & RXf_USE_INTUIT) {
1795 s = re_intuit_start(prog, sv, s + 1, strend, flags, NULL);
1806 if (*s++ == '\n') { /* don't need PL_utf8skip here */
1807 if (regtry(®info, &s))
1814 } else if (RXf_GPOS_CHECK == (prog->extflags & RXf_GPOS_CHECK))
1816 /* the warning about reginfo.ganch being used without intialization
1817 is bogus -- we set it above, when prog->extflags & RXf_GPOS_SEEN
1818 and we only enter this block when the same bit is set. */
1819 char *tmp_s = reginfo.ganch - prog->gofs;
1820 if (regtry(®info, &tmp_s))
1825 /* Messy cases: unanchored match. */
1826 if ((prog->anchored_substr || prog->anchored_utf8) && prog->intflags & PREGf_SKIP) {
1827 /* we have /x+whatever/ */
1828 /* it must be a one character string (XXXX Except UTF?) */
1833 if (!(do_utf8 ? prog->anchored_utf8 : prog->anchored_substr))
1834 do_utf8 ? to_utf8_substr(prog) : to_byte_substr(prog);
1835 ch = SvPVX_const(do_utf8 ? prog->anchored_utf8 : prog->anchored_substr)[0];
1840 DEBUG_EXECUTE_r( did_match = 1 );
1841 if (regtry(®info, &s)) goto got_it;
1843 while (s < strend && *s == ch)
1851 DEBUG_EXECUTE_r( did_match = 1 );
1852 if (regtry(®info, &s)) goto got_it;
1854 while (s < strend && *s == ch)
1859 DEBUG_EXECUTE_r(if (!did_match)
1860 PerlIO_printf(Perl_debug_log,
1861 "Did not find anchored character...\n")
1864 else if (prog->anchored_substr != NULL
1865 || prog->anchored_utf8 != NULL
1866 || ((prog->float_substr != NULL || prog->float_utf8 != NULL)
1867 && prog->float_max_offset < strend - s)) {
1872 char *last1; /* Last position checked before */
1876 if (prog->anchored_substr || prog->anchored_utf8) {
1877 if (!(do_utf8 ? prog->anchored_utf8 : prog->anchored_substr))
1878 do_utf8 ? to_utf8_substr(prog) : to_byte_substr(prog);
1879 must = do_utf8 ? prog->anchored_utf8 : prog->anchored_substr;
1880 back_max = back_min = prog->anchored_offset;
1882 if (!(do_utf8 ? prog->float_utf8 : prog->float_substr))
1883 do_utf8 ? to_utf8_substr(prog) : to_byte_substr(prog);
1884 must = do_utf8 ? prog->float_utf8 : prog->float_substr;
1885 back_max = prog->float_max_offset;
1886 back_min = prog->float_min_offset;
1890 if (must == &PL_sv_undef)
1891 /* could not downgrade utf8 check substring, so must fail */
1897 last = HOP3c(strend, /* Cannot start after this */
1898 -(I32)(CHR_SVLEN(must)
1899 - (SvTAIL(must) != 0) + back_min), strbeg);
1902 last1 = HOPc(s, -1);
1904 last1 = s - 1; /* bogus */
1906 /* XXXX check_substr already used to find "s", can optimize if
1907 check_substr==must. */
1909 dontbother = end_shift;
1910 strend = HOPc(strend, -dontbother);
1911 while ( (s <= last) &&
1912 ((flags & REXEC_SCREAM)
1913 ? (s = screaminstr(sv, must, HOP3c(s, back_min, (back_min<0 ? strbeg : strend)) - strbeg,
1914 end_shift, &scream_pos, 0))
1915 : (s = fbm_instr((unsigned char*)HOP3(s, back_min, (back_min<0 ? strbeg : strend)),
1916 (unsigned char*)strend, must,
1917 multiline ? FBMrf_MULTILINE : 0))) ) {
1918 /* we may be pointing at the wrong string */
1919 if ((flags & REXEC_SCREAM) && RX_MATCH_COPIED(prog))
1920 s = strbeg + (s - SvPVX_const(sv));
1921 DEBUG_EXECUTE_r( did_match = 1 );
1922 if (HOPc(s, -back_max) > last1) {
1923 last1 = HOPc(s, -back_min);
1924 s = HOPc(s, -back_max);
1927 char * const t = (last1 >= PL_bostr) ? HOPc(last1, 1) : last1 + 1;
1929 last1 = HOPc(s, -back_min);
1933 while (s <= last1) {
1934 if (regtry(®info, &s))
1940 while (s <= last1) {
1941 if (regtry(®info, &s))
1947 DEBUG_EXECUTE_r(if (!did_match) {
1948 RE_PV_QUOTED_DECL(quoted, do_utf8, PERL_DEBUG_PAD_ZERO(0),
1949 SvPVX_const(must), RE_SV_DUMPLEN(must), 30);
1950 PerlIO_printf(Perl_debug_log, "Did not find %s substr %s%s...\n",
1951 ((must == prog->anchored_substr || must == prog->anchored_utf8)
1952 ? "anchored" : "floating"),
1953 quoted, RE_SV_TAIL(must));
1957 else if ( (c = progi->regstclass) ) {
1959 const OPCODE op = OP(progi->regstclass);
1960 /* don't bother with what can't match */
1961 if (PL_regkind[op] != EXACT && op != CANY && PL_regkind[op] != TRIE)
1962 strend = HOPc(strend, -(minlen - 1));
1965 SV * const prop = sv_newmortal();
1966 regprop(prog, prop, c);
1968 RE_PV_QUOTED_DECL(quoted,UTF,PERL_DEBUG_PAD_ZERO(1),
1970 PerlIO_printf(Perl_debug_log,
1971 "Matching stclass %.*s against %s (%d chars)\n",
1972 (int)SvCUR(prop), SvPVX_const(prop),
1973 quoted, (int)(strend - s));
1976 if (find_byclass(prog, c, s, strend, ®info))
1978 DEBUG_EXECUTE_r(PerlIO_printf(Perl_debug_log, "Contradicts stclass... [regexec_flags]\n"));
1982 if (prog->float_substr != NULL || prog->float_utf8 != NULL) {
1987 if (!(do_utf8 ? prog->float_utf8 : prog->float_substr))
1988 do_utf8 ? to_utf8_substr(prog) : to_byte_substr(prog);
1989 float_real = do_utf8 ? prog->float_utf8 : prog->float_substr;
1991 if (flags & REXEC_SCREAM) {
1992 last = screaminstr(sv, float_real, s - strbeg,
1993 end_shift, &scream_pos, 1); /* last one */
1995 last = scream_olds; /* Only one occurrence. */
1996 /* we may be pointing at the wrong string */
1997 else if (RX_MATCH_COPIED(prog))
1998 s = strbeg + (s - SvPVX_const(sv));
2002 const char * const little = SvPV_const(float_real, len);
2004 if (SvTAIL(float_real)) {
2005 if (memEQ(strend - len + 1, little, len - 1))
2006 last = strend - len + 1;
2007 else if (!multiline)
2008 last = memEQ(strend - len, little, len)
2009 ? strend - len : NULL;
2015 last = rninstr(s, strend, little, little + len);
2017 last = strend; /* matching "$" */
2022 PerlIO_printf(Perl_debug_log,
2023 "%sCan't trim the tail, match fails (should not happen)%s\n",
2024 PL_colors[4], PL_colors[5]));
2025 goto phooey; /* Should not happen! */
2027 dontbother = strend - last + prog->float_min_offset;
2029 if (minlen && (dontbother < minlen))
2030 dontbother = minlen - 1;
2031 strend -= dontbother; /* this one's always in bytes! */
2032 /* We don't know much -- general case. */
2035 if (regtry(®info, &s))
2044 if (regtry(®info, &s))
2046 } while (s++ < strend);
2054 RX_MATCH_TAINTED_set(prog, PL_reg_flags & RF_tainted);
2056 if (PL_reg_eval_set) {
2057 /* Preserve the current value of $^R */
2058 if (oreplsv != GvSV(PL_replgv))
2059 sv_setsv(oreplsv, GvSV(PL_replgv));/* So that when GvSV(replgv) is
2060 restored, the value remains
2062 restore_pos(aTHX_ prog);
2064 if (prog->paren_names)
2065 (void)hv_iterinit(prog->paren_names);
2067 /* make sure $`, $&, $', and $digit will work later */
2068 if ( !(flags & REXEC_NOT_FIRST) ) {
2069 RX_MATCH_COPY_FREE(prog);
2070 if (flags & REXEC_COPY_STR) {
2071 const I32 i = PL_regeol - startpos + (stringarg - strbeg);
2072 #ifdef PERL_OLD_COPY_ON_WRITE
2074 || (SvFLAGS(sv) & CAN_COW_MASK) == CAN_COW_FLAGS)) {
2076 PerlIO_printf(Perl_debug_log,
2077 "Copy on write: regexp capture, type %d\n",
2080 prog->saved_copy = sv_setsv_cow(prog->saved_copy, sv);
2081 prog->subbeg = (char *)SvPVX_const(prog->saved_copy);
2082 assert (SvPOKp(prog->saved_copy));
2086 RX_MATCH_COPIED_on(prog);
2087 s = savepvn(strbeg, i);
2093 prog->subbeg = strbeg;
2094 prog->sublen = PL_regeol - strbeg; /* strend may have been modified */
2101 DEBUG_EXECUTE_r(PerlIO_printf(Perl_debug_log, "%sMatch failed%s\n",
2102 PL_colors[4], PL_colors[5]));
2103 if (PL_reg_eval_set)
2104 restore_pos(aTHX_ prog);
2106 /* we failed :-( roll it back */
2108 t = progi->swap->startp;
2109 progi->swap->startp = prog->startp;
2111 t = progi->swap->endp;
2112 progi->swap->endp = prog->endp;
2120 - regtry - try match at specific point
2122 STATIC I32 /* 0 failure, 1 success */
2123 S_regtry(pTHX_ regmatch_info *reginfo, char **startpos)
2129 regexp *prog = reginfo->prog;
2130 RXi_GET_DECL(prog,progi);
2131 GET_RE_DEBUG_FLAGS_DECL;
2132 reginfo->cutpoint=NULL;
2134 if ((prog->extflags & RXf_EVAL_SEEN) && !PL_reg_eval_set) {
2137 PL_reg_eval_set = RS_init;
2138 DEBUG_EXECUTE_r(DEBUG_s(
2139 PerlIO_printf(Perl_debug_log, " setting stack tmpbase at %"IVdf"\n",
2140 (IV)(PL_stack_sp - PL_stack_base));
2143 cxstack[cxstack_ix].blk_oldsp = PL_stack_sp - PL_stack_base;
2144 /* Otherwise OP_NEXTSTATE will free whatever on stack now. */
2146 /* Apparently this is not needed, judging by wantarray. */
2147 /* SAVEI8(cxstack[cxstack_ix].blk_gimme);
2148 cxstack[cxstack_ix].blk_gimme = G_SCALAR; */
2151 /* Make $_ available to executed code. */
2152 if (reginfo->sv != DEFSV) {
2154 DEFSV = reginfo->sv;
2157 if (!(SvTYPE(reginfo->sv) >= SVt_PVMG && SvMAGIC(reginfo->sv)
2158 && (mg = mg_find(reginfo->sv, PERL_MAGIC_regex_global)))) {
2159 /* prepare for quick setting of pos */
2160 #ifdef PERL_OLD_COPY_ON_WRITE
2161 if (SvIsCOW(reginfo->sv))
2162 sv_force_normal_flags(reginfo->sv, 0);
2164 mg = sv_magicext(reginfo->sv, NULL, PERL_MAGIC_regex_global,
2165 &PL_vtbl_mglob, NULL, 0);
2169 PL_reg_oldpos = mg->mg_len;
2170 SAVEDESTRUCTOR_X(restore_pos, prog);
2172 if (!PL_reg_curpm) {
2173 Newxz(PL_reg_curpm, 1, PMOP);
2176 SV* const repointer = newSViv(0);
2177 /* so we know which PL_regex_padav element is PL_reg_curpm */
2178 SvFLAGS(repointer) |= SVf_BREAK;
2179 av_push(PL_regex_padav,repointer);
2180 PL_reg_curpm->op_pmoffset = av_len(PL_regex_padav);
2181 PL_regex_pad = AvARRAY(PL_regex_padav);
2185 PM_SETRE(PL_reg_curpm, prog);
2186 PL_reg_oldcurpm = PL_curpm;
2187 PL_curpm = PL_reg_curpm;
2188 if (RX_MATCH_COPIED(prog)) {
2189 /* Here is a serious problem: we cannot rewrite subbeg,
2190 since it may be needed if this match fails. Thus
2191 $` inside (?{}) could fail... */
2192 PL_reg_oldsaved = prog->subbeg;
2193 PL_reg_oldsavedlen = prog->sublen;
2194 #ifdef PERL_OLD_COPY_ON_WRITE
2195 PL_nrs = prog->saved_copy;
2197 RX_MATCH_COPIED_off(prog);
2200 PL_reg_oldsaved = NULL;
2201 prog->subbeg = PL_bostr;
2202 prog->sublen = PL_regeol - PL_bostr; /* strend may have been modified */
2204 DEBUG_EXECUTE_r(PL_reg_starttry = *startpos);
2205 prog->startp[0] = *startpos - PL_bostr;
2206 PL_reginput = *startpos;
2207 PL_reglastparen = &prog->lastparen;
2208 PL_reglastcloseparen = &prog->lastcloseparen;
2209 prog->lastparen = 0;
2210 prog->lastcloseparen = 0;
2212 PL_regstartp = prog->startp;
2213 PL_regendp = prog->endp;
2214 if (PL_reg_start_tmpl <= prog->nparens) {
2215 PL_reg_start_tmpl = prog->nparens*3/2 + 3;
2216 if(PL_reg_start_tmp)
2217 Renew(PL_reg_start_tmp, PL_reg_start_tmpl, char*);
2219 Newx(PL_reg_start_tmp, PL_reg_start_tmpl, char*);
2222 /* XXXX What this code is doing here?!!! There should be no need
2223 to do this again and again, PL_reglastparen should take care of
2226 /* Tests pat.t#187 and split.t#{13,14} seem to depend on this code.
2227 * Actually, the code in regcppop() (which Ilya may be meaning by
2228 * PL_reglastparen), is not needed at all by the test suite
2229 * (op/regexp, op/pat, op/split), but that code is needed, oddly
2230 * enough, for building DynaLoader, or otherwise this
2231 * "Error: '*' not in typemap in DynaLoader.xs, line 164"
2232 * will happen. Meanwhile, this code *is* needed for the
2233 * above-mentioned test suite tests to succeed. The common theme
2234 * on those tests seems to be returning null fields from matches.
2239 if (prog->nparens) {
2241 for (i = prog->nparens; i > (I32)*PL_reglastparen; i--) {
2248 if (regmatch(reginfo, progi->program + 1)) {
2249 PL_regendp[0] = PL_reginput - PL_bostr;
2252 if (reginfo->cutpoint)
2253 *startpos= reginfo->cutpoint;
2254 REGCP_UNWIND(lastcp);
2259 #define sayYES goto yes
2260 #define sayNO goto no
2261 #define sayNO_SILENT goto no_silent
2263 /* we dont use STMT_START/END here because it leads to
2264 "unreachable code" warnings, which are bogus, but distracting. */
2265 #define CACHEsayNO \
2266 if (ST.cache_mask) \
2267 PL_reg_poscache[ST.cache_offset] |= ST.cache_mask; \
2270 /* this is used to determine how far from the left messages like
2271 'failed...' are printed. It should be set such that messages
2272 are inline with the regop output that created them.
2274 #define REPORT_CODE_OFF 32
2277 /* Make sure there is a test for this +1 options in re_tests */
2278 #define TRIE_INITAL_ACCEPT_BUFFLEN 4;
2280 #define CHRTEST_UNINIT -1001 /* c1/c2 haven't been calculated yet */
2281 #define CHRTEST_VOID -1000 /* the c1/c2 "next char" test should be skipped */
2283 #define SLAB_FIRST(s) (&(s)->states[0])
2284 #define SLAB_LAST(s) (&(s)->states[PERL_REGMATCH_SLAB_SLOTS-1])
2286 /* grab a new slab and return the first slot in it */
2288 STATIC regmatch_state *
2291 #if PERL_VERSION < 9
2294 regmatch_slab *s = PL_regmatch_slab->next;
2296 Newx(s, 1, regmatch_slab);
2297 s->prev = PL_regmatch_slab;
2299 PL_regmatch_slab->next = s;
2301 PL_regmatch_slab = s;
2302 return SLAB_FIRST(s);
2306 /* push a new state then goto it */
2308 #define PUSH_STATE_GOTO(state, node) \
2310 st->resume_state = state; \
2313 /* push a new state with success backtracking, then goto it */
2315 #define PUSH_YES_STATE_GOTO(state, node) \
2317 st->resume_state = state; \
2318 goto push_yes_state;
2324 regmatch() - main matching routine
2326 This is basically one big switch statement in a loop. We execute an op,
2327 set 'next' to point the next op, and continue. If we come to a point which
2328 we may need to backtrack to on failure such as (A|B|C), we push a
2329 backtrack state onto the backtrack stack. On failure, we pop the top
2330 state, and re-enter the loop at the state indicated. If there are no more
2331 states to pop, we return failure.
2333 Sometimes we also need to backtrack on success; for example /A+/, where
2334 after successfully matching one A, we need to go back and try to
2335 match another one; similarly for lookahead assertions: if the assertion
2336 completes successfully, we backtrack to the state just before the assertion
2337 and then carry on. In these cases, the pushed state is marked as
2338 'backtrack on success too'. This marking is in fact done by a chain of
2339 pointers, each pointing to the previous 'yes' state. On success, we pop to
2340 the nearest yes state, discarding any intermediate failure-only states.
2341 Sometimes a yes state is pushed just to force some cleanup code to be
2342 called at the end of a successful match or submatch; e.g. (??{$re}) uses
2343 it to free the inner regex.
2345 Note that failure backtracking rewinds the cursor position, while
2346 success backtracking leaves it alone.
2348 A pattern is complete when the END op is executed, while a subpattern
2349 such as (?=foo) is complete when the SUCCESS op is executed. Both of these
2350 ops trigger the "pop to last yes state if any, otherwise return true"
2353 A common convention in this function is to use A and B to refer to the two
2354 subpatterns (or to the first nodes thereof) in patterns like /A*B/: so A is
2355 the subpattern to be matched possibly multiple times, while B is the entire
2356 rest of the pattern. Variable and state names reflect this convention.
2358 The states in the main switch are the union of ops and failure/success of
2359 substates associated with with that op. For example, IFMATCH is the op
2360 that does lookahead assertions /(?=A)B/ and so the IFMATCH state means
2361 'execute IFMATCH'; while IFMATCH_A is a state saying that we have just
2362 successfully matched A and IFMATCH_A_fail is a state saying that we have
2363 just failed to match A. Resume states always come in pairs. The backtrack
2364 state we push is marked as 'IFMATCH_A', but when that is popped, we resume
2365 at IFMATCH_A or IFMATCH_A_fail, depending on whether we are backtracking
2366 on success or failure.
2368 The struct that holds a backtracking state is actually a big union, with
2369 one variant for each major type of op. The variable st points to the
2370 top-most backtrack struct. To make the code clearer, within each
2371 block of code we #define ST to alias the relevant union.
2373 Here's a concrete example of a (vastly oversimplified) IFMATCH
2379 #define ST st->u.ifmatch
2381 case IFMATCH: // we are executing the IFMATCH op, (?=A)B
2382 ST.foo = ...; // some state we wish to save
2384 // push a yes backtrack state with a resume value of
2385 // IFMATCH_A/IFMATCH_A_fail, then continue execution at the
2387 PUSH_YES_STATE_GOTO(IFMATCH_A, A);
2390 case IFMATCH_A: // we have successfully executed A; now continue with B
2392 bar = ST.foo; // do something with the preserved value
2395 case IFMATCH_A_fail: // A failed, so the assertion failed
2396 ...; // do some housekeeping, then ...
2397 sayNO; // propagate the failure
2404 For any old-timers reading this who are familiar with the old recursive
2405 approach, the code above is equivalent to:
2407 case IFMATCH: // we are executing the IFMATCH op, (?=A)B
2416 ...; // do some housekeeping, then ...
2417 sayNO; // propagate the failure
2420 The topmost backtrack state, pointed to by st, is usually free. If you
2421 want to claim it, populate any ST.foo fields in it with values you wish to
2422 save, then do one of
2424 PUSH_STATE_GOTO(resume_state, node);
2425 PUSH_YES_STATE_GOTO(resume_state, node);
2427 which sets that backtrack state's resume value to 'resume_state', pushes a
2428 new free entry to the top of the backtrack stack, then goes to 'node'.
2429 On backtracking, the free slot is popped, and the saved state becomes the
2430 new free state. An ST.foo field in this new top state can be temporarily
2431 accessed to retrieve values, but once the main loop is re-entered, it
2432 becomes available for reuse.
2434 Note that the depth of the backtrack stack constantly increases during the
2435 left-to-right execution of the pattern, rather than going up and down with
2436 the pattern nesting. For example the stack is at its maximum at Z at the
2437 end of the pattern, rather than at X in the following:
2439 /(((X)+)+)+....(Y)+....Z/
2441 The only exceptions to this are lookahead/behind assertions and the cut,
2442 (?>A), which pop all the backtrack states associated with A before
2445 Bascktrack state structs are allocated in slabs of about 4K in size.
2446 PL_regmatch_state and st always point to the currently active state,
2447 and PL_regmatch_slab points to the slab currently containing
2448 PL_regmatch_state. The first time regmatch() is called, the first slab is
2449 allocated, and is never freed until interpreter destruction. When the slab
2450 is full, a new one is allocated and chained to the end. At exit from
2451 regmatch(), slabs allocated since entry are freed.
2456 #define DEBUG_STATE_pp(pp) \
2458 DUMP_EXEC_POS(locinput, scan, do_utf8); \
2459 PerlIO_printf(Perl_debug_log, \
2460 " %*s"pp" %s%s%s%s%s\n", \
2462 reg_name[st->resume_state], \
2463 ((st==yes_state||st==mark_state) ? "[" : ""), \
2464 ((st==yes_state) ? "Y" : ""), \
2465 ((st==mark_state) ? "M" : ""), \
2466 ((st==yes_state||st==mark_state) ? "]" : "") \
2471 #define REG_NODE_NUM(x) ((x) ? (int)((x)-prog) : -1)
2476 S_debug_start_match(pTHX_ const regexp *prog, const bool do_utf8,
2477 const char *start, const char *end, const char *blurb)
2479 const bool utf8_pat= prog->extflags & RXf_UTF8 ? 1 : 0;
2483 RE_PV_QUOTED_DECL(s0, utf8_pat, PERL_DEBUG_PAD_ZERO(0),
2484 prog->precomp, prog->prelen, 60);
2486 RE_PV_QUOTED_DECL(s1, do_utf8, PERL_DEBUG_PAD_ZERO(1),
2487 start, end - start, 60);
2489 PerlIO_printf(Perl_debug_log,
2490 "%s%s REx%s %s against %s\n",
2491 PL_colors[4], blurb, PL_colors[5], s0, s1);
2493 if (do_utf8||utf8_pat)
2494 PerlIO_printf(Perl_debug_log, "UTF-8 %s%s%s...\n",
2495 utf8_pat ? "pattern" : "",
2496 utf8_pat && do_utf8 ? " and " : "",
2497 do_utf8 ? "string" : ""
2503 S_dump_exec_pos(pTHX_ const char *locinput,
2504 const regnode *scan,
2505 const char *loc_regeol,
2506 const char *loc_bostr,
2507 const char *loc_reg_starttry,
2510 const int docolor = *PL_colors[0] || *PL_colors[2] || *PL_colors[4];
2511 const int taill = (docolor ? 10 : 7); /* 3 chars for "> <" */
2512 int l = (loc_regeol - locinput) > taill ? taill : (loc_regeol - locinput);
2513 /* The part of the string before starttry has one color
2514 (pref0_len chars), between starttry and current
2515 position another one (pref_len - pref0_len chars),
2516 after the current position the third one.
2517 We assume that pref0_len <= pref_len, otherwise we
2518 decrease pref0_len. */
2519 int pref_len = (locinput - loc_bostr) > (5 + taill) - l
2520 ? (5 + taill) - l : locinput - loc_bostr;
2523 while (do_utf8 && UTF8_IS_CONTINUATION(*(U8*)(locinput - pref_len)))
2525 pref0_len = pref_len - (locinput - loc_reg_starttry);
2526 if (l + pref_len < (5 + taill) && l < loc_regeol - locinput)
2527 l = ( loc_regeol - locinput > (5 + taill) - pref_len
2528 ? (5 + taill) - pref_len : loc_regeol - locinput);
2529 while (do_utf8 && UTF8_IS_CONTINUATION(*(U8*)(locinput + l)))
2533 if (pref0_len > pref_len)
2534 pref0_len = pref_len;
2536 const int is_uni = (do_utf8 && OP(scan) != CANY) ? 1 : 0;
2538 RE_PV_COLOR_DECL(s0,len0,is_uni,PERL_DEBUG_PAD(0),
2539 (locinput - pref_len),pref0_len, 60, 4, 5);
2541 RE_PV_COLOR_DECL(s1,len1,is_uni,PERL_DEBUG_PAD(1),
2542 (locinput - pref_len + pref0_len),
2543 pref_len - pref0_len, 60, 2, 3);
2545 RE_PV_COLOR_DECL(s2,len2,is_uni,PERL_DEBUG_PAD(2),
2546 locinput, loc_regeol - locinput, 10, 0, 1);
2548 const STRLEN tlen=len0+len1+len2;
2549 PerlIO_printf(Perl_debug_log,
2550 "%4"IVdf" <%.*s%.*s%s%.*s>%*s|",
2551 (IV)(locinput - loc_bostr),
2554 (docolor ? "" : "> <"),
2556 (int)(tlen > 19 ? 0 : 19 - tlen),
2563 /* reg_check_named_buff_matched()
2564 * Checks to see if a named buffer has matched. The data array of
2565 * buffer numbers corresponding to the buffer is expected to reside
2566 * in the regexp->data->data array in the slot stored in the ARG() of
2567 * node involved. Note that this routine doesn't actually care about the
2568 * name, that information is not preserved from compilation to execution.
2569 * Returns the index of the leftmost defined buffer with the given name
2570 * or 0 if non of the buffers matched.
2573 S_reg_check_named_buff_matched(pTHX_ const regexp *rex, const regnode *scan) {
2575 RXi_GET_DECL(rex,rexi);
2576 SV *sv_dat=(SV*)rexi->data->data[ ARG( scan ) ];
2577 I32 *nums=(I32*)SvPVX(sv_dat);
2578 for ( n=0; n<SvIVX(sv_dat); n++ ) {
2579 if ((I32)*PL_reglastparen >= nums[n] &&
2580 PL_regendp[nums[n]] != -1)
2588 STATIC I32 /* 0 failure, 1 success */
2589 S_regmatch(pTHX_ regmatch_info *reginfo, regnode *prog)
2591 #if PERL_VERSION < 9
2595 register const bool do_utf8 = PL_reg_match_utf8;
2596 const U32 uniflags = UTF8_ALLOW_DEFAULT;
2598 regexp *rex = reginfo->prog;
2599 RXi_GET_DECL(rex,rexi);
2601 regmatch_slab *orig_slab;
2602 regmatch_state *orig_state;
2604 /* the current state. This is a cached copy of PL_regmatch_state */
2605 register regmatch_state *st;
2607 /* cache heavy used fields of st in registers */
2608 register regnode *scan;
2609 register regnode *next;
2610 register U32 n = 0; /* general value; init to avoid compiler warning */
2611 register I32 ln = 0; /* len or last; init to avoid compiler warning */
2612 register char *locinput = PL_reginput;
2613 register I32 nextchr; /* is always set to UCHARAT(locinput) */
2615 bool result = 0; /* return value of S_regmatch */
2616 int depth = 0; /* depth of backtrack stack */
2617 int nochange_depth = 0; /* depth of GOSUB recursion with nochange*/
2618 regmatch_state *yes_state = NULL; /* state to pop to on success of
2620 /* mark_state piggy backs on the yes_state logic so that when we unwind
2621 the stack on success we can update the mark_state as we go */
2622 regmatch_state *mark_state = NULL; /* last mark state we have seen */
2624 regmatch_state *cur_eval = NULL; /* most recent EVAL_AB state */
2625 struct regmatch_state *cur_curlyx = NULL; /* most recent curlyx */
2627 bool no_final = 0; /* prevent failure from backtracking? */
2628 bool do_cutgroup = 0; /* no_final only until next branch/trie entry */
2629 char *startpoint = PL_reginput;
2630 SV *popmark = NULL; /* are we looking for a mark? */
2631 SV *sv_commit = NULL; /* last mark name seen in failure */
2632 SV *sv_yes_mark = NULL; /* last mark name we have seen
2633 during a successfull match */
2634 U32 lastopen = 0; /* last open we saw */
2635 bool has_cutgroup = RX_HAS_CUTGROUP(rex) ? 1 : 0;
2638 /* these three flags are set by various ops to signal information to
2639 * the very next op. They have a useful lifetime of exactly one loop
2640 * iteration, and are not preserved or restored by state pushes/pops
2642 bool sw = 0; /* the condition value in (?(cond)a|b) */
2643 bool minmod = 0; /* the next "{n,m}" is a "{n,m}?" */
2644 int logical = 0; /* the following EVAL is:
2648 or the following IFMATCH/UNLESSM is:
2649 false: plain (?=foo)
2650 true: used as a condition: (?(?=foo))
2654 GET_RE_DEBUG_FLAGS_DECL;
2658 PerlIO_printf(Perl_debug_log,"regmatch start\n");
2660 /* on first ever call to regmatch, allocate first slab */
2661 if (!PL_regmatch_slab) {
2662 Newx(PL_regmatch_slab, 1, regmatch_slab);
2663 PL_regmatch_slab->prev = NULL;
2664 PL_regmatch_slab->next = NULL;
2665 PL_regmatch_state = SLAB_FIRST(PL_regmatch_slab);
2668 /* remember current high-water mark for exit */
2669 /* XXX this should be done with SAVE* instead */
2670 orig_slab = PL_regmatch_slab;
2671 orig_state = PL_regmatch_state;
2673 /* grab next free state slot */
2674 st = ++PL_regmatch_state;
2675 if (st > SLAB_LAST(PL_regmatch_slab))
2676 st = PL_regmatch_state = S_push_slab(aTHX);
2678 /* Note that nextchr is a byte even in UTF */
2679 nextchr = UCHARAT(locinput);
2681 while (scan != NULL) {
2684 SV * const prop = sv_newmortal();
2685 regnode *rnext=regnext(scan);
2686 DUMP_EXEC_POS( locinput, scan, do_utf8 );
2687 regprop(rex, prop, scan);
2689 PerlIO_printf(Perl_debug_log,
2690 "%3"IVdf":%*s%s(%"IVdf")\n",
2691 (IV)(scan - rexi->program), depth*2, "",
2693 (PL_regkind[OP(scan)] == END || !rnext) ?
2694 0 : (IV)(rnext - rexi->program));
2697 next = scan + NEXT_OFF(scan);
2700 state_num = OP(scan);
2703 switch (state_num) {
2705 if (locinput == PL_bostr)
2707 /* reginfo->till = reginfo->bol; */
2712 if (locinput == PL_bostr ||
2713 ((nextchr || locinput < PL_regeol) && locinput[-1] == '\n'))
2719 if (locinput == PL_bostr)
2723 if (locinput == reginfo->ganch)
2729 if ((nextchr || locinput < PL_regeol) && nextchr != '\n')
2734 if ((nextchr || locinput < PL_regeol) && nextchr != '\n')
2736 if (PL_regeol - locinput > 1)
2740 if (PL_regeol != locinput)
2744 if (!nextchr && locinput >= PL_regeol)
2747 locinput += PL_utf8skip[nextchr];
2748 if (locinput > PL_regeol)
2750 nextchr = UCHARAT(locinput);
2753 nextchr = UCHARAT(++locinput);
2756 if (!nextchr && locinput >= PL_regeol)
2758 nextchr = UCHARAT(++locinput);
2761 if ((!nextchr && locinput >= PL_regeol) || nextchr == '\n')
2764 locinput += PL_utf8skip[nextchr];
2765 if (locinput > PL_regeol)
2767 nextchr = UCHARAT(locinput);
2770 nextchr = UCHARAT(++locinput);
2774 #define ST st->u.trie
2776 /* In this case the charclass data is available inline so
2777 we can fail fast without a lot of extra overhead.
2779 if (scan->flags == EXACT || !do_utf8) {
2780 if(!ANYOF_BITMAP_TEST(scan, *locinput)) {
2782 PerlIO_printf(Perl_debug_log,
2783 "%*s %sfailed to match trie start class...%s\n",
2784 REPORT_CODE_OFF+depth*2, "", PL_colors[4], PL_colors[5])
2793 /* what type of TRIE am I? (utf8 makes this contextual) */
2794 const enum { trie_plain, trie_utf8, trie_utf8_fold }
2795 trie_type = do_utf8 ?
2796 (scan->flags == EXACT ? trie_utf8 : trie_utf8_fold)
2799 /* what trie are we using right now */
2800 reg_trie_data * const trie
2801 = (reg_trie_data*)rexi->data->data[ ARG( scan ) ];
2802 U32 state = trie->startstate;
2804 if (trie->bitmap && trie_type != trie_utf8_fold &&
2805 !TRIE_BITMAP_TEST(trie,*locinput)
2807 if (trie->states[ state ].wordnum) {
2809 PerlIO_printf(Perl_debug_log,
2810 "%*s %smatched empty string...%s\n",
2811 REPORT_CODE_OFF+depth*2, "", PL_colors[4], PL_colors[5])
2816 PerlIO_printf(Perl_debug_log,
2817 "%*s %sfailed to match trie start class...%s\n",
2818 REPORT_CODE_OFF+depth*2, "", PL_colors[4], PL_colors[5])
2825 U8 *uc = ( U8* )locinput;
2829 U8 *uscan = (U8*)NULL;
2831 SV *sv_accept_buff = NULL;
2832 U8 foldbuf[ UTF8_MAXBYTES_CASE + 1 ];
2834 ST.accepted = 0; /* how many accepting states we have seen */
2836 ST.jump = trie->jump;
2840 traverse the TRIE keeping track of all accepting states
2841 we transition through until we get to a failing node.
2844 while ( state && uc <= (U8*)PL_regeol ) {
2845 U32 base = trie->states[ state ].trans.base;
2848 /* We use charid to hold the wordnum as we don't use it
2849 for charid until after we have done the wordnum logic.
2850 We define an alias just so that the wordnum logic reads
2853 #define got_wordnum charid
2854 got_wordnum = trie->states[ state ].wordnum;
2856 if ( got_wordnum ) {
2857 if ( ! ST.accepted ) {
2860 bufflen = TRIE_INITAL_ACCEPT_BUFFLEN;
2861 sv_accept_buff=newSV(bufflen *
2862 sizeof(reg_trie_accepted) - 1);
2863 SvCUR_set(sv_accept_buff, 0);
2864 SvPOK_on(sv_accept_buff);
2865 sv_2mortal(sv_accept_buff);
2868 (reg_trie_accepted*)SvPV_nolen(sv_accept_buff );
2871 if (ST.accepted >= bufflen) {
2873 ST.accept_buff =(reg_trie_accepted*)
2874 SvGROW(sv_accept_buff,
2875 bufflen * sizeof(reg_trie_accepted));
2877 SvCUR_set(sv_accept_buff,SvCUR(sv_accept_buff)
2878 + sizeof(reg_trie_accepted));
2881 ST.accept_buff[ST.accepted].wordnum = got_wordnum;
2882 ST.accept_buff[ST.accepted].endpos = uc;
2884 } while (trie->nextword && (got_wordnum= trie->nextword[got_wordnum]));
2888 DEBUG_TRIE_EXECUTE_r({
2889 DUMP_EXEC_POS( (char *)uc, scan, do_utf8 );
2890 PerlIO_printf( Perl_debug_log,
2891 "%*s %sState: %4"UVxf" Accepted: %4"UVxf" ",
2892 2+depth * 2, "", PL_colors[4],
2893 (UV)state, (UV)ST.accepted );
2897 REXEC_TRIE_READ_CHAR(trie_type, trie, uc, uscan, len,
2898 uvc, charid, foldlen, foldbuf, uniflags);
2901 (base + charid > trie->uniquecharcount )
2902 && (base + charid - 1 - trie->uniquecharcount
2904 && trie->trans[base + charid - 1 -
2905 trie->uniquecharcount].check == state)
2907 state = trie->trans[base + charid - 1 -
2908 trie->uniquecharcount ].next;
2919 DEBUG_TRIE_EXECUTE_r(
2920 PerlIO_printf( Perl_debug_log,
2921 "Charid:%3x CP:%4"UVxf" After State: %4"UVxf"%s\n",
2922 charid, uvc, (UV)state, PL_colors[5] );
2929 PerlIO_printf( Perl_debug_log,
2930 "%*s %sgot %"IVdf" possible matches%s\n",
2931 REPORT_CODE_OFF + depth * 2, "",
2932 PL_colors[4], (IV)ST.accepted, PL_colors[5] );
2937 case TRIE_next_fail: /* we failed - try next alterative */
2942 if ( ST.accepted == 1 ) {
2943 /* only one choice left - just continue */
2945 reg_trie_data * const trie
2946 = (reg_trie_data*)rexi->data->data[ ARG(ST.me) ];
2947 SV ** const tmp = av_fetch( trie->words,
2948 ST.accept_buff[ 0 ].wordnum-1, 0 );
2949 SV *sv= tmp ? sv_newmortal() : NULL;
2951 PerlIO_printf( Perl_debug_log,
2952 "%*s %sonly one match left: #%d <%s>%s\n",
2953 REPORT_CODE_OFF+depth*2, "", PL_colors[4],
2954 ST.accept_buff[ 0 ].wordnum,
2955 tmp ? pv_pretty(sv, SvPV_nolen_const(*tmp), SvCUR(*tmp), 0,
2956 PL_colors[0], PL_colors[1],
2957 (SvUTF8(*tmp) ? PERL_PV_ESCAPE_UNI : 0)
2959 : "not compiled under -Dr",
2962 PL_reginput = (char *)ST.accept_buff[ 0 ].endpos;
2963 /* in this case we free tmps/leave before we call regmatch
2964 as we wont be using accept_buff again. */
2966 locinput = PL_reginput;
2967 nextchr = UCHARAT(locinput);
2968 if ( !ST.jump || !ST.jump[ST.accept_buff[0].wordnum])
2971 scan = ST.me + ST.jump[ST.accept_buff[0].wordnum];
2972 if (!has_cutgroup) {
2977 PUSH_YES_STATE_GOTO(TRIE_next, scan);
2980 continue; /* execute rest of RE */
2983 if (!ST.accepted-- ) {
2985 PerlIO_printf( Perl_debug_log,
2986 "%*s %sTRIE failed...%s\n",
2987 REPORT_CODE_OFF+depth*2, "",
2997 There are at least two accepting states left. Presumably
2998 the number of accepting states is going to be low,
2999 typically two. So we simply scan through to find the one
3000 with lowest wordnum. Once we find it, we swap the last
3001 state into its place and decrement the size. We then try to
3002 match the rest of the pattern at the point where the word
3003 ends. If we succeed, control just continues along the
3004 regex; if we fail we return here to try the next accepting
3011 for( cur = 1 ; cur <= ST.accepted ; cur++ ) {
3012 DEBUG_TRIE_EXECUTE_r(
3013 PerlIO_printf( Perl_debug_log,
3014 "%*s %sgot %"IVdf" (%d) as best, looking at %"IVdf" (%d)%s\n",
3015 REPORT_CODE_OFF + depth * 2, "", PL_colors[4],
3016 (IV)best, ST.accept_buff[ best ].wordnum, (IV)cur,
3017 ST.accept_buff[ cur ].wordnum, PL_colors[5] );
3020 if (ST.accept_buff[cur].wordnum <
3021 ST.accept_buff[best].wordnum)
3026 reg_trie_data * const trie
3027 = (reg_trie_data*)rexi->data->data[ ARG(ST.me) ];
3028 SV ** const tmp = av_fetch( trie->words,
3029 ST.accept_buff[ best ].wordnum - 1, 0 );
3030 regnode *nextop=(!ST.jump || !ST.jump[ST.accept_buff[best].wordnum]) ?
3032 ST.me + ST.jump[ST.accept_buff[best].wordnum];
3033 SV *sv= tmp ? sv_newmortal() : NULL;
3035 PerlIO_printf( Perl_debug_log,
3036 "%*s %strying alternation #%d <%s> at node #%d %s\n",
3037 REPORT_CODE_OFF+depth*2, "", PL_colors[4],
3038 ST.accept_buff[best].wordnum,
3039 tmp ? pv_pretty(sv, SvPV_nolen_const(*tmp), SvCUR(*tmp), 0,
3040 PL_colors[0], PL_colors[1],
3041 (SvUTF8(*tmp) ? PERL_PV_ESCAPE_UNI : 0)
3042 ) : "not compiled under -Dr",
3043 REG_NODE_NUM(nextop),
3047 if ( best<ST.accepted ) {
3048 reg_trie_accepted tmp = ST.accept_buff[ best ];
3049 ST.accept_buff[ best ] = ST.accept_buff[ ST.accepted ];
3050 ST.accept_buff[ ST.accepted ] = tmp;
3053 PL_reginput = (char *)ST.accept_buff[ best ].endpos;
3054 if ( !ST.jump || !ST.jump[ST.accept_buff[best].wordnum]) {
3058 scan = ST.me + ST.jump[ST.accept_buff[best].wordnum];
3062 PUSH_YES_STATE_GOTO(TRIE_next, scan);
3065 PUSH_STATE_GOTO(TRIE_next, scan);
3078 char *s = STRING(scan);
3080 if (do_utf8 != UTF) {
3081 /* The target and the pattern have differing utf8ness. */
3083 const char * const e = s + ln;
3086 /* The target is utf8, the pattern is not utf8. */
3091 if (NATIVE_TO_UNI(*(U8*)s) !=
3092 utf8n_to_uvuni((U8*)l, UTF8_MAXBYTES, &ulen,
3100 /* The target is not utf8, the pattern is utf8. */
3105 if (NATIVE_TO_UNI(*((U8*)l)) !=
3106 utf8n_to_uvuni((U8*)s, UTF8_MAXBYTES, &ulen,
3114 nextchr = UCHARAT(locinput);
3117 /* The target and the pattern have the same utf8ness. */
3118 /* Inline the first character, for speed. */
3119 if (UCHARAT(s) != nextchr)
3121 if (PL_regeol - locinput < ln)
3123 if (ln > 1 && memNE(s, locinput, ln))
3126 nextchr = UCHARAT(locinput);
3130 PL_reg_flags |= RF_tainted;
3133 char * const s = STRING(scan);
3136 if (do_utf8 || UTF) {
3137 /* Either target or the pattern are utf8. */
3138 const char * const l = locinput;
3139 char *e = PL_regeol;
3141 if (ibcmp_utf8(s, 0, ln, (bool)UTF,
3142 l, &e, 0, do_utf8)) {
3143 /* One more case for the sharp s:
3144 * pack("U0U*", 0xDF) =~ /ss/i,
3145 * the 0xC3 0x9F are the UTF-8
3146 * byte sequence for the U+00DF. */
3148 toLOWER(s[0]) == 's' &&
3150 toLOWER(s[1]) == 's' &&
3157 nextchr = UCHARAT(locinput);
3161 /* Neither the target and the pattern are utf8. */
3163 /* Inline the first character, for speed. */
3164 if (UCHARAT(s) != nextchr &&
3165 UCHARAT(s) != ((OP(scan) == EXACTF)
3166 ? PL_fold : PL_fold_locale)[nextchr])
3168 if (PL_regeol - locinput < ln)
3170 if (ln > 1 && (OP(scan) == EXACTF
3171 ? ibcmp(s, locinput, ln)
3172 : ibcmp_locale(s, locinput, ln)))
3175 nextchr = UCHARAT(locinput);
3180 STRLEN inclasslen = PL_regeol - locinput;
3182 if (!reginclass(rex, scan, (U8*)locinput, &inclasslen, do_utf8))
3184 if (locinput >= PL_regeol)
3186 locinput += inclasslen ? inclasslen : UTF8SKIP(locinput);
3187 nextchr = UCHARAT(locinput);
3192 nextchr = UCHARAT(locinput);
3193 if (!REGINCLASS(rex, scan, (U8*)locinput))
3195 if (!nextchr && locinput >= PL_regeol)
3197 nextchr = UCHARAT(++locinput);
3201 /* If we might have the case of the German sharp s
3202 * in a casefolding Unicode character class. */
3204 if (ANYOF_FOLD_SHARP_S(scan, locinput, PL_regeol)) {
3205 locinput += SHARP_S_SKIP;
3206 nextchr = UCHARAT(locinput);
3212 PL_reg_flags |= RF_tainted;
3218 LOAD_UTF8_CHARCLASS_ALNUM();
3219 if (!(OP(scan) == ALNUM
3220 ? (bool)swash_fetch(PL_utf8_alnum, (U8*)locinput, do_utf8)
3221 : isALNUM_LC_utf8((U8*)locinput)))
3225 locinput += PL_utf8skip[nextchr];
3226 nextchr = UCHARAT(locinput);
3229 if (!(OP(scan) == ALNUM
3230 ? isALNUM(nextchr) : isALNUM_LC(nextchr)))
3232 nextchr = UCHARAT(++locinput);
3235 PL_reg_flags |= RF_tainted;
3238 if (!nextchr && locinput >= PL_regeol)
3241 LOAD_UTF8_CHARCLASS_ALNUM();
3242 if (OP(scan) == NALNUM
3243 ? (bool)swash_fetch(PL_utf8_alnum, (U8*)locinput, do_utf8)
3244 : isALNUM_LC_utf8((U8*)locinput))
3248 locinput += PL_utf8skip[nextchr];
3249 nextchr = UCHARAT(locinput);
3252 if (OP(scan) == NALNUM
3253 ? isALNUM(nextchr) : isALNUM_LC(nextchr))
3255 nextchr = UCHARAT(++locinput);
3259 PL_reg_flags |= RF_tainted;
3263 /* was last char in word? */
3265 if (locinput == PL_bostr)
3268 const U8 * const r = reghop3((U8*)locinput, -1, (U8*)PL_bostr);
3270 ln = utf8n_to_uvchr(r, UTF8SKIP(r), 0, uniflags);
3272 if (OP(scan) == BOUND || OP(scan) == NBOUND) {
3273 ln = isALNUM_uni(ln);
3274 LOAD_UTF8_CHARCLASS_ALNUM();
3275 n = swash_fetch(PL_utf8_alnum, (U8*)locinput, do_utf8);
3278 ln = isALNUM_LC_uvchr(UNI_TO_NATIVE(ln));
3279 n = isALNUM_LC_utf8((U8*)locinput);
3283 ln = (locinput != PL_bostr) ?
3284 UCHARAT(locinput - 1) : '\n';
3285 if (OP(scan) == BOUND || OP(scan) == NBOUND) {
3287 n = isALNUM(nextchr);
3290 ln = isALNUM_LC(ln);
3291 n = isALNUM_LC(nextchr);
3294 if (((!ln) == (!n)) == (OP(scan) == BOUND ||
3295 OP(scan) == BOUNDL))
3299 PL_reg_flags |= RF_tainted;
3305 if (UTF8_IS_CONTINUED(nextchr)) {
3306 LOAD_UTF8_CHARCLASS_SPACE();
3307 if (!(OP(scan) == SPACE
3308 ? (bool)swash_fetch(PL_utf8_space, (U8*)locinput, do_utf8)
3309 : isSPACE_LC_utf8((U8*)locinput)))
3313 locinput += PL_utf8skip[nextchr];
3314 nextchr = UCHARAT(locinput);
3317 if (!(OP(scan) == SPACE
3318 ? isSPACE(nextchr) : isSPACE_LC(nextchr)))
3320 nextchr = UCHARAT(++locinput);
3323 if (!(OP(scan) == SPACE
3324 ? isSPACE(nextchr) : isSPACE_LC(nextchr)))
3326 nextchr = UCHARAT(++locinput);
3330 PL_reg_flags |= RF_tainted;
3333 if (!nextchr && locinput >= PL_regeol)
3336 LOAD_UTF8_CHARCLASS_SPACE();
3337 if (OP(scan) == NSPACE
3338 ? (bool)swash_fetch(PL_utf8_space, (U8*)locinput, do_utf8)
3339 : isSPACE_LC_utf8((U8*)locinput))
3343 locinput += PL_utf8skip[nextchr];
3344 nextchr = UCHARAT(locinput);
3347 if (OP(scan) == NSPACE
3348 ? isSPACE(nextchr) : isSPACE_LC(nextchr))
3350 nextchr = UCHARAT(++locinput);
3353 PL_reg_flags |= RF_tainted;
3359 LOAD_UTF8_CHARCLASS_DIGIT();
3360 if (!(OP(scan) == DIGIT
3361 ? (bool)swash_fetch(PL_utf8_digit, (U8*)locinput, do_utf8)
3362 : isDIGIT_LC_utf8((U8*)locinput)))
3366 locinput += PL_utf8skip[nextchr];
3367 nextchr = UCHARAT(locinput);
3370 if (!(OP(scan) == DIGIT
3371 ? isDIGIT(nextchr) : isDIGIT_LC(nextchr)))
3373 nextchr = UCHARAT(++locinput);
3376 PL_reg_flags |= RF_tainted;
3379 if (!nextchr && locinput >= PL_regeol)
3382 LOAD_UTF8_CHARCLASS_DIGIT();
3383 if (OP(scan) == NDIGIT
3384 ? (bool)swash_fetch(PL_utf8_digit, (U8*)locinput, do_utf8)
3385 : isDIGIT_LC_utf8((U8*)locinput))
3389 locinput += PL_utf8skip[nextchr];
3390 nextchr = UCHARAT(locinput);
3393 if (OP(scan) == NDIGIT
3394 ? isDIGIT(nextchr) : isDIGIT_LC(nextchr))
3396 nextchr = UCHARAT(++locinput);
3399 if (locinput >= PL_regeol)
3402 LOAD_UTF8_CHARCLASS_MARK();
3403 if (swash_fetch(PL_utf8_mark,(U8*)locinput, do_utf8))
3405 locinput += PL_utf8skip[nextchr];
3406 while (locinput < PL_regeol &&
3407 swash_fetch(PL_utf8_mark,(U8*)locinput, do_utf8))
3408 locinput += UTF8SKIP(locinput);
3409 if (locinput > PL_regeol)
3414 nextchr = UCHARAT(locinput);
3421 PL_reg_flags |= RF_tainted;
3426 n = reg_check_named_buff_matched(rex,scan);
3429 type = REF + ( type - NREF );
3436 PL_reg_flags |= RF_tainted;
3440 n = ARG(scan); /* which paren pair */
3443 ln = PL_regstartp[n];
3444 PL_reg_leftiter = PL_reg_maxiter; /* Void cache */
3445 if (*PL_reglastparen < n || ln == -1)
3446 sayNO; /* Do not match unless seen CLOSEn. */
3447 if (ln == PL_regendp[n])
3451 if (do_utf8 && type != REF) { /* REF can do byte comparison */
3453 const char *e = PL_bostr + PL_regendp[n];
3455 * Note that we can't do the "other character" lookup trick as
3456 * in the 8-bit case (no pun intended) because in Unicode we
3457 * have to map both upper and title case to lower case.
3461 STRLEN ulen1, ulen2;
3462 U8 tmpbuf1[UTF8_MAXBYTES_CASE+1];
3463 U8 tmpbuf2[UTF8_MAXBYTES_CASE+1];
3467 toLOWER_utf8((U8*)s, tmpbuf1, &ulen1);
3468 toLOWER_utf8((U8*)l, tmpbuf2, &ulen2);
3469 if (ulen1 != ulen2 || memNE((char *)tmpbuf1, (char *)tmpbuf2, ulen1))
3476 nextchr = UCHARAT(locinput);
3480 /* Inline the first character, for speed. */
3481 if (UCHARAT(s) != nextchr &&
3483 (UCHARAT(s) != (type == REFF
3484 ? PL_fold : PL_fold_locale)[nextchr])))
3486 ln = PL_regendp[n] - ln;
3487 if (locinput + ln > PL_regeol)
3489 if (ln > 1 && (type == REF
3490 ? memNE(s, locinput, ln)
3492 ? ibcmp(s, locinput, ln)
3493 : ibcmp_locale(s, locinput, ln))))
3496 nextchr = UCHARAT(locinput);
3506 #define ST st->u.eval
3510 regexp_internal *rei;
3511 regnode *startpoint;
3514 case GOSUB: /* /(...(?1))/ */
3515 if (cur_eval && cur_eval->locinput==locinput) {
3516 if (cur_eval->u.eval.close_paren == (U32)ARG(scan))
3517 Perl_croak(aTHX_ "Infinite recursion in regex");
3518 if ( ++nochange_depth > MAX_RECURSE_EVAL_NOCHANGE_DEPTH )
3520 "Pattern subroutine nesting without pos change"
3521 " exceeded limit in regex");
3527 (void)ReREFCNT_inc(rex);
3528 if (OP(scan)==GOSUB) {
3529 startpoint = scan + ARG2L(scan);
3530 ST.close_paren = ARG(scan);
3532 startpoint = rei->program+1;
3535 goto eval_recurse_doit;
3537 case EVAL: /* /(?{A})B/ /(??{A})B/ and /(?(?{A})X|Y)B/ */
3538 if (cur_eval && cur_eval->locinput==locinput) {
3539 if ( ++nochange_depth > MAX_RECURSE_EVAL_NOCHANGE_DEPTH )
3540 Perl_croak(aTHX_ "EVAL without pos change exceeded limit in regex");
3545 /* execute the code in the {...} */
3547 SV ** const before = SP;
3548 OP_4tree * const oop = PL_op;
3549 COP * const ocurcop = PL_curcop;
3553 PL_op = (OP_4tree*)rexi->data->data[n];
3554 DEBUG_STATE_r( PerlIO_printf(Perl_debug_log,
3555 " re_eval 0x%"UVxf"\n", PTR2UV(PL_op)) );
3556 PAD_SAVE_LOCAL(old_comppad, (PAD*)rexi->data->data[n + 2]);
3557 PL_regendp[0] = PL_reg_magic->mg_len = locinput - PL_bostr;
3559 CALLRUNOPS(aTHX); /* Scalar context. */
3562 ret = &PL_sv_undef; /* protect against empty (?{}) blocks. */
3569 PAD_RESTORE_LOCAL(old_comppad);
3570 PL_curcop = ocurcop;
3573 sv_setsv(save_scalar(PL_replgv), ret);
3577 if (logical == 2) { /* Postponed subexpression: /(??{...})/ */
3580 /* extract RE object from returned value; compiling if
3585 if(SvROK(ret) && SvSMAGICAL(sv = SvRV(ret)))
3586 mg = mg_find(sv, PERL_MAGIC_qr);
3587 else if (SvSMAGICAL(ret)) {
3588 if (SvGMAGICAL(ret))
3589 sv_unmagic(ret, PERL_MAGIC_qr);
3591 mg = mg_find(ret, PERL_MAGIC_qr);
3595 re = (regexp *)mg->mg_obj;
3596 (void)ReREFCNT_inc(re);
3600 const char * const t = SvPV_const(ret, len);
3602 const I32 osize = PL_regsize;
3605 if (DO_UTF8(ret)) pm.op_pmdynflags |= PMdf_DYN_UTF8;
3606 re = CALLREGCOMP((char*)t, (char*)t + len, &pm);
3608 & (SVs_TEMP | SVs_PADTMP | SVf_READONLY
3610 sv_magic(ret,(SV*)ReREFCNT_inc(re),
3617 debug_start_match(re, do_utf8, locinput, PL_regeol,
3618 "Matching embedded");
3620 startpoint = rei->program + 1;
3621 ST.close_paren = 0; /* only used for GOSUB */
3622 /* borrowed from regtry */
3623 if (PL_reg_start_tmpl <= re->nparens) {
3624 PL_reg_start_tmpl = re->nparens*3/2 + 3;
3625 if(PL_reg_start_tmp)
3626 Renew(PL_reg_start_tmp, PL_reg_start_tmpl, char*);
3628 Newx(PL_reg_start_tmp, PL_reg_start_tmpl, char*);
3631 eval_recurse_doit: /* Share code with GOSUB below this line */
3632 /* run the pattern returned from (??{...}) */
3633 ST.cp = regcppush(0); /* Save *all* the positions. */
3634 REGCP_SET(ST.lastcp);
3636 PL_regstartp = re->startp; /* essentially NOOP on GOSUB */
3637 PL_regendp = re->endp; /* essentially NOOP on GOSUB */
3639 *PL_reglastparen = 0;
3640 *PL_reglastcloseparen = 0;
3641 PL_reginput = locinput;
3644 /* XXXX This is too dramatic a measure... */
3647 ST.toggle_reg_flags = PL_reg_flags;
3648 if (re->extflags & RXf_UTF8)
3649 PL_reg_flags |= RF_utf8;
3651 PL_reg_flags &= ~RF_utf8;
3652 ST.toggle_reg_flags ^= PL_reg_flags; /* diff of old and new */
3655 ST.prev_curlyx = cur_curlyx;
3660 ST.prev_eval = cur_eval;
3662 /* now continue from first node in postoned RE */
3663 PUSH_YES_STATE_GOTO(EVAL_AB, startpoint);
3666 /* logical is 1, /(?(?{...})X|Y)/ */
3667 sw = (bool)SvTRUE(ret);
3672 case EVAL_AB: /* cleanup after a successful (??{A})B */
3673 /* note: this is called twice; first after popping B, then A */
3674 PL_reg_flags ^= ST.toggle_reg_flags;
3677 rexi = RXi_GET(rex);
3679 cur_eval = ST.prev_eval;
3680 cur_curlyx = ST.prev_curlyx;
3681 /* XXXX This is too dramatic a measure... */
3686 case EVAL_AB_fail: /* unsuccessfully ran A or B in (??{A})B */
3687 /* note: this is called twice; first after popping B, then A */
3688 PL_reg_flags ^= ST.toggle_reg_flags;
3691 rexi = RXi_GET(rex);
3692 PL_reginput = locinput;
3693 REGCP_UNWIND(ST.lastcp);
3695 cur_eval = ST.prev_eval;
3696 cur_curlyx = ST.prev_curlyx;
3697 /* XXXX This is too dramatic a measure... */
3703 n = ARG(scan); /* which paren pair */
3704 PL_reg_start_tmp[n] = locinput;
3710 n = ARG(scan); /* which paren pair */
3711 PL_regstartp[n] = PL_reg_start_tmp[n] - PL_bostr;
3712 PL_regendp[n] = locinput - PL_bostr;
3713 /*if (n > PL_regsize)
3715 if (n > *PL_reglastparen)
3716 *PL_reglastparen = n;
3717 *PL_reglastcloseparen = n;
3718 if (cur_eval && cur_eval->u.eval.close_paren == n) {
3726 cursor && OP(cursor)!=END;
3727 cursor=regnext(cursor))
3729 if ( OP(cursor)==CLOSE ){
3731 if ( n <= lastopen ) {
3732 PL_regstartp[n] = PL_reg_start_tmp[n] - PL_bostr;
3733 PL_regendp[n] = locinput - PL_bostr;
3734 /*if (n > PL_regsize)
3736 if (n > *PL_reglastparen)
3737 *PL_reglastparen = n;
3738 *PL_reglastcloseparen = n;
3739 if ( n == ARG(scan) || (cur_eval &&
3740 cur_eval->u.eval.close_paren == n))
3749 n = ARG(scan); /* which paren pair */
3750 sw = (bool)(*PL_reglastparen >= n && PL_regendp[n] != -1);
3753 /* reg_check_named_buff_matched returns 0 for no match */
3754 sw = (bool)(0 < reg_check_named_buff_matched(rex,scan));
3758 sw = (cur_eval && (!n || cur_eval->u.eval.close_paren == n));
3764 PL_reg_leftiter = PL_reg_maxiter; /* Void cache */
3766 next = NEXTOPER(NEXTOPER(scan));
3768 next = scan + ARG(scan);
3769 if (OP(next) == IFTHEN) /* Fake one. */
3770 next = NEXTOPER(NEXTOPER(next));
3774 logical = scan->flags;
3777 /*******************************************************************
3779 The CURLYX/WHILEM pair of ops handle the most generic case of the /A*B/
3780 pattern, where A and B are subpatterns. (For simple A, CURLYM or
3781 STAR/PLUS/CURLY/CURLYN are used instead.)
3783 A*B is compiled as <CURLYX><A><WHILEM><B>
3785 On entry to the subpattern, CURLYX is called. This pushes a CURLYX
3786 state, which contains the current count, initialised to -1. It also sets
3787 cur_curlyx to point to this state, with any previous value saved in the
3790 CURLYX then jumps straight to the WHILEM op, rather than executing A,
3791 since the pattern may possibly match zero times (i.e. it's a while {} loop
3792 rather than a do {} while loop).
3794 Each entry to WHILEM represents a successful match of A. The count in the
3795 CURLYX block is incremented, another WHILEM state is pushed, and execution
3796 passes to A or B depending on greediness and the current count.
3798 For example, if matching against the string a1a2a3b (where the aN are
3799 substrings that match /A/), then the match progresses as follows: (the
3800 pushed states are interspersed with the bits of strings matched so far):
3803 <CURLYX cnt=0><WHILEM>
3804 <CURLYX cnt=1><WHILEM> a1 <WHILEM>
3805 <CURLYX cnt=2><WHILEM> a1 <WHILEM> a2 <WHILEM>
3806 <CURLYX cnt=3><WHILEM> a1 <WHILEM> a2 <WHILEM> a3 <WHILEM>
3807 <CURLYX cnt=3><WHILEM> a1 <WHILEM> a2 <WHILEM> a3 <WHILEM> b
3809 (Contrast this with something like CURLYM, which maintains only a single
3813 a1 <CURLYM cnt=1> a2
3814 a1 a2 <CURLYM cnt=2> a3
3815 a1 a2 a3 <CURLYM cnt=3> b
3818 Each WHILEM state block marks a point to backtrack to upon partial failure
3819 of A or B, and also contains some minor state data related to that
3820 iteration. The CURLYX block, pointed to by cur_curlyx, contains the
3821 overall state, such as the count, and pointers to the A and B ops.
3823 This is complicated slightly by nested CURLYX/WHILEM's. Since cur_curlyx
3824 must always point to the *current* CURLYX block, the rules are:
3826 When executing CURLYX, save the old cur_curlyx in the CURLYX state block,
3827 and set cur_curlyx to point the new block.
3829 When popping the CURLYX block after a successful or unsuccessful match,
3830 restore the previous cur_curlyx.
3832 When WHILEM is about to execute B, save the current cur_curlyx, and set it
3833 to the outer one saved in the CURLYX block.
3835 When popping the WHILEM block after a successful or unsuccessful B match,
3836 restore the previous cur_curlyx.
3838 Here's an example for the pattern (AI* BI)*BO
3839 I and O refer to inner and outer, C and W refer to CURLYX and WHILEM:
3842 curlyx backtrack stack
3843 ------ ---------------
3845 CO <CO prev=NULL> <WO>
3846 CI <CO prev=NULL> <WO> <CI prev=CO> <WI> ai
3847 CO <CO prev=NULL> <WO> <CI prev=CO> <WI> ai <WI prev=CI> bi
3848 NULL <CO prev=NULL> <WO> <CI prev=CO> <WI> ai <WI prev=CI> bi <WO prev=CO> bo
3850 At this point the pattern succeeds, and we work back down the stack to
3851 clean up, restoring as we go:
3853 CO <CO prev=NULL> <WO> <CI prev=CO> <WI> ai <WI prev=CI> bi
3854 CI <CO prev=NULL> <WO> <CI prev=CO> <WI> ai
3855 CO <CO prev=NULL> <WO>
3858 *******************************************************************/
3860 #define ST st->u.curlyx
3862 case CURLYX: /* start of /A*B/ (for complex A) */
3864 /* No need to save/restore up to this paren */
3865 I32 parenfloor = scan->flags;
3867 assert(next); /* keep Coverity happy */
3868 if (OP(PREVOPER(next)) == NOTHING) /* LONGJMP */
3871 /* XXXX Probably it is better to teach regpush to support
3872 parenfloor > PL_regsize... */
3873 if (parenfloor > (I32)*PL_reglastparen)
3874 parenfloor = *PL_reglastparen; /* Pessimization... */
3876 ST.prev_curlyx= cur_curlyx;
3878 ST.cp = PL_savestack_ix;
3880 /* these fields contain the state of the current curly.
3881 * they are accessed by subsequent WHILEMs */
3882 ST.parenfloor = parenfloor;
3883 ST.min = ARG1(scan);
3884 ST.max = ARG2(scan);
3885 ST.A = NEXTOPER(scan) + EXTRA_STEP_2ARGS;
3889 ST.count = -1; /* this will be updated by WHILEM */
3890 ST.lastloc = NULL; /* this will be updated by WHILEM */
3892 PL_reginput = locinput;
3893 PUSH_YES_STATE_GOTO(CURLYX_end, PREVOPER(next));
3897 case CURLYX_end: /* just finished matching all of A*B */
3898 if (PL_reg_eval_set){
3899 SV *pres= GvSV(PL_replgv);
3902 sv_setsv(GvSV(PL_replgv), pres);
3907 cur_curlyx = ST.prev_curlyx;
3911 case CURLYX_end_fail: /* just failed to match all of A*B */
3913 cur_curlyx = ST.prev_curlyx;
3919 #define ST st->u.whilem
3921 case WHILEM: /* just matched an A in /A*B/ (for complex A) */
3923 /* see the discussion above about CURLYX/WHILEM */
3925 assert(cur_curlyx); /* keep Coverity happy */
3926 n = ++cur_curlyx->u.curlyx.count; /* how many A's matched */
3927 ST.save_lastloc = cur_curlyx->u.curlyx.lastloc;
3928 ST.cache_offset = 0;
3931 PL_reginput = locinput;
3933 DEBUG_EXECUTE_r( PerlIO_printf(Perl_debug_log,
3934 "%*s whilem: matched %ld out of %ld..%ld\n",
3935 REPORT_CODE_OFF+depth*2, "", (long)n,
3936 (long)cur_curlyx->u.curlyx.min,
3937 (long)cur_curlyx->u.curlyx.max)
3940 /* First just match a string of min A's. */
3942 if (n < cur_curlyx->u.curlyx.min) {
3943 cur_curlyx->u.curlyx.lastloc = locinput;
3944 PUSH_STATE_GOTO(WHILEM_A_pre, cur_curlyx->u.curlyx.A);
3948 /* If degenerate A matches "", assume A done. */
3950 if (locinput == cur_curlyx->u.curlyx.lastloc) {
3951 DEBUG_EXECUTE_r( PerlIO_printf(Perl_debug_log,
3952 "%*s whilem: empty match detected, trying continuation...\n",
3953 REPORT_CODE_OFF+depth*2, "")
3955 goto do_whilem_B_max;
3958 /* super-linear cache processing */
3962 if (!PL_reg_maxiter) {
3963 /* start the countdown: Postpone detection until we
3964 * know the match is not *that* much linear. */
3965 PL_reg_maxiter = (PL_regeol - PL_bostr + 1) * (scan->flags>>4);
3966 /* possible overflow for long strings and many CURLYX's */
3967 if (PL_reg_maxiter < 0)
3968 PL_reg_maxiter = I32_MAX;
3969 PL_reg_leftiter = PL_reg_maxiter;
3972 if (PL_reg_leftiter-- == 0) {
3973 /* initialise cache */
3974 const I32 size = (PL_reg_maxiter + 7)/8;
3975 if (PL_reg_poscache) {
3976 if ((I32)PL_reg_poscache_size < size) {
3977 Renew(PL_reg_poscache, size, char);
3978 PL_reg_poscache_size = size;
3980 Zero(PL_reg_poscache, size, char);
3983 PL_reg_poscache_size = size;
3984 Newxz(PL_reg_poscache, size, char);
3986 DEBUG_EXECUTE_r( PerlIO_printf(Perl_debug_log,
3987 "%swhilem: Detected a super-linear match, switching on caching%s...\n",
3988 PL_colors[4], PL_colors[5])
3992 if (PL_reg_leftiter < 0) {
3993 /* have we already failed at this position? */
3995 offset = (scan->flags & 0xf) - 1
3996 + (locinput - PL_bostr) * (scan->flags>>4);
3997 mask = 1 << (offset % 8);
3999 if (PL_reg_poscache[offset] & mask) {
4000 DEBUG_EXECUTE_r( PerlIO_printf(Perl_debug_log,
4001 "%*s whilem: (cache) already tried at this position...\n",
4002 REPORT_CODE_OFF+depth*2, "")
4004 sayNO; /* cache records failure */
4006 ST.cache_offset = offset;
4007 ST.cache_mask = mask;
4011 /* Prefer B over A for minimal matching. */
4013 if (cur_curlyx->u.curlyx.minmod) {
4014 ST.save_curlyx = cur_curlyx;
4015 cur_curlyx = cur_curlyx->u.curlyx.prev_curlyx;
4016 ST.cp = regcppush(ST.save_curlyx->u.curlyx.parenfloor);
4017 REGCP_SET(ST.lastcp);
4018 PUSH_YES_STATE_GOTO(WHILEM_B_min, ST.save_curlyx->u.curlyx.B);
4022 /* Prefer A over B for maximal matching. */
4024 if (n < cur_curlyx->u.curlyx.max) { /* More greed allowed? */
4025 ST.cp = regcppush(cur_curlyx->u.curlyx.parenfloor);
4026 cur_curlyx->u.curlyx.lastloc = locinput;
4027 REGCP_SET(ST.lastcp);
4028 PUSH_STATE_GOTO(WHILEM_A_max, cur_curlyx->u.curlyx.A);
4031 goto do_whilem_B_max;
4035 case WHILEM_B_min: /* just matched B in a minimal match */
4036 case WHILEM_B_max: /* just matched B in a maximal match */
4037 cur_curlyx = ST.save_curlyx;
4041 case WHILEM_B_max_fail: /* just failed to match B in a maximal match */
4042 cur_curlyx = ST.save_curlyx;
4043 cur_curlyx->u.curlyx.lastloc = ST.save_lastloc;
4044 cur_curlyx->u.curlyx.count--;
4048 case WHILEM_A_min_fail: /* just failed to match A in a minimal match */
4049 REGCP_UNWIND(ST.lastcp);
4052 case WHILEM_A_pre_fail: /* just failed to match even minimal A */
4053 cur_curlyx->u.curlyx.lastloc = ST.save_lastloc;
4054 cur_curlyx->u.curlyx.count--;
4058 case WHILEM_A_max_fail: /* just failed to match A in a maximal match */
4059 REGCP_UNWIND(ST.lastcp);
4060 regcppop(rex); /* Restore some previous $<digit>s? */
4061 PL_reginput = locinput;
4062 DEBUG_EXECUTE_r(PerlIO_printf(Perl_debug_log,
4063 "%*s whilem: failed, trying continuation...\n",
4064 REPORT_CODE_OFF+depth*2, "")
4067 if (cur_curlyx->u.curlyx.count >= REG_INFTY
4068 && ckWARN(WARN_REGEXP)
4069 && !(PL_reg_flags & RF_warned))
4071 PL_reg_flags |= RF_warned;
4072 Perl_warner(aTHX_ packWARN(WARN_REGEXP), "%s limit (%d) exceeded",
4073 "Complex regular subexpression recursion",
4078 ST.save_curlyx = cur_curlyx;
4079 cur_curlyx = cur_curlyx->u.curlyx.prev_curlyx;
4080 PUSH_YES_STATE_GOTO(WHILEM_B_max, ST.save_curlyx->u.curlyx.B);
4083 case WHILEM_B_min_fail: /* just failed to match B in a minimal match */
4084 cur_curlyx = ST.save_curlyx;
4085 REGCP_UNWIND(ST.lastcp);
4088 if (cur_curlyx->u.curlyx.count >= cur_curlyx->u.curlyx.max) {
4089 /* Maximum greed exceeded */
4090 if (cur_curlyx->u.curlyx.count >= REG_INFTY
4091 && ckWARN(WARN_REGEXP)
4092 && !(PL_reg_flags & RF_warned))
4094 PL_reg_flags |= RF_warned;
4095 Perl_warner(aTHX_ packWARN(WARN_REGEXP),
4096 "%s limit (%d) exceeded",
4097 "Complex regular subexpression recursion",
4100 cur_curlyx->u.curlyx.count--;
4104 DEBUG_EXECUTE_r(PerlIO_printf(Perl_debug_log,
4105 "%*s trying longer...\n", REPORT_CODE_OFF+depth*2, "")
4107 /* Try grabbing another A and see if it helps. */
4108 PL_reginput = locinput;
4109 cur_curlyx->u.curlyx.lastloc = locinput;
4110 ST.cp = regcppush(cur_curlyx->u.curlyx.parenfloor);
4111 REGCP_SET(ST.lastcp);
4112 PUSH_STATE_GOTO(WHILEM_A_min, ST.save_curlyx->u.curlyx.A);
4116 #define ST st->u.branch
4118 case BRANCHJ: /* /(...|A|...)/ with long next pointer */
4119 next = scan + ARG(scan);
4122 scan = NEXTOPER(scan);
4125 case BRANCH: /* /(...|A|...)/ */
4126 scan = NEXTOPER(scan); /* scan now points to inner node */
4127 if ((!next || (OP(next) != BRANCH && OP(next) != BRANCHJ))
4130 /* last branch; skip state push and jump direct to node */
4133 ST.lastparen = *PL_reglastparen;
4134 ST.next_branch = next;
4136 PL_reginput = locinput;
4138 /* Now go into the branch */
4140 PUSH_YES_STATE_GOTO(BRANCH_next, scan);
4142 PUSH_STATE_GOTO(BRANCH_next, scan);
4146 PL_reginput = locinput;
4147 sv_yes_mark = st->u.mark.mark_name = scan->flags ? NULL :
4148 (SV*)rexi->data->data[ ARG( scan ) ];
4149 PUSH_STATE_GOTO(CUTGROUP_next,next);
4151 case CUTGROUP_next_fail:
4154 if (st->u.mark.mark_name)
4155 sv_commit = st->u.mark.mark_name;
4161 case BRANCH_next_fail: /* that branch failed; try the next, if any */
4166 REGCP_UNWIND(ST.cp);
4167 for (n = *PL_reglastparen; n > ST.lastparen; n--)
4169 *PL_reglastparen = n;
4170 /*dmq: *PL_reglastcloseparen = n; */
4171 scan = ST.next_branch;
4172 /* no more branches? */
4173 if (!scan || (OP(scan) != BRANCH && OP(scan) != BRANCHJ)) {
4175 PerlIO_printf( Perl_debug_log,
4176 "%*s %sBRANCH failed...%s\n",
4177 REPORT_CODE_OFF+depth*2, "",
4183 continue; /* execute next BRANCH[J] op */
4191 #define ST st->u.curlym
4193 case CURLYM: /* /A{m,n}B/ where A is fixed-length */
4195 /* This is an optimisation of CURLYX that enables us to push
4196 * only a single backtracking state, no matter now many matches
4197 * there are in {m,n}. It relies on the pattern being constant
4198 * length, with no parens to influence future backrefs
4202 scan = NEXTOPER(scan) + NODE_STEP_REGNODE;
4204 /* if paren positive, emulate an OPEN/CLOSE around A */
4206 U32 paren = ST.me->flags;
4207 if (paren > PL_regsize)
4209 if (paren > *PL_reglastparen)
4210 *PL_reglastparen = paren;
4211 scan += NEXT_OFF(scan); /* Skip former OPEN. */
4219 ST.c1 = CHRTEST_UNINIT;
4222 if (!(ST.minmod ? ARG1(ST.me) : ARG2(ST.me))) /* min/max */
4225 curlym_do_A: /* execute the A in /A{m,n}B/ */
4226 PL_reginput = locinput;
4227 PUSH_YES_STATE_GOTO(CURLYM_A, ST.A); /* match A */
4230 case CURLYM_A: /* we've just matched an A */
4231 locinput = st->locinput;
4232 nextchr = UCHARAT(locinput);
4235 /* after first match, determine A's length: u.curlym.alen */
4236 if (ST.count == 1) {
4237 if (PL_reg_match_utf8) {
4239 while (s < PL_reginput) {
4245 ST.alen = PL_reginput - locinput;
4248 ST.count = ST.minmod ? ARG1(ST.me) : ARG2(ST.me);
4251 PerlIO_printf(Perl_debug_log,
4252 "%*s CURLYM now matched %"IVdf" times, len=%"IVdf"...\n",
4253 (int)(REPORT_CODE_OFF+(depth*2)), "",
4254 (IV) ST.count, (IV)ST.alen)
4257 locinput = PL_reginput;
4259 if (cur_eval && cur_eval->u.eval.close_paren &&
4260 cur_eval->u.eval.close_paren == (U32)ST.me->flags)
4263 if ( ST.count < (ST.minmod ? ARG1(ST.me) : ARG2(ST.me)) )
4264 goto curlym_do_A; /* try to match another A */
4265 goto curlym_do_B; /* try to match B */
4267 case CURLYM_A_fail: /* just failed to match an A */
4268 REGCP_UNWIND(ST.cp);
4270 if (ST.minmod || ST.count < ARG1(ST.me) /* min*/
4271 || (cur_eval && cur_eval->u.eval.close_paren &&
4272 cur_eval->u.eval.close_paren == (U32)ST.me->flags))
4275 curlym_do_B: /* execute the B in /A{m,n}B/ */
4276 PL_reginput = locinput;
4277 if (ST.c1 == CHRTEST_UNINIT) {
4278 /* calculate c1 and c2 for possible match of 1st char
4279 * following curly */
4280 ST.c1 = ST.c2 = CHRTEST_VOID;
4281 if (HAS_TEXT(ST.B) || JUMPABLE(ST.B)) {
4282 regnode *text_node = ST.B;
4283 if (! HAS_TEXT(text_node))
4284 FIND_NEXT_IMPT(text_node);
4285 if (HAS_TEXT(text_node)
4286 && PL_regkind[OP(text_node)] != REF)
4288 ST.c1 = (U8)*STRING(text_node);
4290 (OP(text_node) == EXACTF || OP(text_node) == REFF)
4292 : (OP(text_node) == EXACTFL || OP(text_node) == REFFL)
4293 ? PL_fold_locale[ST.c1]
4300 PerlIO_printf(Perl_debug_log,
4301 "%*s CURLYM trying tail with matches=%"IVdf"...\n",
4302 (int)(REPORT_CODE_OFF+(depth*2)),
4305 if (ST.c1 != CHRTEST_VOID
4306 && UCHARAT(PL_reginput) != ST.c1
4307 && UCHARAT(PL_reginput) != ST.c2)
4309 /* simulate B failing */
4310 state_num = CURLYM_B_fail;
4311 goto reenter_switch;
4315 /* mark current A as captured */
4316 I32 paren = ST.me->flags;
4319 = HOPc(PL_reginput, -ST.alen) - PL_bostr;
4320 PL_regendp[paren] = PL_reginput - PL_bostr;
4321 /*dmq: *PL_reglastcloseparen = paren; */
4324 PL_regendp[paren] = -1;
4325 if (cur_eval && cur_eval->u.eval.close_paren &&
4326 cur_eval->u.eval.close_paren == (U32)ST.me->flags)
4335 PUSH_STATE_GOTO(CURLYM_B, ST.B); /* match B */
4338 case CURLYM_B_fail: /* just failed to match a B */
4339 REGCP_UNWIND(ST.cp);
4341 if (ST.count == ARG2(ST.me) /* max */)
4343 goto curlym_do_A; /* try to match a further A */
4345 /* backtrack one A */
4346 if (ST.count == ARG1(ST.me) /* min */)
4349 locinput = HOPc(locinput, -ST.alen);
4350 goto curlym_do_B; /* try to match B */
4353 #define ST st->u.curly
4355 #define CURLY_SETPAREN(paren, success) \
4358 PL_regstartp[paren] = HOPc(locinput, -1) - PL_bostr; \
4359 PL_regendp[paren] = locinput - PL_bostr; \
4360 *PL_reglastcloseparen = paren; \
4363 PL_regendp[paren] = -1; \
4366 case STAR: /* /A*B/ where A is width 1 */
4370 scan = NEXTOPER(scan);
4372 case PLUS: /* /A+B/ where A is width 1 */
4376 scan = NEXTOPER(scan);
4378 case CURLYN: /* /(A){m,n}B/ where A is width 1 */
4379 ST.paren = scan->flags; /* Which paren to set */
4380 if (ST.paren > PL_regsize)
4381 PL_regsize = ST.paren;
4382 if (ST.paren > *PL_reglastparen)
4383 *PL_reglastparen = ST.paren;
4384 ST.min = ARG1(scan); /* min to match */
4385 ST.max = ARG2(scan); /* max to match */
4386 if (cur_eval && cur_eval->u.eval.close_paren &&
4387 cur_eval->u.eval.close_paren == (U32)ST.paren) {
4391 scan = regnext(NEXTOPER(scan) + NODE_STEP_REGNODE);
4393 case CURLY: /* /A{m,n}B/ where A is width 1 */
4395 ST.min = ARG1(scan); /* min to match */
4396 ST.max = ARG2(scan); /* max to match */
4397 scan = NEXTOPER(scan) + NODE_STEP_REGNODE;
4400 * Lookahead to avoid useless match attempts
4401 * when we know what character comes next.
4403 * Used to only do .*x and .*?x, but now it allows
4404 * for )'s, ('s and (?{ ... })'s to be in the way
4405 * of the quantifier and the EXACT-like node. -- japhy
4408 if (ST.min > ST.max) /* XXX make this a compile-time check? */
4410 if (HAS_TEXT(next) || JUMPABLE(next)) {
4412 regnode *text_node = next;
4414 if (! HAS_TEXT(text_node))
4415 FIND_NEXT_IMPT(text_node);
4417 if (! HAS_TEXT(text_node))
4418 ST.c1 = ST.c2 = CHRTEST_VOID;
4420 if (PL_regkind[OP(text_node)] == REF) {
4421 ST.c1 = ST.c2 = CHRTEST_VOID;
4422 goto assume_ok_easy;
4425 s = (U8*)STRING(text_node);
4429 if (OP(text_node) == EXACTF || OP(text_node) == REFF)
4430 ST.c2 = PL_fold[ST.c1];
4431 else if (OP(text_node) == EXACTFL || OP(text_node) == REFFL)
4432 ST.c2 = PL_fold_locale[ST.c1];
4435 if (OP(text_node) == EXACTF || OP(text_node) == REFF) {
4436 STRLEN ulen1, ulen2;
4437 U8 tmpbuf1[UTF8_MAXBYTES_CASE+1];
4438 U8 tmpbuf2[UTF8_MAXBYTES_CASE+1];
4440 to_utf8_lower((U8*)s, tmpbuf1, &ulen1);
4441 to_utf8_upper((U8*)s, tmpbuf2, &ulen2);
4443 ST.c1 = utf8n_to_uvchr(tmpbuf1, UTF8_MAXLEN, 0,
4445 0 : UTF8_ALLOW_ANY);
4446 ST.c2 = utf8n_to_uvchr(tmpbuf2, UTF8_MAXLEN, 0,
4448 0 : UTF8_ALLOW_ANY);
4450 ST.c1 = utf8n_to_uvuni(tmpbuf1, UTF8_MAXBYTES, 0,
4452 ST.c2 = utf8n_to_uvuni(tmpbuf2, UTF8_MAXBYTES, 0,
4457 ST.c2 = ST.c1 = utf8n_to_uvchr(s, UTF8_MAXBYTES, 0,
4464 ST.c1 = ST.c2 = CHRTEST_VOID;
4469 PL_reginput = locinput;
4472 if (ST.min && regrepeat(rex, ST.A, ST.min, depth) < ST.min)
4475 locinput = PL_reginput;
4477 if (ST.c1 == CHRTEST_VOID)
4478 goto curly_try_B_min;
4480 ST.oldloc = locinput;
4482 /* set ST.maxpos to the furthest point along the
4483 * string that could possibly match */
4484 if (ST.max == REG_INFTY) {
4485 ST.maxpos = PL_regeol - 1;
4487 while (UTF8_IS_CONTINUATION(*(U8*)ST.maxpos))
4491 int m = ST.max - ST.min;
4492 for (ST.maxpos = locinput;
4493 m >0 && ST.maxpos + UTF8SKIP(ST.maxpos) <= PL_regeol; m--)
4494 ST.maxpos += UTF8SKIP(ST.maxpos);
4497 ST.maxpos = locinput + ST.max - ST.min;
4498 if (ST.maxpos >= PL_regeol)
4499 ST.maxpos = PL_regeol - 1;
4501 goto curly_try_B_min_known;
4505 ST.count = regrepeat(rex, ST.A, ST.max, depth);
4506 locinput = PL_reginput;
4507 if (ST.count < ST.min)
4509 if ((ST.count > ST.min)
4510 && (PL_regkind[OP(ST.B)] == EOL) && (OP(ST.B) != MEOL))
4512 /* A{m,n} must come at the end of the string, there's
4513 * no point in backing off ... */
4515 /* ...except that $ and \Z can match before *and* after
4516 newline at the end. Consider "\n\n" =~ /\n+\Z\n/.
4517 We may back off by one in this case. */
4518 if (UCHARAT(PL_reginput - 1) == '\n' && OP(ST.B) != EOS)
4522 goto curly_try_B_max;
4527 case CURLY_B_min_known_fail:
4528 /* failed to find B in a non-greedy match where c1,c2 valid */
4529 if (ST.paren && ST.count)
4530 PL_regendp[ST.paren] = -1;
4532 PL_reginput = locinput; /* Could be reset... */
4533 REGCP_UNWIND(ST.cp);
4534 /* Couldn't or didn't -- move forward. */
4535 ST.oldloc = locinput;
4537 locinput += UTF8SKIP(locinput);
4541 curly_try_B_min_known:
4542 /* find the next place where 'B' could work, then call B */
4546 n = (ST.oldloc == locinput) ? 0 : 1;
4547 if (ST.c1 == ST.c2) {
4549 /* set n to utf8_distance(oldloc, locinput) */
4550 while (locinput <= ST.maxpos &&
4551 utf8n_to_uvchr((U8*)locinput,
4552 UTF8_MAXBYTES, &len,
4553 uniflags) != (UV)ST.c1) {
4559 /* set n to utf8_distance(oldloc, locinput) */
4560 while (locinput <= ST.maxpos) {
4562 const UV c = utf8n_to_uvchr((U8*)locinput,
4563 UTF8_MAXBYTES, &len,
4565 if (c == (UV)ST.c1 || c == (UV)ST.c2)
4573 if (ST.c1 == ST.c2) {
4574 while (locinput <= ST.maxpos &&
4575 UCHARAT(locinput) != ST.c1)
4579 while (locinput <= ST.maxpos
4580 && UCHARAT(locinput) != ST.c1
4581 && UCHARAT(locinput) != ST.c2)
4584 n = locinput - ST.oldloc;
4586 if (locinput > ST.maxpos)
4588 /* PL_reginput == oldloc now */
4591 if (regrepeat(rex, ST.A, n, depth) < n)
4594 PL_reginput = locinput;
4595 CURLY_SETPAREN(ST.paren, ST.count);
4596 if (cur_eval && cur_eval->u.eval.close_paren &&
4597 cur_eval->u.eval.close_paren == (U32)ST.paren) {
4600 PUSH_STATE_GOTO(CURLY_B_min_known, ST.B);
4605 case CURLY_B_min_fail:
4606 /* failed to find B in a non-greedy match where c1,c2 invalid */
4607 if (ST.paren && ST.count)
4608 PL_regendp[ST.paren] = -1;
4610 REGCP_UNWIND(ST.cp);
4611 /* failed -- move forward one */
4612 PL_reginput = locinput;
4613 if (regrepeat(rex, ST.A, 1, depth)) {
4615 locinput = PL_reginput;
4616 if (ST.count <= ST.max || (ST.max == REG_INFTY &&
4617 ST.count > 0)) /* count overflow ? */
4620 CURLY_SETPAREN(ST.paren, ST.count);
4621 if (cur_eval && cur_eval->u.eval.close_paren &&
4622 cur_eval->u.eval.close_paren == (U32)ST.paren) {
4625 PUSH_STATE_GOTO(CURLY_B_min, ST.B);
4633 /* a successful greedy match: now try to match B */
4634 if (cur_eval && cur_eval->u.eval.close_paren &&
4635 cur_eval->u.eval.close_paren == (U32)ST.paren) {
4640 if (ST.c1 != CHRTEST_VOID)
4641 c = do_utf8 ? utf8n_to_uvchr((U8*)PL_reginput,
4642 UTF8_MAXBYTES, 0, uniflags)
4643 : (UV) UCHARAT(PL_reginput);
4644 /* If it could work, try it. */
4645 if (ST.c1 == CHRTEST_VOID || c == (UV)ST.c1 || c == (UV)ST.c2) {
4646 CURLY_SETPAREN(ST.paren, ST.count);
4647 PUSH_STATE_GOTO(CURLY_B_max, ST.B);
4652 case CURLY_B_max_fail:
4653 /* failed to find B in a greedy match */
4654 if (ST.paren && ST.count)
4655 PL_regendp[ST.paren] = -1;
4657 REGCP_UNWIND(ST.cp);
4659 if (--ST.count < ST.min)
4661 PL_reginput = locinput = HOPc(locinput, -1);
4662 goto curly_try_B_max;
4669 /* we've just finished A in /(??{A})B/; now continue with B */
4673 st->u.eval.toggle_reg_flags
4674 = cur_eval->u.eval.toggle_reg_flags;
4675 PL_reg_flags ^= st->u.eval.toggle_reg_flags;
4677 st->u.eval.prev_rex = rex; /* inner */
4678 rex = cur_eval->u.eval.prev_rex; /* outer */
4679 rexi = RXi_GET(rex);
4680 cur_curlyx = cur_eval->u.eval.prev_curlyx;
4682 st->u.eval.cp = regcppush(0); /* Save *all* the positions. */
4683 REGCP_SET(st->u.eval.lastcp);
4684 PL_reginput = locinput;
4686 /* Restore parens of the outer rex without popping the
4688 tmpix = PL_savestack_ix;
4689 PL_savestack_ix = cur_eval->u.eval.lastcp;
4691 PL_savestack_ix = tmpix;
4693 st->u.eval.prev_eval = cur_eval;
4694 cur_eval = cur_eval->u.eval.prev_eval;
4696 PerlIO_printf(Perl_debug_log, "%*s EVAL trying tail ... %"UVxf"\n",
4697 REPORT_CODE_OFF+depth*2, "",PTR2UV(cur_eval)););
4698 PUSH_YES_STATE_GOTO(EVAL_AB,
4699 st->u.eval.prev_eval->u.eval.B); /* match B */
4702 if (locinput < reginfo->till) {
4703 DEBUG_EXECUTE_r(PerlIO_printf(Perl_debug_log,
4704 "%sMatch possible, but length=%ld is smaller than requested=%ld, failing!%s\n",
4706 (long)(locinput - PL_reg_starttry),
4707 (long)(reginfo->till - PL_reg_starttry),
4710 sayNO_SILENT; /* Cannot match: too short. */
4712 PL_reginput = locinput; /* put where regtry can find it */
4713 sayYES; /* Success! */
4715 case SUCCEED: /* successful SUSPEND/UNLESSM/IFMATCH/CURLYM */
4717 PerlIO_printf(Perl_debug_log,
4718 "%*s %ssubpattern success...%s\n",
4719 REPORT_CODE_OFF+depth*2, "", PL_colors[4], PL_colors[5]));
4720 PL_reginput = locinput; /* put where regtry can find it */
4721 sayYES; /* Success! */
4724 #define ST st->u.ifmatch
4726 case SUSPEND: /* (?>A) */
4728 PL_reginput = locinput;
4731 case UNLESSM: /* -ve lookaround: (?!A), or with flags, (?<!A) */
4733 goto ifmatch_trivial_fail_test;
4735 case IFMATCH: /* +ve lookaround: (?=A), or with flags, (?<=A) */
4737 ifmatch_trivial_fail_test:
4739 char * const s = HOPBACKc(locinput, scan->flags);
4744 sw = 1 - (bool)ST.wanted;
4748 next = scan + ARG(scan);
4756 PL_reginput = locinput;
4760 ST.logical = logical;
4761 /* execute body of (?...A) */
4762 PUSH_YES_STATE_GOTO(IFMATCH_A, NEXTOPER(NEXTOPER(scan)));
4765 case IFMATCH_A_fail: /* body of (?...A) failed */
4766 ST.wanted = !ST.wanted;
4769 case IFMATCH_A: /* body of (?...A) succeeded */
4771 sw = (bool)ST.wanted;
4773 else if (!ST.wanted)
4776 if (OP(ST.me) == SUSPEND)
4777 locinput = PL_reginput;
4779 locinput = PL_reginput = st->locinput;
4780 nextchr = UCHARAT(locinput);
4782 scan = ST.me + ARG(ST.me);
4785 continue; /* execute B */
4790 next = scan + ARG(scan);
4795 reginfo->cutpoint = PL_regeol;
4798 PL_reginput = locinput;
4800 sv_yes_mark = sv_commit = (SV*)rexi->data->data[ ARG( scan ) ];
4801 PUSH_STATE_GOTO(COMMIT_next,next);
4803 case COMMIT_next_fail:
4810 #define ST st->u.mark
4812 ST.prev_mark = mark_state;
4813 ST.mark_name = sv_commit = sv_yes_mark
4814 = (SV*)rexi->data->data[ ARG( scan ) ];
4816 ST.mark_loc = PL_reginput = locinput;
4817 PUSH_YES_STATE_GOTO(MARKPOINT_next,next);
4819 case MARKPOINT_next:
4820 mark_state = ST.prev_mark;
4823 case MARKPOINT_next_fail:
4824 if (popmark && sv_eq(ST.mark_name,popmark))
4826 if (ST.mark_loc > startpoint)
4827 reginfo->cutpoint = HOPBACKc(ST.mark_loc, 1);
4828 popmark = NULL; /* we found our mark */
4829 sv_commit = ST.mark_name;
4832 PerlIO_printf(Perl_debug_log,
4833 "%*s %ssetting cutpoint to mark:%"SVf"...%s\n",
4834 REPORT_CODE_OFF+depth*2, "",
4835 PL_colors[4], sv_commit, PL_colors[5]);
4838 mark_state = ST.prev_mark;
4839 sv_yes_mark = mark_state ?
4840 mark_state->u.mark.mark_name : NULL;
4844 PL_reginput = locinput;
4846 /* (*CUT) : if we fail we cut here*/
4847 ST.mark_name = NULL;
4848 ST.mark_loc = locinput;
4849 PUSH_STATE_GOTO(SKIP_next,next);
4851 /* (*CUT:NAME) : if there is a (*MARK:NAME) fail where it was,
4852 otherwise do nothing. Meaning we need to scan
4854 regmatch_state *cur = mark_state;
4855 SV *find = (SV*)rexi->data->data[ ARG( scan ) ];
4858 if ( sv_eq( cur->u.mark.mark_name,
4861 ST.mark_name = find;
4862 PUSH_STATE_GOTO( SKIP_next, next );
4864 cur = cur->u.mark.prev_mark;
4867 /* Didn't find our (*MARK:NAME) so ignore this (*CUT:NAME) */
4869 case SKIP_next_fail:
4871 /* (*CUT:NAME) - Set up to search for the name as we
4872 collapse the stack*/
4873 popmark = ST.mark_name;
4875 /* (*CUT) - No name, we cut here.*/
4876 if (ST.mark_loc > startpoint)
4877 reginfo->cutpoint = HOPBACKc(ST.mark_loc, 1);
4878 /* but we set sv_commit to latest mark_name if there
4879 is one so they can test to see how things lead to this
4882 sv_commit=mark_state->u.mark.mark_name;
4890 PerlIO_printf(Perl_error_log, "%"UVxf" %d\n",
4891 PTR2UV(scan), OP(scan));
4892 Perl_croak(aTHX_ "regexp memory corruption");
4896 /* switch break jumps here */
4897 scan = next; /* prepare to execute the next op and ... */
4898 continue; /* ... jump back to the top, reusing st */
4902 /* push a state that backtracks on success */
4903 st->u.yes.prev_yes_state = yes_state;
4907 /* push a new regex state, then continue at scan */
4909 regmatch_state *newst;
4912 regmatch_state *cur = st;
4913 regmatch_state *curyes = yes_state;
4915 regmatch_slab *slab = PL_regmatch_slab;
4916 for (;curd > -1;cur--,curd--) {
4917 if (cur < SLAB_FIRST(slab)) {
4919 cur = SLAB_LAST(slab);
4921 PerlIO_printf(Perl_error_log, "%*s#%-3d %-10s %s\n",
4922 REPORT_CODE_OFF + 2 + depth * 2,"",
4923 curd, reg_name[cur->resume_state],
4924 (curyes == cur) ? "yes" : ""
4927 curyes = cur->u.yes.prev_yes_state;
4930 DEBUG_STATE_pp("push")
4933 st->locinput = locinput;
4935 if (newst > SLAB_LAST(PL_regmatch_slab))
4936 newst = S_push_slab(aTHX);
4937 PL_regmatch_state = newst;
4939 locinput = PL_reginput;
4940 nextchr = UCHARAT(locinput);
4948 * We get here only if there's trouble -- normally "case END" is
4949 * the terminating point.
4951 Perl_croak(aTHX_ "corrupted regexp pointers");
4957 /* we have successfully completed a subexpression, but we must now
4958 * pop to the state marked by yes_state and continue from there */
4959 assert(st != yes_state);
4961 while (st != yes_state) {
4963 if (st < SLAB_FIRST(PL_regmatch_slab)) {
4964 PL_regmatch_slab = PL_regmatch_slab->prev;
4965 st = SLAB_LAST(PL_regmatch_slab);
4969 DEBUG_STATE_pp("pop (no final)");
4971 DEBUG_STATE_pp("pop (yes)");
4977 while (yes_state < SLAB_FIRST(PL_regmatch_slab)
4978 || yes_state > SLAB_LAST(PL_regmatch_slab))
4980 /* not in this slab, pop slab */
4981 depth -= (st - SLAB_FIRST(PL_regmatch_slab) + 1);
4982 PL_regmatch_slab = PL_regmatch_slab->prev;
4983 st = SLAB_LAST(PL_regmatch_slab);
4985 depth -= (st - yes_state);
4988 yes_state = st->u.yes.prev_yes_state;
4989 PL_regmatch_state = st;
4992 locinput= st->locinput;
4993 nextchr = UCHARAT(locinput);
4995 state_num = st->resume_state + no_final;
4996 goto reenter_switch;
4999 DEBUG_EXECUTE_r(PerlIO_printf(Perl_debug_log, "%sMatch successful!%s\n",
5000 PL_colors[4], PL_colors[5]));
5007 PerlIO_printf(Perl_debug_log,
5008 "%*s %sfailed...%s\n",
5009 REPORT_CODE_OFF+depth*2, "",
5010 PL_colors[4], PL_colors[5])
5022 /* there's a previous state to backtrack to */
5024 if (st < SLAB_FIRST(PL_regmatch_slab)) {
5025 PL_regmatch_slab = PL_regmatch_slab->prev;
5026 st = SLAB_LAST(PL_regmatch_slab);
5028 PL_regmatch_state = st;
5029 locinput= st->locinput;
5030 nextchr = UCHARAT(locinput);
5032 DEBUG_STATE_pp("pop");
5034 if (yes_state == st)
5035 yes_state = st->u.yes.prev_yes_state;
5037 state_num = st->resume_state + 1; /* failure = success + 1 */
5038 goto reenter_switch;
5043 if (rex->intflags & PREGf_VERBARG_SEEN) {
5044 SV *sv_err = get_sv("REGERROR", 1);
5045 SV *sv_mrk = get_sv("REGMARK", 1);
5047 sv_commit = &PL_sv_no;
5049 sv_yes_mark = &PL_sv_yes;
5052 sv_commit = &PL_sv_yes;
5053 sv_yes_mark = &PL_sv_no;
5055 sv_setsv(sv_err, sv_commit);
5056 sv_setsv(sv_mrk, sv_yes_mark);
5058 /* restore original high-water mark */
5059 PL_regmatch_slab = orig_slab;
5060 PL_regmatch_state = orig_state;
5062 /* free all slabs above current one */
5063 if (orig_slab->next) {
5064 regmatch_slab *sl = orig_slab->next;
5065 orig_slab->next = NULL;
5067 regmatch_slab * const osl = sl;
5077 - regrepeat - repeatedly match something simple, report how many
5080 * [This routine now assumes that it will only match on things of length 1.
5081 * That was true before, but now we assume scan - reginput is the count,
5082 * rather than incrementing count on every character. [Er, except utf8.]]
5085 S_regrepeat(pTHX_ const regexp *prog, const regnode *p, I32 max, int depth)
5088 register char *scan;
5090 register char *loceol = PL_regeol;
5091 register I32 hardcount = 0;
5092 register bool do_utf8 = PL_reg_match_utf8;
5095 if (max == REG_INFTY)
5097 else if (max < loceol - scan)
5098 loceol = scan + max;
5103 while (scan < loceol && hardcount < max && *scan != '\n') {
5104 scan += UTF8SKIP(scan);
5108 while (scan < loceol && *scan != '\n')
5115 while (scan < loceol && hardcount < max) {
5116 scan += UTF8SKIP(scan);
5126 case EXACT: /* length of string is 1 */
5128 while (scan < loceol && UCHARAT(scan) == c)
5131 case EXACTF: /* length of string is 1 */
5133 while (scan < loceol &&
5134 (UCHARAT(scan) == c || UCHARAT(scan) == PL_fold[c]))
5137 case EXACTFL: /* length of string is 1 */
5138 PL_reg_flags |= RF_tainted;
5140 while (scan < loceol &&
5141 (UCHARAT(scan) == c || UCHARAT(scan) == PL_fold_locale[c]))
5147 while (hardcount < max && scan < loceol &&
5148 reginclass(prog, p, (U8*)scan, 0, do_utf8)) {
5149 scan += UTF8SKIP(scan);
5153 while (scan < loceol && REGINCLASS(prog, p, (U8*)scan))
5160 LOAD_UTF8_CHARCLASS_ALNUM();
5161 while (hardcount < max && scan < loceol &&
5162 swash_fetch(PL_utf8_alnum, (U8*)scan, do_utf8)) {
5163 scan += UTF8SKIP(scan);
5167 while (scan < loceol && isALNUM(*scan))
5172 PL_reg_flags |= RF_tainted;
5175 while (hardcount < max && scan < loceol &&
5176 isALNUM_LC_utf8((U8*)scan)) {
5177 scan += UTF8SKIP(scan);
5181 while (scan < loceol && isALNUM_LC(*scan))
5188 LOAD_UTF8_CHARCLASS_ALNUM();
5189 while (hardcount < max && scan < loceol &&
5190 !swash_fetch(PL_utf8_alnum, (U8*)scan, do_utf8)) {
5191 scan += UTF8SKIP(scan);
5195 while (scan < loceol && !isALNUM(*scan))
5200 PL_reg_flags |= RF_tainted;
5203 while (hardcount < max && scan < loceol &&
5204 !isALNUM_LC_utf8((U8*)scan)) {
5205 scan += UTF8SKIP(scan);
5209 while (scan < loceol && !isALNUM_LC(*scan))
5216 LOAD_UTF8_CHARCLASS_SPACE();
5217 while (hardcount < max && scan < loceol &&
5219 swash_fetch(PL_utf8_space,(U8*)scan, do_utf8))) {
5220 scan += UTF8SKIP(scan);
5224 while (scan < loceol && isSPACE(*scan))
5229 PL_reg_flags |= RF_tainted;
5232 while (hardcount < max && scan < loceol &&
5233 (*scan == ' ' || isSPACE_LC_utf8((U8*)scan))) {
5234 scan += UTF8SKIP(scan);
5238 while (scan < loceol && isSPACE_LC(*scan))
5245 LOAD_UTF8_CHARCLASS_SPACE();
5246 while (hardcount < max && scan < loceol &&
5248 swash_fetch(PL_utf8_space,(U8*)scan, do_utf8))) {
5249 scan += UTF8SKIP(scan);
5253 while (scan < loceol && !isSPACE(*scan))
5258 PL_reg_flags |= RF_tainted;
5261 while (hardcount < max && scan < loceol &&
5262 !(*scan == ' ' || isSPACE_LC_utf8((U8*)scan))) {
5263 scan += UTF8SKIP(scan);
5267 while (scan < loceol && !isSPACE_LC(*scan))
5274 LOAD_UTF8_CHARCLASS_DIGIT();
5275 while (hardcount < max && scan < loceol &&
5276 swash_fetch(PL_utf8_digit, (U8*)scan, do_utf8)) {
5277 scan += UTF8SKIP(scan);
5281 while (scan < loceol && isDIGIT(*scan))
5288 LOAD_UTF8_CHARCLASS_DIGIT();
5289 while (hardcount < max && scan < loceol &&
5290 !swash_fetch(PL_utf8_digit, (U8*)scan, do_utf8)) {
5291 scan += UTF8SKIP(scan);
5295 while (scan < loceol && !isDIGIT(*scan))
5299 default: /* Called on something of 0 width. */
5300 break; /* So match right here or not at all. */
5306 c = scan - PL_reginput;
5310 GET_RE_DEBUG_FLAGS_DECL;
5312 SV * const prop = sv_newmortal();
5313 regprop(prog, prop, p);
5314 PerlIO_printf(Perl_debug_log,
5315 "%*s %s can match %"IVdf" times out of %"IVdf"...\n",
5316 REPORT_CODE_OFF + depth*2, "", SvPVX_const(prop),(IV)c,(IV)max);
5324 #if !defined(PERL_IN_XSUB_RE) || defined(PLUGGABLE_RE_EXTENSION)
5326 - regclass_swash - prepare the utf8 swash
5330 Perl_regclass_swash(pTHX_ const regexp *prog, register const regnode* node, bool doinit, SV** listsvp, SV **altsvp)
5336 RXi_GET_DECL(prog,progi);
5337 const struct reg_data * const data = prog ? progi->data : NULL;
5339 if (data && data->count) {
5340 const U32 n = ARG(node);
5342 if (data->what[n] == 's') {
5343 SV * const rv = (SV*)data->data[n];
5344 AV * const av = (AV*)SvRV((SV*)rv);
5345 SV **const ary = AvARRAY(av);
5348 /* See the end of regcomp.c:S_regclass() for
5349 * documentation of these array elements. */
5352 a = SvROK(ary[1]) ? &ary[1] : 0;
5353 b = SvTYPE(ary[2]) == SVt_PVAV ? &ary[2] : 0;
5357 else if (si && doinit) {
5358 sw = swash_init("utf8", "", si, 1, 0);
5359 (void)av_store(av, 1, sw);
5376 - reginclass - determine if a character falls into a character class
5378 The n is the ANYOF regnode, the p is the target string, lenp
5379 is pointer to the maximum length of how far to go in the p
5380 (if the lenp is zero, UTF8SKIP(p) is used),
5381 do_utf8 tells whether the target string is in UTF-8.
5386 S_reginclass(pTHX_ const regexp *prog, register const regnode *n, register const U8* p, STRLEN* lenp, register bool do_utf8)
5389 const char flags = ANYOF_FLAGS(n);
5395 if (do_utf8 && !UTF8_IS_INVARIANT(c)) {
5396 c = utf8n_to_uvchr(p, UTF8_MAXBYTES, &len,
5397 (UTF8_ALLOW_DEFAULT & UTF8_ALLOW_ANYUV) | UTF8_CHECK_ONLY);
5398 /* see [perl #37836] for UTF8_ALLOW_ANYUV */
5399 if (len == (STRLEN)-1)
5400 Perl_croak(aTHX_ "Malformed UTF-8 character (fatal)");
5403 plen = lenp ? *lenp : UNISKIP(NATIVE_TO_UNI(c));
5404 if (do_utf8 || (flags & ANYOF_UNICODE)) {
5407 if (do_utf8 && !ANYOF_RUNTIME(n)) {
5408 if (len != (STRLEN)-1 && c < 256 && ANYOF_BITMAP_TEST(n, c))
5411 if (!match && do_utf8 && (flags & ANYOF_UNICODE_ALL) && c >= 256)
5415 SV * const sw = regclass_swash(prog, n, TRUE, 0, (SV**)&av);
5418 if (swash_fetch(sw, p, do_utf8))
5420 else if (flags & ANYOF_FOLD) {
5421 if (!match && lenp && av) {
5423 for (i = 0; i <= av_len(av); i++) {
5424 SV* const sv = *av_fetch(av, i, FALSE);
5426 const char * const s = SvPV_const(sv, len);
5428 if (len <= plen && memEQ(s, (char*)p, len)) {
5436 U8 tmpbuf[UTF8_MAXBYTES_CASE+1];
5439 to_utf8_fold(p, tmpbuf, &tmplen);
5440 if (swash_fetch(sw, tmpbuf, do_utf8))
5446 if (match && lenp && *lenp == 0)
5447 *lenp = UNISKIP(NATIVE_TO_UNI(c));
5449 if (!match && c < 256) {
5450 if (ANYOF_BITMAP_TEST(n, c))
5452 else if (flags & ANYOF_FOLD) {
5455 if (flags & ANYOF_LOCALE) {
5456 PL_reg_flags |= RF_tainted;
5457 f = PL_fold_locale[c];
5461 if (f != c && ANYOF_BITMAP_TEST(n, f))
5465 if (!match && (flags & ANYOF_CLASS)) {
5466 PL_reg_flags |= RF_tainted;
5468 (ANYOF_CLASS_TEST(n, ANYOF_ALNUM) && isALNUM_LC(c)) ||
5469 (ANYOF_CLASS_TEST(n, ANYOF_NALNUM) && !isALNUM_LC(c)) ||
5470 (ANYOF_CLASS_TEST(n, ANYOF_SPACE) && isSPACE_LC(c)) ||
5471 (ANYOF_CLASS_TEST(n, ANYOF_NSPACE) && !isSPACE_LC(c)) ||
5472 (ANYOF_CLASS_TEST(n, ANYOF_DIGIT) && isDIGIT_LC(c)) ||
5473 (ANYOF_CLASS_TEST(n, ANYOF_NDIGIT) && !isDIGIT_LC(c)) ||
5474 (ANYOF_CLASS_TEST(n, ANYOF_ALNUMC) && isALNUMC_LC(c)) ||
5475 (ANYOF_CLASS_TEST(n, ANYOF_NALNUMC) && !isALNUMC_LC(c)) ||
5476 (ANYOF_CLASS_TEST(n, ANYOF_ALPHA) && isALPHA_LC(c)) ||
5477 (ANYOF_CLASS_TEST(n, ANYOF_NALPHA) && !isALPHA_LC(c)) ||
5478 (ANYOF_CLASS_TEST(n, ANYOF_ASCII) && isASCII(c)) ||
5479 (ANYOF_CLASS_TEST(n, ANYOF_NASCII) && !isASCII(c)) ||
5480 (ANYOF_CLASS_TEST(n, ANYOF_CNTRL) && isCNTRL_LC(c)) ||
5481 (ANYOF_CLASS_TEST(n, ANYOF_NCNTRL) && !isCNTRL_LC(c)) ||
5482 (ANYOF_CLASS_TEST(n, ANYOF_GRAPH) && isGRAPH_LC(c)) ||
5483 (ANYOF_CLASS_TEST(n, ANYOF_NGRAPH) && !isGRAPH_LC(c)) ||
5484 (ANYOF_CLASS_TEST(n, ANYOF_LOWER) && isLOWER_LC(c)) ||
5485 (ANYOF_CLASS_TEST(n, ANYOF_NLOWER) && !isLOWER_LC(c)) ||
5486 (ANYOF_CLASS_TEST(n, ANYOF_PRINT) && isPRINT_LC(c)) ||
5487 (ANYOF_CLASS_TEST(n, ANYOF_NPRINT) && !isPRINT_LC(c)) ||
5488 (ANYOF_CLASS_TEST(n, ANYOF_PUNCT) && isPUNCT_LC(c)) ||
5489 (ANYOF_CLASS_TEST(n, ANYOF_NPUNCT) && !isPUNCT_LC(c)) ||
5490 (ANYOF_CLASS_TEST(n, ANYOF_UPPER) && isUPPER_LC(c)) ||
5491 (ANYOF_CLASS_TEST(n, ANYOF_NUPPER) && !isUPPER_LC(c)) ||
5492 (ANYOF_CLASS_TEST(n, ANYOF_XDIGIT) && isXDIGIT(c)) ||
5493 (ANYOF_CLASS_TEST(n, ANYOF_NXDIGIT) && !isXDIGIT(c)) ||
5494 (ANYOF_CLASS_TEST(n, ANYOF_PSXSPC) && isPSXSPC(c)) ||
5495 (ANYOF_CLASS_TEST(n, ANYOF_NPSXSPC) && !isPSXSPC(c)) ||
5496 (ANYOF_CLASS_TEST(n, ANYOF_BLANK) && isBLANK(c)) ||
5497 (ANYOF_CLASS_TEST(n, ANYOF_NBLANK) && !isBLANK(c))
5498 ) /* How's that for a conditional? */
5505 return (flags & ANYOF_INVERT) ? !match : match;
5509 S_reghop3(U8 *s, I32 off, const U8* lim)
5513 while (off-- && s < lim) {
5514 /* XXX could check well-formedness here */
5519 while (off++ && s > lim) {
5521 if (UTF8_IS_CONTINUED(*s)) {
5522 while (s > lim && UTF8_IS_CONTINUATION(*s))
5525 /* XXX could check well-formedness here */
5532 /* there are a bunch of places where we use two reghop3's that should
5533 be replaced with this routine. but since thats not done yet
5534 we ifdef it out - dmq
5537 S_reghop4(U8 *s, I32 off, const U8* llim, const U8* rlim)
5541 while (off-- && s < rlim) {
5542 /* XXX could check well-formedness here */
5547 while (off++ && s > llim) {
5549 if (UTF8_IS_CONTINUED(*s)) {
5550 while (s > llim && UTF8_IS_CONTINUATION(*s))
5553 /* XXX could check well-formedness here */
5561 S_reghopmaybe3(U8* s, I32 off, const U8* lim)
5565 while (off-- && s < lim) {
5566 /* XXX could check well-formedness here */
5573 while (off++ && s > lim) {
5575 if (UTF8_IS_CONTINUED(*s)) {
5576 while (s > lim && UTF8_IS_CONTINUATION(*s))
5579 /* XXX could check well-formedness here */
5588 restore_pos(pTHX_ void *arg)
5591 regexp * const rex = (regexp *)arg;
5592 if (PL_reg_eval_set) {
5593 if (PL_reg_oldsaved) {
5594 rex->subbeg = PL_reg_oldsaved;
5595 rex->sublen = PL_reg_oldsavedlen;
5596 #ifdef PERL_OLD_COPY_ON_WRITE
5597 rex->saved_copy = PL_nrs;
5599 RX_MATCH_COPIED_on(rex);
5601 PL_reg_magic->mg_len = PL_reg_oldpos;
5602 PL_reg_eval_set = 0;
5603 PL_curpm = PL_reg_oldcurpm;
5608 S_to_utf8_substr(pTHX_ register regexp *prog)
5610 if (prog->float_substr && !prog->float_utf8) {
5611 SV* const sv = newSVsv(prog->float_substr);
5612 prog->float_utf8 = sv;
5613 sv_utf8_upgrade(sv);
5614 if (SvTAIL(prog->float_substr))
5616 if (prog->float_substr == prog->check_substr)
5617 prog->check_utf8 = sv;
5619 if (prog->anchored_substr && !prog->anchored_utf8) {
5620 SV* const sv = newSVsv(prog->anchored_substr);
5621 prog->anchored_utf8 = sv;
5622 sv_utf8_upgrade(sv);
5623 if (SvTAIL(prog->anchored_substr))
5625 if (prog->anchored_substr == prog->check_substr)
5626 prog->check_utf8 = sv;
5631 S_to_byte_substr(pTHX_ register regexp *prog)
5634 if (prog->float_utf8 && !prog->float_substr) {
5635 SV* sv = newSVsv(prog->float_utf8);
5636 prog->float_substr = sv;
5637 if (sv_utf8_downgrade(sv, TRUE)) {
5638 if (SvTAIL(prog->float_utf8))
5642 prog->float_substr = sv = &PL_sv_undef;
5644 if (prog->float_utf8 == prog->check_utf8)
5645 prog->check_substr = sv;
5647 if (prog->anchored_utf8 && !prog->anchored_substr) {
5648 SV* sv = newSVsv(prog->anchored_utf8);
5649 prog->anchored_substr = sv;
5650 if (sv_utf8_downgrade(sv, TRUE)) {
5651 if (SvTAIL(prog->anchored_utf8))
5655 prog->anchored_substr = sv = &PL_sv_undef;
5657 if (prog->anchored_utf8 == prog->check_utf8)
5658 prog->check_substr = sv;
5664 * c-indentation-style: bsd
5666 * indent-tabs-mode: t
5669 * ex: set ts=8 sts=4 sw=4 noet: