5 * One Ring to rule them all, One Ring to find them
7 * [p.v of _The Lord of the Rings_, opening poem]
8 * [p.50 of _The Lord of the Rings_, I/iii: "The Shadow of the Past"]
9 * [p.254 of _The Lord of the Rings_, II/ii: "The Council of Elrond"]
12 /* This file contains functions for executing a regular expression. See
13 * also regcomp.c which funnily enough, contains functions for compiling
14 * a regular expression.
16 * This file is also copied at build time to ext/re/re_exec.c, where
17 * it's built with -DPERL_EXT_RE_BUILD -DPERL_EXT_RE_DEBUG -DPERL_EXT.
18 * This causes the main functions to be compiled under new names and with
19 * debugging support added, which makes "use re 'debug'" work.
22 /* NOTE: this is derived from Henry Spencer's regexp code, and should not
23 * confused with the original package (see point 3 below). Thanks, Henry!
26 /* Additional note: this code is very heavily munged from Henry's version
27 * in places. In some spots I've traded clarity for efficiency, so don't
28 * blame Henry for some of the lack of readability.
31 /* The names of the functions have been changed from regcomp and
32 * regexec to pregcomp and pregexec in order to avoid conflicts
33 * with the POSIX routines of the same names.
36 #ifdef PERL_EXT_RE_BUILD
41 * pregcomp and pregexec -- regsub and regerror are not used in perl
43 * Copyright (c) 1986 by University of Toronto.
44 * Written by Henry Spencer. Not derived from licensed software.
46 * Permission is granted to anyone to use this software for any
47 * purpose on any computer system, and to redistribute it freely,
48 * subject to the following restrictions:
50 * 1. The author is not responsible for the consequences of use of
51 * this software, no matter how awful, even if they arise
54 * 2. The origin of this software must not be misrepresented, either
55 * by explicit claim or by omission.
57 * 3. Altered versions must be plainly marked as such, and must not
58 * be misrepresented as being the original software.
60 **** Alterations to Henry's code are...
62 **** Copyright (C) 1991, 1992, 1993, 1994, 1995, 1996, 1997, 1998, 1999,
63 **** 2000, 2001, 2002, 2003, 2004, 2005, 2006, 2007, 2008
64 **** by Larry Wall and others
66 **** You may distribute under the terms of either the GNU General Public
67 **** License or the Artistic License, as specified in the README file.
69 * Beware that some of this code is subtly aware of the way operator
70 * precedence is structured in regular expressions. Serious changes in
71 * regular-expression syntax might require a total rethink.
74 #define PERL_IN_REGEXEC_C
77 #ifdef PERL_IN_XSUB_RE
83 #include "invlist_inline.h"
84 #include "unicode_constants.h"
86 #define B_ON_NON_UTF8_LOCALE_IS_WRONG \
87 "Use of \\b{} or \\B{} for non-UTF-8 locale is wrong. Assuming a UTF-8 locale"
89 static const char utf8_locale_required[] =
90 "Use of (?[ ]) for non-UTF-8 locale is wrong. Assuming a UTF-8 locale";
93 /* At least one required character in the target string is expressible only in
95 static const char* const non_utf8_target_but_utf8_required
96 = "Can't match, because target string needs to be in UTF-8\n";
99 #define NON_UTF8_TARGET_BUT_UTF8_REQUIRED(target) STMT_START { \
100 DEBUG_EXECUTE_r(PerlIO_printf(Perl_debug_log, "%s", non_utf8_target_but_utf8_required));\
104 #define HAS_NONLATIN1_FOLD_CLOSURE(i) _HAS_NONLATIN1_FOLD_CLOSURE_ONLY_FOR_USE_BY_REGCOMP_DOT_C_AND_REGEXEC_DOT_C(i)
107 #define STATIC static
110 /* Valid only for non-utf8 strings: avoids the reginclass
111 * call if there are no complications: i.e., if everything matchable is
112 * straight forward in the bitmap */
113 #define REGINCLASS(prog,p,c) (ANYOF_FLAGS(p) ? reginclass(prog,p,c,c+1,0) \
114 : ANYOF_BITMAP_TEST(p,*(c)))
120 #define CHR_SVLEN(sv) (utf8_target ? sv_len_utf8(sv) : SvCUR(sv))
121 #define CHR_DIST(a,b) (reginfo->is_utf8_target ? utf8_distance(a,b) : a - b)
123 #define HOPc(pos,off) \
124 (char *)(reginfo->is_utf8_target \
125 ? reghop3((U8*)pos, off, \
126 (U8*)(off >= 0 ? reginfo->strend : reginfo->strbeg)) \
129 #define HOPBACKc(pos, off) \
130 (char*)(reginfo->is_utf8_target \
131 ? reghopmaybe3((U8*)pos, -off, (U8*)(reginfo->strbeg)) \
132 : (pos - off >= reginfo->strbeg) \
136 #define HOP3(pos,off,lim) (reginfo->is_utf8_target ? reghop3((U8*)(pos), off, (U8*)(lim)) : (U8*)(pos + off))
137 #define HOP3c(pos,off,lim) ((char*)HOP3(pos,off,lim))
139 /* lim must be +ve. Returns NULL on overshoot */
140 #define HOPMAYBE3(pos,off,lim) \
141 (reginfo->is_utf8_target \
142 ? reghopmaybe3((U8*)pos, off, (U8*)(lim)) \
143 : ((U8*)pos + off <= lim) \
147 /* like HOP3, but limits the result to <= lim even for the non-utf8 case.
148 * off must be >=0; args should be vars rather than expressions */
149 #define HOP3lim(pos,off,lim) (reginfo->is_utf8_target \
150 ? reghop3((U8*)(pos), off, (U8*)(lim)) \
151 : (U8*)((pos + off) > lim ? lim : (pos + off)))
153 #define HOP4(pos,off,llim, rlim) (reginfo->is_utf8_target \
154 ? reghop4((U8*)(pos), off, (U8*)(llim), (U8*)(rlim)) \
156 #define HOP4c(pos,off,llim, rlim) ((char*)HOP4(pos,off,llim, rlim))
158 #define NEXTCHR_EOS -10 /* nextchr has fallen off the end */
159 #define NEXTCHR_IS_EOS (nextchr < 0)
161 #define SET_nextchr \
162 nextchr = ((locinput < reginfo->strend) ? UCHARAT(locinput) : NEXTCHR_EOS)
164 #define SET_locinput(p) \
169 #define LOAD_UTF8_CHARCLASS(swash_ptr, property_name, invlist) STMT_START { \
171 U8 flags = _CORE_SWASH_INIT_ACCEPT_INVLIST; \
172 swash_ptr = _core_swash_init("utf8", property_name, &PL_sv_undef, \
173 1, 0, invlist, &flags); \
178 /* If in debug mode, we test that a known character properly matches */
180 # define LOAD_UTF8_CHARCLASS_DEBUG_TEST(swash_ptr, \
183 utf8_char_in_property) \
184 LOAD_UTF8_CHARCLASS(swash_ptr, property_name, invlist); \
185 assert(swash_fetch(swash_ptr, (U8 *) utf8_char_in_property, TRUE));
187 # define LOAD_UTF8_CHARCLASS_DEBUG_TEST(swash_ptr, \
190 utf8_char_in_property) \
191 LOAD_UTF8_CHARCLASS(swash_ptr, property_name, invlist)
194 #define LOAD_UTF8_CHARCLASS_ALNUM() LOAD_UTF8_CHARCLASS_DEBUG_TEST( \
195 PL_utf8_swash_ptrs[_CC_WORDCHAR], \
197 PL_XPosix_ptrs[_CC_WORDCHAR], \
198 LATIN_SMALL_LIGATURE_LONG_S_T_UTF8);
200 #define PLACEHOLDER /* Something for the preprocessor to grab onto */
201 /* TODO: Combine JUMPABLE and HAS_TEXT to cache OP(rn) */
203 /* for use after a quantifier and before an EXACT-like node -- japhy */
204 /* it would be nice to rework regcomp.sym to generate this stuff. sigh
206 * NOTE that *nothing* that affects backtracking should be in here, specifically
207 * VERBS must NOT be included. JUMPABLE is used to determine if we can ignore a
208 * node that is in between two EXACT like nodes when ascertaining what the required
209 * "follow" character is. This should probably be moved to regex compile time
210 * although it may be done at run time beause of the REF possibility - more
211 * investigation required. -- demerphq
213 #define JUMPABLE(rn) ( \
215 (OP(rn) == CLOSE && (!cur_eval || cur_eval->u.eval.close_paren != ARG(rn))) || \
217 OP(rn) == SUSPEND || OP(rn) == IFMATCH || \
218 OP(rn) == PLUS || OP(rn) == MINMOD || \
220 (PL_regkind[OP(rn)] == CURLY && ARG1(rn) > 0) \
222 #define IS_EXACT(rn) (PL_regkind[OP(rn)] == EXACT)
224 #define HAS_TEXT(rn) ( IS_EXACT(rn) || PL_regkind[OP(rn)] == REF )
227 /* Currently these are only used when PL_regkind[OP(rn)] == EXACT so
228 we don't need this definition. XXX These are now out-of-sync*/
229 #define IS_TEXT(rn) ( OP(rn)==EXACT || OP(rn)==REF || OP(rn)==NREF )
230 #define IS_TEXTF(rn) ( OP(rn)==EXACTFU || OP(rn)==EXACTFU_SS || OP(rn)==EXACTFA || OP(rn)==EXACTFA_NO_TRIE || OP(rn)==EXACTF || OP(rn)==REFF || OP(rn)==NREFF )
231 #define IS_TEXTFL(rn) ( OP(rn)==EXACTFL || OP(rn)==REFFL || OP(rn)==NREFFL )
234 /* ... so we use this as its faster. */
235 #define IS_TEXT(rn) ( OP(rn)==EXACT || OP(rn)==EXACTL )
236 #define IS_TEXTFU(rn) ( OP(rn)==EXACTFU || OP(rn)==EXACTFLU8 || OP(rn)==EXACTFU_SS || OP(rn) == EXACTFA || OP(rn) == EXACTFA_NO_TRIE)
237 #define IS_TEXTF(rn) ( OP(rn)==EXACTF )
238 #define IS_TEXTFL(rn) ( OP(rn)==EXACTFL )
243 Search for mandatory following text node; for lookahead, the text must
244 follow but for lookbehind (rn->flags != 0) we skip to the next step.
246 #define FIND_NEXT_IMPT(rn) STMT_START { \
247 while (JUMPABLE(rn)) { \
248 const OPCODE type = OP(rn); \
249 if (type == SUSPEND || PL_regkind[type] == CURLY) \
250 rn = NEXTOPER(NEXTOPER(rn)); \
251 else if (type == PLUS) \
253 else if (type == IFMATCH) \
254 rn = (rn->flags == 0) ? NEXTOPER(NEXTOPER(rn)) : rn + ARG(rn); \
255 else rn += NEXT_OFF(rn); \
259 #define SLAB_FIRST(s) (&(s)->states[0])
260 #define SLAB_LAST(s) (&(s)->states[PERL_REGMATCH_SLAB_SLOTS-1])
262 static void S_setup_eval_state(pTHX_ regmatch_info *const reginfo);
263 static void S_cleanup_regmatch_info_aux(pTHX_ void *arg);
264 static regmatch_state * S_push_slab(pTHX);
266 #define REGCP_PAREN_ELEMS 3
267 #define REGCP_OTHER_ELEMS 3
268 #define REGCP_FRAME_ELEMS 1
269 /* REGCP_FRAME_ELEMS are not part of the REGCP_OTHER_ELEMS and
270 * are needed for the regexp context stack bookkeeping. */
273 S_regcppush(pTHX_ const regexp *rex, I32 parenfloor, U32 maxopenparen)
275 const int retval = PL_savestack_ix;
276 const int paren_elems_to_push =
277 (maxopenparen - parenfloor) * REGCP_PAREN_ELEMS;
278 const UV total_elems = paren_elems_to_push + REGCP_OTHER_ELEMS;
279 const UV elems_shifted = total_elems << SAVE_TIGHT_SHIFT;
281 GET_RE_DEBUG_FLAGS_DECL;
283 PERL_ARGS_ASSERT_REGCPPUSH;
285 if (paren_elems_to_push < 0)
286 Perl_croak(aTHX_ "panic: paren_elems_to_push, %i < 0, maxopenparen: %i parenfloor: %i REGCP_PAREN_ELEMS: %u",
287 (int)paren_elems_to_push, (int)maxopenparen,
288 (int)parenfloor, (unsigned)REGCP_PAREN_ELEMS);
290 if ((elems_shifted >> SAVE_TIGHT_SHIFT) != total_elems)
291 Perl_croak(aTHX_ "panic: paren_elems_to_push offset %"UVuf
292 " out of range (%lu-%ld)",
294 (unsigned long)maxopenparen,
297 SSGROW(total_elems + REGCP_FRAME_ELEMS);
300 if ((int)maxopenparen > (int)parenfloor)
301 PerlIO_printf(Perl_debug_log,
302 "rex=0x%"UVxf" offs=0x%"UVxf": saving capture indices:\n",
307 for (p = parenfloor+1; p <= (I32)maxopenparen; p++) {
308 /* REGCP_PARENS_ELEMS are pushed per pairs of parentheses. */
309 SSPUSHIV(rex->offs[p].end);
310 SSPUSHIV(rex->offs[p].start);
311 SSPUSHINT(rex->offs[p].start_tmp);
312 DEBUG_BUFFERS_r(PerlIO_printf(Perl_debug_log,
313 " \\%"UVuf": %"IVdf"(%"IVdf")..%"IVdf"\n",
315 (IV)rex->offs[p].start,
316 (IV)rex->offs[p].start_tmp,
320 /* REGCP_OTHER_ELEMS are pushed in any case, parentheses or no. */
321 SSPUSHINT(maxopenparen);
322 SSPUSHINT(rex->lastparen);
323 SSPUSHINT(rex->lastcloseparen);
324 SSPUSHUV(SAVEt_REGCONTEXT | elems_shifted); /* Magic cookie. */
329 /* These are needed since we do not localize EVAL nodes: */
330 #define REGCP_SET(cp) \
332 PerlIO_printf(Perl_debug_log, \
333 " Setting an EVAL scope, savestack=%"IVdf"\n", \
334 (IV)PL_savestack_ix)); \
337 #define REGCP_UNWIND(cp) \
339 if (cp != PL_savestack_ix) \
340 PerlIO_printf(Perl_debug_log, \
341 " Clearing an EVAL scope, savestack=%"IVdf"..%"IVdf"\n", \
342 (IV)(cp), (IV)PL_savestack_ix)); \
345 #define UNWIND_PAREN(lp, lcp) \
346 for (n = rex->lastparen; n > lp; n--) \
347 rex->offs[n].end = -1; \
348 rex->lastparen = n; \
349 rex->lastcloseparen = lcp;
353 S_regcppop(pTHX_ regexp *rex, U32 *maxopenparen_p)
357 GET_RE_DEBUG_FLAGS_DECL;
359 PERL_ARGS_ASSERT_REGCPPOP;
361 /* Pop REGCP_OTHER_ELEMS before the parentheses loop starts. */
363 assert((i & SAVE_MASK) == SAVEt_REGCONTEXT); /* Check that the magic cookie is there. */
364 i >>= SAVE_TIGHT_SHIFT; /* Parentheses elements to pop. */
365 rex->lastcloseparen = SSPOPINT;
366 rex->lastparen = SSPOPINT;
367 *maxopenparen_p = SSPOPINT;
369 i -= REGCP_OTHER_ELEMS;
370 /* Now restore the parentheses context. */
372 if (i || rex->lastparen + 1 <= rex->nparens)
373 PerlIO_printf(Perl_debug_log,
374 "rex=0x%"UVxf" offs=0x%"UVxf": restoring capture indices to:\n",
379 paren = *maxopenparen_p;
380 for ( ; i > 0; i -= REGCP_PAREN_ELEMS) {
382 rex->offs[paren].start_tmp = SSPOPINT;
383 rex->offs[paren].start = SSPOPIV;
385 if (paren <= rex->lastparen)
386 rex->offs[paren].end = tmps;
387 DEBUG_BUFFERS_r( PerlIO_printf(Perl_debug_log,
388 " \\%"UVuf": %"IVdf"(%"IVdf")..%"IVdf"%s\n",
390 (IV)rex->offs[paren].start,
391 (IV)rex->offs[paren].start_tmp,
392 (IV)rex->offs[paren].end,
393 (paren > rex->lastparen ? "(skipped)" : ""));
398 /* It would seem that the similar code in regtry()
399 * already takes care of this, and in fact it is in
400 * a better location to since this code can #if 0-ed out
401 * but the code in regtry() is needed or otherwise tests
402 * requiring null fields (pat.t#187 and split.t#{13,14}
403 * (as of patchlevel 7877) will fail. Then again,
404 * this code seems to be necessary or otherwise
405 * this erroneously leaves $1 defined: "1" =~ /^(?:(\d)x)?\d$/
406 * --jhi updated by dapm */
407 for (i = rex->lastparen + 1; i <= rex->nparens; i++) {
408 if (i > *maxopenparen_p)
409 rex->offs[i].start = -1;
410 rex->offs[i].end = -1;
411 DEBUG_BUFFERS_r( PerlIO_printf(Perl_debug_log,
412 " \\%"UVuf": %s ..-1 undeffing\n",
414 (i > *maxopenparen_p) ? "-1" : " "
420 /* restore the parens and associated vars at savestack position ix,
421 * but without popping the stack */
424 S_regcp_restore(pTHX_ regexp *rex, I32 ix, U32 *maxopenparen_p)
426 I32 tmpix = PL_savestack_ix;
427 PL_savestack_ix = ix;
428 regcppop(rex, maxopenparen_p);
429 PL_savestack_ix = tmpix;
432 #define regcpblow(cp) LEAVE_SCOPE(cp) /* Ignores regcppush()ed data. */
435 S_isFOO_lc(pTHX_ const U8 classnum, const U8 character)
437 /* Returns a boolean as to whether or not 'character' is a member of the
438 * Posix character class given by 'classnum' that should be equivalent to a
439 * value in the typedef '_char_class_number'.
441 * Ideally this could be replaced by a just an array of function pointers
442 * to the C library functions that implement the macros this calls.
443 * However, to compile, the precise function signatures are required, and
444 * these may vary from platform to to platform. To avoid having to figure
445 * out what those all are on each platform, I (khw) am using this method,
446 * which adds an extra layer of function call overhead (unless the C
447 * optimizer strips it away). But we don't particularly care about
448 * performance with locales anyway. */
450 switch ((_char_class_number) classnum) {
451 case _CC_ENUM_ALPHANUMERIC: return isALPHANUMERIC_LC(character);
452 case _CC_ENUM_ALPHA: return isALPHA_LC(character);
453 case _CC_ENUM_ASCII: return isASCII_LC(character);
454 case _CC_ENUM_BLANK: return isBLANK_LC(character);
455 case _CC_ENUM_CASED: return isLOWER_LC(character)
456 || isUPPER_LC(character);
457 case _CC_ENUM_CNTRL: return isCNTRL_LC(character);
458 case _CC_ENUM_DIGIT: return isDIGIT_LC(character);
459 case _CC_ENUM_GRAPH: return isGRAPH_LC(character);
460 case _CC_ENUM_LOWER: return isLOWER_LC(character);
461 case _CC_ENUM_PRINT: return isPRINT_LC(character);
462 case _CC_ENUM_PUNCT: return isPUNCT_LC(character);
463 case _CC_ENUM_SPACE: return isSPACE_LC(character);
464 case _CC_ENUM_UPPER: return isUPPER_LC(character);
465 case _CC_ENUM_WORDCHAR: return isWORDCHAR_LC(character);
466 case _CC_ENUM_XDIGIT: return isXDIGIT_LC(character);
467 default: /* VERTSPACE should never occur in locales */
468 Perl_croak(aTHX_ "panic: isFOO_lc() has an unexpected character class '%d'", classnum);
471 NOT_REACHED; /* NOTREACHED */
476 S_isFOO_utf8_lc(pTHX_ const U8 classnum, const U8* character)
478 /* Returns a boolean as to whether or not the (well-formed) UTF-8-encoded
479 * 'character' is a member of the Posix character class given by 'classnum'
480 * that should be equivalent to a value in the typedef
481 * '_char_class_number'.
483 * This just calls isFOO_lc on the code point for the character if it is in
484 * the range 0-255. Outside that range, all characters use Unicode
485 * rules, ignoring any locale. So use the Unicode function if this class
486 * requires a swash, and use the Unicode macro otherwise. */
488 PERL_ARGS_ASSERT_ISFOO_UTF8_LC;
490 if (UTF8_IS_INVARIANT(*character)) {
491 return isFOO_lc(classnum, *character);
493 else if (UTF8_IS_DOWNGRADEABLE_START(*character)) {
494 return isFOO_lc(classnum,
495 EIGHT_BIT_UTF8_TO_NATIVE(*character, *(character + 1)));
498 _CHECK_AND_OUTPUT_WIDE_LOCALE_UTF8_MSG(character, character + UTF8SKIP(character));
500 if (classnum < _FIRST_NON_SWASH_CC) {
502 /* Initialize the swash unless done already */
503 if (! PL_utf8_swash_ptrs[classnum]) {
504 U8 flags = _CORE_SWASH_INIT_ACCEPT_INVLIST;
505 PL_utf8_swash_ptrs[classnum] =
506 _core_swash_init("utf8",
509 PL_XPosix_ptrs[classnum], &flags);
512 return cBOOL(swash_fetch(PL_utf8_swash_ptrs[classnum], (U8 *)
514 TRUE /* is UTF */ ));
517 switch ((_char_class_number) classnum) {
518 case _CC_ENUM_SPACE: return is_XPERLSPACE_high(character);
519 case _CC_ENUM_BLANK: return is_HORIZWS_high(character);
520 case _CC_ENUM_XDIGIT: return is_XDIGIT_high(character);
521 case _CC_ENUM_VERTSPACE: return is_VERTWS_high(character);
525 return FALSE; /* Things like CNTRL are always below 256 */
529 * pregexec and friends
532 #ifndef PERL_IN_XSUB_RE
534 - pregexec - match a regexp against a string
537 Perl_pregexec(pTHX_ REGEXP * const prog, char* stringarg, char *strend,
538 char *strbeg, SSize_t minend, SV *screamer, U32 nosave)
539 /* stringarg: the point in the string at which to begin matching */
540 /* strend: pointer to null at end of string */
541 /* strbeg: real beginning of string */
542 /* minend: end of match must be >= minend bytes after stringarg. */
543 /* screamer: SV being matched: only used for utf8 flag, pos() etc; string
544 * itself is accessed via the pointers above */
545 /* nosave: For optimizations. */
547 PERL_ARGS_ASSERT_PREGEXEC;
550 regexec_flags(prog, stringarg, strend, strbeg, minend, screamer, NULL,
551 nosave ? 0 : REXEC_COPY_STR);
557 /* re_intuit_start():
559 * Based on some optimiser hints, try to find the earliest position in the
560 * string where the regex could match.
562 * rx: the regex to match against
563 * sv: the SV being matched: only used for utf8 flag; the string
564 * itself is accessed via the pointers below. Note that on
565 * something like an overloaded SV, SvPOK(sv) may be false
566 * and the string pointers may point to something unrelated to
568 * strbeg: real beginning of string
569 * strpos: the point in the string at which to begin matching
570 * strend: pointer to the byte following the last char of the string
571 * flags currently unused; set to 0
572 * data: currently unused; set to NULL
574 * The basic idea of re_intuit_start() is to use some known information
575 * about the pattern, namely:
577 * a) the longest known anchored substring (i.e. one that's at a
578 * constant offset from the beginning of the pattern; but not
579 * necessarily at a fixed offset from the beginning of the
581 * b) the longest floating substring (i.e. one that's not at a constant
582 * offset from the beginning of the pattern);
583 * c) Whether the pattern is anchored to the string; either
584 * an absolute anchor: /^../, or anchored to \n: /^.../m,
585 * or anchored to pos(): /\G/;
586 * d) A start class: a real or synthetic character class which
587 * represents which characters are legal at the start of the pattern;
589 * to either quickly reject the match, or to find the earliest position
590 * within the string at which the pattern might match, thus avoiding
591 * running the full NFA engine at those earlier locations, only to
592 * eventually fail and retry further along.
594 * Returns NULL if the pattern can't match, or returns the address within
595 * the string which is the earliest place the match could occur.
597 * The longest of the anchored and floating substrings is called 'check'
598 * and is checked first. The other is called 'other' and is checked
599 * second. The 'other' substring may not be present. For example,
601 * /(abc|xyz)ABC\d{0,3}DEFG/
605 * check substr (float) = "DEFG", offset 6..9 chars
606 * other substr (anchored) = "ABC", offset 3..3 chars
609 * Be aware that during the course of this function, sometimes 'anchored'
610 * refers to a substring being anchored relative to the start of the
611 * pattern, and sometimes to the pattern itself being anchored relative to
612 * the string. For example:
614 * /\dabc/: "abc" is anchored to the pattern;
615 * /^\dabc/: "abc" is anchored to the pattern and the string;
616 * /\d+abc/: "abc" is anchored to neither the pattern nor the string;
617 * /^\d+abc/: "abc" is anchored to neither the pattern nor the string,
618 * but the pattern is anchored to the string.
622 Perl_re_intuit_start(pTHX_
625 const char * const strbeg,
629 re_scream_pos_data *data)
631 struct regexp *const prog = ReANY(rx);
632 SSize_t start_shift = prog->check_offset_min;
633 /* Should be nonnegative! */
634 SSize_t end_shift = 0;
635 /* current lowest pos in string where the regex can start matching */
636 char *rx_origin = strpos;
638 const bool utf8_target = (sv && SvUTF8(sv)) ? 1 : 0; /* if no sv we have to assume bytes */
639 U8 other_ix = 1 - prog->substrs->check_ix;
641 char *other_last = strpos;/* latest pos 'other' substr already checked to */
642 char *check_at = NULL; /* check substr found at this pos */
643 const I32 multiline = prog->extflags & RXf_PMf_MULTILINE;
644 RXi_GET_DECL(prog,progi);
645 regmatch_info reginfo_buf; /* create some info to pass to find_byclass */
646 regmatch_info *const reginfo = ®info_buf;
647 GET_RE_DEBUG_FLAGS_DECL;
649 PERL_ARGS_ASSERT_RE_INTUIT_START;
650 PERL_UNUSED_ARG(flags);
651 PERL_UNUSED_ARG(data);
653 DEBUG_EXECUTE_r(PerlIO_printf(Perl_debug_log,
654 "Intuit: trying to determine minimum start position...\n"));
656 /* for now, assume that all substr offsets are positive. If at some point
657 * in the future someone wants to do clever things with look-behind and
658 * -ve offsets, they'll need to fix up any code in this function
659 * which uses these offsets. See the thread beginning
660 * <20140113145929.GF27210@iabyn.com>
662 assert(prog->substrs->data[0].min_offset >= 0);
663 assert(prog->substrs->data[0].max_offset >= 0);
664 assert(prog->substrs->data[1].min_offset >= 0);
665 assert(prog->substrs->data[1].max_offset >= 0);
666 assert(prog->substrs->data[2].min_offset >= 0);
667 assert(prog->substrs->data[2].max_offset >= 0);
669 /* for now, assume that if both present, that the floating substring
670 * doesn't start before the anchored substring.
671 * If you break this assumption (e.g. doing better optimisations
672 * with lookahead/behind), then you'll need to audit the code in this
673 * function carefully first
676 ! ( (prog->anchored_utf8 || prog->anchored_substr)
677 && (prog->float_utf8 || prog->float_substr))
678 || (prog->float_min_offset >= prog->anchored_offset));
680 /* byte rather than char calculation for efficiency. It fails
681 * to quickly reject some cases that can't match, but will reject
682 * them later after doing full char arithmetic */
683 if (prog->minlen > strend - strpos) {
684 DEBUG_EXECUTE_r(PerlIO_printf(Perl_debug_log,
685 " String too short...\n"));
689 RX_MATCH_UTF8_set(rx,utf8_target);
690 reginfo->is_utf8_target = cBOOL(utf8_target);
691 reginfo->info_aux = NULL;
692 reginfo->strbeg = strbeg;
693 reginfo->strend = strend;
694 reginfo->is_utf8_pat = cBOOL(RX_UTF8(rx));
696 /* not actually used within intuit, but zero for safety anyway */
697 reginfo->poscache_maxiter = 0;
700 if (!prog->check_utf8 && prog->check_substr)
701 to_utf8_substr(prog);
702 check = prog->check_utf8;
704 if (!prog->check_substr && prog->check_utf8) {
705 if (! to_byte_substr(prog)) {
706 NON_UTF8_TARGET_BUT_UTF8_REQUIRED(fail);
709 check = prog->check_substr;
712 /* dump the various substring data */
713 DEBUG_OPTIMISE_MORE_r({
715 for (i=0; i<=2; i++) {
716 SV *sv = (utf8_target ? prog->substrs->data[i].utf8_substr
717 : prog->substrs->data[i].substr);
721 PerlIO_printf(Perl_debug_log,
722 " substrs[%d]: min=%"IVdf" max=%"IVdf" end shift=%"IVdf
723 " useful=%"IVdf" utf8=%d [%s]\n",
725 (IV)prog->substrs->data[i].min_offset,
726 (IV)prog->substrs->data[i].max_offset,
727 (IV)prog->substrs->data[i].end_shift,
734 if (prog->intflags & PREGf_ANCH) { /* Match at \G, beg-of-str or after \n */
736 /* ml_anch: check after \n?
738 * A note about PREGf_IMPLICIT: on an un-anchored pattern beginning
739 * with /.*.../, these flags will have been added by the
741 * /.*abc/, /.*abc/m: PREGf_IMPLICIT | PREGf_ANCH_MBOL
742 * /.*abc/s: PREGf_IMPLICIT | PREGf_ANCH_SBOL
744 ml_anch = (prog->intflags & PREGf_ANCH_MBOL)
745 && !(prog->intflags & PREGf_IMPLICIT);
747 if (!ml_anch && !(prog->intflags & PREGf_IMPLICIT)) {
748 /* we are only allowed to match at BOS or \G */
750 /* trivially reject if there's a BOS anchor and we're not at BOS.
752 * Note that we don't try to do a similar quick reject for
753 * \G, since generally the caller will have calculated strpos
754 * based on pos() and gofs, so the string is already correctly
755 * anchored by definition; and handling the exceptions would
756 * be too fiddly (e.g. REXEC_IGNOREPOS).
758 if ( strpos != strbeg
759 && (prog->intflags & PREGf_ANCH_SBOL))
761 DEBUG_EXECUTE_r(PerlIO_printf(Perl_debug_log,
762 " Not at start...\n"));
766 /* in the presence of an anchor, the anchored (relative to the
767 * start of the regex) substr must also be anchored relative
768 * to strpos. So quickly reject if substr isn't found there.
769 * This works for \G too, because the caller will already have
770 * subtracted gofs from pos, and gofs is the offset from the
771 * \G to the start of the regex. For example, in /.abc\Gdef/,
772 * where substr="abcdef", pos()=3, gofs=4, offset_min=1:
773 * caller will have set strpos=pos()-4; we look for the substr
774 * at position pos()-4+1, which lines up with the "a" */
776 if (prog->check_offset_min == prog->check_offset_max) {
777 /* Substring at constant offset from beg-of-str... */
778 SSize_t slen = SvCUR(check);
779 char *s = HOP3c(strpos, prog->check_offset_min, strend);
781 DEBUG_EXECUTE_r(PerlIO_printf(Perl_debug_log,
782 " Looking for check substr at fixed offset %"IVdf"...\n",
783 (IV)prog->check_offset_min));
786 /* In this case, the regex is anchored at the end too.
787 * Unless it's a multiline match, the lengths must match
788 * exactly, give or take a \n. NB: slen >= 1 since
789 * the last char of check is \n */
791 && ( strend - s > slen
792 || strend - s < slen - 1
793 || (strend - s == slen && strend[-1] != '\n')))
795 DEBUG_EXECUTE_r(PerlIO_printf(Perl_debug_log,
796 " String too long...\n"));
799 /* Now should match s[0..slen-2] */
802 if (slen && (*SvPVX_const(check) != *s
803 || (slen > 1 && memNE(SvPVX_const(check), s, slen))))
805 DEBUG_EXECUTE_r(PerlIO_printf(Perl_debug_log,
806 " String not equal...\n"));
811 goto success_at_start;
816 end_shift = prog->check_end_shift;
818 #ifdef DEBUGGING /* 7/99: reports of failure (with the older version) */
820 Perl_croak(aTHX_ "panic: end_shift: %"IVdf" pattern:\n%s\n ",
821 (IV)end_shift, RX_PRECOMP(prog));
826 /* This is the (re)entry point of the main loop in this function.
827 * The goal of this loop is to:
828 * 1) find the "check" substring in the region rx_origin..strend
829 * (adjusted by start_shift / end_shift). If not found, reject
831 * 2) If it exists, look for the "other" substr too if defined; for
832 * example, if the check substr maps to the anchored substr, then
833 * check the floating substr, and vice-versa. If not found, go
834 * back to (1) with rx_origin suitably incremented.
835 * 3) If we find an rx_origin position that doesn't contradict
836 * either of the substrings, then check the possible additional
837 * constraints on rx_origin of /^.../m or a known start class.
838 * If these fail, then depending on which constraints fail, jump
839 * back to here, or to various other re-entry points further along
840 * that skip some of the first steps.
841 * 4) If we pass all those tests, update the BmUSEFUL() count on the
842 * substring. If the start position was determined to be at the
843 * beginning of the string - so, not rejected, but not optimised,
844 * since we have to run regmatch from position 0 - decrement the
845 * BmUSEFUL() count. Otherwise increment it.
849 /* first, look for the 'check' substring */
855 DEBUG_OPTIMISE_MORE_r({
856 PerlIO_printf(Perl_debug_log,
857 " At restart: rx_origin=%"IVdf" Check offset min: %"IVdf
858 " Start shift: %"IVdf" End shift %"IVdf
859 " Real end Shift: %"IVdf"\n",
860 (IV)(rx_origin - strbeg),
861 (IV)prog->check_offset_min,
864 (IV)prog->check_end_shift);
867 end_point = HOP3(strend, -end_shift, strbeg);
868 start_point = HOPMAYBE3(rx_origin, start_shift, end_point);
873 /* If the regex is absolutely anchored to either the start of the
874 * string (SBOL) or to pos() (ANCH_GPOS), then
875 * check_offset_max represents an upper bound on the string where
876 * the substr could start. For the ANCH_GPOS case, we assume that
877 * the caller of intuit will have already set strpos to
878 * pos()-gofs, so in this case strpos + offset_max will still be
879 * an upper bound on the substr.
882 && prog->intflags & PREGf_ANCH
883 && prog->check_offset_max != SSize_t_MAX)
885 SSize_t len = SvCUR(check) - !!SvTAIL(check);
886 const char * const anchor =
887 (prog->intflags & PREGf_ANCH_GPOS ? strpos : strbeg);
889 /* do a bytes rather than chars comparison. It's conservative;
890 * so it skips doing the HOP if the result can't possibly end
891 * up earlier than the old value of end_point.
893 if ((char*)end_point - anchor > prog->check_offset_max) {
894 end_point = HOP3lim((U8*)anchor,
895 prog->check_offset_max,
901 check_at = fbm_instr( start_point, end_point,
902 check, multiline ? FBMrf_MULTILINE : 0);
904 DEBUG_EXECUTE_r(PerlIO_printf(Perl_debug_log,
905 " doing 'check' fbm scan, [%"IVdf"..%"IVdf"] gave %"IVdf"\n",
906 (IV)((char*)start_point - strbeg),
907 (IV)((char*)end_point - strbeg),
908 (IV)(check_at ? check_at - strbeg : -1)
911 /* Update the count-of-usability, remove useless subpatterns,
915 RE_PV_QUOTED_DECL(quoted, utf8_target, PERL_DEBUG_PAD_ZERO(0),
916 SvPVX_const(check), RE_SV_DUMPLEN(check), 30);
917 PerlIO_printf(Perl_debug_log, " %s %s substr %s%s%s",
918 (check_at ? "Found" : "Did not find"),
919 (check == (utf8_target ? prog->anchored_utf8 : prog->anchored_substr)
920 ? "anchored" : "floating"),
923 (check_at ? " at offset " : "...\n") );
928 /* set rx_origin to the minimum position where the regex could start
929 * matching, given the constraint of the just-matched check substring.
930 * But don't set it lower than previously.
933 if (check_at - rx_origin > prog->check_offset_max)
934 rx_origin = HOP3c(check_at, -prog->check_offset_max, rx_origin);
935 /* Finish the diagnostic message */
936 DEBUG_EXECUTE_r(PerlIO_printf(Perl_debug_log,
937 "%ld (rx_origin now %"IVdf")...\n",
938 (long)(check_at - strbeg),
939 (IV)(rx_origin - strbeg)
944 /* now look for the 'other' substring if defined */
946 if (utf8_target ? prog->substrs->data[other_ix].utf8_substr
947 : prog->substrs->data[other_ix].substr)
949 /* Take into account the "other" substring. */
953 struct reg_substr_datum *other;
956 other = &prog->substrs->data[other_ix];
958 /* if "other" is anchored:
959 * we've previously found a floating substr starting at check_at.
960 * This means that the regex origin must lie somewhere
961 * between min (rx_origin): HOP3(check_at, -check_offset_max)
962 * and max: HOP3(check_at, -check_offset_min)
963 * (except that min will be >= strpos)
964 * So the fixed substr must lie somewhere between
965 * HOP3(min, anchored_offset)
966 * HOP3(max, anchored_offset) + SvCUR(substr)
969 /* if "other" is floating
970 * Calculate last1, the absolute latest point where the
971 * floating substr could start in the string, ignoring any
972 * constraints from the earlier fixed match. It is calculated
975 * strend - prog->minlen (in chars) is the absolute latest
976 * position within the string where the origin of the regex
977 * could appear. The latest start point for the floating
978 * substr is float_min_offset(*) on from the start of the
979 * regex. last1 simply combines thee two offsets.
981 * (*) You might think the latest start point should be
982 * float_max_offset from the regex origin, and technically
983 * you'd be correct. However, consider
985 * Here, float min, max are 3,5 and minlen is 7.
986 * This can match either
990 * In the first case, the regex matches minlen chars; in the
991 * second, minlen+1, in the third, minlen+2.
992 * In the first case, the floating offset is 3 (which equals
993 * float_min), in the second, 4, and in the third, 5 (which
994 * equals float_max). In all cases, the floating string bcd
995 * can never start more than 4 chars from the end of the
996 * string, which equals minlen - float_min. As the substring
997 * starts to match more than float_min from the start of the
998 * regex, it makes the regex match more than minlen chars,
999 * and the two cancel each other out. So we can always use
1000 * float_min - minlen, rather than float_max - minlen for the
1001 * latest position in the string.
1003 * Note that -minlen + float_min_offset is equivalent (AFAIKT)
1004 * to CHR_SVLEN(must) - !!SvTAIL(must) + prog->float_end_shift
1007 assert(prog->minlen >= other->min_offset);
1008 last1 = HOP3c(strend,
1009 other->min_offset - prog->minlen, strbeg);
1011 if (other_ix) {/* i.e. if (other-is-float) */
1012 /* last is the latest point where the floating substr could
1013 * start, *given* any constraints from the earlier fixed
1014 * match. This constraint is that the floating string starts
1015 * <= float_max_offset chars from the regex origin (rx_origin).
1016 * If this value is less than last1, use it instead.
1018 assert(rx_origin <= last1);
1020 /* this condition handles the offset==infinity case, and
1021 * is a short-cut otherwise. Although it's comparing a
1022 * byte offset to a char length, it does so in a safe way,
1023 * since 1 char always occupies 1 or more bytes,
1024 * so if a string range is (last1 - rx_origin) bytes,
1025 * it will be less than or equal to (last1 - rx_origin)
1026 * chars; meaning it errs towards doing the accurate HOP3
1027 * rather than just using last1 as a short-cut */
1028 (last1 - rx_origin) < other->max_offset
1030 : (char*)HOP3lim(rx_origin, other->max_offset, last1);
1033 assert(strpos + start_shift <= check_at);
1034 last = HOP4c(check_at, other->min_offset - start_shift,
1038 s = HOP3c(rx_origin, other->min_offset, strend);
1039 if (s < other_last) /* These positions already checked */
1042 must = utf8_target ? other->utf8_substr : other->substr;
1043 assert(SvPOK(must));
1046 char *to = last + SvCUR(must) - (SvTAIL(must)!=0);
1050 DEBUG_EXECUTE_r(PerlIO_printf(Perl_debug_log,
1051 " skipping 'other' fbm scan: %"IVdf" > %"IVdf"\n",
1052 (IV)(from - strbeg),
1058 (unsigned char*)from,
1061 multiline ? FBMrf_MULTILINE : 0
1063 DEBUG_EXECUTE_r(PerlIO_printf(Perl_debug_log,
1064 " doing 'other' fbm scan, [%"IVdf"..%"IVdf"] gave %"IVdf"\n",
1065 (IV)(from - strbeg),
1067 (IV)(s ? s - strbeg : -1)
1073 RE_PV_QUOTED_DECL(quoted, utf8_target, PERL_DEBUG_PAD_ZERO(0),
1074 SvPVX_const(must), RE_SV_DUMPLEN(must), 30);
1075 PerlIO_printf(Perl_debug_log, " %s %s substr %s%s",
1076 s ? "Found" : "Contradicts",
1077 other_ix ? "floating" : "anchored",
1078 quoted, RE_SV_TAIL(must));
1083 /* last1 is latest possible substr location. If we didn't
1084 * find it before there, we never will */
1085 if (last >= last1) {
1086 DEBUG_EXECUTE_r(PerlIO_printf(Perl_debug_log,
1087 "; giving up...\n"));
1091 /* try to find the check substr again at a later
1092 * position. Maybe next time we'll find the "other" substr
1094 other_last = HOP3c(last, 1, strend) /* highest failure */;
1096 other_ix /* i.e. if other-is-float */
1097 ? HOP3c(rx_origin, 1, strend)
1098 : HOP4c(last, 1 - other->min_offset, strbeg, strend);
1099 DEBUG_EXECUTE_r(PerlIO_printf(Perl_debug_log,
1100 "; about to retry %s at offset %ld (rx_origin now %"IVdf")...\n",
1101 (other_ix ? "floating" : "anchored"),
1102 (long)(HOP3c(check_at, 1, strend) - strbeg),
1103 (IV)(rx_origin - strbeg)
1108 if (other_ix) { /* if (other-is-float) */
1109 /* other_last is set to s, not s+1, since its possible for
1110 * a floating substr to fail first time, then succeed
1111 * second time at the same floating position; e.g.:
1112 * "-AB--AABZ" =~ /\wAB\d*Z/
1113 * The first time round, anchored and float match at
1114 * "-(AB)--AAB(Z)" then fail on the initial \w character
1115 * class. Second time round, they match at "-AB--A(AB)(Z)".
1120 rx_origin = HOP3c(s, -other->min_offset, strbeg);
1121 other_last = HOP3c(s, 1, strend);
1123 DEBUG_EXECUTE_r(PerlIO_printf(Perl_debug_log,
1124 " at offset %ld (rx_origin now %"IVdf")...\n",
1126 (IV)(rx_origin - strbeg)
1132 DEBUG_OPTIMISE_MORE_r(
1133 PerlIO_printf(Perl_debug_log,
1134 " Check-only match: offset min:%"IVdf" max:%"IVdf
1135 " check_at:%"IVdf" rx_origin:%"IVdf" rx_origin-check_at:%"IVdf
1136 " strend:%"IVdf"\n",
1137 (IV)prog->check_offset_min,
1138 (IV)prog->check_offset_max,
1139 (IV)(check_at-strbeg),
1140 (IV)(rx_origin-strbeg),
1141 (IV)(rx_origin-check_at),
1147 postprocess_substr_matches:
1149 /* handle the extra constraint of /^.../m if present */
1151 if (ml_anch && rx_origin != strbeg && rx_origin[-1] != '\n') {
1154 DEBUG_EXECUTE_r(PerlIO_printf(Perl_debug_log,
1155 " looking for /^/m anchor"));
1157 /* we have failed the constraint of a \n before rx_origin.
1158 * Find the next \n, if any, even if it's beyond the current
1159 * anchored and/or floating substrings. Whether we should be
1160 * scanning ahead for the next \n or the next substr is debatable.
1161 * On the one hand you'd expect rare substrings to appear less
1162 * often than \n's. On the other hand, searching for \n means
1163 * we're effectively flipping between check_substr and "\n" on each
1164 * iteration as the current "rarest" string candidate, which
1165 * means for example that we'll quickly reject the whole string if
1166 * hasn't got a \n, rather than trying every substr position
1170 s = HOP3c(strend, - prog->minlen, strpos);
1171 if (s <= rx_origin ||
1172 ! ( rx_origin = (char *)memchr(rx_origin, '\n', s - rx_origin)))
1174 DEBUG_EXECUTE_r(PerlIO_printf(Perl_debug_log,
1175 " Did not find /%s^%s/m...\n",
1176 PL_colors[0], PL_colors[1]));
1180 /* earliest possible origin is 1 char after the \n.
1181 * (since *rx_origin == '\n', it's safe to ++ here rather than
1182 * HOP(rx_origin, 1)) */
1185 if (prog->substrs->check_ix == 0 /* check is anchored */
1186 || rx_origin >= HOP3c(check_at, - prog->check_offset_min, strpos))
1188 /* Position contradicts check-string; either because
1189 * check was anchored (and thus has no wiggle room),
1190 * or check was float and rx_origin is above the float range */
1191 DEBUG_EXECUTE_r(PerlIO_printf(Perl_debug_log,
1192 " Found /%s^%s/m, about to restart lookup for check-string with rx_origin %ld...\n",
1193 PL_colors[0], PL_colors[1], (long)(rx_origin - strbeg)));
1197 /* if we get here, the check substr must have been float,
1198 * is in range, and we may or may not have had an anchored
1199 * "other" substr which still contradicts */
1200 assert(prog->substrs->check_ix); /* check is float */
1202 if (utf8_target ? prog->anchored_utf8 : prog->anchored_substr) {
1203 /* whoops, the anchored "other" substr exists, so we still
1204 * contradict. On the other hand, the float "check" substr
1205 * didn't contradict, so just retry the anchored "other"
1207 DEBUG_EXECUTE_r(PerlIO_printf(Perl_debug_log,
1208 " Found /%s^%s/m, rescanning for anchored from offset %"IVdf" (rx_origin now %"IVdf")...\n",
1209 PL_colors[0], PL_colors[1],
1210 (IV)(rx_origin - strbeg + prog->anchored_offset),
1211 (IV)(rx_origin - strbeg)
1213 goto do_other_substr;
1216 /* success: we don't contradict the found floating substring
1217 * (and there's no anchored substr). */
1218 DEBUG_EXECUTE_r(PerlIO_printf(Perl_debug_log,
1219 " Found /%s^%s/m with rx_origin %ld...\n",
1220 PL_colors[0], PL_colors[1], (long)(rx_origin - strbeg)));
1223 DEBUG_EXECUTE_r(PerlIO_printf(Perl_debug_log,
1224 " (multiline anchor test skipped)\n"));
1230 /* if we have a starting character class, then test that extra constraint.
1231 * (trie stclasses are too expensive to use here, we are better off to
1232 * leave it to regmatch itself) */
1234 if (progi->regstclass && PL_regkind[OP(progi->regstclass)]!=TRIE) {
1235 const U8* const str = (U8*)STRING(progi->regstclass);
1237 /* XXX this value could be pre-computed */
1238 const int cl_l = (PL_regkind[OP(progi->regstclass)] == EXACT
1239 ? (reginfo->is_utf8_pat
1240 ? utf8_distance(str + STR_LEN(progi->regstclass), str)
1241 : STR_LEN(progi->regstclass))
1245 /* latest pos that a matching float substr constrains rx start to */
1246 char *rx_max_float = NULL;
1248 /* if the current rx_origin is anchored, either by satisfying an
1249 * anchored substring constraint, or a /^.../m constraint, then we
1250 * can reject the current origin if the start class isn't found
1251 * at the current position. If we have a float-only match, then
1252 * rx_origin is constrained to a range; so look for the start class
1253 * in that range. if neither, then look for the start class in the
1254 * whole rest of the string */
1256 /* XXX DAPM it's not clear what the minlen test is for, and why
1257 * it's not used in the floating case. Nothing in the test suite
1258 * causes minlen == 0 here. See <20140313134639.GS12844@iabyn.com>.
1259 * Here are some old comments, which may or may not be correct:
1261 * minlen == 0 is possible if regstclass is \b or \B,
1262 * and the fixed substr is ''$.
1263 * Since minlen is already taken into account, rx_origin+1 is
1264 * before strend; accidentally, minlen >= 1 guaranties no false
1265 * positives at rx_origin + 1 even for \b or \B. But (minlen? 1 :
1266 * 0) below assumes that regstclass does not come from lookahead...
1267 * If regstclass takes bytelength more than 1: If charlength==1, OK.
1268 * This leaves EXACTF-ish only, which are dealt with in
1272 if (prog->anchored_substr || prog->anchored_utf8 || ml_anch)
1273 endpos= HOP3c(rx_origin, (prog->minlen ? cl_l : 0), strend);
1274 else if (prog->float_substr || prog->float_utf8) {
1275 rx_max_float = HOP3c(check_at, -start_shift, strbeg);
1276 endpos= HOP3c(rx_max_float, cl_l, strend);
1281 DEBUG_EXECUTE_r(PerlIO_printf(Perl_debug_log,
1282 " looking for class: start_shift: %"IVdf" check_at: %"IVdf
1283 " rx_origin: %"IVdf" endpos: %"IVdf"\n",
1284 (IV)start_shift, (IV)(check_at - strbeg),
1285 (IV)(rx_origin - strbeg), (IV)(endpos - strbeg)));
1287 s = find_byclass(prog, progi->regstclass, rx_origin, endpos,
1290 if (endpos == strend) {
1291 DEBUG_EXECUTE_r( PerlIO_printf(Perl_debug_log,
1292 " Could not match STCLASS...\n") );
1295 DEBUG_EXECUTE_r( PerlIO_printf(Perl_debug_log,
1296 " This position contradicts STCLASS...\n") );
1297 if ((prog->intflags & PREGf_ANCH) && !ml_anch
1298 && !(prog->intflags & PREGf_IMPLICIT))
1301 /* Contradict one of substrings */
1302 if (prog->anchored_substr || prog->anchored_utf8) {
1303 if (prog->substrs->check_ix == 1) { /* check is float */
1304 /* Have both, check_string is floating */
1305 assert(rx_origin + start_shift <= check_at);
1306 if (rx_origin + start_shift != check_at) {
1307 /* not at latest position float substr could match:
1308 * Recheck anchored substring, but not floating.
1309 * The condition above is in bytes rather than
1310 * chars for efficiency. It's conservative, in
1311 * that it errs on the side of doing 'goto
1312 * do_other_substr'. In this case, at worst,
1313 * an extra anchored search may get done, but in
1314 * practice the extra fbm_instr() is likely to
1315 * get skipped anyway. */
1316 DEBUG_EXECUTE_r( PerlIO_printf(Perl_debug_log,
1317 " about to retry anchored at offset %ld (rx_origin now %"IVdf")...\n",
1318 (long)(other_last - strbeg),
1319 (IV)(rx_origin - strbeg)
1321 goto do_other_substr;
1329 /* In the presence of ml_anch, we might be able to
1330 * find another \n without breaking the current float
1333 /* strictly speaking this should be HOP3c(..., 1, ...),
1334 * but since we goto a block of code that's going to
1335 * search for the next \n if any, its safe here */
1337 DEBUG_EXECUTE_r( PerlIO_printf(Perl_debug_log,
1338 " about to look for /%s^%s/m starting at rx_origin %ld...\n",
1339 PL_colors[0], PL_colors[1],
1340 (long)(rx_origin - strbeg)) );
1341 goto postprocess_substr_matches;
1344 /* strictly speaking this can never be true; but might
1345 * be if we ever allow intuit without substrings */
1346 if (!(utf8_target ? prog->float_utf8 : prog->float_substr))
1349 rx_origin = rx_max_float;
1352 /* at this point, any matching substrings have been
1353 * contradicted. Start again... */
1355 rx_origin = HOP3c(rx_origin, 1, strend);
1357 /* uses bytes rather than char calculations for efficiency.
1358 * It's conservative: it errs on the side of doing 'goto restart',
1359 * where there is code that does a proper char-based test */
1360 if (rx_origin + start_shift + end_shift > strend) {
1361 DEBUG_EXECUTE_r( PerlIO_printf(Perl_debug_log,
1362 " Could not match STCLASS...\n") );
1365 DEBUG_EXECUTE_r( PerlIO_printf(Perl_debug_log,
1366 " about to look for %s substr starting at offset %ld (rx_origin now %"IVdf")...\n",
1367 (prog->substrs->check_ix ? "floating" : "anchored"),
1368 (long)(rx_origin + start_shift - strbeg),
1369 (IV)(rx_origin - strbeg)
1376 if (rx_origin != s) {
1377 DEBUG_EXECUTE_r(PerlIO_printf(Perl_debug_log,
1378 " By STCLASS: moving %ld --> %ld\n",
1379 (long)(rx_origin - strbeg), (long)(s - strbeg))
1383 DEBUG_EXECUTE_r(PerlIO_printf(Perl_debug_log,
1384 " Does not contradict STCLASS...\n");
1389 /* Decide whether using the substrings helped */
1391 if (rx_origin != strpos) {
1392 /* Fixed substring is found far enough so that the match
1393 cannot start at strpos. */
1395 DEBUG_EXECUTE_r(PerlIO_printf(Perl_debug_log, " try at offset...\n"));
1396 ++BmUSEFUL(utf8_target ? prog->check_utf8 : prog->check_substr); /* hooray/5 */
1399 /* The found rx_origin position does not prohibit matching at
1400 * strpos, so calling intuit didn't gain us anything. Decrement
1401 * the BmUSEFUL() count on the check substring, and if we reach
1403 if (!(prog->intflags & PREGf_NAUGHTY)
1405 prog->check_utf8 /* Could be deleted already */
1406 && --BmUSEFUL(prog->check_utf8) < 0
1407 && (prog->check_utf8 == prog->float_utf8)
1409 prog->check_substr /* Could be deleted already */
1410 && --BmUSEFUL(prog->check_substr) < 0
1411 && (prog->check_substr == prog->float_substr)
1414 /* If flags & SOMETHING - do not do it many times on the same match */
1415 DEBUG_EXECUTE_r(PerlIO_printf(Perl_debug_log, " ... Disabling check substring...\n"));
1416 /* XXX Does the destruction order has to change with utf8_target? */
1417 SvREFCNT_dec(utf8_target ? prog->check_utf8 : prog->check_substr);
1418 SvREFCNT_dec(utf8_target ? prog->check_substr : prog->check_utf8);
1419 prog->check_substr = prog->check_utf8 = NULL; /* disable */
1420 prog->float_substr = prog->float_utf8 = NULL; /* clear */
1421 check = NULL; /* abort */
1422 /* XXXX This is a remnant of the old implementation. It
1423 looks wasteful, since now INTUIT can use many
1424 other heuristics. */
1425 prog->extflags &= ~RXf_USE_INTUIT;
1429 DEBUG_EXECUTE_r(PerlIO_printf(Perl_debug_log,
1430 "Intuit: %sSuccessfully guessed:%s match at offset %ld\n",
1431 PL_colors[4], PL_colors[5], (long)(rx_origin - strbeg)) );
1435 fail_finish: /* Substring not found */
1436 if (prog->check_substr || prog->check_utf8) /* could be removed already */
1437 BmUSEFUL(utf8_target ? prog->check_utf8 : prog->check_substr) += 5; /* hooray */
1439 DEBUG_EXECUTE_r(PerlIO_printf(Perl_debug_log, "%sMatch rejected by optimizer%s\n",
1440 PL_colors[4], PL_colors[5]));
1445 #define DECL_TRIE_TYPE(scan) \
1446 const enum { trie_plain, trie_utf8, trie_utf8_fold, trie_latin_utf8_fold, \
1447 trie_utf8_exactfa_fold, trie_latin_utf8_exactfa_fold, \
1448 trie_utf8l, trie_flu8 } \
1449 trie_type = ((scan->flags == EXACT) \
1450 ? (utf8_target ? trie_utf8 : trie_plain) \
1451 : (scan->flags == EXACTL) \
1452 ? (utf8_target ? trie_utf8l : trie_plain) \
1453 : (scan->flags == EXACTFA) \
1455 ? trie_utf8_exactfa_fold \
1456 : trie_latin_utf8_exactfa_fold) \
1457 : (scan->flags == EXACTFLU8 \
1461 : trie_latin_utf8_fold)))
1463 #define REXEC_TRIE_READ_CHAR(trie_type, trie, widecharmap, uc, uscan, len, uvc, charid, foldlen, foldbuf, uniflags) \
1466 U8 flags = FOLD_FLAGS_FULL; \
1467 switch (trie_type) { \
1469 _CHECK_AND_WARN_PROBLEMATIC_LOCALE; \
1470 if (utf8_target && UTF8_IS_ABOVE_LATIN1(*uc)) { \
1471 _CHECK_AND_OUTPUT_WIDE_LOCALE_UTF8_MSG(uc, uc + UTF8SKIP(uc)); \
1473 goto do_trie_utf8_fold; \
1474 case trie_utf8_exactfa_fold: \
1475 flags |= FOLD_FLAGS_NOMIX_ASCII; \
1477 case trie_utf8_fold: \
1478 do_trie_utf8_fold: \
1479 if ( foldlen>0 ) { \
1480 uvc = utf8n_to_uvchr( (const U8*) uscan, UTF8_MAXLEN, &len, uniflags ); \
1485 uvc = _to_utf8_fold_flags( (const U8*) uc, foldbuf, &foldlen, flags); \
1486 len = UTF8SKIP(uc); \
1487 skiplen = UVCHR_SKIP( uvc ); \
1488 foldlen -= skiplen; \
1489 uscan = foldbuf + skiplen; \
1492 case trie_latin_utf8_exactfa_fold: \
1493 flags |= FOLD_FLAGS_NOMIX_ASCII; \
1495 case trie_latin_utf8_fold: \
1496 if ( foldlen>0 ) { \
1497 uvc = utf8n_to_uvchr( (const U8*) uscan, UTF8_MAXLEN, &len, uniflags ); \
1503 uvc = _to_fold_latin1( (U8) *uc, foldbuf, &foldlen, flags); \
1504 skiplen = UVCHR_SKIP( uvc ); \
1505 foldlen -= skiplen; \
1506 uscan = foldbuf + skiplen; \
1510 _CHECK_AND_WARN_PROBLEMATIC_LOCALE; \
1511 if (utf8_target && UTF8_IS_ABOVE_LATIN1(*uc)) { \
1512 _CHECK_AND_OUTPUT_WIDE_LOCALE_UTF8_MSG(uc, uc + UTF8SKIP(uc)); \
1516 uvc = utf8n_to_uvchr( (const U8*) uc, UTF8_MAXLEN, &len, uniflags ); \
1523 charid = trie->charmap[ uvc ]; \
1527 if (widecharmap) { \
1528 SV** const svpp = hv_fetch(widecharmap, \
1529 (char*)&uvc, sizeof(UV), 0); \
1531 charid = (U16)SvIV(*svpp); \
1536 #define DUMP_EXEC_POS(li,s,doutf8) \
1537 dump_exec_pos(li,s,(reginfo->strend),(reginfo->strbeg), \
1540 #define REXEC_FBC_EXACTISH_SCAN(COND) \
1544 && (ln == 1 || folder(s, pat_string, ln)) \
1545 && (reginfo->intuit || regtry(reginfo, &s)) )\
1551 #define REXEC_FBC_UTF8_SCAN(CODE) \
1553 while (s < strend) { \
1559 #define REXEC_FBC_SCAN(CODE) \
1561 while (s < strend) { \
1567 #define REXEC_FBC_UTF8_CLASS_SCAN(COND) \
1568 REXEC_FBC_UTF8_SCAN( /* Loops while (s < strend) */ \
1570 if (tmp && (reginfo->intuit || regtry(reginfo, &s))) \
1579 #define REXEC_FBC_CLASS_SCAN(COND) \
1580 REXEC_FBC_SCAN( /* Loops while (s < strend) */ \
1582 if (tmp && (reginfo->intuit || regtry(reginfo, &s))) \
1591 #define REXEC_FBC_CSCAN(CONDUTF8,COND) \
1592 if (utf8_target) { \
1593 REXEC_FBC_UTF8_CLASS_SCAN(CONDUTF8); \
1596 REXEC_FBC_CLASS_SCAN(COND); \
1599 /* The three macros below are slightly different versions of the same logic.
1601 * The first is for /a and /aa when the target string is UTF-8. This can only
1602 * match ascii, but it must advance based on UTF-8. The other two handle the
1603 * non-UTF-8 and the more generic UTF-8 cases. In all three, we are looking
1604 * for the boundary (or non-boundary) between a word and non-word character.
1605 * The utf8 and non-utf8 cases have the same logic, but the details must be
1606 * different. Find the "wordness" of the character just prior to this one, and
1607 * compare it with the wordness of this one. If they differ, we have a
1608 * boundary. At the beginning of the string, pretend that the previous
1609 * character was a new-line.
1611 * All these macros uncleanly have side-effects with each other and outside
1612 * variables. So far it's been too much trouble to clean-up
1614 * TEST_NON_UTF8 is the macro or function to call to test if its byte input is
1615 * a word character or not.
1616 * IF_SUCCESS is code to do if it finds that we are at a boundary between
1618 * IF_FAIL is code to do if we aren't at a boundary between word/non-word
1620 * Exactly one of the two IF_FOO parameters is a no-op, depending on whether we
1621 * are looking for a boundary or for a non-boundary. If we are looking for a
1622 * boundary, we want IF_FAIL to be the no-op, and for IF_SUCCESS to go out and
1623 * see if this tentative match actually works, and if so, to quit the loop
1624 * here. And vice-versa if we are looking for a non-boundary.
1626 * 'tmp' below in the next three macros in the REXEC_FBC_SCAN and
1627 * REXEC_FBC_UTF8_SCAN loops is a loop invariant, a bool giving the return of
1628 * TEST_NON_UTF8(s-1). To see this, note that that's what it is defined to be
1629 * at entry to the loop, and to get to the IF_FAIL branch, tmp must equal
1630 * TEST_NON_UTF8(s), and in the opposite branch, IF_SUCCESS, tmp is that
1631 * complement. But in that branch we complement tmp, meaning that at the
1632 * bottom of the loop tmp is always going to be equal to TEST_NON_UTF8(s),
1633 * which means at the top of the loop in the next iteration, it is
1634 * TEST_NON_UTF8(s-1) */
1635 #define FBC_UTF8_A(TEST_NON_UTF8, IF_SUCCESS, IF_FAIL) \
1636 tmp = (s != reginfo->strbeg) ? UCHARAT(s - 1) : '\n'; \
1637 tmp = TEST_NON_UTF8(tmp); \
1638 REXEC_FBC_UTF8_SCAN( /* advances s while s < strend */ \
1639 if (tmp == ! TEST_NON_UTF8((U8) *s)) { \
1641 IF_SUCCESS; /* Is a boundary if values for s-1 and s differ */ \
1648 /* Like FBC_UTF8_A, but TEST_UV is a macro which takes a UV as its input, and
1649 * TEST_UTF8 is a macro that for the same input code points returns identically
1650 * to TEST_UV, but takes a pointer to a UTF-8 encoded string instead */
1651 #define FBC_UTF8(TEST_UV, TEST_UTF8, IF_SUCCESS, IF_FAIL) \
1652 if (s == reginfo->strbeg) { \
1655 else { /* Back-up to the start of the previous character */ \
1656 U8 * const r = reghop3((U8*)s, -1, (U8*)reginfo->strbeg); \
1657 tmp = utf8n_to_uvchr(r, (U8*) reginfo->strend - r, \
1658 0, UTF8_ALLOW_DEFAULT); \
1660 tmp = TEST_UV(tmp); \
1661 LOAD_UTF8_CHARCLASS_ALNUM(); \
1662 REXEC_FBC_UTF8_SCAN( /* advances s while s < strend */ \
1663 if (tmp == ! (TEST_UTF8((U8 *) s))) { \
1672 /* Like the above two macros. UTF8_CODE is the complete code for handling
1673 * UTF-8. Common to the BOUND and NBOUND cases, set-up by the FBC_BOUND, etc
1675 #define FBC_BOUND_COMMON(UTF8_CODE, TEST_NON_UTF8, IF_SUCCESS, IF_FAIL) \
1676 if (utf8_target) { \
1679 else { /* Not utf8 */ \
1680 tmp = (s != reginfo->strbeg) ? UCHARAT(s - 1) : '\n'; \
1681 tmp = TEST_NON_UTF8(tmp); \
1682 REXEC_FBC_SCAN( /* advances s while s < strend */ \
1683 if (tmp == ! TEST_NON_UTF8((U8) *s)) { \
1692 /* Here, things have been set up by the previous code so that tmp is the \
1693 * return of TEST_NON_UTF(s-1) or TEST_UTF8(s-1) (depending on the \
1694 * utf8ness of the target). We also have to check if this matches against \
1695 * the EOS, which we treat as a \n (which is the same value in both UTF-8 \
1696 * or non-UTF8, so can use the non-utf8 test condition even for a UTF-8 \
1698 if (tmp == ! TEST_NON_UTF8('\n')) { \
1705 /* This is the macro to use when we want to see if something that looks like it
1706 * could match, actually does, and if so exits the loop */
1707 #define REXEC_FBC_TRYIT \
1708 if ((reginfo->intuit || regtry(reginfo, &s))) \
1711 /* The only difference between the BOUND and NBOUND cases is that
1712 * REXEC_FBC_TRYIT is called when matched in BOUND, and when non-matched in
1713 * NBOUND. This is accomplished by passing it as either the if or else clause,
1714 * with the other one being empty (PLACEHOLDER is defined as empty).
1716 * The TEST_FOO parameters are for operating on different forms of input, but
1717 * all should be ones that return identically for the same underlying code
1719 #define FBC_BOUND(TEST_NON_UTF8, TEST_UV, TEST_UTF8) \
1721 FBC_UTF8(TEST_UV, TEST_UTF8, REXEC_FBC_TRYIT, PLACEHOLDER), \
1722 TEST_NON_UTF8, REXEC_FBC_TRYIT, PLACEHOLDER)
1724 #define FBC_BOUND_A(TEST_NON_UTF8) \
1726 FBC_UTF8_A(TEST_NON_UTF8, REXEC_FBC_TRYIT, PLACEHOLDER), \
1727 TEST_NON_UTF8, REXEC_FBC_TRYIT, PLACEHOLDER)
1729 #define FBC_NBOUND(TEST_NON_UTF8, TEST_UV, TEST_UTF8) \
1731 FBC_UTF8(TEST_UV, TEST_UTF8, PLACEHOLDER, REXEC_FBC_TRYIT), \
1732 TEST_NON_UTF8, PLACEHOLDER, REXEC_FBC_TRYIT)
1734 #define FBC_NBOUND_A(TEST_NON_UTF8) \
1736 FBC_UTF8_A(TEST_NON_UTF8, PLACEHOLDER, REXEC_FBC_TRYIT), \
1737 TEST_NON_UTF8, PLACEHOLDER, REXEC_FBC_TRYIT)
1739 /* Takes a pointer to an inversion list, a pointer to its corresponding
1740 * inversion map, and a code point, and returns the code point's value
1741 * according to the two arrays. It assumes that all code points have a value.
1742 * This is used as the base macro for macros for particular properties */
1743 #define _generic_GET_BREAK_VAL_CP(invlist, invmap, cp) \
1744 invmap[_invlist_search(invlist, cp)]
1746 /* Same as above, but takes begin, end ptrs to a UTF-8 encoded string instead
1747 * of a code point, returning the value for the first code point in the string.
1748 * And it takes the particular macro name that finds the desired value given a
1749 * code point. Merely convert the UTF-8 to code point and call the cp macro */
1750 #define _generic_GET_BREAK_VAL_UTF8(cp_macro, pos, strend) \
1751 (__ASSERT_(pos < strend) \
1752 /* Note assumes is valid UTF-8 */ \
1753 (cp_macro(utf8_to_uvchr_buf((pos), (strend), NULL))))
1755 /* Returns the GCB value for the input code point */
1756 #define getGCB_VAL_CP(cp) \
1757 _generic_GET_BREAK_VAL_CP( \
1762 /* Returns the GCB value for the first code point in the UTF-8 encoded string
1763 * bounded by pos and strend */
1764 #define getGCB_VAL_UTF8(pos, strend) \
1765 _generic_GET_BREAK_VAL_UTF8(getGCB_VAL_CP, pos, strend)
1768 /* Returns the SB value for the input code point */
1769 #define getSB_VAL_CP(cp) \
1770 _generic_GET_BREAK_VAL_CP( \
1775 /* Returns the SB value for the first code point in the UTF-8 encoded string
1776 * bounded by pos and strend */
1777 #define getSB_VAL_UTF8(pos, strend) \
1778 _generic_GET_BREAK_VAL_UTF8(getSB_VAL_CP, pos, strend)
1780 /* Returns the WB value for the input code point */
1781 #define getWB_VAL_CP(cp) \
1782 _generic_GET_BREAK_VAL_CP( \
1787 /* Returns the WB value for the first code point in the UTF-8 encoded string
1788 * bounded by pos and strend */
1789 #define getWB_VAL_UTF8(pos, strend) \
1790 _generic_GET_BREAK_VAL_UTF8(getWB_VAL_CP, pos, strend)
1792 /* We know what class REx starts with. Try to find this position... */
1793 /* if reginfo->intuit, its a dryrun */
1794 /* annoyingly all the vars in this routine have different names from their counterparts
1795 in regmatch. /grrr */
1797 S_find_byclass(pTHX_ regexp * prog, const regnode *c, char *s,
1798 const char *strend, regmatch_info *reginfo)
1801 const I32 doevery = (prog->intflags & PREGf_SKIP) == 0;
1802 char *pat_string; /* The pattern's exactish string */
1803 char *pat_end; /* ptr to end char of pat_string */
1804 re_fold_t folder; /* Function for computing non-utf8 folds */
1805 const U8 *fold_array; /* array for folding ords < 256 */
1811 I32 tmp = 1; /* Scratch variable? */
1812 const bool utf8_target = reginfo->is_utf8_target;
1813 UV utf8_fold_flags = 0;
1814 const bool is_utf8_pat = reginfo->is_utf8_pat;
1815 bool to_complement = FALSE; /* Invert the result? Taking the xor of this
1816 with a result inverts that result, as 0^1 =
1818 _char_class_number classnum;
1820 RXi_GET_DECL(prog,progi);
1822 PERL_ARGS_ASSERT_FIND_BYCLASS;
1824 /* We know what class it must start with. */
1827 _CHECK_AND_WARN_PROBLEMATIC_LOCALE;
1829 if ((FLAGS(c) & ANYOF_LOC_REQ_UTF8) && ! IN_UTF8_CTYPE_LOCALE) {
1830 Perl_ck_warner(aTHX_ packWARN(WARN_LOCALE), utf8_locale_required);
1837 REXEC_FBC_UTF8_CLASS_SCAN(
1838 reginclass(prog, c, (U8*)s, (U8*) strend, utf8_target));
1841 REXEC_FBC_CLASS_SCAN(REGINCLASS(prog, c, (U8*)s));
1845 case EXACTFA_NO_TRIE: /* This node only generated for non-utf8 patterns */
1846 assert(! is_utf8_pat);
1849 if (is_utf8_pat || utf8_target) {
1850 utf8_fold_flags = FOLDEQ_UTF8_NOMIX_ASCII;
1851 goto do_exactf_utf8;
1853 fold_array = PL_fold_latin1; /* Latin1 folds are not affected by */
1854 folder = foldEQ_latin1; /* /a, except the sharp s one which */
1855 goto do_exactf_non_utf8; /* isn't dealt with by these */
1857 case EXACTF: /* This node only generated for non-utf8 patterns */
1858 assert(! is_utf8_pat);
1860 utf8_fold_flags = 0;
1861 goto do_exactf_utf8;
1863 fold_array = PL_fold;
1865 goto do_exactf_non_utf8;
1868 _CHECK_AND_WARN_PROBLEMATIC_LOCALE;
1869 if (is_utf8_pat || utf8_target || IN_UTF8_CTYPE_LOCALE) {
1870 utf8_fold_flags = FOLDEQ_LOCALE;
1871 goto do_exactf_utf8;
1873 fold_array = PL_fold_locale;
1874 folder = foldEQ_locale;
1875 goto do_exactf_non_utf8;
1879 utf8_fold_flags = FOLDEQ_S2_ALREADY_FOLDED;
1881 goto do_exactf_utf8;
1884 if (! utf8_target) { /* All code points in this node require
1885 UTF-8 to express. */
1888 utf8_fold_flags = FOLDEQ_LOCALE | FOLDEQ_S2_ALREADY_FOLDED
1889 | FOLDEQ_S2_FOLDS_SANE;
1890 goto do_exactf_utf8;
1893 if (is_utf8_pat || utf8_target) {
1894 utf8_fold_flags = is_utf8_pat ? FOLDEQ_S2_ALREADY_FOLDED : 0;
1895 goto do_exactf_utf8;
1898 /* Any 'ss' in the pattern should have been replaced by regcomp,
1899 * so we don't have to worry here about this single special case
1900 * in the Latin1 range */
1901 fold_array = PL_fold_latin1;
1902 folder = foldEQ_latin1;
1906 do_exactf_non_utf8: /* Neither pattern nor string are UTF8, and there
1907 are no glitches with fold-length differences
1908 between the target string and pattern */
1910 /* The idea in the non-utf8 EXACTF* cases is to first find the
1911 * first character of the EXACTF* node and then, if necessary,
1912 * case-insensitively compare the full text of the node. c1 is the
1913 * first character. c2 is its fold. This logic will not work for
1914 * Unicode semantics and the german sharp ss, which hence should
1915 * not be compiled into a node that gets here. */
1916 pat_string = STRING(c);
1917 ln = STR_LEN(c); /* length to match in octets/bytes */
1919 /* We know that we have to match at least 'ln' bytes (which is the
1920 * same as characters, since not utf8). If we have to match 3
1921 * characters, and there are only 2 availabe, we know without
1922 * trying that it will fail; so don't start a match past the
1923 * required minimum number from the far end */
1924 e = HOP3c(strend, -((SSize_t)ln), s);
1926 if (reginfo->intuit && e < s) {
1927 e = s; /* Due to minlen logic of intuit() */
1931 c2 = fold_array[c1];
1932 if (c1 == c2) { /* If char and fold are the same */
1933 REXEC_FBC_EXACTISH_SCAN(*(U8*)s == c1);
1936 REXEC_FBC_EXACTISH_SCAN(*(U8*)s == c1 || *(U8*)s == c2);
1944 /* If one of the operands is in utf8, we can't use the simpler folding
1945 * above, due to the fact that many different characters can have the
1946 * same fold, or portion of a fold, or different- length fold */
1947 pat_string = STRING(c);
1948 ln = STR_LEN(c); /* length to match in octets/bytes */
1949 pat_end = pat_string + ln;
1950 lnc = is_utf8_pat /* length to match in characters */
1951 ? utf8_length((U8 *) pat_string, (U8 *) pat_end)
1954 /* We have 'lnc' characters to match in the pattern, but because of
1955 * multi-character folding, each character in the target can match
1956 * up to 3 characters (Unicode guarantees it will never exceed
1957 * this) if it is utf8-encoded; and up to 2 if not (based on the
1958 * fact that the Latin 1 folds are already determined, and the
1959 * only multi-char fold in that range is the sharp-s folding to
1960 * 'ss'. Thus, a pattern character can match as little as 1/3 of a
1961 * string character. Adjust lnc accordingly, rounding up, so that
1962 * if we need to match at least 4+1/3 chars, that really is 5. */
1963 expansion = (utf8_target) ? UTF8_MAX_FOLD_CHAR_EXPAND : 2;
1964 lnc = (lnc + expansion - 1) / expansion;
1966 /* As in the non-UTF8 case, if we have to match 3 characters, and
1967 * only 2 are left, it's guaranteed to fail, so don't start a
1968 * match that would require us to go beyond the end of the string
1970 e = HOP3c(strend, -((SSize_t)lnc), s);
1972 if (reginfo->intuit && e < s) {
1973 e = s; /* Due to minlen logic of intuit() */
1976 /* XXX Note that we could recalculate e to stop the loop earlier,
1977 * as the worst case expansion above will rarely be met, and as we
1978 * go along we would usually find that e moves further to the left.
1979 * This would happen only after we reached the point in the loop
1980 * where if there were no expansion we should fail. Unclear if
1981 * worth the expense */
1984 char *my_strend= (char *)strend;
1985 if (foldEQ_utf8_flags(s, &my_strend, 0, utf8_target,
1986 pat_string, NULL, ln, is_utf8_pat, utf8_fold_flags)
1987 && (reginfo->intuit || regtry(reginfo, &s)) )
1991 s += (utf8_target) ? UTF8SKIP(s) : 1;
1997 _CHECK_AND_WARN_PROBLEMATIC_LOCALE;
1998 if (FLAGS(c) != TRADITIONAL_BOUND) {
1999 if (! IN_UTF8_CTYPE_LOCALE) {
2000 Perl_ck_warner(aTHX_ packWARN(WARN_LOCALE),
2001 B_ON_NON_UTF8_LOCALE_IS_WRONG);
2006 FBC_BOUND(isWORDCHAR_LC, isWORDCHAR_LC_uvchr, isWORDCHAR_LC_utf8);
2010 _CHECK_AND_WARN_PROBLEMATIC_LOCALE;
2011 if (FLAGS(c) != TRADITIONAL_BOUND) {
2012 if (! IN_UTF8_CTYPE_LOCALE) {
2013 Perl_ck_warner(aTHX_ packWARN(WARN_LOCALE),
2014 B_ON_NON_UTF8_LOCALE_IS_WRONG);
2019 FBC_NBOUND(isWORDCHAR_LC, isWORDCHAR_LC_uvchr, isWORDCHAR_LC_utf8);
2022 case BOUND: /* regcomp.c makes sure that this only has the traditional \b
2024 assert(FLAGS(c) == TRADITIONAL_BOUND);
2026 FBC_BOUND(isWORDCHAR, isWORDCHAR_uni, isWORDCHAR_utf8);
2029 case BOUNDA: /* regcomp.c makes sure that this only has the traditional \b
2031 assert(FLAGS(c) == TRADITIONAL_BOUND);
2033 FBC_BOUND_A(isWORDCHAR_A);
2036 case NBOUND: /* regcomp.c makes sure that this only has the traditional \b
2038 assert(FLAGS(c) == TRADITIONAL_BOUND);
2040 FBC_NBOUND(isWORDCHAR, isWORDCHAR_uni, isWORDCHAR_utf8);
2043 case NBOUNDA: /* regcomp.c makes sure that this only has the traditional \b
2045 assert(FLAGS(c) == TRADITIONAL_BOUND);
2047 FBC_NBOUND_A(isWORDCHAR_A);
2051 if ((bound_type) FLAGS(c) == TRADITIONAL_BOUND) {
2052 FBC_NBOUND(isWORDCHAR_L1, isWORDCHAR_uni, isWORDCHAR_utf8);
2063 switch((bound_type) FLAGS(c)) {
2064 case TRADITIONAL_BOUND:
2065 FBC_BOUND(isWORDCHAR_L1, isWORDCHAR_uni, isWORDCHAR_utf8);
2068 if (s == reginfo->strbeg) {
2069 if (reginfo->intuit || regtry(reginfo, &s))
2074 /* Didn't match. Try at the next position (if there is one) */
2075 s += (utf8_target) ? UTF8SKIP(s) : 1;
2076 if (UNLIKELY(s >= reginfo->strend)) {
2082 GCB_enum before = getGCB_VAL_UTF8(
2084 (U8*)(reginfo->strbeg)),
2085 (U8*) reginfo->strend);
2086 while (s < strend) {
2087 GCB_enum after = getGCB_VAL_UTF8((U8*) s,
2088 (U8*) reginfo->strend);
2089 if ( (to_complement ^ isGCB(before, after))
2090 && (reginfo->intuit || regtry(reginfo, &s)))
2098 else { /* Not utf8. Everything is a GCB except between CR and
2100 while (s < strend) {
2101 if ((to_complement ^ ( UCHARAT(s - 1) != '\r'
2102 || UCHARAT(s) != '\n'))
2103 && (reginfo->intuit || regtry(reginfo, &s)))
2111 /* And, since this is a bound, it can match after the final
2112 * character in the string */
2113 if ((reginfo->intuit || regtry(reginfo, &s))) {
2119 if (s == reginfo->strbeg) {
2120 if (reginfo->intuit || regtry(reginfo, &s)) {
2123 s += (utf8_target) ? UTF8SKIP(s) : 1;
2124 if (UNLIKELY(s >= reginfo->strend)) {
2130 SB_enum before = getSB_VAL_UTF8(reghop3((U8*)s,
2132 (U8*)(reginfo->strbeg)),
2133 (U8*) reginfo->strend);
2134 while (s < strend) {
2135 SB_enum after = getSB_VAL_UTF8((U8*) s,
2136 (U8*) reginfo->strend);
2137 if ((to_complement ^ isSB(before,
2139 (U8*) reginfo->strbeg,
2141 (U8*) reginfo->strend,
2143 && (reginfo->intuit || regtry(reginfo, &s)))
2151 else { /* Not utf8. */
2152 SB_enum before = getSB_VAL_CP((U8) *(s -1));
2153 while (s < strend) {
2154 SB_enum after = getSB_VAL_CP((U8) *s);
2155 if ((to_complement ^ isSB(before,
2157 (U8*) reginfo->strbeg,
2159 (U8*) reginfo->strend,
2161 && (reginfo->intuit || regtry(reginfo, &s)))
2170 /* Here are at the final position in the target string. The SB
2171 * value is always true here, so matches, depending on other
2173 if (reginfo->intuit || regtry(reginfo, &s)) {
2180 if (s == reginfo->strbeg) {
2181 if (reginfo->intuit || regtry(reginfo, &s)) {
2184 s += (utf8_target) ? UTF8SKIP(s) : 1;
2185 if (UNLIKELY(s >= reginfo->strend)) {
2191 /* We are at a boundary between char_sub_0 and char_sub_1.
2192 * We also keep track of the value for char_sub_-1 as we
2193 * loop through the line. Context may be needed to make a
2194 * determination, and if so, this can save having to
2196 WB_enum previous = WB_UNKNOWN;
2197 WB_enum before = getWB_VAL_UTF8(
2200 (U8*)(reginfo->strbeg)),
2201 (U8*) reginfo->strend);
2202 while (s < strend) {
2203 WB_enum after = getWB_VAL_UTF8((U8*) s,
2204 (U8*) reginfo->strend);
2205 if ((to_complement ^ isWB(previous,
2208 (U8*) reginfo->strbeg,
2210 (U8*) reginfo->strend,
2212 && (reginfo->intuit || regtry(reginfo, &s)))
2221 else { /* Not utf8. */
2222 WB_enum previous = WB_UNKNOWN;
2223 WB_enum before = getWB_VAL_CP((U8) *(s -1));
2224 while (s < strend) {
2225 WB_enum after = getWB_VAL_CP((U8) *s);
2226 if ((to_complement ^ isWB(previous,
2229 (U8*) reginfo->strbeg,
2231 (U8*) reginfo->strend,
2233 && (reginfo->intuit || regtry(reginfo, &s)))
2243 if (reginfo->intuit || regtry(reginfo, &s)) {
2250 REXEC_FBC_CSCAN(is_LNBREAK_utf8_safe(s, strend),
2251 is_LNBREAK_latin1_safe(s, strend)
2255 /* The argument to all the POSIX node types is the class number to pass to
2256 * _generic_isCC() to build a mask for searching in PL_charclass[] */
2263 _CHECK_AND_WARN_PROBLEMATIC_LOCALE;
2264 REXEC_FBC_CSCAN(to_complement ^ cBOOL(isFOO_utf8_lc(FLAGS(c), (U8 *) s)),
2265 to_complement ^ cBOOL(isFOO_lc(FLAGS(c), *s)));
2280 /* The complement of something that matches only ASCII matches all
2281 * non-ASCII, plus everything in ASCII that isn't in the class. */
2282 REXEC_FBC_UTF8_CLASS_SCAN(! isASCII_utf8(s)
2283 || ! _generic_isCC_A(*s, FLAGS(c)));
2292 /* Don't need to worry about utf8, as it can match only a single
2293 * byte invariant character. */
2294 REXEC_FBC_CLASS_SCAN(
2295 to_complement ^ cBOOL(_generic_isCC_A(*s, FLAGS(c))));
2303 if (! utf8_target) {
2304 REXEC_FBC_CLASS_SCAN(to_complement ^ cBOOL(_generic_isCC(*s,
2310 classnum = (_char_class_number) FLAGS(c);
2311 if (classnum < _FIRST_NON_SWASH_CC) {
2312 while (s < strend) {
2314 /* We avoid loading in the swash as long as possible, but
2315 * should we have to, we jump to a separate loop. This
2316 * extra 'if' statement is what keeps this code from being
2317 * just a call to REXEC_FBC_UTF8_CLASS_SCAN() */
2318 if (UTF8_IS_ABOVE_LATIN1(*s)) {
2319 goto found_above_latin1;
2321 if ((UTF8_IS_INVARIANT(*s)
2322 && to_complement ^ cBOOL(_generic_isCC((U8) *s,
2324 || (UTF8_IS_DOWNGRADEABLE_START(*s)
2325 && to_complement ^ cBOOL(
2326 _generic_isCC(EIGHT_BIT_UTF8_TO_NATIVE(*s,
2330 if (tmp && (reginfo->intuit || regtry(reginfo, &s)))
2342 else switch (classnum) { /* These classes are implemented as
2344 case _CC_ENUM_SPACE:
2345 REXEC_FBC_UTF8_CLASS_SCAN(
2346 to_complement ^ cBOOL(isSPACE_utf8(s)));
2349 case _CC_ENUM_BLANK:
2350 REXEC_FBC_UTF8_CLASS_SCAN(
2351 to_complement ^ cBOOL(isBLANK_utf8(s)));
2354 case _CC_ENUM_XDIGIT:
2355 REXEC_FBC_UTF8_CLASS_SCAN(
2356 to_complement ^ cBOOL(isXDIGIT_utf8(s)));
2359 case _CC_ENUM_VERTSPACE:
2360 REXEC_FBC_UTF8_CLASS_SCAN(
2361 to_complement ^ cBOOL(isVERTWS_utf8(s)));
2364 case _CC_ENUM_CNTRL:
2365 REXEC_FBC_UTF8_CLASS_SCAN(
2366 to_complement ^ cBOOL(isCNTRL_utf8(s)));
2370 Perl_croak(aTHX_ "panic: find_byclass() node %d='%s' has an unexpected character class '%d'", OP(c), PL_reg_name[OP(c)], classnum);
2371 NOT_REACHED; /* NOTREACHED */
2376 found_above_latin1: /* Here we have to load a swash to get the result
2377 for the current code point */
2378 if (! PL_utf8_swash_ptrs[classnum]) {
2379 U8 flags = _CORE_SWASH_INIT_ACCEPT_INVLIST;
2380 PL_utf8_swash_ptrs[classnum] =
2381 _core_swash_init("utf8",
2384 PL_XPosix_ptrs[classnum], &flags);
2387 /* This is a copy of the loop above for swash classes, though using the
2388 * FBC macro instead of being expanded out. Since we've loaded the
2389 * swash, we don't have to check for that each time through the loop */
2390 REXEC_FBC_UTF8_CLASS_SCAN(
2391 to_complement ^ cBOOL(_generic_utf8(
2394 swash_fetch(PL_utf8_swash_ptrs[classnum],
2402 /* what trie are we using right now */
2403 reg_ac_data *aho = (reg_ac_data*)progi->data->data[ ARG( c ) ];
2404 reg_trie_data *trie = (reg_trie_data*)progi->data->data[ aho->trie ];
2405 HV *widecharmap = MUTABLE_HV(progi->data->data[ aho->trie + 1 ]);
2407 const char *last_start = strend - trie->minlen;
2409 const char *real_start = s;
2411 STRLEN maxlen = trie->maxlen;
2413 U8 **points; /* map of where we were in the input string
2414 when reading a given char. For ASCII this
2415 is unnecessary overhead as the relationship
2416 is always 1:1, but for Unicode, especially
2417 case folded Unicode this is not true. */
2418 U8 foldbuf[ UTF8_MAXBYTES_CASE + 1 ];
2422 GET_RE_DEBUG_FLAGS_DECL;
2424 /* We can't just allocate points here. We need to wrap it in
2425 * an SV so it gets freed properly if there is a croak while
2426 * running the match */
2429 sv_points=newSV(maxlen * sizeof(U8 *));
2430 SvCUR_set(sv_points,
2431 maxlen * sizeof(U8 *));
2432 SvPOK_on(sv_points);
2433 sv_2mortal(sv_points);
2434 points=(U8**)SvPV_nolen(sv_points );
2435 if ( trie_type != trie_utf8_fold
2436 && (trie->bitmap || OP(c)==AHOCORASICKC) )
2439 bitmap=(U8*)trie->bitmap;
2441 bitmap=(U8*)ANYOF_BITMAP(c);
2443 /* this is the Aho-Corasick algorithm modified a touch
2444 to include special handling for long "unknown char" sequences.
2445 The basic idea being that we use AC as long as we are dealing
2446 with a possible matching char, when we encounter an unknown char
2447 (and we have not encountered an accepting state) we scan forward
2448 until we find a legal starting char.
2449 AC matching is basically that of trie matching, except that when
2450 we encounter a failing transition, we fall back to the current
2451 states "fail state", and try the current char again, a process
2452 we repeat until we reach the root state, state 1, or a legal
2453 transition. If we fail on the root state then we can either
2454 terminate if we have reached an accepting state previously, or
2455 restart the entire process from the beginning if we have not.
2458 while (s <= last_start) {
2459 const U32 uniflags = UTF8_ALLOW_DEFAULT;
2467 U8 *uscan = (U8*)NULL;
2468 U8 *leftmost = NULL;
2470 U32 accepted_word= 0;
2474 while ( state && uc <= (U8*)strend ) {
2476 U32 word = aho->states[ state ].wordnum;
2480 DEBUG_TRIE_EXECUTE_r(
2481 if ( uc <= (U8*)last_start && !BITMAP_TEST(bitmap,*uc) ) {
2482 dump_exec_pos( (char *)uc, c, strend, real_start,
2483 (char *)uc, utf8_target );
2484 PerlIO_printf( Perl_debug_log,
2485 " Scanning for legal start char...\n");
2489 while ( uc <= (U8*)last_start && !BITMAP_TEST(bitmap,*uc) ) {
2493 while ( uc <= (U8*)last_start && !BITMAP_TEST(bitmap,*uc) ) {
2499 if (uc >(U8*)last_start) break;
2503 U8 *lpos= points[ (pointpos - trie->wordinfo[word].len) % maxlen ];
2504 if (!leftmost || lpos < leftmost) {
2505 DEBUG_r(accepted_word=word);
2511 points[pointpos++ % maxlen]= uc;
2512 if (foldlen || uc < (U8*)strend) {
2513 REXEC_TRIE_READ_CHAR(trie_type, trie,
2515 uscan, len, uvc, charid, foldlen,
2517 DEBUG_TRIE_EXECUTE_r({
2518 dump_exec_pos( (char *)uc, c, strend,
2519 real_start, s, utf8_target);
2520 PerlIO_printf(Perl_debug_log,
2521 " Charid:%3u CP:%4"UVxf" ",
2533 word = aho->states[ state ].wordnum;
2535 base = aho->states[ state ].trans.base;
2537 DEBUG_TRIE_EXECUTE_r({
2539 dump_exec_pos( (char *)uc, c, strend, real_start,
2541 PerlIO_printf( Perl_debug_log,
2542 "%sState: %4"UVxf", word=%"UVxf,
2543 failed ? " Fail transition to " : "",
2544 (UV)state, (UV)word);
2550 ( ((offset = base + charid
2551 - 1 - trie->uniquecharcount)) >= 0)
2552 && ((U32)offset < trie->lasttrans)
2553 && trie->trans[offset].check == state
2554 && (tmp=trie->trans[offset].next))
2556 DEBUG_TRIE_EXECUTE_r(
2557 PerlIO_printf( Perl_debug_log," - legal\n"));
2562 DEBUG_TRIE_EXECUTE_r(
2563 PerlIO_printf( Perl_debug_log," - fail\n"));
2565 state = aho->fail[state];
2569 /* we must be accepting here */
2570 DEBUG_TRIE_EXECUTE_r(
2571 PerlIO_printf( Perl_debug_log," - accepting\n"));
2580 if (!state) state = 1;
2583 if ( aho->states[ state ].wordnum ) {
2584 U8 *lpos = points[ (pointpos - trie->wordinfo[aho->states[ state ].wordnum].len) % maxlen ];
2585 if (!leftmost || lpos < leftmost) {
2586 DEBUG_r(accepted_word=aho->states[ state ].wordnum);
2591 s = (char*)leftmost;
2592 DEBUG_TRIE_EXECUTE_r({
2594 Perl_debug_log,"Matches word #%"UVxf" at position %"IVdf". Trying full pattern...\n",
2595 (UV)accepted_word, (IV)(s - real_start)
2598 if (reginfo->intuit || regtry(reginfo, &s)) {
2604 DEBUG_TRIE_EXECUTE_r({
2605 PerlIO_printf( Perl_debug_log,"Pattern failed. Looking for new start point...\n");
2608 DEBUG_TRIE_EXECUTE_r(
2609 PerlIO_printf( Perl_debug_log,"No match.\n"));
2618 Perl_croak(aTHX_ "panic: unknown regstclass %d", (int)OP(c));
2625 /* set RX_SAVED_COPY, RX_SUBBEG etc.
2626 * flags have same meanings as with regexec_flags() */
2629 S_reg_set_capture_string(pTHX_ REGEXP * const rx,
2636 struct regexp *const prog = ReANY(rx);
2638 if (flags & REXEC_COPY_STR) {
2642 PerlIO_printf(Perl_debug_log,
2643 "Copy on write: regexp capture, type %d\n",
2646 /* Create a new COW SV to share the match string and store
2647 * in saved_copy, unless the current COW SV in saved_copy
2648 * is valid and suitable for our purpose */
2649 if (( prog->saved_copy
2650 && SvIsCOW(prog->saved_copy)
2651 && SvPOKp(prog->saved_copy)
2654 && SvPVX(sv) == SvPVX(prog->saved_copy)))
2656 /* just reuse saved_copy SV */
2657 if (RXp_MATCH_COPIED(prog)) {
2658 Safefree(prog->subbeg);
2659 RXp_MATCH_COPIED_off(prog);
2663 /* create new COW SV to share string */
2664 RX_MATCH_COPY_FREE(rx);
2665 prog->saved_copy = sv_setsv_cow(prog->saved_copy, sv);
2667 prog->subbeg = (char *)SvPVX_const(prog->saved_copy);
2668 assert (SvPOKp(prog->saved_copy));
2669 prog->sublen = strend - strbeg;
2670 prog->suboffset = 0;
2671 prog->subcoffset = 0;
2676 SSize_t max = strend - strbeg;
2679 if ( (flags & REXEC_COPY_SKIP_POST)
2680 && !(prog->extflags & RXf_PMf_KEEPCOPY) /* //p */
2681 && !(PL_sawampersand & SAWAMPERSAND_RIGHT)
2682 ) { /* don't copy $' part of string */
2685 /* calculate the right-most part of the string covered
2686 * by a capture. Due to look-ahead, this may be to
2687 * the right of $&, so we have to scan all captures */
2688 while (n <= prog->lastparen) {
2689 if (prog->offs[n].end > max)
2690 max = prog->offs[n].end;
2694 max = (PL_sawampersand & SAWAMPERSAND_LEFT)
2695 ? prog->offs[0].start
2697 assert(max >= 0 && max <= strend - strbeg);
2700 if ( (flags & REXEC_COPY_SKIP_PRE)
2701 && !(prog->extflags & RXf_PMf_KEEPCOPY) /* //p */
2702 && !(PL_sawampersand & SAWAMPERSAND_LEFT)
2703 ) { /* don't copy $` part of string */
2706 /* calculate the left-most part of the string covered
2707 * by a capture. Due to look-behind, this may be to
2708 * the left of $&, so we have to scan all captures */
2709 while (min && n <= prog->lastparen) {
2710 if ( prog->offs[n].start != -1
2711 && prog->offs[n].start < min)
2713 min = prog->offs[n].start;
2717 if ((PL_sawampersand & SAWAMPERSAND_RIGHT)
2718 && min > prog->offs[0].end
2720 min = prog->offs[0].end;
2724 assert(min >= 0 && min <= max && min <= strend - strbeg);
2727 if (RX_MATCH_COPIED(rx)) {
2728 if (sublen > prog->sublen)
2730 (char*)saferealloc(prog->subbeg, sublen+1);
2733 prog->subbeg = (char*)safemalloc(sublen+1);
2734 Copy(strbeg + min, prog->subbeg, sublen, char);
2735 prog->subbeg[sublen] = '\0';
2736 prog->suboffset = min;
2737 prog->sublen = sublen;
2738 RX_MATCH_COPIED_on(rx);
2740 prog->subcoffset = prog->suboffset;
2741 if (prog->suboffset && utf8_target) {
2742 /* Convert byte offset to chars.
2743 * XXX ideally should only compute this if @-/@+
2744 * has been seen, a la PL_sawampersand ??? */
2746 /* If there's a direct correspondence between the
2747 * string which we're matching and the original SV,
2748 * then we can use the utf8 len cache associated with
2749 * the SV. In particular, it means that under //g,
2750 * sv_pos_b2u() will use the previously cached
2751 * position to speed up working out the new length of
2752 * subcoffset, rather than counting from the start of
2753 * the string each time. This stops
2754 * $x = "\x{100}" x 1E6; 1 while $x =~ /(.)/g;
2755 * from going quadratic */
2756 if (SvPOKp(sv) && SvPVX(sv) == strbeg)
2757 prog->subcoffset = sv_pos_b2u_flags(sv, prog->subcoffset,
2758 SV_GMAGIC|SV_CONST_RETURN);
2760 prog->subcoffset = utf8_length((U8*)strbeg,
2761 (U8*)(strbeg+prog->suboffset));
2765 RX_MATCH_COPY_FREE(rx);
2766 prog->subbeg = strbeg;
2767 prog->suboffset = 0;
2768 prog->subcoffset = 0;
2769 prog->sublen = strend - strbeg;
2777 - regexec_flags - match a regexp against a string
2780 Perl_regexec_flags(pTHX_ REGEXP * const rx, char *stringarg, char *strend,
2781 char *strbeg, SSize_t minend, SV *sv, void *data, U32 flags)
2782 /* stringarg: the point in the string at which to begin matching */
2783 /* strend: pointer to null at end of string */
2784 /* strbeg: real beginning of string */
2785 /* minend: end of match must be >= minend bytes after stringarg. */
2786 /* sv: SV being matched: only used for utf8 flag, pos() etc; string
2787 * itself is accessed via the pointers above */
2788 /* data: May be used for some additional optimizations.
2789 Currently unused. */
2790 /* flags: For optimizations. See REXEC_* in regexp.h */
2793 struct regexp *const prog = ReANY(rx);
2797 SSize_t minlen; /* must match at least this many chars */
2798 SSize_t dontbother = 0; /* how many characters not to try at end */
2799 const bool utf8_target = cBOOL(DO_UTF8(sv));
2801 RXi_GET_DECL(prog,progi);
2802 regmatch_info reginfo_buf; /* create some info to pass to regtry etc */
2803 regmatch_info *const reginfo = ®info_buf;
2804 regexp_paren_pair *swap = NULL;
2806 GET_RE_DEBUG_FLAGS_DECL;
2808 PERL_ARGS_ASSERT_REGEXEC_FLAGS;
2809 PERL_UNUSED_ARG(data);
2811 /* Be paranoid... */
2813 Perl_croak(aTHX_ "NULL regexp parameter");
2817 debug_start_match(rx, utf8_target, stringarg, strend,
2821 startpos = stringarg;
2823 if (prog->intflags & PREGf_GPOS_SEEN) {
2826 /* set reginfo->ganch, the position where \G can match */
2829 (flags & REXEC_IGNOREPOS)
2830 ? stringarg /* use start pos rather than pos() */
2831 : ((mg = mg_find_mglob(sv)) && mg->mg_len >= 0)
2832 /* Defined pos(): */
2833 ? strbeg + MgBYTEPOS(mg, sv, strbeg, strend-strbeg)
2834 : strbeg; /* pos() not defined; use start of string */
2836 DEBUG_GPOS_r(PerlIO_printf(Perl_debug_log,
2837 "GPOS ganch set to strbeg[%"IVdf"]\n", (IV)(reginfo->ganch - strbeg)));
2839 /* in the presence of \G, we may need to start looking earlier in
2840 * the string than the suggested start point of stringarg:
2841 * if prog->gofs is set, then that's a known, fixed minimum
2844 * /ab|c\G/: gofs = 1
2845 * or if the minimum offset isn't known, then we have to go back
2846 * to the start of the string, e.g. /w+\G/
2849 if (prog->intflags & PREGf_ANCH_GPOS) {
2850 startpos = reginfo->ganch - prog->gofs;
2852 ((flags & REXEC_FAIL_ON_UNDERFLOW) ? stringarg : strbeg))
2854 DEBUG_r(PerlIO_printf(Perl_debug_log,
2855 "fail: ganch-gofs before earliest possible start\n"));
2859 else if (prog->gofs) {
2860 if (startpos - prog->gofs < strbeg)
2863 startpos -= prog->gofs;
2865 else if (prog->intflags & PREGf_GPOS_FLOAT)
2869 minlen = prog->minlen;
2870 if ((startpos + minlen) > strend || startpos < strbeg) {
2871 DEBUG_r(PerlIO_printf(Perl_debug_log,
2872 "Regex match can't succeed, so not even tried\n"));
2876 /* at the end of this function, we'll do a LEAVE_SCOPE(oldsave),
2877 * which will call destuctors to reset PL_regmatch_state, free higher
2878 * PL_regmatch_slabs, and clean up regmatch_info_aux and
2879 * regmatch_info_aux_eval */
2881 oldsave = PL_savestack_ix;
2885 if ((prog->extflags & RXf_USE_INTUIT)
2886 && !(flags & REXEC_CHECKED))
2888 s = re_intuit_start(rx, sv, strbeg, startpos, strend,
2893 if (prog->extflags & RXf_CHECK_ALL) {
2894 /* we can match based purely on the result of INTUIT.
2895 * Set up captures etc just for $& and $-[0]
2896 * (an intuit-only match wont have $1,$2,..) */
2897 assert(!prog->nparens);
2899 /* s/// doesn't like it if $& is earlier than where we asked it to
2900 * start searching (which can happen on something like /.\G/) */
2901 if ( (flags & REXEC_FAIL_ON_UNDERFLOW)
2904 /* this should only be possible under \G */
2905 assert(prog->intflags & PREGf_GPOS_SEEN);
2906 DEBUG_EXECUTE_r(PerlIO_printf(Perl_debug_log,
2907 "matched, but failing for REXEC_FAIL_ON_UNDERFLOW\n"));
2911 /* match via INTUIT shouldn't have any captures.
2912 * Let @-, @+, $^N know */
2913 prog->lastparen = prog->lastcloseparen = 0;
2914 RX_MATCH_UTF8_set(rx, utf8_target);
2915 prog->offs[0].start = s - strbeg;
2916 prog->offs[0].end = utf8_target
2917 ? (char*)utf8_hop((U8*)s, prog->minlenret) - strbeg
2918 : s - strbeg + prog->minlenret;
2919 if ( !(flags & REXEC_NOT_FIRST) )
2920 S_reg_set_capture_string(aTHX_ rx,
2922 sv, flags, utf8_target);
2928 multiline = prog->extflags & RXf_PMf_MULTILINE;
2930 if (strend - s < (minlen+(prog->check_offset_min<0?prog->check_offset_min:0))) {
2931 DEBUG_EXECUTE_r(PerlIO_printf(Perl_debug_log,
2932 "String too short [regexec_flags]...\n"));
2936 /* Check validity of program. */
2937 if (UCHARAT(progi->program) != REG_MAGIC) {
2938 Perl_croak(aTHX_ "corrupted regexp program");
2941 RX_MATCH_TAINTED_off(rx);
2942 RX_MATCH_UTF8_set(rx, utf8_target);
2944 reginfo->prog = rx; /* Yes, sorry that this is confusing. */
2945 reginfo->intuit = 0;
2946 reginfo->is_utf8_target = cBOOL(utf8_target);
2947 reginfo->is_utf8_pat = cBOOL(RX_UTF8(rx));
2948 reginfo->warned = FALSE;
2949 reginfo->strbeg = strbeg;
2951 reginfo->poscache_maxiter = 0; /* not yet started a countdown */
2952 reginfo->strend = strend;
2953 /* see how far we have to get to not match where we matched before */
2954 reginfo->till = stringarg + minend;
2956 if (prog->extflags & RXf_EVAL_SEEN && SvPADTMP(sv)) {
2957 /* SAVEFREESV, not sv_mortalcopy, as this SV must last until after
2958 S_cleanup_regmatch_info_aux has executed (registered by
2959 SAVEDESTRUCTOR_X below). S_cleanup_regmatch_info_aux modifies
2960 magic belonging to this SV.
2961 Not newSVsv, either, as it does not COW.
2963 reginfo->sv = newSV(0);
2964 SvSetSV_nosteal(reginfo->sv, sv);
2965 SAVEFREESV(reginfo->sv);
2968 /* reserve next 2 or 3 slots in PL_regmatch_state:
2969 * slot N+0: may currently be in use: skip it
2970 * slot N+1: use for regmatch_info_aux struct
2971 * slot N+2: use for regmatch_info_aux_eval struct if we have (?{})'s
2972 * slot N+3: ready for use by regmatch()
2976 regmatch_state *old_regmatch_state;
2977 regmatch_slab *old_regmatch_slab;
2978 int i, max = (prog->extflags & RXf_EVAL_SEEN) ? 2 : 1;
2980 /* on first ever match, allocate first slab */
2981 if (!PL_regmatch_slab) {
2982 Newx(PL_regmatch_slab, 1, regmatch_slab);
2983 PL_regmatch_slab->prev = NULL;
2984 PL_regmatch_slab->next = NULL;
2985 PL_regmatch_state = SLAB_FIRST(PL_regmatch_slab);
2988 old_regmatch_state = PL_regmatch_state;
2989 old_regmatch_slab = PL_regmatch_slab;
2991 for (i=0; i <= max; i++) {
2993 reginfo->info_aux = &(PL_regmatch_state->u.info_aux);
2995 reginfo->info_aux_eval =
2996 reginfo->info_aux->info_aux_eval =
2997 &(PL_regmatch_state->u.info_aux_eval);
2999 if (++PL_regmatch_state > SLAB_LAST(PL_regmatch_slab))
3000 PL_regmatch_state = S_push_slab(aTHX);
3003 /* note initial PL_regmatch_state position; at end of match we'll
3004 * pop back to there and free any higher slabs */
3006 reginfo->info_aux->old_regmatch_state = old_regmatch_state;
3007 reginfo->info_aux->old_regmatch_slab = old_regmatch_slab;
3008 reginfo->info_aux->poscache = NULL;
3010 SAVEDESTRUCTOR_X(S_cleanup_regmatch_info_aux, reginfo->info_aux);
3012 if ((prog->extflags & RXf_EVAL_SEEN))
3013 S_setup_eval_state(aTHX_ reginfo);
3015 reginfo->info_aux_eval = reginfo->info_aux->info_aux_eval = NULL;
3018 /* If there is a "must appear" string, look for it. */
3020 if (PL_curpm && (PM_GETRE(PL_curpm) == rx)) {
3021 /* We have to be careful. If the previous successful match
3022 was from this regex we don't want a subsequent partially
3023 successful match to clobber the old results.
3024 So when we detect this possibility we add a swap buffer
3025 to the re, and switch the buffer each match. If we fail,
3026 we switch it back; otherwise we leave it swapped.
3029 /* do we need a save destructor here for eval dies? */
3030 Newxz(prog->offs, (prog->nparens + 1), regexp_paren_pair);
3031 DEBUG_BUFFERS_r(PerlIO_printf(Perl_debug_log,
3032 "rex=0x%"UVxf" saving offs: orig=0x%"UVxf" new=0x%"UVxf"\n",
3039 /* Simplest case: anchored match need be tried only once, or with
3040 * MBOL, only at the beginning of each line.
3042 * Note that /.*.../ sets PREGf_IMPLICIT|MBOL, while /.*.../s sets
3043 * PREGf_IMPLICIT|SBOL. The idea is that with /.*.../s, if it doesn't
3044 * match at the start of the string then it won't match anywhere else
3045 * either; while with /.*.../, if it doesn't match at the beginning,
3046 * the earliest it could match is at the start of the next line */
3048 if (prog->intflags & (PREGf_ANCH & ~PREGf_ANCH_GPOS)) {
3051 if (regtry(reginfo, &s))
3054 if (!(prog->intflags & PREGf_ANCH_MBOL))
3057 /* didn't match at start, try at other newline positions */
3060 dontbother = minlen - 1;
3061 end = HOP3c(strend, -dontbother, strbeg) - 1;
3063 /* skip to next newline */
3065 while (s <= end) { /* note it could be possible to match at the end of the string */
3066 /* NB: newlines are the same in unicode as they are in latin */
3069 if (prog->check_substr || prog->check_utf8) {
3070 /* note that with PREGf_IMPLICIT, intuit can only fail
3071 * or return the start position, so it's of limited utility.
3072 * Nevertheless, I made the decision that the potential for
3073 * quick fail was still worth it - DAPM */
3074 s = re_intuit_start(rx, sv, strbeg, s, strend, flags, NULL);
3078 if (regtry(reginfo, &s))
3082 } /* end anchored search */
3084 if (prog->intflags & PREGf_ANCH_GPOS)
3086 /* PREGf_ANCH_GPOS should never be true if PREGf_GPOS_SEEN is not true */
3087 assert(prog->intflags & PREGf_GPOS_SEEN);
3088 /* For anchored \G, the only position it can match from is
3089 * (ganch-gofs); we already set startpos to this above; if intuit
3090 * moved us on from there, we can't possibly succeed */
3091 assert(startpos == reginfo->ganch - prog->gofs);
3092 if (s == startpos && regtry(reginfo, &s))
3097 /* Messy cases: unanchored match. */
3098 if ((prog->anchored_substr || prog->anchored_utf8) && prog->intflags & PREGf_SKIP) {
3099 /* we have /x+whatever/ */
3100 /* it must be a one character string (XXXX Except is_utf8_pat?) */
3106 if (! prog->anchored_utf8) {
3107 to_utf8_substr(prog);
3109 ch = SvPVX_const(prog->anchored_utf8)[0];
3112 DEBUG_EXECUTE_r( did_match = 1 );
3113 if (regtry(reginfo, &s)) goto got_it;
3115 while (s < strend && *s == ch)
3122 if (! prog->anchored_substr) {
3123 if (! to_byte_substr(prog)) {
3124 NON_UTF8_TARGET_BUT_UTF8_REQUIRED(phooey);
3127 ch = SvPVX_const(prog->anchored_substr)[0];
3130 DEBUG_EXECUTE_r( did_match = 1 );
3131 if (regtry(reginfo, &s)) goto got_it;
3133 while (s < strend && *s == ch)
3138 DEBUG_EXECUTE_r(if (!did_match)
3139 PerlIO_printf(Perl_debug_log,
3140 "Did not find anchored character...\n")
3143 else if (prog->anchored_substr != NULL
3144 || prog->anchored_utf8 != NULL
3145 || ((prog->float_substr != NULL || prog->float_utf8 != NULL)
3146 && prog->float_max_offset < strend - s)) {
3151 char *last1; /* Last position checked before */
3155 if (prog->anchored_substr || prog->anchored_utf8) {
3157 if (! prog->anchored_utf8) {
3158 to_utf8_substr(prog);
3160 must = prog->anchored_utf8;
3163 if (! prog->anchored_substr) {
3164 if (! to_byte_substr(prog)) {
3165 NON_UTF8_TARGET_BUT_UTF8_REQUIRED(phooey);
3168 must = prog->anchored_substr;
3170 back_max = back_min = prog->anchored_offset;
3173 if (! prog->float_utf8) {
3174 to_utf8_substr(prog);
3176 must = prog->float_utf8;
3179 if (! prog->float_substr) {
3180 if (! to_byte_substr(prog)) {
3181 NON_UTF8_TARGET_BUT_UTF8_REQUIRED(phooey);
3184 must = prog->float_substr;
3186 back_max = prog->float_max_offset;
3187 back_min = prog->float_min_offset;
3193 last = HOP3c(strend, /* Cannot start after this */
3194 -(SSize_t)(CHR_SVLEN(must)
3195 - (SvTAIL(must) != 0) + back_min), strbeg);
3197 if (s > reginfo->strbeg)
3198 last1 = HOPc(s, -1);
3200 last1 = s - 1; /* bogus */
3202 /* XXXX check_substr already used to find "s", can optimize if
3203 check_substr==must. */
3205 strend = HOPc(strend, -dontbother);
3206 while ( (s <= last) &&
3207 (s = fbm_instr((unsigned char*)HOP4c(s, back_min, strbeg, strend),
3208 (unsigned char*)strend, must,
3209 multiline ? FBMrf_MULTILINE : 0)) ) {
3210 DEBUG_EXECUTE_r( did_match = 1 );
3211 if (HOPc(s, -back_max) > last1) {
3212 last1 = HOPc(s, -back_min);
3213 s = HOPc(s, -back_max);
3216 char * const t = (last1 >= reginfo->strbeg)
3217 ? HOPc(last1, 1) : last1 + 1;
3219 last1 = HOPc(s, -back_min);
3223 while (s <= last1) {
3224 if (regtry(reginfo, &s))
3227 s++; /* to break out of outer loop */
3234 while (s <= last1) {
3235 if (regtry(reginfo, &s))
3241 DEBUG_EXECUTE_r(if (!did_match) {
3242 RE_PV_QUOTED_DECL(quoted, utf8_target, PERL_DEBUG_PAD_ZERO(0),
3243 SvPVX_const(must), RE_SV_DUMPLEN(must), 30);
3244 PerlIO_printf(Perl_debug_log, "Did not find %s substr %s%s...\n",
3245 ((must == prog->anchored_substr || must == prog->anchored_utf8)
3246 ? "anchored" : "floating"),
3247 quoted, RE_SV_TAIL(must));
3251 else if ( (c = progi->regstclass) ) {
3253 const OPCODE op = OP(progi->regstclass);
3254 /* don't bother with what can't match */
3255 if (PL_regkind[op] != EXACT && PL_regkind[op] != TRIE)
3256 strend = HOPc(strend, -(minlen - 1));
3259 SV * const prop = sv_newmortal();
3260 regprop(prog, prop, c, reginfo, NULL);
3262 RE_PV_QUOTED_DECL(quoted,utf8_target,PERL_DEBUG_PAD_ZERO(1),
3264 PerlIO_printf(Perl_debug_log,
3265 "Matching stclass %.*s against %s (%d bytes)\n",
3266 (int)SvCUR(prop), SvPVX_const(prop),
3267 quoted, (int)(strend - s));
3270 if (find_byclass(prog, c, s, strend, reginfo))
3272 DEBUG_EXECUTE_r(PerlIO_printf(Perl_debug_log, "Contradicts stclass... [regexec_flags]\n"));
3276 if (prog->float_substr != NULL || prog->float_utf8 != NULL) {
3284 if (! prog->float_utf8) {
3285 to_utf8_substr(prog);
3287 float_real = prog->float_utf8;
3290 if (! prog->float_substr) {
3291 if (! to_byte_substr(prog)) {
3292 NON_UTF8_TARGET_BUT_UTF8_REQUIRED(phooey);
3295 float_real = prog->float_substr;
3298 little = SvPV_const(float_real, len);
3299 if (SvTAIL(float_real)) {
3300 /* This means that float_real contains an artificial \n on
3301 * the end due to the presence of something like this:
3302 * /foo$/ where we can match both "foo" and "foo\n" at the
3303 * end of the string. So we have to compare the end of the
3304 * string first against the float_real without the \n and
3305 * then against the full float_real with the string. We
3306 * have to watch out for cases where the string might be
3307 * smaller than the float_real or the float_real without
3309 char *checkpos= strend - len;
3311 PerlIO_printf(Perl_debug_log,
3312 "%sChecking for float_real.%s\n",
3313 PL_colors[4], PL_colors[5]));
3314 if (checkpos + 1 < strbeg) {
3315 /* can't match, even if we remove the trailing \n
3316 * string is too short to match */
3318 PerlIO_printf(Perl_debug_log,
3319 "%sString shorter than required trailing substring, cannot match.%s\n",
3320 PL_colors[4], PL_colors[5]));
3322 } else if (memEQ(checkpos + 1, little, len - 1)) {
3323 /* can match, the end of the string matches without the
3325 last = checkpos + 1;
3326 } else if (checkpos < strbeg) {
3327 /* cant match, string is too short when the "\n" is
3330 PerlIO_printf(Perl_debug_log,
3331 "%sString does not contain required trailing substring, cannot match.%s\n",
3332 PL_colors[4], PL_colors[5]));
3334 } else if (!multiline) {
3335 /* non multiline match, so compare with the "\n" at the
3336 * end of the string */
3337 if (memEQ(checkpos, little, len)) {
3341 PerlIO_printf(Perl_debug_log,
3342 "%sString does not contain required trailing substring, cannot match.%s\n",
3343 PL_colors[4], PL_colors[5]));
3347 /* multiline match, so we have to search for a place
3348 * where the full string is located */
3354 last = rninstr(s, strend, little, little + len);
3356 last = strend; /* matching "$" */
3359 /* at one point this block contained a comment which was
3360 * probably incorrect, which said that this was a "should not
3361 * happen" case. Even if it was true when it was written I am
3362 * pretty sure it is not anymore, so I have removed the comment
3363 * and replaced it with this one. Yves */
3365 PerlIO_printf(Perl_debug_log,
3366 "%sString does not contain required substring, cannot match.%s\n",
3367 PL_colors[4], PL_colors[5]
3371 dontbother = strend - last + prog->float_min_offset;
3373 if (minlen && (dontbother < minlen))
3374 dontbother = minlen - 1;
3375 strend -= dontbother; /* this one's always in bytes! */
3376 /* We don't know much -- general case. */
3379 if (regtry(reginfo, &s))
3388 if (regtry(reginfo, &s))
3390 } while (s++ < strend);
3398 /* s/// doesn't like it if $& is earlier than where we asked it to
3399 * start searching (which can happen on something like /.\G/) */
3400 if ( (flags & REXEC_FAIL_ON_UNDERFLOW)
3401 && (prog->offs[0].start < stringarg - strbeg))
3403 /* this should only be possible under \G */
3404 assert(prog->intflags & PREGf_GPOS_SEEN);
3405 DEBUG_EXECUTE_r(PerlIO_printf(Perl_debug_log,
3406 "matched, but failing for REXEC_FAIL_ON_UNDERFLOW\n"));
3412 PerlIO_printf(Perl_debug_log,
3413 "rex=0x%"UVxf" freeing offs: 0x%"UVxf"\n",
3420 /* clean up; this will trigger destructors that will free all slabs
3421 * above the current one, and cleanup the regmatch_info_aux
3422 * and regmatch_info_aux_eval sructs */
3424 LEAVE_SCOPE(oldsave);
3426 if (RXp_PAREN_NAMES(prog))
3427 (void)hv_iterinit(RXp_PAREN_NAMES(prog));
3429 /* make sure $`, $&, $', and $digit will work later */
3430 if ( !(flags & REXEC_NOT_FIRST) )
3431 S_reg_set_capture_string(aTHX_ rx,
3432 strbeg, reginfo->strend,
3433 sv, flags, utf8_target);
3438 DEBUG_EXECUTE_r(PerlIO_printf(Perl_debug_log, "%sMatch failed%s\n",
3439 PL_colors[4], PL_colors[5]));
3441 /* clean up; this will trigger destructors that will free all slabs
3442 * above the current one, and cleanup the regmatch_info_aux
3443 * and regmatch_info_aux_eval sructs */
3445 LEAVE_SCOPE(oldsave);
3448 /* we failed :-( roll it back */
3449 DEBUG_BUFFERS_r(PerlIO_printf(Perl_debug_log,
3450 "rex=0x%"UVxf" rolling back offs: freeing=0x%"UVxf" restoring=0x%"UVxf"\n",
3455 Safefree(prog->offs);
3462 /* Set which rex is pointed to by PL_reg_curpm, handling ref counting.
3463 * Do inc before dec, in case old and new rex are the same */
3464 #define SET_reg_curpm(Re2) \
3465 if (reginfo->info_aux_eval) { \
3466 (void)ReREFCNT_inc(Re2); \
3467 ReREFCNT_dec(PM_GETRE(PL_reg_curpm)); \
3468 PM_SETRE((PL_reg_curpm), (Re2)); \
3473 - regtry - try match at specific point
3475 STATIC bool /* 0 failure, 1 success */
3476 S_regtry(pTHX_ regmatch_info *reginfo, char **startposp)
3479 REGEXP *const rx = reginfo->prog;
3480 regexp *const prog = ReANY(rx);
3482 RXi_GET_DECL(prog,progi);
3483 GET_RE_DEBUG_FLAGS_DECL;
3485 PERL_ARGS_ASSERT_REGTRY;
3487 reginfo->cutpoint=NULL;
3489 prog->offs[0].start = *startposp - reginfo->strbeg;
3490 prog->lastparen = 0;
3491 prog->lastcloseparen = 0;
3493 /* XXXX What this code is doing here?!!! There should be no need
3494 to do this again and again, prog->lastparen should take care of
3497 /* Tests pat.t#187 and split.t#{13,14} seem to depend on this code.
3498 * Actually, the code in regcppop() (which Ilya may be meaning by
3499 * prog->lastparen), is not needed at all by the test suite
3500 * (op/regexp, op/pat, op/split), but that code is needed otherwise
3501 * this erroneously leaves $1 defined: "1" =~ /^(?:(\d)x)?\d$/
3502 * Meanwhile, this code *is* needed for the
3503 * above-mentioned test suite tests to succeed. The common theme
3504 * on those tests seems to be returning null fields from matches.
3505 * --jhi updated by dapm */
3507 if (prog->nparens) {
3508 regexp_paren_pair *pp = prog->offs;
3510 for (i = prog->nparens; i > (I32)prog->lastparen; i--) {
3518 result = regmatch(reginfo, *startposp, progi->program + 1);
3520 prog->offs[0].end = result;
3523 if (reginfo->cutpoint)
3524 *startposp= reginfo->cutpoint;
3525 REGCP_UNWIND(lastcp);
3530 #define sayYES goto yes
3531 #define sayNO goto no
3532 #define sayNO_SILENT goto no_silent
3534 /* we dont use STMT_START/END here because it leads to
3535 "unreachable code" warnings, which are bogus, but distracting. */
3536 #define CACHEsayNO \
3537 if (ST.cache_mask) \
3538 reginfo->info_aux->poscache[ST.cache_offset] |= ST.cache_mask; \
3541 /* this is used to determine how far from the left messages like
3542 'failed...' are printed. It should be set such that messages
3543 are inline with the regop output that created them.
3545 #define REPORT_CODE_OFF 32
3548 #define CHRTEST_UNINIT -1001 /* c1/c2 haven't been calculated yet */
3549 #define CHRTEST_VOID -1000 /* the c1/c2 "next char" test should be skipped */
3550 #define CHRTEST_NOT_A_CP_1 -999
3551 #define CHRTEST_NOT_A_CP_2 -998
3553 /* grab a new slab and return the first slot in it */
3555 STATIC regmatch_state *
3558 #if PERL_VERSION < 9 && !defined(PERL_CORE)
3561 regmatch_slab *s = PL_regmatch_slab->next;
3563 Newx(s, 1, regmatch_slab);
3564 s->prev = PL_regmatch_slab;
3566 PL_regmatch_slab->next = s;
3568 PL_regmatch_slab = s;
3569 return SLAB_FIRST(s);
3573 /* push a new state then goto it */
3575 #define PUSH_STATE_GOTO(state, node, input) \
3576 pushinput = input; \
3578 st->resume_state = state; \
3581 /* push a new state with success backtracking, then goto it */
3583 #define PUSH_YES_STATE_GOTO(state, node, input) \
3584 pushinput = input; \
3586 st->resume_state = state; \
3587 goto push_yes_state;
3594 regmatch() - main matching routine
3596 This is basically one big switch statement in a loop. We execute an op,
3597 set 'next' to point the next op, and continue. If we come to a point which
3598 we may need to backtrack to on failure such as (A|B|C), we push a
3599 backtrack state onto the backtrack stack. On failure, we pop the top
3600 state, and re-enter the loop at the state indicated. If there are no more
3601 states to pop, we return failure.
3603 Sometimes we also need to backtrack on success; for example /A+/, where
3604 after successfully matching one A, we need to go back and try to
3605 match another one; similarly for lookahead assertions: if the assertion
3606 completes successfully, we backtrack to the state just before the assertion
3607 and then carry on. In these cases, the pushed state is marked as
3608 'backtrack on success too'. This marking is in fact done by a chain of
3609 pointers, each pointing to the previous 'yes' state. On success, we pop to
3610 the nearest yes state, discarding any intermediate failure-only states.
3611 Sometimes a yes state is pushed just to force some cleanup code to be
3612 called at the end of a successful match or submatch; e.g. (??{$re}) uses
3613 it to free the inner regex.
3615 Note that failure backtracking rewinds the cursor position, while
3616 success backtracking leaves it alone.
3618 A pattern is complete when the END op is executed, while a subpattern
3619 such as (?=foo) is complete when the SUCCESS op is executed. Both of these
3620 ops trigger the "pop to last yes state if any, otherwise return true"
3623 A common convention in this function is to use A and B to refer to the two
3624 subpatterns (or to the first nodes thereof) in patterns like /A*B/: so A is
3625 the subpattern to be matched possibly multiple times, while B is the entire
3626 rest of the pattern. Variable and state names reflect this convention.
3628 The states in the main switch are the union of ops and failure/success of
3629 substates associated with with that op. For example, IFMATCH is the op
3630 that does lookahead assertions /(?=A)B/ and so the IFMATCH state means
3631 'execute IFMATCH'; while IFMATCH_A is a state saying that we have just
3632 successfully matched A and IFMATCH_A_fail is a state saying that we have
3633 just failed to match A. Resume states always come in pairs. The backtrack
3634 state we push is marked as 'IFMATCH_A', but when that is popped, we resume
3635 at IFMATCH_A or IFMATCH_A_fail, depending on whether we are backtracking
3636 on success or failure.
3638 The struct that holds a backtracking state is actually a big union, with
3639 one variant for each major type of op. The variable st points to the
3640 top-most backtrack struct. To make the code clearer, within each
3641 block of code we #define ST to alias the relevant union.
3643 Here's a concrete example of a (vastly oversimplified) IFMATCH
3649 #define ST st->u.ifmatch
3651 case IFMATCH: // we are executing the IFMATCH op, (?=A)B
3652 ST.foo = ...; // some state we wish to save
3654 // push a yes backtrack state with a resume value of
3655 // IFMATCH_A/IFMATCH_A_fail, then continue execution at the
3657 PUSH_YES_STATE_GOTO(IFMATCH_A, A, newinput);
3660 case IFMATCH_A: // we have successfully executed A; now continue with B
3662 bar = ST.foo; // do something with the preserved value
3665 case IFMATCH_A_fail: // A failed, so the assertion failed
3666 ...; // do some housekeeping, then ...
3667 sayNO; // propagate the failure
3674 For any old-timers reading this who are familiar with the old recursive
3675 approach, the code above is equivalent to:
3677 case IFMATCH: // we are executing the IFMATCH op, (?=A)B
3686 ...; // do some housekeeping, then ...
3687 sayNO; // propagate the failure
3690 The topmost backtrack state, pointed to by st, is usually free. If you
3691 want to claim it, populate any ST.foo fields in it with values you wish to
3692 save, then do one of
3694 PUSH_STATE_GOTO(resume_state, node, newinput);
3695 PUSH_YES_STATE_GOTO(resume_state, node, newinput);
3697 which sets that backtrack state's resume value to 'resume_state', pushes a
3698 new free entry to the top of the backtrack stack, then goes to 'node'.
3699 On backtracking, the free slot is popped, and the saved state becomes the
3700 new free state. An ST.foo field in this new top state can be temporarily
3701 accessed to retrieve values, but once the main loop is re-entered, it
3702 becomes available for reuse.
3704 Note that the depth of the backtrack stack constantly increases during the
3705 left-to-right execution of the pattern, rather than going up and down with
3706 the pattern nesting. For example the stack is at its maximum at Z at the
3707 end of the pattern, rather than at X in the following:
3709 /(((X)+)+)+....(Y)+....Z/
3711 The only exceptions to this are lookahead/behind assertions and the cut,
3712 (?>A), which pop all the backtrack states associated with A before
3715 Backtrack state structs are allocated in slabs of about 4K in size.
3716 PL_regmatch_state and st always point to the currently active state,
3717 and PL_regmatch_slab points to the slab currently containing
3718 PL_regmatch_state. The first time regmatch() is called, the first slab is
3719 allocated, and is never freed until interpreter destruction. When the slab
3720 is full, a new one is allocated and chained to the end. At exit from
3721 regmatch(), slabs allocated since entry are freed.
3726 #define DEBUG_STATE_pp(pp) \
3728 DUMP_EXEC_POS(locinput, scan, utf8_target); \
3729 PerlIO_printf(Perl_debug_log, \
3730 " %*s"pp" %s%s%s%s%s\n", \
3732 PL_reg_name[st->resume_state], \
3733 ((st==yes_state||st==mark_state) ? "[" : ""), \
3734 ((st==yes_state) ? "Y" : ""), \
3735 ((st==mark_state) ? "M" : ""), \
3736 ((st==yes_state||st==mark_state) ? "]" : "") \
3741 #define REG_NODE_NUM(x) ((x) ? (int)((x)-prog) : -1)
3746 S_debug_start_match(pTHX_ const REGEXP *prog, const bool utf8_target,
3747 const char *start, const char *end, const char *blurb)
3749 const bool utf8_pat = RX_UTF8(prog) ? 1 : 0;
3751 PERL_ARGS_ASSERT_DEBUG_START_MATCH;
3756 RE_PV_QUOTED_DECL(s0, utf8_pat, PERL_DEBUG_PAD_ZERO(0),
3757 RX_PRECOMP_const(prog), RX_PRELEN(prog), 60);
3759 RE_PV_QUOTED_DECL(s1, utf8_target, PERL_DEBUG_PAD_ZERO(1),
3760 start, end - start, 60);
3762 PerlIO_printf(Perl_debug_log,
3763 "%s%s REx%s %s against %s\n",
3764 PL_colors[4], blurb, PL_colors[5], s0, s1);
3766 if (utf8_target||utf8_pat)
3767 PerlIO_printf(Perl_debug_log, "UTF-8 %s%s%s...\n",
3768 utf8_pat ? "pattern" : "",
3769 utf8_pat && utf8_target ? " and " : "",
3770 utf8_target ? "string" : ""
3776 S_dump_exec_pos(pTHX_ const char *locinput,
3777 const regnode *scan,
3778 const char *loc_regeol,
3779 const char *loc_bostr,
3780 const char *loc_reg_starttry,
3781 const bool utf8_target)
3783 const int docolor = *PL_colors[0] || *PL_colors[2] || *PL_colors[4];
3784 const int taill = (docolor ? 10 : 7); /* 3 chars for "> <" */
3785 int l = (loc_regeol - locinput) > taill ? taill : (loc_regeol - locinput);
3786 /* The part of the string before starttry has one color
3787 (pref0_len chars), between starttry and current
3788 position another one (pref_len - pref0_len chars),
3789 after the current position the third one.
3790 We assume that pref0_len <= pref_len, otherwise we
3791 decrease pref0_len. */
3792 int pref_len = (locinput - loc_bostr) > (5 + taill) - l
3793 ? (5 + taill) - l : locinput - loc_bostr;
3796 PERL_ARGS_ASSERT_DUMP_EXEC_POS;
3798 while (utf8_target && UTF8_IS_CONTINUATION(*(U8*)(locinput - pref_len)))
3800 pref0_len = pref_len - (locinput - loc_reg_starttry);
3801 if (l + pref_len < (5 + taill) && l < loc_regeol - locinput)
3802 l = ( loc_regeol - locinput > (5 + taill) - pref_len
3803 ? (5 + taill) - pref_len : loc_regeol - locinput);
3804 while (utf8_target && UTF8_IS_CONTINUATION(*(U8*)(locinput + l)))
3808 if (pref0_len > pref_len)
3809 pref0_len = pref_len;
3811 const int is_uni = utf8_target ? 1 : 0;
3813 RE_PV_COLOR_DECL(s0,len0,is_uni,PERL_DEBUG_PAD(0),
3814 (locinput - pref_len),pref0_len, 60, 4, 5);
3816 RE_PV_COLOR_DECL(s1,len1,is_uni,PERL_DEBUG_PAD(1),
3817 (locinput - pref_len + pref0_len),
3818 pref_len - pref0_len, 60, 2, 3);
3820 RE_PV_COLOR_DECL(s2,len2,is_uni,PERL_DEBUG_PAD(2),
3821 locinput, loc_regeol - locinput, 10, 0, 1);
3823 const STRLEN tlen=len0+len1+len2;
3824 PerlIO_printf(Perl_debug_log,
3825 "%4"IVdf" <%.*s%.*s%s%.*s>%*s|",
3826 (IV)(locinput - loc_bostr),
3829 (docolor ? "" : "> <"),
3831 (int)(tlen > 19 ? 0 : 19 - tlen),
3838 /* reg_check_named_buff_matched()
3839 * Checks to see if a named buffer has matched. The data array of
3840 * buffer numbers corresponding to the buffer is expected to reside
3841 * in the regexp->data->data array in the slot stored in the ARG() of
3842 * node involved. Note that this routine doesn't actually care about the
3843 * name, that information is not preserved from compilation to execution.
3844 * Returns the index of the leftmost defined buffer with the given name
3845 * or 0 if non of the buffers matched.
3848 S_reg_check_named_buff_matched(const regexp *rex, const regnode *scan)
3851 RXi_GET_DECL(rex,rexi);
3852 SV *sv_dat= MUTABLE_SV(rexi->data->data[ ARG( scan ) ]);
3853 I32 *nums=(I32*)SvPVX(sv_dat);
3855 PERL_ARGS_ASSERT_REG_CHECK_NAMED_BUFF_MATCHED;
3857 for ( n=0; n<SvIVX(sv_dat); n++ ) {
3858 if ((I32)rex->lastparen >= nums[n] &&
3859 rex->offs[nums[n]].end != -1)
3869 S_setup_EXACTISH_ST_c1_c2(pTHX_ const regnode * const text_node, int *c1p,
3870 U8* c1_utf8, int *c2p, U8* c2_utf8, regmatch_info *reginfo)
3872 /* This function determines if there are one or two characters that match
3873 * the first character of the passed-in EXACTish node <text_node>, and if
3874 * so, returns them in the passed-in pointers.
3876 * If it determines that no possible character in the target string can
3877 * match, it returns FALSE; otherwise TRUE. (The FALSE situation occurs if
3878 * the first character in <text_node> requires UTF-8 to represent, and the
3879 * target string isn't in UTF-8.)
3881 * If there are more than two characters that could match the beginning of
3882 * <text_node>, or if more context is required to determine a match or not,
3883 * it sets both *<c1p> and *<c2p> to CHRTEST_VOID.
3885 * The motiviation behind this function is to allow the caller to set up
3886 * tight loops for matching. If <text_node> is of type EXACT, there is
3887 * only one possible character that can match its first character, and so
3888 * the situation is quite simple. But things get much more complicated if
3889 * folding is involved. It may be that the first character of an EXACTFish
3890 * node doesn't participate in any possible fold, e.g., punctuation, so it
3891 * can be matched only by itself. The vast majority of characters that are
3892 * in folds match just two things, their lower and upper-case equivalents.
3893 * But not all are like that; some have multiple possible matches, or match
3894 * sequences of more than one character. This function sorts all that out.
3896 * Consider the patterns A*B or A*?B where A and B are arbitrary. In a
3897 * loop of trying to match A*, we know we can't exit where the thing
3898 * following it isn't a B. And something can't be a B unless it is the
3899 * beginning of B. By putting a quick test for that beginning in a tight
3900 * loop, we can rule out things that can't possibly be B without having to
3901 * break out of the loop, thus avoiding work. Similarly, if A is a single
3902 * character, we can make a tight loop matching A*, using the outputs of
3905 * If the target string to match isn't in UTF-8, and there aren't
3906 * complications which require CHRTEST_VOID, *<c1p> and *<c2p> are set to
3907 * the one or two possible octets (which are characters in this situation)
3908 * that can match. In all cases, if there is only one character that can
3909 * match, *<c1p> and *<c2p> will be identical.
3911 * If the target string is in UTF-8, the buffers pointed to by <c1_utf8>
3912 * and <c2_utf8> will contain the one or two UTF-8 sequences of bytes that
3913 * can match the beginning of <text_node>. They should be declared with at
3914 * least length UTF8_MAXBYTES+1. (If the target string isn't in UTF-8, it is
3915 * undefined what these contain.) If one or both of the buffers are
3916 * invariant under UTF-8, *<c1p>, and *<c2p> will also be set to the
3917 * corresponding invariant. If variant, the corresponding *<c1p> and/or
3918 * *<c2p> will be set to a negative number(s) that shouldn't match any code
3919 * point (unless inappropriately coerced to unsigned). *<c1p> will equal
3920 * *<c2p> if and only if <c1_utf8> and <c2_utf8> are the same. */
3922 const bool utf8_target = reginfo->is_utf8_target;
3924 UV c1 = (UV)CHRTEST_NOT_A_CP_1;
3925 UV c2 = (UV)CHRTEST_NOT_A_CP_2;
3926 bool use_chrtest_void = FALSE;
3927 const bool is_utf8_pat = reginfo->is_utf8_pat;
3929 /* Used when we have both utf8 input and utf8 output, to avoid converting
3930 * to/from code points */
3931 bool utf8_has_been_setup = FALSE;
3935 U8 *pat = (U8*)STRING(text_node);
3936 U8 folded[UTF8_MAX_FOLD_CHAR_EXPAND * UTF8_MAXBYTES_CASE + 1] = { '\0' };
3938 if (OP(text_node) == EXACT || OP(text_node) == EXACTL) {
3940 /* In an exact node, only one thing can be matched, that first
3941 * character. If both the pat and the target are UTF-8, we can just
3942 * copy the input to the output, avoiding finding the code point of
3947 else if (utf8_target) {
3948 Copy(pat, c1_utf8, UTF8SKIP(pat), U8);
3949 Copy(pat, c2_utf8, UTF8SKIP(pat), U8);
3950 utf8_has_been_setup = TRUE;
3953 c2 = c1 = valid_utf8_to_uvchr(pat, NULL);
3956 else { /* an EXACTFish node */
3957 U8 *pat_end = pat + STR_LEN(text_node);
3959 /* An EXACTFL node has at least some characters unfolded, because what
3960 * they match is not known until now. So, now is the time to fold
3961 * the first few of them, as many as are needed to determine 'c1' and
3962 * 'c2' later in the routine. If the pattern isn't UTF-8, we only need
3963 * to fold if in a UTF-8 locale, and then only the Sharp S; everything
3964 * else is 1-1 and isn't assumed to be folded. In a UTF-8 pattern, we
3965 * need to fold as many characters as a single character can fold to,
3966 * so that later we can check if the first ones are such a multi-char
3967 * fold. But, in such a pattern only locale-problematic characters
3968 * aren't folded, so we can skip this completely if the first character
3969 * in the node isn't one of the tricky ones */
3970 if (OP(text_node) == EXACTFL) {
3972 if (! is_utf8_pat) {
3973 if (IN_UTF8_CTYPE_LOCALE && *pat == LATIN_SMALL_LETTER_SHARP_S)
3975 folded[0] = folded[1] = 's';
3977 pat_end = folded + 2;
3980 else if (is_PROBLEMATIC_LOCALE_FOLDEDS_START_utf8(pat)) {
3985 for (i = 0; i < UTF8_MAX_FOLD_CHAR_EXPAND && s < pat_end; i++) {
3987 *(d++) = (U8) toFOLD_LC(*s);
3992 _to_utf8_fold_flags(s,
3995 FOLD_FLAGS_FULL | FOLD_FLAGS_LOCALE);
4006 if ((is_utf8_pat && is_MULTI_CHAR_FOLD_utf8_safe(pat, pat_end))
4007 || (!is_utf8_pat && is_MULTI_CHAR_FOLD_latin1_safe(pat, pat_end)))
4009 /* Multi-character folds require more context to sort out. Also
4010 * PL_utf8_foldclosures used below doesn't handle them, so have to
4011 * be handled outside this routine */
4012 use_chrtest_void = TRUE;
4014 else { /* an EXACTFish node which doesn't begin with a multi-char fold */
4015 c1 = is_utf8_pat ? valid_utf8_to_uvchr(pat, NULL) : *pat;
4017 /* Load the folds hash, if not already done */
4019 if (! PL_utf8_foldclosures) {
4020 _load_PL_utf8_foldclosures();
4023 /* The fold closures data structure is a hash with the keys
4024 * being the UTF-8 of every character that is folded to, like
4025 * 'k', and the values each an array of all code points that
4026 * fold to its key. e.g. [ 'k', 'K', KELVIN_SIGN ].
4027 * Multi-character folds are not included */
4028 if ((! (listp = hv_fetch(PL_utf8_foldclosures,
4033 /* Not found in the hash, therefore there are no folds
4034 * containing it, so there is only a single character that
4038 else { /* Does participate in folds */
4039 AV* list = (AV*) *listp;
4040 if (av_tindex(list) != 1) {
4042 /* If there aren't exactly two folds to this, it is
4043 * outside the scope of this function */
4044 use_chrtest_void = TRUE;
4046 else { /* There are two. Get them */
4047 SV** c_p = av_fetch(list, 0, FALSE);
4049 Perl_croak(aTHX_ "panic: invalid PL_utf8_foldclosures structure");
4053 c_p = av_fetch(list, 1, FALSE);
4055 Perl_croak(aTHX_ "panic: invalid PL_utf8_foldclosures structure");
4059 /* Folds that cross the 255/256 boundary are forbidden
4060 * if EXACTFL (and isnt a UTF8 locale), or EXACTFA and
4061 * one is ASCIII. Since the pattern character is above
4062 * 255, and its only other match is below 256, the only
4063 * legal match will be to itself. We have thrown away
4064 * the original, so have to compute which is the one
4066 if ((c1 < 256) != (c2 < 256)) {
4067 if ((OP(text_node) == EXACTFL
4068 && ! IN_UTF8_CTYPE_LOCALE)
4069 || ((OP(text_node) == EXACTFA
4070 || OP(text_node) == EXACTFA_NO_TRIE)
4071 && (isASCII(c1) || isASCII(c2))))
4084 else /* Here, c1 is <= 255 */
4086 && HAS_NONLATIN1_FOLD_CLOSURE(c1)
4087 && ( ! (OP(text_node) == EXACTFL && ! IN_UTF8_CTYPE_LOCALE))
4088 && ((OP(text_node) != EXACTFA
4089 && OP(text_node) != EXACTFA_NO_TRIE)
4092 /* Here, there could be something above Latin1 in the target
4093 * which folds to this character in the pattern. All such
4094 * cases except LATIN SMALL LETTER Y WITH DIAERESIS have more
4095 * than two characters involved in their folds, so are outside
4096 * the scope of this function */
4097 if (UNLIKELY(c1 == LATIN_SMALL_LETTER_Y_WITH_DIAERESIS)) {
4098 c2 = LATIN_CAPITAL_LETTER_Y_WITH_DIAERESIS;
4101 use_chrtest_void = TRUE;
4104 else { /* Here nothing above Latin1 can fold to the pattern
4106 switch (OP(text_node)) {
4108 case EXACTFL: /* /l rules */
4109 c2 = PL_fold_locale[c1];
4112 case EXACTF: /* This node only generated for non-utf8
4114 assert(! is_utf8_pat);
4115 if (! utf8_target) { /* /d rules */
4120 /* /u rules for all these. This happens to work for
4121 * EXACTFA as nothing in Latin1 folds to ASCII */
4122 case EXACTFA_NO_TRIE: /* This node only generated for
4123 non-utf8 patterns */
4124 assert(! is_utf8_pat);
4129 c2 = PL_fold_latin1[c1];
4133 Perl_croak(aTHX_ "panic: Unexpected op %u", OP(text_node));
4134 NOT_REACHED; /* NOTREACHED */
4140 /* Here have figured things out. Set up the returns */
4141 if (use_chrtest_void) {
4142 *c2p = *c1p = CHRTEST_VOID;
4144 else if (utf8_target) {
4145 if (! utf8_has_been_setup) { /* Don't have the utf8; must get it */
4146 uvchr_to_utf8(c1_utf8, c1);
4147 uvchr_to_utf8(c2_utf8, c2);
4150 /* Invariants are stored in both the utf8 and byte outputs; Use
4151 * negative numbers otherwise for the byte ones. Make sure that the
4152 * byte ones are the same iff the utf8 ones are the same */
4153 *c1p = (UTF8_IS_INVARIANT(*c1_utf8)) ? *c1_utf8 : CHRTEST_NOT_A_CP_1;
4154 *c2p = (UTF8_IS_INVARIANT(*c2_utf8))
4157 ? CHRTEST_NOT_A_CP_1
4158 : CHRTEST_NOT_A_CP_2;
4160 else if (c1 > 255) {
4161 if (c2 > 255) { /* both possibilities are above what a non-utf8 string
4166 *c1p = *c2p = c2; /* c2 is the only representable value */
4168 else { /* c1 is representable; see about c2 */
4170 *c2p = (c2 < 256) ? c2 : c1;
4176 /* This creates a single number by combining two, with 'before' being like the
4177 * 10's digit, but this isn't necessarily base 10; it is base however many
4178 * elements of the enum there are */
4179 #define GCBcase(before, after) ((GCB_ENUM_COUNT * before) + after)
4182 S_isGCB(const GCB_enum before, const GCB_enum after)
4184 /* returns a boolean indicating if there is a Grapheme Cluster Boundary
4185 * between the inputs. See http://www.unicode.org/reports/tr29/ */
4187 switch (GCBcase(before, after)) {
4189 /* Break at the start and end of text.
4193 Break before and after controls except between CR and LF
4194 GB4. ( Control | CR | LF ) ÷
4195 GB5. ÷ ( Control | CR | LF )
4197 Otherwise, break everywhere.
4202 /* Do not break between a CR and LF.
4204 case GCBcase(GCB_CR, GCB_LF):
4207 /* Do not break Hangul syllable sequences.
4208 GB6. L × ( L | V | LV | LVT ) */
4209 case GCBcase(GCB_L, GCB_L):
4210 case GCBcase(GCB_L, GCB_V):
4211 case GCBcase(GCB_L, GCB_LV):
4212 case GCBcase(GCB_L, GCB_LVT):
4215 /* GB7. ( LV | V ) × ( V | T ) */
4216 case GCBcase(GCB_LV, GCB_V):
4217 case GCBcase(GCB_LV, GCB_T):
4218 case GCBcase(GCB_V, GCB_V):
4219 case GCBcase(GCB_V, GCB_T):
4222 /* GB8. ( LVT | T) × T */
4223 case GCBcase(GCB_LVT, GCB_T):
4224 case GCBcase(GCB_T, GCB_T):
4227 /* Do not break between regional indicator symbols.
4228 GB8a. Regional_Indicator × Regional_Indicator */
4229 case GCBcase(GCB_Regional_Indicator, GCB_Regional_Indicator):
4232 /* Do not break before extending characters.
4234 case GCBcase(GCB_Other, GCB_Extend):
4235 case GCBcase(GCB_Extend, GCB_Extend):
4236 case GCBcase(GCB_L, GCB_Extend):
4237 case GCBcase(GCB_LV, GCB_Extend):
4238 case GCBcase(GCB_LVT, GCB_Extend):
4239 case GCBcase(GCB_Prepend, GCB_Extend):
4240 case GCBcase(GCB_Regional_Indicator, GCB_Extend):
4241 case GCBcase(GCB_SpacingMark, GCB_Extend):
4242 case GCBcase(GCB_T, GCB_Extend):
4243 case GCBcase(GCB_V, GCB_Extend):
4246 /* Do not break before SpacingMarks, or after Prepend characters.
4247 GB9a. × SpacingMark */
4248 case GCBcase(GCB_Other, GCB_SpacingMark):
4249 case GCBcase(GCB_Extend, GCB_SpacingMark):
4250 case GCBcase(GCB_L, GCB_SpacingMark):
4251 case GCBcase(GCB_LV, GCB_SpacingMark):
4252 case GCBcase(GCB_LVT, GCB_SpacingMark):
4253 case GCBcase(GCB_Prepend, GCB_SpacingMark):
4254 case GCBcase(GCB_Regional_Indicator, GCB_SpacingMark):
4255 case GCBcase(GCB_SpacingMark, GCB_SpacingMark):
4256 case GCBcase(GCB_T, GCB_SpacingMark):
4257 case GCBcase(GCB_V, GCB_SpacingMark):
4260 /* GB9b. Prepend × */
4261 case GCBcase(GCB_Prepend, GCB_Other):
4262 case GCBcase(GCB_Prepend, GCB_L):
4263 case GCBcase(GCB_Prepend, GCB_LV):
4264 case GCBcase(GCB_Prepend, GCB_LVT):
4265 case GCBcase(GCB_Prepend, GCB_Prepend):
4266 case GCBcase(GCB_Prepend, GCB_Regional_Indicator):
4267 case GCBcase(GCB_Prepend, GCB_T):
4268 case GCBcase(GCB_Prepend, GCB_V):
4272 NOT_REACHED; /* NOTREACHED */
4275 #define SBcase(before, after) ((SB_ENUM_COUNT * before) + after)
4278 S_isSB(pTHX_ SB_enum before,
4280 const U8 * const strbeg,
4281 const U8 * const curpos,
4282 const U8 * const strend,
4283 const bool utf8_target)
4285 /* returns a boolean indicating if there is a Sentence Boundary Break
4286 * between the inputs. See http://www.unicode.org/reports/tr29/ */
4288 U8 * lpos = (U8 *) curpos;
4292 PERL_ARGS_ASSERT_ISSB;
4294 /* Break at the start and end of text.
4297 if (before == SB_EDGE || after == SB_EDGE) {
4301 /* SB 3: Do not break within CRLF. */
4302 if (before == SB_CR && after == SB_LF) {
4306 /* Break after paragraph separators. (though why CR and LF are considered
4307 * so is beyond me (khw)
4308 SB4. Sep | CR | LF ÷ */
4309 if (before == SB_Sep || before == SB_CR || before == SB_LF) {
4313 /* Ignore Format and Extend characters, except after sot, Sep, CR, or LF.
4314 * (See Section 6.2, Replacing Ignore Rules.)
4315 SB5. X (Extend | Format)* → X */
4316 if (after == SB_Extend || after == SB_Format) {
4320 if (before == SB_Extend || before == SB_Format) {
4321 before = backup_one_SB(strbeg, &lpos, utf8_target);
4324 /* Do not break after ambiguous terminators like period, if they are
4325 * immediately followed by a number or lowercase letter, if they are
4326 * between uppercase letters, if the first following letter (optionally
4327 * after certain punctuation) is lowercase, or if they are followed by
4328 * "continuation" punctuation such as comma, colon, or semicolon. For
4329 * example, a period may be an abbreviation or numeric period, and thus may
4330 * not mark the end of a sentence.
4332 * SB6. ATerm × Numeric */
4333 if (before == SB_ATerm && after == SB_Numeric) {
4337 /* SB7. (Upper | Lower) ATerm × Upper */
4338 if (before == SB_ATerm && after == SB_Upper) {
4340 backup = backup_one_SB(strbeg, &temp_pos, utf8_target);
4341 if (backup == SB_Upper || backup == SB_Lower) {
4346 /* SB8a. (STerm | ATerm) Close* Sp* × (SContinue | STerm | ATerm)
4347 * SB10. (STerm | ATerm) Close* Sp* × ( Sp | Sep | CR | LF ) */
4350 while (backup == SB_Sp) {
4351 backup = backup_one_SB(strbeg, &temp_pos, utf8_target);
4353 while (backup == SB_Close) {
4354 backup = backup_one_SB(strbeg, &temp_pos, utf8_target);
4356 if ((backup == SB_STerm || backup == SB_ATerm)
4357 && ( after == SB_SContinue
4358 || after == SB_STerm
4359 || after == SB_ATerm
4368 /* SB8. ATerm Close* Sp* × ( ¬(OLetter | Upper | Lower | Sep | CR | LF |
4369 * STerm | ATerm) )* Lower */
4370 if (backup == SB_ATerm) {
4371 U8 * rpos = (U8 *) curpos;
4372 SB_enum later = after;
4374 while ( later != SB_OLetter
4375 && later != SB_Upper
4376 && later != SB_Lower
4380 && later != SB_STerm
4381 && later != SB_ATerm
4382 && later != SB_EDGE)
4384 later = advance_one_SB(&rpos, strend, utf8_target);
4386 if (later == SB_Lower) {
4391 /* Break after sentence terminators, but include closing punctuation,
4392 * trailing spaces, and a paragraph separator (if present). [See note
4394 * SB9. ( STerm | ATerm ) Close* × ( Close | Sp | Sep | CR | LF ) */
4397 while (backup == SB_Close) {
4398 backup = backup_one_SB(strbeg, &temp_pos, utf8_target);
4400 if ((backup == SB_STerm || backup == SB_ATerm)
4401 && ( after == SB_Close
4411 /* SB11. ( STerm | ATerm ) Close* Sp* ( Sep | CR | LF )? ÷ */
4413 backup = backup_one_SB(strbeg, &temp_pos, utf8_target);
4414 if ( backup == SB_Sep
4423 while (backup == SB_Sp) {
4424 backup = backup_one_SB(strbeg, &lpos, utf8_target);
4426 while (backup == SB_Close) {
4427 backup = backup_one_SB(strbeg, &lpos, utf8_target);
4429 if (backup == SB_STerm || backup == SB_ATerm) {
4433 /* Otherwise, do not break.
4440 S_advance_one_SB(pTHX_ U8 ** curpos, const U8 * const strend, const bool utf8_target)
4444 PERL_ARGS_ASSERT_ADVANCE_ONE_SB;
4446 if (*curpos >= strend) {
4452 *curpos += UTF8SKIP(*curpos);
4453 if (*curpos >= strend) {
4456 sb = getSB_VAL_UTF8(*curpos, strend);
4457 } while (sb == SB_Extend || sb == SB_Format);
4462 if (*curpos >= strend) {
4465 sb = getSB_VAL_CP(**curpos);
4466 } while (sb == SB_Extend || sb == SB_Format);
4473 S_backup_one_SB(pTHX_ const U8 * const strbeg, U8 ** curpos, const bool utf8_target)
4477 PERL_ARGS_ASSERT_BACKUP_ONE_SB;
4479 if (*curpos < strbeg) {
4484 U8 * prev_char_pos = reghopmaybe3(*curpos, -1, strbeg);
4485 if (! prev_char_pos) {
4489 /* Back up over Extend and Format. curpos is always just to the right
4490 * of the characater whose value we are getting */
4492 U8 * prev_prev_char_pos;
4493 if ((prev_prev_char_pos = reghopmaybe3((U8 *) prev_char_pos, -1,
4496 sb = getSB_VAL_UTF8(prev_prev_char_pos, prev_char_pos);
4497 *curpos = prev_char_pos;
4498 prev_char_pos = prev_prev_char_pos;
4501 *curpos = (U8 *) strbeg;
4504 } while (sb == SB_Extend || sb == SB_Format);
4508 if (*curpos - 2 < strbeg) {
4509 *curpos = (U8 *) strbeg;
4513 sb = getSB_VAL_CP(*(*curpos - 1));
4514 } while (sb == SB_Extend || sb == SB_Format);
4520 #define WBcase(before, after) ((WB_ENUM_COUNT * before) + after)
4523 S_isWB(pTHX_ WB_enum previous,
4526 const U8 * const strbeg,
4527 const U8 * const curpos,
4528 const U8 * const strend,
4529 const bool utf8_target)
4531 /* Return a boolean as to if the boundary between 'before' and 'after' is
4532 * a Unicode word break, using their published algorithm. Context may be
4533 * needed to make this determination. If the value for the character
4534 * before 'before' is known, it is passed as 'previous'; otherwise that
4535 * should be set to WB_UNKNOWN. The other input parameters give the
4536 * boundaries and current position in the matching of the string. That
4537 * is, 'curpos' marks the position where the character whose wb value is
4538 * 'after' begins. See http://www.unicode.org/reports/tr29/ */
4540 U8 * before_pos = (U8 *) curpos;
4541 U8 * after_pos = (U8 *) curpos;
4543 PERL_ARGS_ASSERT_ISWB;
4545 /* WB1 and WB2: Break at the start and end of text. */
4546 if (before == WB_EDGE || after == WB_EDGE) {
4550 /* WB 3: Do not break within CRLF. */
4551 if (before == WB_CR && after == WB_LF) {
4555 /* WB 3a and WB 3b: Otherwise break before and after Newlines (including CR
4557 if ( before == WB_CR || before == WB_LF || before == WB_Newline
4558 || after == WB_CR || after == WB_LF || after == WB_Newline)
4563 /* Ignore Format and Extend characters, except when they appear at the
4564 * beginning of a region of text.
4565 * WB4. X (Extend | Format)* → X. */
4567 if (after == WB_Extend || after == WB_Format) {
4571 if (before == WB_Extend || before == WB_Format) {
4572 before = backup_one_WB(&previous, strbeg, &before_pos, utf8_target);
4575 switch (WBcase(before, after)) {
4576 /* Otherwise, break everywhere (including around ideographs).
4581 /* Do not break between most letters.
4582 WB5. (ALetter | Hebrew_Letter) × (ALetter | Hebrew_Letter) */
4583 case WBcase(WB_ALetter, WB_ALetter):
4584 case WBcase(WB_ALetter, WB_Hebrew_Letter):
4585 case WBcase(WB_Hebrew_Letter, WB_ALetter):
4586 case WBcase(WB_Hebrew_Letter, WB_Hebrew_Letter):
4589 /* Do not break letters across certain punctuation.
4590 WB6. (ALetter | Hebrew_Letter)
4591 × (MidLetter | MidNumLet | Single_Quote) (ALetter
4593 case WBcase(WB_ALetter, WB_MidLetter):
4594 case WBcase(WB_ALetter, WB_MidNumLet):
4595 case WBcase(WB_ALetter, WB_Single_Quote):
4596 case WBcase(WB_Hebrew_Letter, WB_MidLetter):
4597 case WBcase(WB_Hebrew_Letter, WB_MidNumLet):
4598 /*case WBcase(WB_Hebrew_Letter, WB_Single_Quote):*/
4599 after = advance_one_WB(&after_pos, strend, utf8_target);
4600 return after != WB_ALetter && after != WB_Hebrew_Letter;
4602 /* WB7. (ALetter | Hebrew_Letter) (MidLetter | MidNumLet |
4603 * Single_Quote) × (ALetter | Hebrew_Letter) */
4604 case WBcase(WB_MidLetter, WB_ALetter):
4605 case WBcase(WB_MidLetter, WB_Hebrew_Letter):
4606 case WBcase(WB_MidNumLet, WB_ALetter):
4607 case WBcase(WB_MidNumLet, WB_Hebrew_Letter):
4608 case WBcase(WB_Single_Quote, WB_ALetter):
4609 case WBcase(WB_Single_Quote, WB_Hebrew_Letter):
4611 = backup_one_WB(&previous, strbeg, &before_pos, utf8_target);
4612 return before != WB_ALetter && before != WB_Hebrew_Letter;
4614 /* WB7a. Hebrew_Letter × Single_Quote */
4615 case WBcase(WB_Hebrew_Letter, WB_Single_Quote):
4618 /* WB7b. Hebrew_Letter × Double_Quote Hebrew_Letter */
4619 case WBcase(WB_Hebrew_Letter, WB_Double_Quote):
4620 return advance_one_WB(&after_pos, strend, utf8_target)
4621 != WB_Hebrew_Letter;
4623 /* WB7c. Hebrew_Letter Double_Quote × Hebrew_Letter */
4624 case WBcase(WB_Double_Quote, WB_Hebrew_Letter):
4625 return backup_one_WB(&previous, strbeg, &before_pos, utf8_target)
4626 != WB_Hebrew_Letter;
4628 /* Do not break within sequences of digits, or digits adjacent to
4629 * letters (“3a”, or “A3”).
4630 WB8. Numeric × Numeric */
4631 case WBcase(WB_Numeric, WB_Numeric):
4634 /* WB9. (ALetter | Hebrew_Letter) × Numeric */
4635 case WBcase(WB_ALetter, WB_Numeric):
4636 case WBcase(WB_Hebrew_Letter, WB_Numeric):
4639 /* WB10. Numeric × (ALetter | Hebrew_Letter) */
4640 case WBcase(WB_Numeric, WB_ALetter):
4641 case WBcase(WB_Numeric, WB_Hebrew_Letter):
4644 /* Do not break within sequences, such as “3.2” or “3,456.789”.
4645 WB11. Numeric (MidNum | MidNumLet | Single_Quote) × Numeric
4647 case WBcase(WB_MidNum, WB_Numeric):
4648 case WBcase(WB_MidNumLet, WB_Numeric):
4649 case WBcase(WB_Single_Quote, WB_Numeric):
4650 return backup_one_WB(&previous, strbeg, &before_pos, utf8_target)
4653 /* WB12. Numeric × (MidNum | MidNumLet | Single_Quote) Numeric
4655 case WBcase(WB_Numeric, WB_MidNum):
4656 case WBcase(WB_Numeric, WB_MidNumLet):
4657 case WBcase(WB_Numeric, WB_Single_Quote):
4658 return advance_one_WB(&after_pos, strend, utf8_target)
4661 /* Do not break between Katakana.
4662 WB13. Katakana × Katakana */
4663 case WBcase(WB_Katakana, WB_Katakana):
4666 /* Do not break from extenders.
4667 WB13a. (ALetter | Hebrew_Letter | Numeric | Katakana |
4668 ExtendNumLet) × ExtendNumLet */
4669 case WBcase(WB_ALetter, WB_ExtendNumLet):
4670 case WBcase(WB_Hebrew_Letter, WB_ExtendNumLet):
4671 case WBcase(WB_Numeric, WB_ExtendNumLet):
4672 case WBcase(WB_Katakana, WB_ExtendNumLet):
4673 case WBcase(WB_ExtendNumLet, WB_ExtendNumLet):
4676 /* WB13b. ExtendNumLet × (ALetter | Hebrew_Letter | Numeric
4678 case WBcase(WB_ExtendNumLet, WB_ALetter):
4679 case WBcase(WB_ExtendNumLet, WB_Hebrew_Letter):
4680 case WBcase(WB_ExtendNumLet, WB_Numeric):
4681 case WBcase(WB_ExtendNumLet, WB_Katakana):
4684 /* Do not break between regional indicator symbols.
4685 WB13c. Regional_Indicator × Regional_Indicator */
4686 case WBcase(WB_Regional_Indicator, WB_Regional_Indicator):
4691 NOT_REACHED; /* NOTREACHED */
4695 S_advance_one_WB(pTHX_ U8 ** curpos, const U8 * const strend, const bool utf8_target)
4699 PERL_ARGS_ASSERT_ADVANCE_ONE_WB;
4701 if (*curpos >= strend) {
4707 /* Advance over Extend and Format */
4709 *curpos += UTF8SKIP(*curpos);
4710 if (*curpos >= strend) {
4713 wb = getWB_VAL_UTF8(*curpos, strend);
4714 } while (wb == WB_Extend || wb == WB_Format);
4719 if (*curpos >= strend) {
4722 wb = getWB_VAL_CP(**curpos);
4723 } while (wb == WB_Extend || wb == WB_Format);
4730 S_backup_one_WB(pTHX_ WB_enum * previous, const U8 * const strbeg, U8 ** curpos, const bool utf8_target)
4734 PERL_ARGS_ASSERT_BACKUP_ONE_WB;
4736 /* If we know what the previous character's break value is, don't have
4738 if (*previous != WB_UNKNOWN) {
4741 /* But we need to move backwards by one */
4743 *curpos = reghopmaybe3(*curpos, -1, strbeg);
4745 *previous = WB_EDGE;
4746 *curpos = (U8 *) strbeg;
4749 *previous = WB_UNKNOWN;
4754 *previous = (*curpos <= strbeg) ? WB_EDGE : WB_UNKNOWN;
4757 /* And we always back up over these two types */
4758 if (wb != WB_Extend && wb != WB_Format) {
4763 if (*curpos < strbeg) {
4768 U8 * prev_char_pos = reghopmaybe3(*curpos, -1, strbeg);
4769 if (! prev_char_pos) {
4773 /* Back up over Extend and Format. curpos is always just to the right
4774 * of the characater whose value we are getting */
4776 U8 * prev_prev_char_pos;
4777 if ((prev_prev_char_pos = reghopmaybe3((U8 *) prev_char_pos,
4781 wb = getWB_VAL_UTF8(prev_prev_char_pos, prev_char_pos);
4782 *curpos = prev_char_pos;
4783 prev_char_pos = prev_prev_char_pos;
4786 *curpos = (U8 *) strbeg;
4789 } while (wb == WB_Extend || wb == WB_Format);
4793 if (*curpos - 2 < strbeg) {
4794 *curpos = (U8 *) strbeg;
4798 wb = getWB_VAL_CP(*(*curpos - 1));
4799 } while (wb == WB_Extend || wb == WB_Format);
4805 /* returns -1 on failure, $+[0] on success */
4807 S_regmatch(pTHX_ regmatch_info *reginfo, char *startpos, regnode *prog)
4809 #if PERL_VERSION < 9 && !defined(PERL_CORE)
4813 const bool utf8_target = reginfo->is_utf8_target;
4814 const U32 uniflags = UTF8_ALLOW_DEFAULT;
4815 REGEXP *rex_sv = reginfo->prog;
4816 regexp *rex = ReANY(rex_sv);
4817 RXi_GET_DECL(rex,rexi);
4818 /* the current state. This is a cached copy of PL_regmatch_state */
4820 /* cache heavy used fields of st in registers */
4823 U32 n = 0; /* general value; init to avoid compiler warning */
4824 SSize_t ln = 0; /* len or last; init to avoid compiler warning */
4825 char *locinput = startpos;
4826 char *pushinput; /* where to continue after a PUSH */
4827 I32 nextchr; /* is always set to UCHARAT(locinput) */
4829 bool result = 0; /* return value of S_regmatch */
4830 int depth = 0; /* depth of backtrack stack */
4831 U32 nochange_depth = 0; /* depth of GOSUB recursion with nochange */
4832 const U32 max_nochange_depth =
4833 (3 * rex->nparens > MAX_RECURSE_EVAL_NOCHANGE_DEPTH) ?
4834 3 * rex->nparens : MAX_RECURSE_EVAL_NOCHANGE_DEPTH;
4835 regmatch_state *yes_state = NULL; /* state to pop to on success of
4837 /* mark_state piggy backs on the yes_state logic so that when we unwind
4838 the stack on success we can update the mark_state as we go */
4839 regmatch_state *mark_state = NULL; /* last mark state we have seen */
4840 regmatch_state *cur_eval = NULL; /* most recent EVAL_AB state */
4841 struct regmatch_state *cur_curlyx = NULL; /* most recent curlyx */
4843 bool no_final = 0; /* prevent failure from backtracking? */
4844 bool do_cutgroup = 0; /* no_final only until next branch/trie entry */
4845 char *startpoint = locinput;
4846 SV *popmark = NULL; /* are we looking for a mark? */
4847 SV *sv_commit = NULL; /* last mark name seen in failure */
4848 SV *sv_yes_mark = NULL; /* last mark name we have seen
4849 during a successful match */
4850 U32 lastopen = 0; /* last open we saw */
4851 bool has_cutgroup = RX_HAS_CUTGROUP(rex) ? 1 : 0;
4852 SV* const oreplsv = GvSVn(PL_replgv);
4853 /* these three flags are set by various ops to signal information to
4854 * the very next op. They have a useful lifetime of exactly one loop
4855 * iteration, and are not preserved or restored by state pushes/pops
4857 bool sw = 0; /* the condition value in (?(cond)a|b) */
4858 bool minmod = 0; /* the next "{n,m}" is a "{n,m}?" */
4859 int logical = 0; /* the following EVAL is:
4863 or the following IFMATCH/UNLESSM is:
4864 false: plain (?=foo)
4865 true: used as a condition: (?(?=foo))
4867 PAD* last_pad = NULL;
4869 I32 gimme = G_SCALAR;
4870 CV *caller_cv = NULL; /* who called us */
4871 CV *last_pushed_cv = NULL; /* most recently called (?{}) CV */
4872 CHECKPOINT runops_cp; /* savestack position before executing EVAL */
4873 U32 maxopenparen = 0; /* max '(' index seen so far */
4874 int to_complement; /* Invert the result? */
4875 _char_class_number classnum;
4876 bool is_utf8_pat = reginfo->is_utf8_pat;
4881 GET_RE_DEBUG_FLAGS_DECL;
4884 /* protect against undef(*^R) */
4885 SAVEFREESV(SvREFCNT_inc_simple_NN(oreplsv));
4887 /* shut up 'may be used uninitialized' compiler warnings for dMULTICALL */
4888 multicall_oldcatch = 0;
4889 multicall_cv = NULL;
4891 PERL_UNUSED_VAR(multicall_cop);
4892 PERL_UNUSED_VAR(newsp);
4895 PERL_ARGS_ASSERT_REGMATCH;
4897 DEBUG_OPTIMISE_r( DEBUG_EXECUTE_r({
4898 PerlIO_printf(Perl_debug_log,"regmatch start\n");
4901 st = PL_regmatch_state;
4903 /* Note that nextchr is a byte even in UTF */
4906 while (scan != NULL) {
4909 SV * const prop = sv_newmortal();
4910 regnode *rnext=regnext(scan);
4911 DUMP_EXEC_POS( locinput, scan, utf8_target );
4912 regprop(rex, prop, scan, reginfo, NULL);
4914 PerlIO_printf(Perl_debug_log,
4915 "%3"IVdf":%*s%s(%"IVdf")\n",
4916 (IV)(scan - rexi->program), depth*2, "",
4918 (PL_regkind[OP(scan)] == END || !rnext) ?
4919 0 : (IV)(rnext - rexi->program));
4922 next = scan + NEXT_OFF(scan);
4925 state_num = OP(scan);
4931 assert(nextchr < 256 && (nextchr >= 0 || nextchr == NEXTCHR_EOS));
4933 switch (state_num) {
4934 case SBOL: /* /^../ and /\A../ */
4935 if (locinput == reginfo->strbeg)
4939 case MBOL: /* /^../m */
4940 if (locinput == reginfo->strbeg ||
4941 (!NEXTCHR_IS_EOS && locinput[-1] == '\n'))
4948 if (locinput == reginfo->ganch)
4952 case KEEPS: /* \K */
4953 /* update the startpoint */
4954 st->u.keeper.val = rex->offs[0].start;
4955 rex->offs[0].start = locinput - reginfo->strbeg;
4956 PUSH_STATE_GOTO(KEEPS_next, next, locinput);
4958 NOT_REACHED; /* NOTREACHED */
4960 case KEEPS_next_fail:
4961 /* rollback the start point change */
4962 rex->offs[0].start = st->u.keeper.val;
4965 NOT_REACHED; /* NOTREACHED */
4967 case MEOL: /* /..$/m */
4968 if (!NEXTCHR_IS_EOS && nextchr != '\n')
4972 case SEOL: /* /..$/ */
4973 if (!NEXTCHR_IS_EOS && nextchr != '\n')
4975 if (reginfo->strend - locinput > 1)
4980 if (!NEXTCHR_IS_EOS)
4984 case SANY: /* /./s */
4987 goto increment_locinput;
4989 case REG_ANY: /* /./ */
4990 if ((NEXTCHR_IS_EOS) || nextchr == '\n')
4992 goto increment_locinput;
4996 #define ST st->u.trie
4997 case TRIEC: /* (ab|cd) with known charclass */
4998 /* In this case the charclass data is available inline so
4999 we can fail fast without a lot of extra overhead.
5001 if(!NEXTCHR_IS_EOS && !ANYOF_BITMAP_TEST(scan, nextchr)) {
5003 PerlIO_printf(Perl_debug_log,
5004 "%*s %sfailed to match trie start class...%s\n",
5005 REPORT_CODE_OFF+depth*2, "", PL_colors[4], PL_colors[5])
5009 NOT_REACHED; /* NOTREACHED */
5012 case TRIE: /* (ab|cd) */
5013 /* the basic plan of execution of the trie is:
5014 * At the beginning, run though all the states, and
5015 * find the longest-matching word. Also remember the position
5016 * of the shortest matching word. For example, this pattern:
5019 * when matched against the string "abcde", will generate
5020 * accept states for all words except 3, with the longest
5021 * matching word being 4, and the shortest being 2 (with
5022 * the position being after char 1 of the string).
5024 * Then for each matching word, in word order (i.e. 1,2,4,5),
5025 * we run the remainder of the pattern; on each try setting
5026 * the current position to the character following the word,
5027 * returning to try the next word on failure.
5029 * We avoid having to build a list of words at runtime by
5030 * using a compile-time structure, wordinfo[].prev, which
5031 * gives, for each word, the previous accepting word (if any).
5032 * In the case above it would contain the mappings 1->2, 2->0,
5033 * 3->0, 4->5, 5->1. We can use this table to generate, from
5034 * the longest word (4 above), a list of all words, by
5035 * following the list of prev pointers; this gives us the
5036 * unordered list 4,5,1,2. Then given the current word we have
5037 * just tried, we can go through the list and find the
5038 * next-biggest word to try (so if we just failed on word 2,
5039 * the next in the list is 4).
5041 * Since at runtime we don't record the matching position in
5042 * the string for each word, we have to work that out for
5043 * each word we're about to process. The wordinfo table holds
5044 * the character length of each word; given that we recorded
5045 * at the start: the position of the shortest word and its
5046 * length in chars, we just need to move the pointer the
5047 * difference between the two char lengths. Depending on
5048 * Unicode status and folding, that's cheap or expensive.
5050 * This algorithm is optimised for the case where are only a
5051 * small number of accept states, i.e. 0,1, or maybe 2.
5052 * With lots of accepts states, and having to try all of them,
5053 * it becomes quadratic on number of accept states to find all
5058 /* what type of TRIE am I? (utf8 makes this contextual) */
5059 DECL_TRIE_TYPE(scan);
5061 /* what trie are we using right now */
5062 reg_trie_data * const trie
5063 = (reg_trie_data*)rexi->data->data[ ARG( scan ) ];
5064 HV * widecharmap = MUTABLE_HV(rexi->data->data[ ARG( scan ) + 1 ]);
5065 U32 state = trie->startstate;
5067 if (scan->flags == EXACTL || scan->flags == EXACTFLU8) {
5068 _CHECK_AND_WARN_PROBLEMATIC_LOCALE;
5070 && UTF8_IS_ABOVE_LATIN1(nextchr)
5071 && scan->flags == EXACTL)
5073 /* We only output for EXACTL, as we let the folder
5074 * output this message for EXACTFLU8 to avoid
5076 _CHECK_AND_OUTPUT_WIDE_LOCALE_UTF8_MSG(locinput,
5081 && (NEXTCHR_IS_EOS || !TRIE_BITMAP_TEST(trie, nextchr)))
5083 if (trie->states[ state ].wordnum) {
5085 PerlIO_printf(Perl_debug_log,
5086 "%*s %smatched empty string...%s\n",
5087 REPORT_CODE_OFF+depth*2, "", PL_colors[4], PL_colors[5])
5093 PerlIO_printf(Perl_debug_log,
5094 "%*s %sfailed to match trie start class...%s\n",
5095 REPORT_CODE_OFF+depth*2, "", PL_colors[4], PL_colors[5])
5102 U8 *uc = ( U8* )locinput;
5106 U8 *uscan = (U8*)NULL;
5107 U8 foldbuf[ UTF8_MAXBYTES_CASE + 1 ];
5108 U32 charcount = 0; /* how many input chars we have matched */
5109 U32 accepted = 0; /* have we seen any accepting states? */
5111 ST.jump = trie->jump;
5114 ST.longfold = FALSE; /* char longer if folded => it's harder */
5117 /* fully traverse the TRIE; note the position of the
5118 shortest accept state and the wordnum of the longest
5121 while ( state && uc <= (U8*)(reginfo->strend) ) {
5122 U32 base = trie->states[ state ].trans.base;
5126 wordnum = trie->states[ state ].wordnum;
5128 if (wordnum) { /* it's an accept state */
5131 /* record first match position */
5133 ST.firstpos = (U8*)locinput;
5138 ST.firstchars = charcount;
5141 if (!ST.nextword || wordnum < ST.nextword)
5142 ST.nextword = wordnum;
5143 ST.topword = wordnum;
5146 DEBUG_TRIE_EXECUTE_r({
5147 DUMP_EXEC_POS( (char *)uc, scan, utf8_target );
5148 PerlIO_printf( Perl_debug_log,
5149 "%*s %sState: %4"UVxf" Accepted: %c ",
5150 2+depth * 2, "", PL_colors[4],
5151 (UV)state, (accepted ? 'Y' : 'N'));
5154 /* read a char and goto next state */
5155 if ( base && (foldlen || uc < (U8*)(reginfo->strend))) {
5157 REXEC_TRIE_READ_CHAR(trie_type, trie, widecharmap, uc,
5158 uscan, len, uvc, charid, foldlen,
5165 base + charid - 1 - trie->uniquecharcount)) >= 0)
5167 && ((U32)offset < trie->lasttrans)
5168 && trie->trans[offset].check == state)
5170 state = trie->trans[offset].next;
5181 DEBUG_TRIE_EXECUTE_r(
5182 PerlIO_printf( Perl_debug_log,
5183 "Charid:%3x CP:%4"UVxf" After State: %4"UVxf"%s\n",
5184 charid, uvc, (UV)state, PL_colors[5] );
5190 /* calculate total number of accept states */
5195 w = trie->wordinfo[w].prev;
5198 ST.accepted = accepted;
5202 PerlIO_printf( Perl_debug_log,
5203 "%*s %sgot %"IVdf" possible matches%s\n",
5204 REPORT_CODE_OFF + depth * 2, "",
5205 PL_colors[4], (IV)ST.accepted, PL_colors[5] );
5207 goto trie_first_try; /* jump into the fail handler */
5210 NOT_REACHED; /* NOTREACHED */
5212 case TRIE_next_fail: /* we failed - try next alternative */
5216 REGCP_UNWIND(ST.cp);
5217 UNWIND_PAREN(ST.lastparen, ST.lastcloseparen);
5219 if (!--ST.accepted) {
5221 PerlIO_printf( Perl_debug_log,
5222 "%*s %sTRIE failed...%s\n",
5223 REPORT_CODE_OFF+depth*2, "",
5230 /* Find next-highest word to process. Note that this code
5231 * is O(N^2) per trie run (O(N) per branch), so keep tight */
5234 U16 const nextword = ST.nextword;
5235 reg_trie_wordinfo * const wordinfo
5236 = ((reg_trie_data*)rexi->data->data[ARG(ST.me)])->wordinfo;
5237 for (word=ST.topword; word; word=wordinfo[word].prev) {
5238 if (word > nextword && (!min || word < min))
5251 ST.lastparen = rex->lastparen;
5252 ST.lastcloseparen = rex->lastcloseparen;
5256 /* find start char of end of current word */
5258 U32 chars; /* how many chars to skip */
5259 reg_trie_data * const trie
5260 = (reg_trie_data*)rexi->data->data[ARG(ST.me)];
5262 assert((trie->wordinfo[ST.nextword].len - trie->prefixlen)
5264 chars = (trie->wordinfo[ST.nextword].len - trie->prefixlen)
5269 /* the hard option - fold each char in turn and find
5270 * its folded length (which may be different */
5271 U8 foldbuf[UTF8_MAXBYTES_CASE + 1];
5279 uvc = utf8n_to_uvchr((U8*)uc, UTF8_MAXLEN, &len,
5287 uvc = to_uni_fold(uvc, foldbuf, &foldlen);
5292 uvc = utf8n_to_uvchr(uscan, UTF8_MAXLEN, &len,
5308 scan = ST.me + ((ST.jump && ST.jump[ST.nextword])
5309 ? ST.jump[ST.nextword]
5313 PerlIO_printf( Perl_debug_log,
5314 "%*s %sTRIE matched word #%d, continuing%s\n",
5315 REPORT_CODE_OFF+depth*2, "",
5322 if (ST.accepted > 1 || has_cutgroup) {
5323 PUSH_STATE_GOTO(TRIE_next, scan, (char*)uc);
5325 NOT_REACHED; /* NOTREACHED */
5327 /* only one choice left - just continue */
5329 AV *const trie_words
5330 = MUTABLE_AV(rexi->data->data[ARG(ST.me)+TRIE_WORDS_OFFSET]);
5331 SV ** const tmp = trie_words
5332 ? av_fetch(trie_words, ST.nextword - 1, 0) : NULL;
5333 SV *sv= tmp ? sv_newmortal() : NULL;
5335 PerlIO_printf( Perl_debug_log,
5336 "%*s %sonly one match left, short-circuiting: #%d <%s>%s\n",
5337 REPORT_CODE_OFF+depth*2, "", PL_colors[4],
5339 tmp ? pv_pretty(sv, SvPV_nolen_const(*tmp), SvCUR(*tmp), 0,
5340 PL_colors[0], PL_colors[1],
5341 (SvUTF8(*tmp) ? PERL_PV_ESCAPE_UNI : 0)|PERL_PV_ESCAPE_NONASCII
5343 : "not compiled under -Dr",
5347 locinput = (char*)uc;
5348 continue; /* execute rest of RE */
5353 case EXACTL: /* /abc/l */
5354 _CHECK_AND_WARN_PROBLEMATIC_LOCALE;
5356 /* Complete checking would involve going through every character
5357 * matched by the string to see if any is above latin1. But the
5358 * comparision otherwise might very well be a fast assembly
5359 * language routine, and I (khw) don't think slowing things down
5360 * just to check for this warning is worth it. So this just checks
5361 * the first character */
5362 if (utf8_target && UTF8_IS_ABOVE_LATIN1(*locinput)) {
5363 _CHECK_AND_OUTPUT_WIDE_LOCALE_UTF8_MSG(locinput, reginfo->strend);
5366 case EXACT: { /* /abc/ */
5367 char *s = STRING(scan);
5369 if (utf8_target != is_utf8_pat) {
5370 /* The target and the pattern have differing utf8ness. */
5372 const char * const e = s + ln;
5375 /* The target is utf8, the pattern is not utf8.
5376 * Above-Latin1 code points can't match the pattern;
5377 * invariants match exactly, and the other Latin1 ones need
5378 * to be downgraded to a single byte in order to do the
5379 * comparison. (If we could be confident that the target
5380 * is not malformed, this could be refactored to have fewer
5381 * tests by just assuming that if the first bytes match, it
5382 * is an invariant, but there are tests in the test suite
5383 * dealing with (??{...}) which violate this) */
5385 if (l >= reginfo->strend
5386 || UTF8_IS_ABOVE_LATIN1(* (U8*) l))
5390 if (UTF8_IS_INVARIANT(*(U8*)l)) {
5397 if (EIGHT_BIT_UTF8_TO_NATIVE(*l, *(l+1)) != * (U8*) s)
5407 /* The target is not utf8, the pattern is utf8. */
5409 if (l >= reginfo->strend
5410 || UTF8_IS_ABOVE_LATIN1(* (U8*) s))
5414 if (UTF8_IS_INVARIANT(*(U8*)s)) {
5421 if (EIGHT_BIT_UTF8_TO_NATIVE(*s, *(s+1)) != * (U8*) l)
5433 /* The target and the pattern have the same utf8ness. */
5434 /* Inline the first character, for speed. */
5435 if (reginfo->strend - locinput < ln
5436 || UCHARAT(s) != nextchr
5437 || (ln > 1 && memNE(s, locinput, ln)))
5446 case EXACTFL: { /* /abc/il */
5448 const U8 * fold_array;
5450 U32 fold_utf8_flags;
5452 _CHECK_AND_WARN_PROBLEMATIC_LOCALE;
5453 folder = foldEQ_locale;
5454 fold_array = PL_fold_locale;
5455 fold_utf8_flags = FOLDEQ_LOCALE;
5458 case EXACTFLU8: /* /abc/il; but all 'abc' are above 255, so
5459 is effectively /u; hence to match, target
5461 if (! utf8_target) {
5464 fold_utf8_flags = FOLDEQ_LOCALE | FOLDEQ_S1_ALREADY_FOLDED
5465 | FOLDEQ_S1_FOLDS_SANE;
5466 folder = foldEQ_latin1;
5467 fold_array = PL_fold_latin1;
5470 case EXACTFU_SS: /* /\x{df}/iu */
5471 case EXACTFU: /* /abc/iu */
5472 folder = foldEQ_latin1;
5473 fold_array = PL_fold_latin1;
5474 fold_utf8_flags = is_utf8_pat ? FOLDEQ_S1_ALREADY_FOLDED : 0;
5477 case EXACTFA_NO_TRIE: /* This node only generated for non-utf8
5479 assert(! is_utf8_pat);
5481 case EXACTFA: /* /abc/iaa */
5482 folder = foldEQ_latin1;
5483 fold_array = PL_fold_latin1;
5484 fold_utf8_flags = FOLDEQ_UTF8_NOMIX_ASCII;
5487 case EXACTF: /* /abc/i This node only generated for
5488 non-utf8 patterns */
5489 assert(! is_utf8_pat);
5491 fold_array = PL_fold;
5492 fold_utf8_flags = 0;
5500 || state_num == EXACTFU_SS
5501 || (state_num == EXACTFL && IN_UTF8_CTYPE_LOCALE))
5503 /* Either target or the pattern are utf8, or has the issue where
5504 * the fold lengths may differ. */
5505 const char * const l = locinput;
5506 char *e = reginfo->strend;
5508 if (! foldEQ_utf8_flags(s, 0, ln, is_utf8_pat,
5509 l, &e, 0, utf8_target, fold_utf8_flags))
5517 /* Neither the target nor the pattern are utf8 */
5518 if (UCHARAT(s) != nextchr
5520 && UCHARAT(s) != fold_array[nextchr])
5524 if (reginfo->strend - locinput < ln)
5526 if (ln > 1 && ! folder(s, locinput, ln))
5532 case NBOUNDL: /* /\B/l */
5536 case BOUNDL: /* /\b/l */
5539 _CHECK_AND_WARN_PROBLEMATIC_LOCALE;
5541 if (FLAGS(scan) != TRADITIONAL_BOUND) {
5542 if (! IN_UTF8_CTYPE_LOCALE) {
5543 Perl_ck_warner(aTHX_ packWARN(WARN_LOCALE),
5544 B_ON_NON_UTF8_LOCALE_IS_WRONG);
5550 if (locinput == reginfo->strbeg)
5551 b1 = isWORDCHAR_LC('\n');
5553 b1 = isWORDCHAR_LC_utf8(reghop3((U8*)locinput, -1,
5554 (U8*)(reginfo->strbeg)));
5556 b2 = (NEXTCHR_IS_EOS)
5557 ? isWORDCHAR_LC('\n')
5558 : isWORDCHAR_LC_utf8((U8*)locinput);
5560 else { /* Here the string isn't utf8 */
5561 b1 = (locinput == reginfo->strbeg)
5562 ? isWORDCHAR_LC('\n')
5563 : isWORDCHAR_LC(UCHARAT(locinput - 1));
5564 b2 = (NEXTCHR_IS_EOS)
5565 ? isWORDCHAR_LC('\n')
5566 : isWORDCHAR_LC(nextchr);
5568 if (to_complement ^ (b1 == b2)) {
5574 case NBOUND: /* /\B/ */
5578 case BOUND: /* /\b/ */
5582 goto bound_ascii_match_only;
5584 case NBOUNDA: /* /\B/a */
5588 case BOUNDA: /* /\b/a */
5592 bound_ascii_match_only:
5593 /* Here the string isn't utf8, or is utf8 and only ascii characters
5594 * are to match \w. In the latter case looking at the byte just
5595 * prior to the current one may be just the final byte of a
5596 * multi-byte character. This is ok. There are two cases:
5597 * 1) it is a single byte character, and then the test is doing
5598 * just what it's supposed to.
5599 * 2) it is a multi-byte character, in which case the final byte is
5600 * never mistakable for ASCII, and so the test will say it is
5601 * not a word character, which is the correct answer. */
5602 b1 = (locinput == reginfo->strbeg)
5603 ? isWORDCHAR_A('\n')
5604 : isWORDCHAR_A(UCHARAT(locinput - 1));
5605 b2 = (NEXTCHR_IS_EOS)
5606 ? isWORDCHAR_A('\n')
5607 : isWORDCHAR_A(nextchr);
5608 if (to_complement ^ (b1 == b2)) {
5614 case NBOUNDU: /* /\B/u */
5618 case BOUNDU: /* /\b/u */
5621 if (UNLIKELY(reginfo->strbeg >= reginfo->strend)) {
5624 else if (utf8_target) {
5626 switch((bound_type) FLAGS(scan)) {
5627 case TRADITIONAL_BOUND:
5630 b1 = (locinput == reginfo->strbeg)
5631 ? 0 /* isWORDCHAR_L1('\n') */
5632 : isWORDCHAR_utf8(reghop3((U8*)locinput, -1,
5633 (U8*)(reginfo->strbeg)));
5634 b2 = (NEXTCHR_IS_EOS)
5635 ? 0 /* isWORDCHAR_L1('\n') */
5636 : isWORDCHAR_utf8((U8*)locinput);
5637 match = cBOOL(b1 != b2);
5641 if (locinput == reginfo->strbeg || NEXTCHR_IS_EOS) {
5642 match = TRUE; /* GCB always matches at begin and
5646 /* Find the gcb values of previous and current
5647 * chars, then see if is a break point */
5648 match = isGCB(getGCB_VAL_UTF8(
5649 reghop3((U8*)locinput,
5651 (U8*)(reginfo->strbeg)),
5652 (U8*) reginfo->strend),
5653 getGCB_VAL_UTF8((U8*) locinput,
5654 (U8*) reginfo->strend));
5658 case SB_BOUND: /* Always matches at begin and end */
5659 if (locinput == reginfo->strbeg || NEXTCHR_IS_EOS) {
5663 match = isSB(getSB_VAL_UTF8(
5664 reghop3((U8*)locinput,
5666 (U8*)(reginfo->strbeg)),
5667 (U8*) reginfo->strend),
5668 getSB_VAL_UTF8((U8*) locinput,
5669 (U8*) reginfo->strend),
5670 (U8*) reginfo->strbeg,
5672 (U8*) reginfo->strend,
5678 if (locinput == reginfo->strbeg || NEXTCHR_IS_EOS) {
5682 match = isWB(WB_UNKNOWN,
5684 reghop3((U8*)locinput,
5686 (U8*)(reginfo->strbeg)),
5687 (U8*) reginfo->strend),
5688 getWB_VAL_UTF8((U8*) locinput,
5689 (U8*) reginfo->strend),
5690 (U8*) reginfo->strbeg,
5692 (U8*) reginfo->strend,
5698 else { /* Not utf8 target */
5699 switch((bound_type) FLAGS(scan)) {
5700 case TRADITIONAL_BOUND:
5703 b1 = (locinput == reginfo->strbeg)
5704 ? 0 /* isWORDCHAR_L1('\n') */
5705 : isWORDCHAR_L1(UCHARAT(locinput - 1));
5706 b2 = (NEXTCHR_IS_EOS)
5707 ? 0 /* isWORDCHAR_L1('\n') */
5708 : isWORDCHAR_L1(nextchr);
5709 match = cBOOL(b1 != b2);
5714 if (locinput == reginfo->strbeg || NEXTCHR_IS_EOS) {
5715 match = TRUE; /* GCB always matches at begin and
5718 else { /* Only CR-LF combo isn't a GCB in 0-255
5720 match = UCHARAT(locinput - 1) != '\r'
5721 || UCHARAT(locinput) != '\n';
5725 case SB_BOUND: /* Always matches at begin and end */
5726 if (locinput == reginfo->strbeg || NEXTCHR_IS_EOS) {
5730 match = isSB(getSB_VAL_CP(UCHARAT(locinput -1)),
5731 getSB_VAL_CP(UCHARAT(locinput)),
5732 (U8*) reginfo->strbeg,
5734 (U8*) reginfo->strend,
5740 if (locinput == reginfo->strbeg || NEXTCHR_IS_EOS) {
5744 match = isWB(WB_UNKNOWN,
5745 getWB_VAL_CP(UCHARAT(locinput -1)),
5746 getWB_VAL_CP(UCHARAT(locinput)),
5747 (U8*) reginfo->strbeg,
5749 (U8*) reginfo->strend,
5756 if (to_complement ^ ! match) {
5761 case ANYOFL: /* /[abc]/l */
5762 _CHECK_AND_WARN_PROBLEMATIC_LOCALE;
5764 if ((FLAGS(scan) & ANYOF_LOC_REQ_UTF8) && ! IN_UTF8_CTYPE_LOCALE)
5766 Perl_ck_warner(aTHX_ packWARN(WARN_LOCALE), utf8_locale_required);
5769 case ANYOFD: /* /[abc]/d */
5770 case ANYOF: /* /[abc]/ */
5774 if (!reginclass(rex, scan, (U8*)locinput, (U8*)reginfo->strend,
5777 locinput += UTF8SKIP(locinput);
5780 if (!REGINCLASS(rex, scan, (U8*)locinput))
5786 /* The argument (FLAGS) to all the POSIX node types is the class number
5789 case NPOSIXL: /* \W or [:^punct:] etc. under /l */
5793 case POSIXL: /* \w or [:punct:] etc. under /l */
5794 _CHECK_AND_WARN_PROBLEMATIC_LOCALE;
5798 /* Use isFOO_lc() for characters within Latin1. (Note that
5799 * UTF8_IS_INVARIANT works even on non-UTF-8 strings, or else
5800 * wouldn't be invariant) */
5801 if (UTF8_IS_INVARIANT(nextchr) || ! utf8_target) {
5802 if (! (to_complement ^ cBOOL(isFOO_lc(FLAGS(scan), (U8) nextchr)))) {
5806 else if (UTF8_IS_DOWNGRADEABLE_START(nextchr)) {
5807 if (! (to_complement ^ cBOOL(isFOO_lc(FLAGS(scan),
5808 (U8) EIGHT_BIT_UTF8_TO_NATIVE(nextchr,
5809 *(locinput + 1))))))
5814 else { /* Here, must be an above Latin-1 code point */
5815 _CHECK_AND_OUTPUT_WIDE_LOCALE_UTF8_MSG(locinput, reginfo->strend);
5816 goto utf8_posix_above_latin1;
5819 /* Here, must be utf8 */
5820 locinput += UTF8SKIP(locinput);
5823 case NPOSIXD: /* \W or [:^punct:] etc. under /d */
5827 case POSIXD: /* \w or [:punct:] etc. under /d */
5833 case NPOSIXA: /* \W or [:^punct:] etc. under /a */
5835 if (NEXTCHR_IS_EOS) {
5839 /* All UTF-8 variants match */
5840 if (! UTF8_IS_INVARIANT(nextchr)) {
5841 goto increment_locinput;
5847 case POSIXA: /* \w or [:punct:] etc. under /a */
5850 /* We get here through POSIXD, NPOSIXD, and NPOSIXA when not in
5851 * UTF-8, and also from NPOSIXA even in UTF-8 when the current
5852 * character is a single byte */
5855 || ! (to_complement ^ cBOOL(_generic_isCC_A(nextchr,
5861 /* Here we are either not in utf8, or we matched a utf8-invariant,
5862 * so the next char is the next byte */
5866 case NPOSIXU: /* \W or [:^punct:] etc. under /u */
5870 case POSIXU: /* \w or [:punct:] etc. under /u */
5872 if (NEXTCHR_IS_EOS) {
5876 /* Use _generic_isCC() for characters within Latin1. (Note that
5877 * UTF8_IS_INVARIANT works even on non-UTF-8 strings, or else
5878 * wouldn't be invariant) */
5879 if (UTF8_IS_INVARIANT(nextchr) || ! utf8_target) {
5880 if (! (to_complement ^ cBOOL(_generic_isCC(nextchr,
5887 else if (UTF8_IS_DOWNGRADEABLE_START(nextchr)) {
5888 if (! (to_complement
5889 ^ cBOOL(_generic_isCC(EIGHT_BIT_UTF8_TO_NATIVE(nextchr,
5897 else { /* Handle above Latin-1 code points */
5898 utf8_posix_above_latin1:
5899 classnum = (_char_class_number) FLAGS(scan);
5900 if (classnum < _FIRST_NON_SWASH_CC) {
5902 /* Here, uses a swash to find such code points. Load if if
5903 * not done already */
5904 if (! PL_utf8_swash_ptrs[classnum]) {
5905 U8 flags = _CORE_SWASH_INIT_ACCEPT_INVLIST;
5906 PL_utf8_swash_ptrs[classnum]
5907 = _core_swash_init("utf8",
5910 PL_XPosix_ptrs[classnum], &flags);
5912 if (! (to_complement
5913 ^ cBOOL(swash_fetch(PL_utf8_swash_ptrs[classnum],
5914 (U8 *) locinput, TRUE))))
5919 else { /* Here, uses macros to find above Latin-1 code points */
5921 case _CC_ENUM_SPACE:
5922 if (! (to_complement
5923 ^ cBOOL(is_XPERLSPACE_high(locinput))))
5928 case _CC_ENUM_BLANK:
5929 if (! (to_complement
5930 ^ cBOOL(is_HORIZWS_high(locinput))))
5935 case _CC_ENUM_XDIGIT:
5936 if (! (to_complement
5937 ^ cBOOL(is_XDIGIT_high(locinput))))
5942 case _CC_ENUM_VERTSPACE:
5943 if (! (to_complement
5944 ^ cBOOL(is_VERTWS_high(locinput))))
5949 default: /* The rest, e.g. [:cntrl:], can't match
5951 if (! to_complement) {
5957 locinput += UTF8SKIP(locinput);
5961 case CLUMP: /* Match \X: logical Unicode character. This is defined as
5962 a Unicode extended Grapheme Cluster */
5965 if (! utf8_target) {
5967 /* Match either CR LF or '.', as all the other possibilities
5969 locinput++; /* Match the . or CR */
5970 if (nextchr == '\r' /* And if it was CR, and the next is LF,
5972 && locinput < reginfo->strend
5973 && UCHARAT(locinput) == '\n')
5980 /* Get the gcb type for the current character */
5981 GCB_enum prev_gcb = getGCB_VAL_UTF8((U8*) locinput,
5982 (U8*) reginfo->strend);
5984 /* Then scan through the input until we get to the first
5985 * character whose type is supposed to be a gcb with the
5986 * current character. (There is always a break at the
5988 locinput += UTF8SKIP(locinput);
5989 while (locinput < reginfo->strend) {
5990 GCB_enum cur_gcb = getGCB_VAL_UTF8((U8*) locinput,
5991 (U8*) reginfo->strend);
5992 if (isGCB(prev_gcb, cur_gcb)) {
5997 locinput += UTF8SKIP(locinput);
6004 case NREFFL: /* /\g{name}/il */
6005 { /* The capture buffer cases. The ones beginning with N for the
6006 named buffers just convert to the equivalent numbered and
6007 pretend they were called as the corresponding numbered buffer
6009 /* don't initialize these in the declaration, it makes C++
6014 const U8 *fold_array;
6017 _CHECK_AND_WARN_PROBLEMATIC_LOCALE;
6018 folder = foldEQ_locale;
6019 fold_array = PL_fold_locale;
6021 utf8_fold_flags = FOLDEQ_LOCALE;
6024 case NREFFA: /* /\g{name}/iaa */
6025 folder = foldEQ_latin1;
6026 fold_array = PL_fold_latin1;
6028 utf8_fold_flags = FOLDEQ_UTF8_NOMIX_ASCII;
6031 case NREFFU: /* /\g{name}/iu */
6032 folder = foldEQ_latin1;
6033 fold_array = PL_fold_latin1;
6035 utf8_fold_flags = 0;
6038 case NREFF: /* /\g{name}/i */
6040 fold_array = PL_fold;
6042 utf8_fold_flags = 0;
6045 case NREF: /* /\g{name}/ */
6049 utf8_fold_flags = 0;
6052 /* For the named back references, find the corresponding buffer
6054 n = reg_check_named_buff_matched(rex,scan);
6059 goto do_nref_ref_common;
6061 case REFFL: /* /\1/il */
6062 _CHECK_AND_WARN_PROBLEMATIC_LOCALE;
6063 folder = foldEQ_locale;
6064 fold_array = PL_fold_locale;
6065 utf8_fold_flags = FOLDEQ_LOCALE;
6068 case REFFA: /* /\1/iaa */
6069 folder = foldEQ_latin1;
6070 fold_array = PL_fold_latin1;
6071 utf8_fold_flags = FOLDEQ_UTF8_NOMIX_ASCII;
6074 case REFFU: /* /\1/iu */
6075 folder = foldEQ_latin1;
6076 fold_array = PL_fold_latin1;
6077 utf8_fold_flags = 0;
6080 case REFF: /* /\1/i */
6082 fold_array = PL_fold;
6083 utf8_fold_flags = 0;
6086 case REF: /* /\1/ */
6089 utf8_fold_flags = 0;
6093 n = ARG(scan); /* which paren pair */
6096 ln = rex->offs[n].start;
6097 reginfo->poscache_iter = reginfo->poscache_maxiter; /* Void cache */
6098 if (rex->lastparen < n || ln == -1)
6099 sayNO; /* Do not match unless seen CLOSEn. */
6100 if (ln == rex->offs[n].end)
6103 s = reginfo->strbeg + ln;
6104 if (type != REF /* REF can do byte comparison */
6105 && (utf8_target || type == REFFU || type == REFFL))
6107 char * limit = reginfo->strend;
6109 /* This call case insensitively compares the entire buffer
6110 * at s, with the current input starting at locinput, but
6111 * not going off the end given by reginfo->strend, and
6112 * returns in <limit> upon success, how much of the
6113 * current input was matched */
6114 if (! foldEQ_utf8_flags(s, NULL, rex->offs[n].end - ln, utf8_target,
6115 locinput, &limit, 0, utf8_target, utf8_fold_flags))
6123 /* Not utf8: Inline the first character, for speed. */
6124 if (!NEXTCHR_IS_EOS &&
6125 UCHARAT(s) != nextchr &&
6127 UCHARAT(s) != fold_array[nextchr]))
6129 ln = rex->offs[n].end - ln;
6130 if (locinput + ln > reginfo->strend)
6132 if (ln > 1 && (type == REF
6133 ? memNE(s, locinput, ln)
6134 : ! folder(s, locinput, ln)))
6140 case NOTHING: /* null op; e.g. the 'nothing' following
6141 * the '*' in m{(a+|b)*}' */
6143 case TAIL: /* placeholder while compiling (A|B|C) */
6147 #define ST st->u.eval
6152 regexp_internal *rei;
6153 regnode *startpoint;
6155 case GOSTART: /* (?R) */
6156 case GOSUB: /* /(...(?1))/ /(...(?&foo))/ */
6157 if (cur_eval && cur_eval->locinput==locinput) {
6158 if (cur_eval->u.eval.close_paren == (U32)ARG(scan))
6159 Perl_croak(aTHX_ "Infinite recursion in regex");
6160 if ( ++nochange_depth > max_nochange_depth )
6162 "Pattern subroutine nesting without pos change"
6163 " exceeded limit in regex");
6170 if (OP(scan)==GOSUB) {
6171 startpoint = scan + ARG2L(scan);
6172 ST.close_paren = ARG(scan);
6174 startpoint = rei->program+1;
6178 /* Save all the positions seen so far. */
6179 ST.cp = regcppush(rex, 0, maxopenparen);
6180 REGCP_SET(ST.lastcp);
6182 /* and then jump to the code we share with EVAL */
6183 goto eval_recurse_doit;
6186 case EVAL: /* /(?{A})B/ /(??{A})B/ and /(?(?{A})X|Y)B/ */
6187 if (cur_eval && cur_eval->locinput==locinput) {
6188 if ( ++nochange_depth > max_nochange_depth )
6189 Perl_croak(aTHX_ "EVAL without pos change exceeded limit in regex");
6194 /* execute the code in the {...} */
6198 OP * const oop = PL_op;
6199 COP * const ocurcop = PL_curcop;
6203 /* save *all* paren positions */
6204 regcppush(rex, 0, maxopenparen);
6205 REGCP_SET(runops_cp);
6208 caller_cv = find_runcv(NULL);
6212 if (rexi->data->what[n] == 'r') { /* code from an external qr */
6214 (REGEXP*)(rexi->data->data[n])
6217 nop = (OP*)rexi->data->data[n+1];
6219 else if (rexi->data->what[n] == 'l') { /* literal code */
6221 nop = (OP*)rexi->data->data[n];
6222 assert(CvDEPTH(newcv));
6225 /* literal with own CV */
6226 assert(rexi->data->what[n] == 'L');
6227 newcv = rex->qr_anoncv;
6228 nop = (OP*)rexi->data->data[n];
6231 /* normally if we're about to execute code from the same
6232 * CV that we used previously, we just use the existing
6233 * CX stack entry. However, its possible that in the
6234 * meantime we may have backtracked, popped from the save
6235 * stack, and undone the SAVECOMPPAD(s) associated with
6236 * PUSH_MULTICALL; in which case PL_comppad no longer
6237 * points to newcv's pad. */
6238 if (newcv != last_pushed_cv || PL_comppad != last_pad)
6240 U8 flags = (CXp_SUB_RE |
6241 ((newcv == caller_cv) ? CXp_SUB_RE_FAKE : 0));
6242 if (last_pushed_cv) {
6243 CHANGE_MULTICALL_FLAGS(newcv, flags);
6246 PUSH_MULTICALL_FLAGS(newcv, flags);
6248 last_pushed_cv = newcv;
6251 /* these assignments are just to silence compiler
6253 multicall_cop = NULL;
6256 last_pad = PL_comppad;
6258 /* the initial nextstate you would normally execute
6259 * at the start of an eval (which would cause error
6260 * messages to come from the eval), may be optimised
6261 * away from the execution path in the regex code blocks;
6262 * so manually set PL_curcop to it initially */
6264 OP *o = cUNOPx(nop)->op_first;
6265 assert(o->op_type == OP_NULL);
6266 if (o->op_targ == OP_SCOPE) {
6267 o = cUNOPo->op_first;
6270 assert(o->op_targ == OP_LEAVE);
6271 o = cUNOPo->op_first;
6272 assert(o->op_type == OP_ENTER);
6276 if (o->op_type != OP_STUB) {
6277 assert( o->op_type == OP_NEXTSTATE
6278 || o->op_type == OP_DBSTATE
6279 || (o->op_type == OP_NULL
6280 && ( o->op_targ == OP_NEXTSTATE
6281 || o->op_targ == OP_DBSTATE
6285 PL_curcop = (COP*)o;
6290 DEBUG_STATE_r( PerlIO_printf(Perl_debug_log,
6291 " re EVAL PL_op=0x%"UVxf"\n", PTR2UV(nop)) );
6293 rex->offs[0].end = locinput - reginfo->strbeg;
6294 if (reginfo->info_aux_eval->pos_magic)
6295 MgBYTEPOS_set(reginfo->info_aux_eval->pos_magic,
6296 reginfo->sv, reginfo->strbeg,
6297 locinput - reginfo->strbeg);
6300 SV *sv_mrk = get_sv("REGMARK", 1);
6301 sv_setsv(sv_mrk, sv_yes_mark);
6304 /* we don't use MULTICALL here as we want to call the
6305 * first op of the block of interest, rather than the
6306 * first op of the sub */
6307 before = (IV)(SP-PL_stack_base);
6309 CALLRUNOPS(aTHX); /* Scalar context. */
6311 if ((IV)(SP-PL_stack_base) == before)
6312 ret = &PL_sv_undef; /* protect against empty (?{}) blocks. */
6318 /* before restoring everything, evaluate the returned
6319 * value, so that 'uninit' warnings don't use the wrong
6320 * PL_op or pad. Also need to process any magic vars
6321 * (e.g. $1) *before* parentheses are restored */
6326 if (logical == 0) /* (?{})/ */
6327 sv_setsv(save_scalar(PL_replgv), ret); /* $^R */
6328 else if (logical == 1) { /* /(?(?{...})X|Y)/ */
6329 sw = cBOOL(SvTRUE(ret));
6332 else { /* /(??{}) */
6333 /* if its overloaded, let the regex compiler handle
6334 * it; otherwise extract regex, or stringify */
6335 if (SvGMAGICAL(ret))
6336 ret = sv_mortalcopy(ret);
6337 if (!SvAMAGIC(ret)) {
6341 if (SvTYPE(sv) == SVt_REGEXP)
6342 re_sv = (REGEXP*) sv;
6343 else if (SvSMAGICAL(ret)) {
6344 MAGIC *mg = mg_find(ret, PERL_MAGIC_qr);
6346 re_sv = (REGEXP *) mg->mg_obj;
6349 /* force any undef warnings here */
6350 if (!re_sv && !SvPOK(ret) && !SvNIOK(ret)) {
6351 ret = sv_mortalcopy(ret);
6352 (void) SvPV_force_nolen(ret);
6358 /* *** Note that at this point we don't restore
6359 * PL_comppad, (or pop the CxSUB) on the assumption it may
6360 * be used again soon. This is safe as long as nothing
6361 * in the regexp code uses the pad ! */
6363 PL_curcop = ocurcop;
6364 S_regcp_restore(aTHX_ rex, runops_cp, &maxopenparen);
6365 PL_curpm = PL_reg_curpm;
6371 /* only /(??{})/ from now on */
6374 /* extract RE object from returned value; compiling if
6378 re_sv = reg_temp_copy(NULL, re_sv);
6383 if (SvUTF8(ret) && IN_BYTES) {
6384 /* In use 'bytes': make a copy of the octet
6385 * sequence, but without the flag on */
6387 const char *const p = SvPV(ret, len);
6388 ret = newSVpvn_flags(p, len, SVs_TEMP);
6390 if (rex->intflags & PREGf_USE_RE_EVAL)
6391 pm_flags |= PMf_USE_RE_EVAL;
6393 /* if we got here, it should be an engine which
6394 * supports compiling code blocks and stuff */
6395 assert(rex->engine && rex->engine->op_comp);
6396 assert(!(scan->flags & ~RXf_PMf_COMPILETIME));
6397 re_sv = rex->engine->op_comp(aTHX_ &ret, 1, NULL,
6398 rex->engine, NULL, NULL,
6399 /* copy /msixn etc to inner pattern */
6404 & (SVs_TEMP | SVs_GMG | SVf_ROK))
6405 && (!SvPADTMP(ret) || SvREADONLY(ret))) {
6406 /* This isn't a first class regexp. Instead, it's
6407 caching a regexp onto an existing, Perl visible
6409 sv_magic(ret, MUTABLE_SV(re_sv), PERL_MAGIC_qr, 0, 0);
6415 RXp_MATCH_COPIED_off(re);
6416 re->subbeg = rex->subbeg;
6417 re->sublen = rex->sublen;
6418 re->suboffset = rex->suboffset;
6419 re->subcoffset = rex->subcoffset;
6421 re->lastcloseparen = 0;
6424 debug_start_match(re_sv, utf8_target, locinput,
6425 reginfo->strend, "Matching embedded");
6427 startpoint = rei->program + 1;
6428 ST.close_paren = 0; /* only used for GOSUB */
6429 /* Save all the seen positions so far. */
6430 ST.cp = regcppush(rex, 0, maxopenparen);
6431 REGCP_SET(ST.lastcp);
6432 /* and set maxopenparen to 0, since we are starting a "fresh" match */
6434 /* run the pattern returned from (??{...}) */
6436 eval_recurse_doit: /* Share code with GOSUB below this line
6437 * At this point we expect the stack context to be
6438 * set up correctly */
6440 /* invalidate the S-L poscache. We're now executing a
6441 * different set of WHILEM ops (and their associated
6442 * indexes) against the same string, so the bits in the
6443 * cache are meaningless. Setting maxiter to zero forces
6444 * the cache to be invalidated and zeroed before reuse.
6445 * XXX This is too dramatic a measure. Ideally we should
6446 * save the old cache and restore when running the outer
6448 reginfo->poscache_maxiter = 0;
6450 /* the new regexp might have a different is_utf8_pat than we do */
6451 is_utf8_pat = reginfo->is_utf8_pat = cBOOL(RX_UTF8(re_sv));
6453 ST.prev_rex = rex_sv;
6454 ST.prev_curlyx = cur_curlyx;
6456 SET_reg_curpm(rex_sv);
6461 ST.prev_eval = cur_eval;
6463 /* now continue from first node in postoned RE */
6464 PUSH_YES_STATE_GOTO(EVAL_AB, startpoint, locinput);
6466 NOT_REACHED; /* NOTREACHED */
6469 case EVAL_AB: /* cleanup after a successful (??{A})B */
6470 /* note: this is called twice; first after popping B, then A */
6471 rex_sv = ST.prev_rex;
6472 is_utf8_pat = reginfo->is_utf8_pat = cBOOL(RX_UTF8(rex_sv));
6473 SET_reg_curpm(rex_sv);
6474 rex = ReANY(rex_sv);
6475 rexi = RXi_GET(rex);
6477 /* preserve $^R across LEAVE's. See Bug 121070. */
6478 SV *save_sv= GvSV(PL_replgv);
6479 SvREFCNT_inc(save_sv);
6480 regcpblow(ST.cp); /* LEAVE in disguise */
6481 sv_setsv(GvSV(PL_replgv), save_sv);
6482 SvREFCNT_dec(save_sv);
6484 cur_eval = ST.prev_eval;
6485 cur_curlyx = ST.prev_curlyx;
6487 /* Invalidate cache. See "invalidate" comment above. */
6488 reginfo->poscache_maxiter = 0;
6489 if ( nochange_depth )
6494 case EVAL_AB_fail: /* unsuccessfully ran A or B in (??{A})B */
6495 /* note: this is called twice; first after popping B, then A */
6496 rex_sv = ST.prev_rex;
6497 is_utf8_pat = reginfo->is_utf8_pat = cBOOL(RX_UTF8(rex_sv));
6498 SET_reg_curpm(rex_sv);
6499 rex = ReANY(rex_sv);
6500 rexi = RXi_GET(rex);
6502 REGCP_UNWIND(ST.lastcp);
6503 regcppop(rex, &maxopenparen);
6504 cur_eval = ST.prev_eval;
6505 cur_curlyx = ST.prev_curlyx;
6506 /* Invalidate cache. See "invalidate" comment above. */
6507 reginfo->poscache_maxiter = 0;
6508 if ( nochange_depth )
6514 n = ARG(scan); /* which paren pair */
6515 rex->offs[n].start_tmp = locinput - reginfo->strbeg;
6516 if (n > maxopenparen)
6518 DEBUG_BUFFERS_r(PerlIO_printf(Perl_debug_log,
6519 "rex=0x%"UVxf" offs=0x%"UVxf": \\%"UVuf": set %"IVdf" tmp; maxopenparen=%"UVuf"\n",
6523 (IV)rex->offs[n].start_tmp,
6529 /* XXX really need to log other places start/end are set too */
6530 #define CLOSE_CAPTURE \
6531 rex->offs[n].start = rex->offs[n].start_tmp; \
6532 rex->offs[n].end = locinput - reginfo->strbeg; \
6533 DEBUG_BUFFERS_r(PerlIO_printf(Perl_debug_log, \
6534 "rex=0x%"UVxf" offs=0x%"UVxf": \\%"UVuf": set %"IVdf"..%"IVdf"\n", \
6536 PTR2UV(rex->offs), \
6538 (IV)rex->offs[n].start, \
6539 (IV)rex->offs[n].end \
6543 n = ARG(scan); /* which paren pair */
6545 if (n > rex->lastparen)
6547 rex->lastcloseparen = n;
6548 if (cur_eval && cur_eval->u.eval.close_paren == n) {
6553 case ACCEPT: /* (*ACCEPT) */
6555 sv_yes_mark = MUTABLE_SV(rexi->data->data[ ARG( scan ) ]);
6559 cursor && OP(cursor)!=END;
6560 cursor=regnext(cursor))
6562 if ( OP(cursor)==CLOSE ){
6564 if ( n <= lastopen ) {
6566 if (n > rex->lastparen)
6568 rex->lastcloseparen = n;
6569 if ( n == ARG(scan) || (cur_eval &&
6570 cur_eval->u.eval.close_paren == n))
6579 case GROUPP: /* (?(1)) */
6580 n = ARG(scan); /* which paren pair */
6581 sw = cBOOL(rex->lastparen >= n && rex->offs[n].end != -1);
6584 case NGROUPP: /* (?(<name>)) */
6585 /* reg_check_named_buff_matched returns 0 for no match */
6586 sw = cBOOL(0 < reg_check_named_buff_matched(rex,scan));
6589 case INSUBP: /* (?(R)) */
6591 sw = (cur_eval && (!n || cur_eval->u.eval.close_paren == n));
6594 case DEFINEP: /* (?(DEFINE)) */
6598 case IFTHEN: /* (?(cond)A|B) */
6599 reginfo->poscache_iter = reginfo->poscache_maxiter; /* Void cache */
6601 next = NEXTOPER(NEXTOPER(scan));
6603 next = scan + ARG(scan);
6604 if (OP(next) == IFTHEN) /* Fake one. */
6605 next = NEXTOPER(NEXTOPER(next));
6609 case LOGICAL: /* modifier for EVAL and IFMATCH */
6610 logical = scan->flags;
6613 /*******************************************************************
6615 The CURLYX/WHILEM pair of ops handle the most generic case of the /A*B/
6616 pattern, where A and B are subpatterns. (For simple A, CURLYM or
6617 STAR/PLUS/CURLY/CURLYN are used instead.)
6619 A*B is compiled as <CURLYX><A><WHILEM><B>
6621 On entry to the subpattern, CURLYX is called. This pushes a CURLYX
6622 state, which contains the current count, initialised to -1. It also sets
6623 cur_curlyx to point to this state, with any previous value saved in the
6626 CURLYX then jumps straight to the WHILEM op, rather than executing A,
6627 since the pattern may possibly match zero times (i.e. it's a while {} loop
6628 rather than a do {} while loop).
6630 Each entry to WHILEM represents a successful match of A. The count in the
6631 CURLYX block is incremented, another WHILEM state is pushed, and execution
6632 passes to A or B depending on greediness and the current count.
6634 For example, if matching against the string a1a2a3b (where the aN are
6635 substrings that match /A/), then the match progresses as follows: (the
6636 pushed states are interspersed with the bits of strings matched so far):
6639 <CURLYX cnt=0><WHILEM>
6640 <CURLYX cnt=1><WHILEM> a1 <WHILEM>
6641 <CURLYX cnt=2><WHILEM> a1 <WHILEM> a2 <WHILEM>
6642 <CURLYX cnt=3><WHILEM> a1 <WHILEM> a2 <WHILEM> a3 <WHILEM>
6643 <CURLYX cnt=3><WHILEM> a1 <WHILEM> a2 <WHILEM> a3 <WHILEM> b
6645 (Contrast this with something like CURLYM, which maintains only a single
6649 a1 <CURLYM cnt=1> a2
6650 a1 a2 <CURLYM cnt=2> a3
6651 a1 a2 a3 <CURLYM cnt=3> b
6654 Each WHILEM state block marks a point to backtrack to upon partial failure
6655 of A or B, and also contains some minor state data related to that
6656 iteration. The CURLYX block, pointed to by cur_curlyx, contains the
6657 overall state, such as the count, and pointers to the A and B ops.
6659 This is complicated slightly by nested CURLYX/WHILEM's. Since cur_curlyx
6660 must always point to the *current* CURLYX block, the rules are:
6662 When executing CURLYX, save the old cur_curlyx in the CURLYX state block,
6663 and set cur_curlyx to point the new block.
6665 When popping the CURLYX block after a successful or unsuccessful match,
6666 restore the previous cur_curlyx.
6668 When WHILEM is about to execute B, save the current cur_curlyx, and set it
6669 to the outer one saved in the CURLYX block.
6671 When popping the WHILEM block after a successful or unsuccessful B match,
6672 restore the previous cur_curlyx.
6674 Here's an example for the pattern (AI* BI)*BO
6675 I and O refer to inner and outer, C and W refer to CURLYX and WHILEM:
6678 curlyx backtrack stack
6679 ------ ---------------
6681 CO <CO prev=NULL> <WO>
6682 CI <CO prev=NULL> <WO> <CI prev=CO> <WI> ai
6683 CO <CO prev=NULL> <WO> <CI prev=CO> <WI> ai <WI prev=CI> bi
6684 NULL <CO prev=NULL> <WO> <CI prev=CO> <WI> ai <WI prev=CI> bi <WO prev=CO> bo
6686 At this point the pattern succeeds, and we work back down the stack to
6687 clean up, restoring as we go:
6689 CO <CO prev=NULL> <WO> <CI prev=CO> <WI> ai <WI prev=CI> bi
6690 CI <CO prev=NULL> <WO> <CI prev=CO> <WI> ai
6691 CO <CO prev=NULL> <WO>
6694 *******************************************************************/
6696 #define ST st->u.curlyx
6698 case CURLYX: /* start of /A*B/ (for complex A) */
6700 /* No need to save/restore up to this paren */
6701 I32 parenfloor = scan->flags;
6703 assert(next); /* keep Coverity happy */
6704 if (OP(PREVOPER(next)) == NOTHING) /* LONGJMP */
6707 /* XXXX Probably it is better to teach regpush to support
6708 parenfloor > maxopenparen ... */
6709 if (parenfloor > (I32)rex->lastparen)
6710 parenfloor = rex->lastparen; /* Pessimization... */
6712 ST.prev_curlyx= cur_curlyx;
6714 ST.cp = PL_savestack_ix;
6716 /* these fields contain the state of the current curly.
6717 * they are accessed by subsequent WHILEMs */
6718 ST.parenfloor = parenfloor;
6723 ST.count = -1; /* this will be updated by WHILEM */
6724 ST.lastloc = NULL; /* this will be updated by WHILEM */
6726 PUSH_YES_STATE_GOTO(CURLYX_end, PREVOPER(next), locinput);
6728 NOT_REACHED; /* NOTREACHED */
6731 case CURLYX_end: /* just finished matching all of A*B */
6732 cur_curlyx = ST.prev_curlyx;
6735 NOT_REACHED; /* NOTREACHED */
6737 case CURLYX_end_fail: /* just failed to match all of A*B */
6739 cur_curlyx = ST.prev_curlyx;
6742 NOT_REACHED; /* NOTREACHED */
6746 #define ST st->u.whilem
6748 case WHILEM: /* just matched an A in /A*B/ (for complex A) */
6750 /* see the discussion above about CURLYX/WHILEM */
6755 assert(cur_curlyx); /* keep Coverity happy */
6757 min = ARG1(cur_curlyx->u.curlyx.me);
6758 max = ARG2(cur_curlyx->u.curlyx.me);
6759 A = NEXTOPER(cur_curlyx->u.curlyx.me) + EXTRA_STEP_2ARGS;
6760 n = ++cur_curlyx->u.curlyx.count; /* how many A's matched */
6761 ST.save_lastloc = cur_curlyx->u.curlyx.lastloc;
6762 ST.cache_offset = 0;
6766 DEBUG_EXECUTE_r( PerlIO_printf(Perl_debug_log,
6767 "%*s whilem: matched %ld out of %d..%d\n",
6768 REPORT_CODE_OFF+depth*2, "", (long)n, min, max)
6771 /* First just match a string of min A's. */
6774 ST.cp = regcppush(rex, cur_curlyx->u.curlyx.parenfloor,
6776 cur_curlyx->u.curlyx.lastloc = locinput;
6777 REGCP_SET(ST.lastcp);
6779 PUSH_STATE_GOTO(WHILEM_A_pre, A, locinput);
6781 NOT_REACHED; /* NOTREACHED */
6784 /* If degenerate A matches "", assume A done. */
6786 if (locinput == cur_curlyx->u.curlyx.lastloc) {
6787 DEBUG_EXECUTE_r( PerlIO_printf(Perl_debug_log,
6788 "%*s whilem: empty match detected, trying continuation...\n",
6789 REPORT_CODE_OFF+depth*2, "")
6791 goto do_whilem_B_max;
6794 /* super-linear cache processing.
6796 * The idea here is that for certain types of CURLYX/WHILEM -
6797 * principally those whose upper bound is infinity (and
6798 * excluding regexes that have things like \1 and other very
6799 * non-regular expresssiony things), then if a pattern like
6800 * /....A*.../ fails and we backtrack to the WHILEM, then we
6801 * make a note that this particular WHILEM op was at string
6802 * position 47 (say) when the rest of pattern failed. Then, if
6803 * we ever find ourselves back at that WHILEM, and at string
6804 * position 47 again, we can just fail immediately rather than
6805 * running the rest of the pattern again.
6807 * This is very handy when patterns start to go
6808 * 'super-linear', like in (a+)*(a+)*(a+)*, where you end up
6809 * with a combinatorial explosion of backtracking.
6811 * The cache is implemented as a bit array, with one bit per
6812 * string byte position per WHILEM op (up to 16) - so its
6813 * between 0.25 and 2x the string size.
6815 * To avoid allocating a poscache buffer every time, we do an
6816 * initially countdown; only after we have executed a WHILEM
6817 * op (string-length x #WHILEMs) times do we allocate the
6820 * The top 4 bits of scan->flags byte say how many different
6821 * relevant CURLLYX/WHILEM op pairs there are, while the
6822 * bottom 4-bits is the identifying index number of this
6828 if (!reginfo->poscache_maxiter) {
6829 /* start the countdown: Postpone detection until we
6830 * know the match is not *that* much linear. */
6831 reginfo->poscache_maxiter
6832 = (reginfo->strend - reginfo->strbeg + 1)
6834 /* possible overflow for long strings and many CURLYX's */
6835 if (reginfo->poscache_maxiter < 0)
6836 reginfo->poscache_maxiter = I32_MAX;
6837 reginfo->poscache_iter = reginfo->poscache_maxiter;
6840 if (reginfo->poscache_iter-- == 0) {
6841 /* initialise cache */
6842 const SSize_t size = (reginfo->poscache_maxiter + 7)/8;
6843 regmatch_info_aux *const aux = reginfo->info_aux;
6844 if (aux->poscache) {
6845 if ((SSize_t)reginfo->poscache_size < size) {
6846 Renew(aux->poscache, size, char);
6847 reginfo->poscache_size = size;
6849 Zero(aux->poscache, size, char);
6852 reginfo->poscache_size = size;
6853 Newxz(aux->poscache, size, char);
6855 DEBUG_EXECUTE_r( PerlIO_printf(Perl_debug_log,
6856 "%swhilem: Detected a super-linear match, switching on caching%s...\n",
6857 PL_colors[4], PL_colors[5])
6861 if (reginfo->poscache_iter < 0) {
6862 /* have we already failed at this position? */
6863 SSize_t offset, mask;
6865 reginfo->poscache_iter = -1; /* stop eventual underflow */
6866 offset = (scan->flags & 0xf) - 1
6867 + (locinput - reginfo->strbeg)
6869 mask = 1 << (offset % 8);
6871 if (reginfo->info_aux->poscache[offset] & mask) {
6872 DEBUG_EXECUTE_r( PerlIO_printf(Perl_debug_log,
6873 "%*s whilem: (cache) already tried at this position...\n",
6874 REPORT_CODE_OFF+depth*2, "")
6876 sayNO; /* cache records failure */
6878 ST.cache_offset = offset;
6879 ST.cache_mask = mask;
6883 /* Prefer B over A for minimal matching. */
6885 if (cur_curlyx->u.curlyx.minmod) {
6886 ST.save_curlyx = cur_curlyx;
6887 cur_curlyx = cur_curlyx->u.curlyx.prev_curlyx;
6888 ST.cp = regcppush(rex, ST.save_curlyx->u.curlyx.parenfloor,
6890 REGCP_SET(ST.lastcp);
6891 PUSH_YES_STATE_GOTO(WHILEM_B_min, ST.save_curlyx->u.curlyx.B,
6894 NOT_REACHED; /* NOTREACHED */
6897 /* Prefer A over B for maximal matching. */
6899 if (n < max) { /* More greed allowed? */
6900 ST.cp = regcppush(rex, cur_curlyx->u.curlyx.parenfloor,
6902 cur_curlyx->u.curlyx.lastloc = locinput;
6903 REGCP_SET(ST.lastcp);
6904 PUSH_STATE_GOTO(WHILEM_A_max, A, locinput);
6906 NOT_REACHED; /* NOTREACHED */
6908 goto do_whilem_B_max;
6911 NOT_REACHED; /* NOTREACHED */
6913 case WHILEM_B_min: /* just matched B in a minimal match */
6914 case WHILEM_B_max: /* just matched B in a maximal match */
6915 cur_curlyx = ST.save_curlyx;
6918 NOT_REACHED; /* NOTREACHED */
6920 case WHILEM_B_max_fail: /* just failed to match B in a maximal match */
6921 cur_curlyx = ST.save_curlyx;
6922 cur_curlyx->u.curlyx.lastloc = ST.save_lastloc;
6923 cur_curlyx->u.curlyx.count--;
6926 NOT_REACHED; /* NOTREACHED */
6928 case WHILEM_A_min_fail: /* just failed to match A in a minimal match */
6930 case WHILEM_A_pre_fail: /* just failed to match even minimal A */
6931 REGCP_UNWIND(ST.lastcp);
6932 regcppop(rex, &maxopenparen);
6933 cur_curlyx->u.curlyx.lastloc = ST.save_lastloc;
6934 cur_curlyx->u.curlyx.count--;
6937 NOT_REACHED; /* NOTREACHED */
6939 case WHILEM_A_max_fail: /* just failed to match A in a maximal match */
6940 REGCP_UNWIND(ST.lastcp);
6941 regcppop(rex, &maxopenparen); /* Restore some previous $<digit>s? */
6942 DEBUG_EXECUTE_r(PerlIO_printf(Perl_debug_log,
6943 "%*s whilem: failed, trying continuation...\n",
6944 REPORT_CODE_OFF+depth*2, "")
6947 if (cur_curlyx->u.curlyx.count >= REG_INFTY
6948 && ckWARN(WARN_REGEXP)
6949 && !reginfo->warned)
6951 reginfo->warned = TRUE;
6952 Perl_warner(aTHX_ packWARN(WARN_REGEXP),
6953 "Complex regular subexpression recursion limit (%d) "
6959 ST.save_curlyx = cur_curlyx;
6960 cur_curlyx = cur_curlyx->u.curlyx.prev_curlyx;
6961 PUSH_YES_STATE_GOTO(WHILEM_B_max, ST.save_curlyx->u.curlyx.B,
6964 NOT_REACHED; /* NOTREACHED */
6966 case WHILEM_B_min_fail: /* just failed to match B in a minimal match */
6967 cur_curlyx = ST.save_curlyx;
6968 REGCP_UNWIND(ST.lastcp);
6969 regcppop(rex, &maxopenparen);
6971 if (cur_curlyx->u.curlyx.count >= /*max*/ARG2(cur_curlyx->u.curlyx.me)) {
6972 /* Maximum greed exceeded */
6973 if (cur_curlyx->u.curlyx.count >= REG_INFTY
6974 && ckWARN(WARN_REGEXP)
6975 && !reginfo->warned)
6977 reginfo->warned = TRUE;
6978 Perl_warner(aTHX_ packWARN(WARN_REGEXP),
6979 "Complex regular subexpression recursion "
6980 "limit (%d) exceeded",
6983 cur_curlyx->u.curlyx.count--;
6987 DEBUG_EXECUTE_r(PerlIO_printf(Perl_debug_log,
6988 "%*s trying longer...\n", REPORT_CODE_OFF+depth*2, "")
6990 /* Try grabbing another A and see if it helps. */
6991 cur_curlyx->u.curlyx.lastloc = locinput;
6992 ST.cp = regcppush(rex, cur_curlyx->u.curlyx.parenfloor,
6994 REGCP_SET(ST.lastcp);
6995 PUSH_STATE_GOTO(WHILEM_A_min,
6996 /*A*/ NEXTOPER(ST.save_curlyx->u.curlyx.me) + EXTRA_STEP_2ARGS,
6999 NOT_REACHED; /* NOTREACHED */
7002 #define ST st->u.branch
7004 case BRANCHJ: /* /(...|A|...)/ with long next pointer */
7005 next = scan + ARG(scan);
7008 scan = NEXTOPER(scan);
7011 case BRANCH: /* /(...|A|...)/ */
7012 scan = NEXTOPER(scan); /* scan now points to inner node */
7013 ST.lastparen = rex->lastparen;
7014 ST.lastcloseparen = rex->lastcloseparen;
7015 ST.next_branch = next;
7018 /* Now go into the branch */
7020 PUSH_YES_STATE_GOTO(BRANCH_next, scan, locinput);
7022 PUSH_STATE_GOTO(BRANCH_next, scan, locinput);
7025 NOT_REACHED; /* NOTREACHED */
7027 case CUTGROUP: /* /(*THEN)/ */
7028 sv_yes_mark = st->u.mark.mark_name = scan->flags
7029 ? MUTABLE_SV(rexi->data->data[ ARG( scan ) ])
7031 PUSH_STATE_GOTO(CUTGROUP_next, next, locinput);
7033 NOT_REACHED; /* NOTREACHED */
7035 case CUTGROUP_next_fail:
7038 if (st->u.mark.mark_name)
7039 sv_commit = st->u.mark.mark_name;
7042 NOT_REACHED; /* NOTREACHED */
7047 NOT_REACHED; /* NOTREACHED */
7049 case BRANCH_next_fail: /* that branch failed; try the next, if any */
7054 REGCP_UNWIND(ST.cp);
7055 UNWIND_PAREN(ST.lastparen, ST.lastcloseparen);
7056 scan = ST.next_branch;
7057 /* no more branches? */
7058 if (!scan || (OP(scan) != BRANCH && OP(scan) != BRANCHJ)) {
7060 PerlIO_printf( Perl_debug_log,
7061 "%*s %sBRANCH failed...%s\n",
7062 REPORT_CODE_OFF+depth*2, "",
7068 continue; /* execute next BRANCH[J] op */
7071 case MINMOD: /* next op will be non-greedy, e.g. A*? */
7076 #define ST st->u.curlym
7078 case CURLYM: /* /A{m,n}B/ where A is fixed-length */
7080 /* This is an optimisation of CURLYX that enables us to push
7081 * only a single backtracking state, no matter how many matches
7082 * there are in {m,n}. It relies on the pattern being constant
7083 * length, with no parens to influence future backrefs
7087 scan = NEXTOPER(scan) + NODE_STEP_REGNODE;
7089 ST.lastparen = rex->lastparen;
7090 ST.lastcloseparen = rex->lastcloseparen;
7092 /* if paren positive, emulate an OPEN/CLOSE around A */
7094 U32 paren = ST.me->flags;
7095 if (paren > maxopenparen)
7096 maxopenparen = paren;
7097 scan += NEXT_OFF(scan); /* Skip former OPEN. */
7105 ST.c1 = CHRTEST_UNINIT;
7108 if (!(ST.minmod ? ARG1(ST.me) : ARG2(ST.me))) /* min/max */
7111 curlym_do_A: /* execute the A in /A{m,n}B/ */
7112 PUSH_YES_STATE_GOTO(CURLYM_A, ST.A, locinput); /* match A */
7114 NOT_REACHED; /* NOTREACHED */
7116 case CURLYM_A: /* we've just matched an A */
7118 /* after first match, determine A's length: u.curlym.alen */
7119 if (ST.count == 1) {
7120 if (reginfo->is_utf8_target) {
7121 char *s = st->locinput;
7122 while (s < locinput) {
7128 ST.alen = locinput - st->locinput;
7131 ST.count = ST.minmod ? ARG1(ST.me) : ARG2(ST.me);
7134 PerlIO_printf(Perl_debug_log,
7135 "%*s CURLYM now matched %"IVdf" times, len=%"IVdf"...\n",
7136 (int)(REPORT_CODE_OFF+(depth*2)), "",
7137 (IV) ST.count, (IV)ST.alen)
7140 if (cur_eval && cur_eval->u.eval.close_paren &&
7141 cur_eval->u.eval.close_paren == (U32)ST.me->flags)
7145 I32 max = (ST.minmod ? ARG1(ST.me) : ARG2(ST.me));
7146 if ( max == REG_INFTY || ST.count < max )
7147 goto curlym_do_A; /* try to match another A */
7149 goto curlym_do_B; /* try to match B */
7151 case CURLYM_A_fail: /* just failed to match an A */
7152 REGCP_UNWIND(ST.cp);
7154 if (ST.minmod || ST.count < ARG1(ST.me) /* min*/
7155 || (cur_eval && cur_eval->u.eval.close_paren &&
7156 cur_eval->u.eval.close_paren == (U32)ST.me->flags))
7159 curlym_do_B: /* execute the B in /A{m,n}B/ */
7160 if (ST.c1 == CHRTEST_UNINIT) {
7161 /* calculate c1 and c2 for possible match of 1st char
7162 * following curly */
7163 ST.c1 = ST.c2 = CHRTEST_VOID;
7165 if (HAS_TEXT(ST.B) || JUMPABLE(ST.B)) {
7166 regnode *text_node = ST.B;
7167 if (! HAS_TEXT(text_node))
7168 FIND_NEXT_IMPT(text_node);
7171 (HAS_TEXT(text_node) && PL_regkind[OP(text_node)] == EXACT)
7173 But the former is redundant in light of the latter.
7175 if this changes back then the macro for
7176 IS_TEXT and friends need to change.
7178 if (PL_regkind[OP(text_node)] == EXACT) {
7179 if (! S_setup_EXACTISH_ST_c1_c2(aTHX_
7180 text_node, &ST.c1, ST.c1_utf8, &ST.c2, ST.c2_utf8,
7190 PerlIO_printf(Perl_debug_log,
7191 "%*s CURLYM trying tail with matches=%"IVdf"...\n",
7192 (int)(REPORT_CODE_OFF+(depth*2)),
7195 if (! NEXTCHR_IS_EOS && ST.c1 != CHRTEST_VOID) {
7196 if (! UTF8_IS_INVARIANT(nextchr) && utf8_target) {
7197 if (memNE(locinput, ST.c1_utf8, UTF8SKIP(locinput))
7198 && memNE(locinput, ST.c2_utf8, UTF8SKIP(locinput)))
7200 /* simulate B failing */
7202 PerlIO_printf(Perl_debug_log,
7203 "%*s CURLYM Fast bail next target=0x%"UVXf" c1=0x%"UVXf" c2=0x%"UVXf"\n",
7204 (int)(REPORT_CODE_OFF+(depth*2)),"",
7205 valid_utf8_to_uvchr((U8 *) locinput, NULL),
7206 valid_utf8_to_uvchr(ST.c1_utf8, NULL),
7207 valid_utf8_to_uvchr(ST.c2_utf8, NULL))
7209 state_num = CURLYM_B_fail;
7210 goto reenter_switch;
7213 else if (nextchr != ST.c1 && nextchr != ST.c2) {
7214 /* simulate B failing */
7216 PerlIO_printf(Perl_debug_log,
7217 "%*s CURLYM Fast bail next target=0x%X c1=0x%X c2=0x%X\n",
7218 (int)(REPORT_CODE_OFF+(depth*2)),"",
7219 (int) nextchr, ST.c1, ST.c2)
7221 state_num = CURLYM_B_fail;
7222 goto reenter_switch;
7227 /* emulate CLOSE: mark current A as captured */
7228 I32 paren = ST.me->flags;
7230 rex->offs[paren].start
7231 = HOPc(locinput, -ST.alen) - reginfo->strbeg;
7232 rex->offs[paren].end = locinput - reginfo->strbeg;
7233 if ((U32)paren > rex->lastparen)
7234 rex->lastparen = paren;
7235 rex->lastcloseparen = paren;
7238 rex->offs[paren].end = -1;
7239 if (cur_eval && cur_eval->u.eval.close_paren &&
7240 cur_eval->u.eval.close_paren == (U32)ST.me->flags)
7249 PUSH_STATE_GOTO(CURLYM_B, ST.B, locinput); /* match B */
7251 NOT_REACHED; /* NOTREACHED */
7253 case CURLYM_B_fail: /* just failed to match a B */
7254 REGCP_UNWIND(ST.cp);
7255 UNWIND_PAREN(ST.lastparen, ST.lastcloseparen);
7257 I32 max = ARG2(ST.me);
7258 if (max != REG_INFTY && ST.count == max)
7260 goto curlym_do_A; /* try to match a further A */
7262 /* backtrack one A */
7263 if (ST.count == ARG1(ST.me) /* min */)
7266 SET_locinput(HOPc(locinput, -ST.alen));
7267 goto curlym_do_B; /* try to match B */
7270 #define ST st->u.curly
7272 #define CURLY_SETPAREN(paren, success) \
7275 rex->offs[paren].start = HOPc(locinput, -1) - reginfo->strbeg; \
7276 rex->offs[paren].end = locinput - reginfo->strbeg; \
7277 if (paren > rex->lastparen) \
7278 rex->lastparen = paren; \
7279 rex->lastcloseparen = paren; \
7282 rex->offs[paren].end = -1; \
7283 rex->lastparen = ST.lastparen; \
7284 rex->lastcloseparen = ST.lastcloseparen; \
7288 case STAR: /* /A*B/ where A is width 1 char */
7292 scan = NEXTOPER(scan);
7295 case PLUS: /* /A+B/ where A is width 1 char */
7299 scan = NEXTOPER(scan);
7302 case CURLYN: /* /(A){m,n}B/ where A is width 1 char */
7303 ST.paren = scan->flags; /* Which paren to set */
7304 ST.lastparen = rex->lastparen;
7305 ST.lastcloseparen = rex->lastcloseparen;
7306 if (ST.paren > maxopenparen)
7307 maxopenparen = ST.paren;
7308 ST.min = ARG1(scan); /* min to match */
7309 ST.max = ARG2(scan); /* max to match */
7310 if (cur_eval && cur_eval->u.eval.close_paren &&
7311 cur_eval->u.eval.close_paren == (U32)ST.paren) {
7315 scan = regnext(NEXTOPER(scan) + NODE_STEP_REGNODE);
7318 case CURLY: /* /A{m,n}B/ where A is width 1 char */
7320 ST.min = ARG1(scan); /* min to match */
7321 ST.max = ARG2(scan); /* max to match */
7322 scan = NEXTOPER(scan) + NODE_STEP_REGNODE;
7325 * Lookahead to avoid useless match attempts
7326 * when we know what character comes next.
7328 * Used to only do .*x and .*?x, but now it allows
7329 * for )'s, ('s and (?{ ... })'s to be in the way
7330 * of the quantifier and the EXACT-like node. -- japhy
7333 assert(ST.min <= ST.max);
7334 if (! HAS_TEXT(next) && ! JUMPABLE(next)) {
7335 ST.c1 = ST.c2 = CHRTEST_VOID;
7338 regnode *text_node = next;
7340 if (! HAS_TEXT(text_node))
7341 FIND_NEXT_IMPT(text_node);
7343 if (! HAS_TEXT(text_node))
7344 ST.c1 = ST.c2 = CHRTEST_VOID;
7346 if ( PL_regkind[OP(text_node)] != EXACT ) {
7347 ST.c1 = ST.c2 = CHRTEST_VOID;
7351 /* Currently we only get here when
7353 PL_rekind[OP(text_node)] == EXACT
7355 if this changes back then the macro for IS_TEXT and
7356 friends need to change. */
7357 if (! S_setup_EXACTISH_ST_c1_c2(aTHX_
7358 text_node, &ST.c1, ST.c1_utf8, &ST.c2, ST.c2_utf8,
7370 char *li = locinput;
7373 regrepeat(rex, &li, ST.A, reginfo, ST.min, depth)
7379 if (ST.c1 == CHRTEST_VOID)
7380 goto curly_try_B_min;
7382 ST.oldloc = locinput;
7384 /* set ST.maxpos to the furthest point along the
7385 * string that could possibly match */
7386 if (ST.max == REG_INFTY) {
7387 ST.maxpos = reginfo->strend - 1;
7389 while (UTF8_IS_CONTINUATION(*(U8*)ST.maxpos))
7392 else if (utf8_target) {
7393 int m = ST.max - ST.min;
7394 for (ST.maxpos = locinput;
7395 m >0 && ST.maxpos < reginfo->strend; m--)
7396 ST.maxpos += UTF8SKIP(ST.maxpos);
7399 ST.maxpos = locinput + ST.max - ST.min;
7400 if (ST.maxpos >= reginfo->strend)
7401 ST.maxpos = reginfo->strend - 1;
7403 goto curly_try_B_min_known;
7407 /* avoid taking address of locinput, so it can remain
7409 char *li = locinput;
7410 ST.count = regrepeat(rex, &li, ST.A, reginfo, ST.max, depth);
7411 if (ST.count < ST.min)
7414 if ((ST.count > ST.min)
7415 && (PL_regkind[OP(ST.B)] == EOL) && (OP(ST.B) != MEOL))
7417 /* A{m,n} must come at the end of the string, there's
7418 * no point in backing off ... */
7420 /* ...except that $ and \Z can match before *and* after
7421 newline at the end. Consider "\n\n" =~ /\n+\Z\n/.
7422 We may back off by one in this case. */
7423 if (UCHARAT(locinput - 1) == '\n' && OP(ST.B) != EOS)
7427 goto curly_try_B_max;
7430 NOT_REACHED; /* NOTREACHED */
7432 case CURLY_B_min_known_fail:
7433 /* failed to find B in a non-greedy match where c1,c2 valid */
7435 REGCP_UNWIND(ST.cp);
7437 UNWIND_PAREN(ST.lastparen, ST.lastcloseparen);
7439 /* Couldn't or didn't -- move forward. */
7440 ST.oldloc = locinput;
7442 locinput += UTF8SKIP(locinput);
7446 curly_try_B_min_known:
7447 /* find the next place where 'B' could work, then call B */
7451 n = (ST.oldloc == locinput) ? 0 : 1;
7452 if (ST.c1 == ST.c2) {
7453 /* set n to utf8_distance(oldloc, locinput) */
7454 while (locinput <= ST.maxpos
7455 && memNE(locinput, ST.c1_utf8, UTF8SKIP(locinput)))
7457 locinput += UTF8SKIP(locinput);
7462 /* set n to utf8_distance(oldloc, locinput) */
7463 while (locinput <= ST.maxpos
7464 && memNE(locinput, ST.c1_utf8, UTF8SKIP(locinput))
7465 && memNE(locinput, ST.c2_utf8, UTF8SKIP(locinput)))
7467 locinput += UTF8SKIP(locinput);
7472 else { /* Not utf8_target */
7473 if (ST.c1 == ST.c2) {
7474 while (locinput <= ST.maxpos &&
7475 UCHARAT(locinput) != ST.c1)
7479 while (locinput <= ST.maxpos
7480 && UCHARAT(locinput) != ST.c1
7481 && UCHARAT(locinput) != ST.c2)
7484 n = locinput - ST.oldloc;
7486 if (locinput > ST.maxpos)
7489 /* In /a{m,n}b/, ST.oldloc is at "a" x m, locinput is
7490 * at b; check that everything between oldloc and
7491 * locinput matches */
7492 char *li = ST.oldloc;
7494 if (regrepeat(rex, &li, ST.A, reginfo, n, depth) < n)
7496 assert(n == REG_INFTY || locinput == li);
7498 CURLY_SETPAREN(ST.paren, ST.count);
7499 if (cur_eval && cur_eval->u.eval.close_paren &&
7500 cur_eval->u.eval.close_paren == (U32)ST.paren) {
7503 PUSH_STATE_GOTO(CURLY_B_min_known, ST.B, locinput);
7506 NOT_REACHED; /* NOTREACHED */
7508 case CURLY_B_min_fail:
7509 /* failed to find B in a non-greedy match where c1,c2 invalid */
7511 REGCP_UNWIND(ST.cp);
7513 UNWIND_PAREN(ST.lastparen, ST.lastcloseparen);
7515 /* failed -- move forward one */
7517 char *li = locinput;
7518 if (!regrepeat(rex, &li, ST.A, reginfo, 1, depth)) {
7525 if (ST.count <= ST.max || (ST.max == REG_INFTY &&
7526 ST.count > 0)) /* count overflow ? */
7529 CURLY_SETPAREN(ST.paren, ST.count);
7530 if (cur_eval && cur_eval->u.eval.close_paren &&
7531 cur_eval->u.eval.close_paren == (U32)ST.paren) {
7534 PUSH_STATE_GOTO(CURLY_B_min, ST.B, locinput);
7539 NOT_REACHED; /* NOTREACHED */
7542 /* a successful greedy match: now try to match B */
7543 if (cur_eval && cur_eval->u.eval.close_paren &&
7544 cur_eval->u.eval.close_paren == (U32)ST.paren) {
7548 bool could_match = locinput < reginfo->strend;
7550 /* If it could work, try it. */
7551 if (ST.c1 != CHRTEST_VOID && could_match) {
7552 if (! UTF8_IS_INVARIANT(UCHARAT(locinput)) && utf8_target)
7554 could_match = memEQ(locinput,
7559 UTF8SKIP(locinput));
7562 could_match = UCHARAT(locinput) == ST.c1
7563 || UCHARAT(locinput) == ST.c2;
7566 if (ST.c1 == CHRTEST_VOID || could_match) {
7567 CURLY_SETPAREN(ST.paren, ST.count);
7568 PUSH_STATE_GOTO(CURLY_B_max, ST.B, locinput);
7570 NOT_REACHED; /* NOTREACHED */
7575 case CURLY_B_max_fail:
7576 /* failed to find B in a greedy match */
7578 REGCP_UNWIND(ST.cp);
7580 UNWIND_PAREN(ST.lastparen, ST.lastcloseparen);
7583 if (--ST.count < ST.min)
7585 locinput = HOPc(locinput, -1);
7586 goto curly_try_B_max;
7590 case END: /* last op of main pattern */
7593 /* we've just finished A in /(??{A})B/; now continue with B */
7595 st->u.eval.prev_rex = rex_sv; /* inner */
7597 /* Save *all* the positions. */
7598 st->u.eval.cp = regcppush(rex, 0, maxopenparen);
7599 rex_sv = cur_eval->u.eval.prev_rex;
7600 is_utf8_pat = reginfo->is_utf8_pat = cBOOL(RX_UTF8(rex_sv));
7601 SET_reg_curpm(rex_sv);
7602 rex = ReANY(rex_sv);
7603 rexi = RXi_GET(rex);
7604 cur_curlyx = cur_eval->u.eval.prev_curlyx;
7606 REGCP_SET(st->u.eval.lastcp);
7608 /* Restore parens of the outer rex without popping the
7610 S_regcp_restore(aTHX_ rex, cur_eval->u.eval.lastcp,
7613 st->u.eval.prev_eval = cur_eval;
7614 cur_eval = cur_eval->u.eval.prev_eval;
7616 PerlIO_printf(Perl_debug_log, "%*s EVAL trying tail ... %"UVxf"\n",
7617 REPORT_CODE_OFF+depth*2, "",PTR2UV(cur_eval)););
7618 if ( nochange_depth )
7621 PUSH_YES_STATE_GOTO(EVAL_AB, st->u.eval.prev_eval->u.eval.B,
7622 locinput); /* match B */
7625 if (locinput < reginfo->till) {
7626 DEBUG_EXECUTE_r(PerlIO_printf(Perl_debug_log,
7627 "%sMatch possible, but length=%ld is smaller than requested=%ld, failing!%s\n",
7629 (long)(locinput - startpos),
7630 (long)(reginfo->till - startpos),
7633 sayNO_SILENT; /* Cannot match: too short. */
7635 sayYES; /* Success! */
7637 case SUCCEED: /* successful SUSPEND/UNLESSM/IFMATCH/CURLYM */
7639 PerlIO_printf(Perl_debug_log,
7640 "%*s %ssubpattern success...%s\n",
7641 REPORT_CODE_OFF+depth*2, "", PL_colors[4], PL_colors[5]));
7642 sayYES; /* Success! */
7645 #define ST st->u.ifmatch
7650 case SUSPEND: /* (?>A) */
7652 newstart = locinput;
7655 case UNLESSM: /* -ve lookaround: (?!A), or with flags, (?<!A) */
7657 goto ifmatch_trivial_fail_test;
7659 case IFMATCH: /* +ve lookaround: (?=A), or with flags, (?<=A) */
7661 ifmatch_trivial_fail_test:
7663 char * const s = HOPBACKc(locinput, scan->flags);
7668 sw = 1 - cBOOL(ST.wanted);
7672 next = scan + ARG(scan);
7680 newstart = locinput;
7684 ST.logical = logical;
7685 logical = 0; /* XXX: reset state of logical once it has been saved into ST */
7687 /* execute body of (?...A) */
7688 PUSH_YES_STATE_GOTO(IFMATCH_A, NEXTOPER(NEXTOPER(scan)), newstart);
7690 NOT_REACHED; /* NOTREACHED */
7693 case IFMATCH_A_fail: /* body of (?...A) failed */
7694 ST.wanted = !ST.wanted;
7697 case IFMATCH_A: /* body of (?...A) succeeded */
7699 sw = cBOOL(ST.wanted);
7701 else if (!ST.wanted)
7704 if (OP(ST.me) != SUSPEND) {
7705 /* restore old position except for (?>...) */
7706 locinput = st->locinput;
7708 scan = ST.me + ARG(ST.me);
7711 continue; /* execute B */
7715 case LONGJMP: /* alternative with many branches compiles to
7716 * (BRANCHJ; EXACT ...; LONGJMP ) x N */
7717 next = scan + ARG(scan);
7722 case COMMIT: /* (*COMMIT) */
7723 reginfo->cutpoint = reginfo->strend;
7726 case PRUNE: /* (*PRUNE) */
7728 sv_yes_mark = sv_commit = MUTABLE_SV(rexi->data->data[ ARG( scan ) ]);
7729 PUSH_STATE_GOTO(COMMIT_next, next, locinput);
7731 NOT_REACHED; /* NOTREACHED */
7733 case COMMIT_next_fail:
7737 NOT_REACHED; /* NOTREACHED */
7739 case OPFAIL: /* (*FAIL) */
7741 sv_commit = MUTABLE_SV(rexi->data->data[ ARG( scan ) ]);
7743 /* deal with (?(?!)X|Y) properly,
7744 * make sure we trigger the no branch
7745 * of the trailing IFTHEN structure*/
7752 NOT_REACHED; /* NOTREACHED */
7754 #define ST st->u.mark
7755 case MARKPOINT: /* (*MARK:foo) */
7756 ST.prev_mark = mark_state;
7757 ST.mark_name = sv_commit = sv_yes_mark
7758 = MUTABLE_SV(rexi->data->data[ ARG( scan ) ]);
7760 ST.mark_loc = locinput;
7761 PUSH_YES_STATE_GOTO(MARKPOINT_next, next, locinput);
7763 NOT_REACHED; /* NOTREACHED */
7765 case MARKPOINT_next:
7766 mark_state = ST.prev_mark;
7769 NOT_REACHED; /* NOTREACHED */
7771 case MARKPOINT_next_fail:
7772 if (popmark && sv_eq(ST.mark_name,popmark))
7774 if (ST.mark_loc > startpoint)
7775 reginfo->cutpoint = HOPBACKc(ST.mark_loc, 1);
7776 popmark = NULL; /* we found our mark */
7777 sv_commit = ST.mark_name;
7780 PerlIO_printf(Perl_debug_log,
7781 "%*s %ssetting cutpoint to mark:%"SVf"...%s\n",
7782 REPORT_CODE_OFF+depth*2, "",
7783 PL_colors[4], SVfARG(sv_commit), PL_colors[5]);
7786 mark_state = ST.prev_mark;
7787 sv_yes_mark = mark_state ?
7788 mark_state->u.mark.mark_name : NULL;
7791 NOT_REACHED; /* NOTREACHED */
7793 case SKIP: /* (*SKIP) */
7795 /* (*SKIP) : if we fail we cut here*/
7796 ST.mark_name = NULL;
7797 ST.mark_loc = locinput;
7798 PUSH_STATE_GOTO(SKIP_next,next, locinput);
7800 /* (*SKIP:NAME) : if there is a (*MARK:NAME) fail where it was,
7801 otherwise do nothing. Meaning we need to scan
7803 regmatch_state *cur = mark_state;
7804 SV *find = MUTABLE_SV(rexi->data->data[ ARG( scan ) ]);
7807 if ( sv_eq( cur->u.mark.mark_name,
7810 ST.mark_name = find;
7811 PUSH_STATE_GOTO( SKIP_next, next, locinput);
7813 cur = cur->u.mark.prev_mark;
7816 /* Didn't find our (*MARK:NAME) so ignore this (*SKIP:NAME) */
7819 case SKIP_next_fail:
7821 /* (*CUT:NAME) - Set up to search for the name as we
7822 collapse the stack*/
7823 popmark = ST.mark_name;
7825 /* (*CUT) - No name, we cut here.*/
7826 if (ST.mark_loc > startpoint)
7827 reginfo->cutpoint = HOPBACKc(ST.mark_loc, 1);
7828 /* but we set sv_commit to latest mark_name if there
7829 is one so they can test to see how things lead to this
7832 sv_commit=mark_state->u.mark.mark_name;
7837 NOT_REACHED; /* NOTREACHED */
7840 case LNBREAK: /* \R */
7841 if ((n=is_LNBREAK_safe(locinput, reginfo->strend, utf8_target))) {
7848 PerlIO_printf(Perl_error_log, "%"UVxf" %d\n",
7849 PTR2UV(scan), OP(scan));
7850 Perl_croak(aTHX_ "regexp memory corruption");
7852 /* this is a point to jump to in order to increment
7853 * locinput by one character */
7855 assert(!NEXTCHR_IS_EOS);
7857 locinput += PL_utf8skip[nextchr];
7858 /* locinput is allowed to go 1 char off the end, but not 2+ */
7859 if (locinput > reginfo->strend)
7868 /* switch break jumps here */
7869 scan = next; /* prepare to execute the next op and ... */
7870 continue; /* ... jump back to the top, reusing st */
7874 /* push a state that backtracks on success */
7875 st->u.yes.prev_yes_state = yes_state;
7879 /* push a new regex state, then continue at scan */
7881 regmatch_state *newst;
7884 regmatch_state *cur = st;
7885 regmatch_state *curyes = yes_state;
7887 regmatch_slab *slab = PL_regmatch_slab;
7888 for (;curd > -1;cur--,curd--) {
7889 if (cur < SLAB_FIRST(slab)) {
7891 cur = SLAB_LAST(slab);
7893 PerlIO_printf(Perl_error_log, "%*s#%-3d %-10s %s\n",
7894 REPORT_CODE_OFF + 2 + depth * 2,"",
7895 curd, PL_reg_name[cur->resume_state],
7896 (curyes == cur) ? "yes" : ""
7899 curyes = cur->u.yes.prev_yes_state;
7902 DEBUG_STATE_pp("push")
7905 st->locinput = locinput;
7907 if (newst > SLAB_LAST(PL_regmatch_slab))
7908 newst = S_push_slab(aTHX);
7909 PL_regmatch_state = newst;
7911 locinput = pushinput;
7919 * We get here only if there's trouble -- normally "case END" is
7920 * the terminating point.
7922 Perl_croak(aTHX_ "corrupted regexp pointers");
7925 NOT_REACHED; /* NOTREACHED */
7929 /* we have successfully completed a subexpression, but we must now
7930 * pop to the state marked by yes_state and continue from there */
7931 assert(st != yes_state);
7933 while (st != yes_state) {
7935 if (st < SLAB_FIRST(PL_regmatch_slab)) {
7936 PL_regmatch_slab = PL_regmatch_slab->prev;
7937 st = SLAB_LAST(PL_regmatch_slab);
7941 DEBUG_STATE_pp("pop (no final)");
7943 DEBUG_STATE_pp("pop (yes)");
7949 while (yes_state < SLAB_FIRST(PL_regmatch_slab)
7950 || yes_state > SLAB_LAST(PL_regmatch_slab))
7952 /* not in this slab, pop slab */
7953 depth -= (st - SLAB_FIRST(PL_regmatch_slab) + 1);
7954 PL_regmatch_slab = PL_regmatch_slab->prev;
7955 st = SLAB_LAST(PL_regmatch_slab);
7957 depth -= (st - yes_state);
7960 yes_state = st->u.yes.prev_yes_state;
7961 PL_regmatch_state = st;
7964 locinput= st->locinput;
7965 state_num = st->resume_state + no_final;
7966 goto reenter_switch;
7969 DEBUG_EXECUTE_r(PerlIO_printf(Perl_debug_log, "%sMatch successful!%s\n",
7970 PL_colors[4], PL_colors[5]));
7972 if (reginfo->info_aux_eval) {
7973 /* each successfully executed (?{...}) block does the equivalent of
7974 * local $^R = do {...}
7975 * When popping the save stack, all these locals would be undone;
7976 * bypass this by setting the outermost saved $^R to the latest
7978 /* I dont know if this is needed or works properly now.
7979 * see code related to PL_replgv elsewhere in this file.
7982 if (oreplsv != GvSV(PL_replgv))
7983 sv_setsv(oreplsv, GvSV(PL_replgv));
7990 PerlIO_printf(Perl_debug_log,
7991 "%*s %sfailed...%s\n",
7992 REPORT_CODE_OFF+depth*2, "",
7993 PL_colors[4], PL_colors[5])
8005 /* there's a previous state to backtrack to */
8007 if (st < SLAB_FIRST(PL_regmatch_slab)) {
8008 PL_regmatch_slab = PL_regmatch_slab->prev;
8009 st = SLAB_LAST(PL_regmatch_slab);
8011 PL_regmatch_state = st;
8012 locinput= st->locinput;
8014 DEBUG_STATE_pp("pop");
8016 if (yes_state == st)
8017 yes_state = st->u.yes.prev_yes_state;
8019 state_num = st->resume_state + 1; /* failure = success + 1 */
8020 goto reenter_switch;
8025 if (rex->intflags & PREGf_VERBARG_SEEN) {
8026 SV *sv_err = get_sv("REGERROR", 1);
8027 SV *sv_mrk = get_sv("REGMARK", 1);
8029 sv_commit = &PL_sv_no;
8031 sv_yes_mark = &PL_sv_yes;
8034 sv_commit = &PL_sv_yes;
8035 sv_yes_mark = &PL_sv_no;
8039 sv_setsv(sv_err, sv_commit);
8040 sv_setsv(sv_mrk, sv_yes_mark);
8044 if (last_pushed_cv) {
8047 PERL_UNUSED_VAR(SP);
8050 assert(!result || locinput - reginfo->strbeg >= 0);
8051 return result ? locinput - reginfo->strbeg : -1;
8055 - regrepeat - repeatedly match something simple, report how many
8057 * What 'simple' means is a node which can be the operand of a quantifier like
8060 * startposp - pointer a pointer to the start position. This is updated
8061 * to point to the byte following the highest successful
8063 * p - the regnode to be repeatedly matched against.
8064 * reginfo - struct holding match state, such as strend
8065 * max - maximum number of things to match.
8066 * depth - (for debugging) backtracking depth.
8069 S_regrepeat(pTHX_ regexp *prog, char **startposp, const regnode *p,
8070 regmatch_info *const reginfo, I32 max, int depth)
8072 char *scan; /* Pointer to current position in target string */
8074 char *loceol = reginfo->strend; /* local version */
8075 I32 hardcount = 0; /* How many matches so far */
8076 bool utf8_target = reginfo->is_utf8_target;
8077 unsigned int to_complement = 0; /* Invert the result? */
8079 _char_class_number classnum;
8081 PERL_UNUSED_ARG(depth);
8084 PERL_ARGS_ASSERT_REGREPEAT;
8087 if (max == REG_INFTY)
8089 else if (! utf8_target && loceol - scan > max)
8090 loceol = scan + max;
8092 /* Here, for the case of a non-UTF-8 target we have adjusted <loceol> down
8093 * to the maximum of how far we should go in it (leaving it set to the real
8094 * end, if the maximum permissible would take us beyond that). This allows
8095 * us to make the loop exit condition that we haven't gone past <loceol> to
8096 * also mean that we haven't exceeded the max permissible count, saving a
8097 * test each time through the loop. But it assumes that the OP matches a
8098 * single byte, which is true for most of the OPs below when applied to a
8099 * non-UTF-8 target. Those relatively few OPs that don't have this
8100 * characteristic will have to compensate.
8102 * There is no adjustment for UTF-8 targets, as the number of bytes per
8103 * character varies. OPs will have to test both that the count is less
8104 * than the max permissible (using <hardcount> to keep track), and that we
8105 * are still within the bounds of the string (using <loceol>. A few OPs
8106 * match a single byte no matter what the encoding. They can omit the max
8107 * test if, for the UTF-8 case, they do the adjustment that was skipped
8110 * Thus, the code above sets things up for the common case; and exceptional
8111 * cases need extra work; the common case is to make sure <scan> doesn't
8112 * go past <loceol>, and for UTF-8 to also use <hardcount> to make sure the
8113 * count doesn't exceed the maximum permissible */
8118 while (scan < loceol && hardcount < max && *scan != '\n') {
8119 scan += UTF8SKIP(scan);
8123 while (scan < loceol && *scan != '\n')
8129 while (scan < loceol && hardcount < max) {
8130 scan += UTF8SKIP(scan);
8138 _CHECK_AND_WARN_PROBLEMATIC_LOCALE;
8139 if (utf8_target && UTF8_IS_ABOVE_LATIN1(*scan)) {
8140 _CHECK_AND_OUTPUT_WIDE_LOCALE_UTF8_MSG(scan, loceol);
8144 assert(STR_LEN(p) == reginfo->is_utf8_pat ? UTF8SKIP(STRING(p)) : 1);
8148 /* Can use a simple loop if the pattern char to match on is invariant
8149 * under UTF-8, or both target and pattern aren't UTF-8. Note that we
8150 * can use UTF8_IS_INVARIANT() even if the pattern isn't UTF-8, as it's
8151 * true iff it doesn't matter if the argument is in UTF-8 or not */
8152 if (UTF8_IS_INVARIANT(c) || (! utf8_target && ! reginfo->is_utf8_pat)) {
8153 if (utf8_target && loceol - scan > max) {
8154 /* We didn't adjust <loceol> because is UTF-8, but ok to do so,
8155 * since here, to match at all, 1 char == 1 byte */
8156 loceol = scan + max;
8158 while (scan < loceol && UCHARAT(scan) == c) {
8162 else if (reginfo->is_utf8_pat) {
8164 STRLEN scan_char_len;
8166 /* When both target and pattern are UTF-8, we have to do
8168 while (hardcount < max
8170 && (scan_char_len = UTF8SKIP(scan)) <= STR_LEN(p)
8171 && memEQ(scan, STRING(p), scan_char_len))
8173 scan += scan_char_len;
8177 else if (! UTF8_IS_ABOVE_LATIN1(c)) {
8179 /* Target isn't utf8; convert the character in the UTF-8
8180 * pattern to non-UTF8, and do a simple loop */
8181 c = EIGHT_BIT_UTF8_TO_NATIVE(c, *(STRING(p) + 1));
8182 while (scan < loceol && UCHARAT(scan) == c) {
8185 } /* else pattern char is above Latin1, can't possibly match the
8190 /* Here, the string must be utf8; pattern isn't, and <c> is
8191 * different in utf8 than not, so can't compare them directly.
8192 * Outside the loop, find the two utf8 bytes that represent c, and
8193 * then look for those in sequence in the utf8 string */
8194 U8 high = UTF8_TWO_BYTE_HI(c);
8195 U8 low = UTF8_TWO_BYTE_LO(c);
8197 while (hardcount < max
8198 && scan + 1 < loceol
8199 && UCHARAT(scan) == high
8200 && UCHARAT(scan + 1) == low)
8208 case EXACTFA_NO_TRIE: /* This node only generated for non-utf8 patterns */
8209 assert(! reginfo->is_utf8_pat);
8212 utf8_flags = FOLDEQ_UTF8_NOMIX_ASCII;
8216 _CHECK_AND_WARN_PROBLEMATIC_LOCALE;
8217 utf8_flags = FOLDEQ_LOCALE;
8220 case EXACTF: /* This node only generated for non-utf8 patterns */
8221 assert(! reginfo->is_utf8_pat);
8226 if (! utf8_target) {
8229 utf8_flags = FOLDEQ_LOCALE | FOLDEQ_S2_ALREADY_FOLDED
8230 | FOLDEQ_S2_FOLDS_SANE;
8235 utf8_flags = reginfo->is_utf8_pat ? FOLDEQ_S2_ALREADY_FOLDED : 0;
8239 U8 c1_utf8[UTF8_MAXBYTES+1], c2_utf8[UTF8_MAXBYTES+1];
8241 assert(STR_LEN(p) == reginfo->is_utf8_pat ? UTF8SKIP(STRING(p)) : 1);
8243 if (S_setup_EXACTISH_ST_c1_c2(aTHX_ p, &c1, c1_utf8, &c2, c2_utf8,
8246 if (c1 == CHRTEST_VOID) {
8247 /* Use full Unicode fold matching */
8248 char *tmpeol = reginfo->strend;
8249 STRLEN pat_len = reginfo->is_utf8_pat ? UTF8SKIP(STRING(p)) : 1;
8250 while (hardcount < max
8251 && foldEQ_utf8_flags(scan, &tmpeol, 0, utf8_target,
8252 STRING(p), NULL, pat_len,
8253 reginfo->is_utf8_pat, utf8_flags))
8256 tmpeol = reginfo->strend;
8260 else if (utf8_target) {
8262 while (scan < loceol
8264 && memEQ(scan, c1_utf8, UTF8SKIP(scan)))
8266 scan += UTF8SKIP(scan);
8271 while (scan < loceol
8273 && (memEQ(scan, c1_utf8, UTF8SKIP(scan))
8274 || memEQ(scan, c2_utf8, UTF8SKIP(scan))))
8276 scan += UTF8SKIP(scan);
8281 else if (c1 == c2) {
8282 while (scan < loceol && UCHARAT(scan) == c1) {
8287 while (scan < loceol &&
8288 (UCHARAT(scan) == c1 || UCHARAT(scan) == c2))
8297 _CHECK_AND_WARN_PROBLEMATIC_LOCALE;
8299 if ((FLAGS(p) & ANYOF_LOC_REQ_UTF8) && ! IN_UTF8_CTYPE_LOCALE) {
8300 Perl_ck_warner(aTHX_ packWARN(WARN_LOCALE), utf8_locale_required);
8306 while (hardcount < max
8308 && reginclass(prog, p, (U8*)scan, (U8*) loceol, utf8_target))
8310 scan += UTF8SKIP(scan);
8314 while (scan < loceol && REGINCLASS(prog, p, (U8*)scan))
8319 /* The argument (FLAGS) to all the POSIX node types is the class number */
8326 _CHECK_AND_WARN_PROBLEMATIC_LOCALE;
8327 if (! utf8_target) {
8328 while (scan < loceol && to_complement ^ cBOOL(isFOO_lc(FLAGS(p),
8334 while (hardcount < max && scan < loceol
8335 && to_complement ^ cBOOL(isFOO_utf8_lc(FLAGS(p),
8338 scan += UTF8SKIP(scan);
8351 if (utf8_target && loceol - scan > max) {
8353 /* We didn't adjust <loceol> at the beginning of this routine
8354 * because is UTF-8, but it is actually ok to do so, since here, to
8355 * match, 1 char == 1 byte. */
8356 loceol = scan + max;
8358 while (scan < loceol && _generic_isCC_A((U8) *scan, FLAGS(p))) {
8371 if (! utf8_target) {
8372 while (scan < loceol && ! _generic_isCC_A((U8) *scan, FLAGS(p))) {
8378 /* The complement of something that matches only ASCII matches all
8379 * non-ASCII, plus everything in ASCII that isn't in the class. */
8380 while (hardcount < max && scan < loceol
8381 && (! isASCII_utf8(scan)
8382 || ! _generic_isCC_A((U8) *scan, FLAGS(p))))
8384 scan += UTF8SKIP(scan);
8395 if (! utf8_target) {
8396 while (scan < loceol && to_complement
8397 ^ cBOOL(_generic_isCC((U8) *scan, FLAGS(p))))
8404 classnum = (_char_class_number) FLAGS(p);
8405 if (classnum < _FIRST_NON_SWASH_CC) {
8407 /* Here, a swash is needed for above-Latin1 code points.
8408 * Process as many Latin1 code points using the built-in rules.
8409 * Go to another loop to finish processing upon encountering
8410 * the first Latin1 code point. We could do that in this loop
8411 * as well, but the other way saves having to test if the swash
8412 * has been loaded every time through the loop: extra space to
8414 while (hardcount < max && scan < loceol) {
8415 if (UTF8_IS_INVARIANT(*scan)) {
8416 if (! (to_complement ^ cBOOL(_generic_isCC((U8) *scan,
8423 else if (UTF8_IS_DOWNGRADEABLE_START(*scan)) {
8424 if (! (to_complement
8425 ^ cBOOL(_generic_isCC(EIGHT_BIT_UTF8_TO_NATIVE(*scan,
8434 goto found_above_latin1;
8441 /* For these character classes, the knowledge of how to handle
8442 * every code point is compiled in to Perl via a macro. This
8443 * code is written for making the loops as tight as possible.
8444 * It could be refactored to save space instead */
8446 case _CC_ENUM_SPACE:
8447 while (hardcount < max
8449 && (to_complement ^ cBOOL(isSPACE_utf8(scan))))
8451 scan += UTF8SKIP(scan);
8455 case _CC_ENUM_BLANK:
8456 while (hardcount < max
8458 && (to_complement ^ cBOOL(isBLANK_utf8(scan))))
8460 scan += UTF8SKIP(scan);
8464 case _CC_ENUM_XDIGIT:
8465 while (hardcount < max
8467 && (to_complement ^ cBOOL(isXDIGIT_utf8(scan))))
8469 scan += UTF8SKIP(scan);
8473 case _CC_ENUM_VERTSPACE:
8474 while (hardcount < max
8476 && (to_complement ^ cBOOL(isVERTWS_utf8(scan))))
8478 scan += UTF8SKIP(scan);
8482 case _CC_ENUM_CNTRL:
8483 while (hardcount < max
8485 && (to_complement ^ cBOOL(isCNTRL_utf8(scan))))
8487 scan += UTF8SKIP(scan);
8492 Perl_croak(aTHX_ "panic: regrepeat() node %d='%s' has an unexpected character class '%d'", OP(p), PL_reg_name[OP(p)], classnum);
8498 found_above_latin1: /* Continuation of POSIXU and NPOSIXU */
8500 /* Load the swash if not already present */
8501 if (! PL_utf8_swash_ptrs[classnum]) {
8502 U8 flags = _CORE_SWASH_INIT_ACCEPT_INVLIST;
8503 PL_utf8_swash_ptrs[classnum] = _core_swash_init(
8507 PL_XPosix_ptrs[classnum], &flags);
8510 while (hardcount < max && scan < loceol
8511 && to_complement ^ cBOOL(_generic_utf8(
8514 swash_fetch(PL_utf8_swash_ptrs[classnum],
8518 scan += UTF8SKIP(scan);
8525 while (hardcount < max && scan < loceol &&
8526 (c=is_LNBREAK_utf8_safe(scan, loceol))) {
8531 /* LNBREAK can match one or two latin chars, which is ok, but we
8532 * have to use hardcount in this situation, and throw away the
8533 * adjustment to <loceol> done before the switch statement */
8534 loceol = reginfo->strend;
8535 while (scan < loceol && (c=is_LNBREAK_latin1_safe(scan, loceol))) {
8544 _CHECK_AND_WARN_PROBLEMATIC_LOCALE;
8558 /* These are all 0 width, so match right here or not at all. */
8562 Perl_croak(aTHX_ "panic: regrepeat() called with unrecognized node type %d='%s'", OP(p), PL_reg_name[OP(p)]);
8564 NOT_REACHED; /* NOTREACHED */
8571 c = scan - *startposp;
8575 GET_RE_DEBUG_FLAGS_DECL;
8577 SV * const prop = sv_newmortal();
8578 regprop(prog, prop, p, reginfo, NULL);
8579 PerlIO_printf(Perl_debug_log,
8580 "%*s %s can match %"IVdf" times out of %"IVdf"...\n",
8581 REPORT_CODE_OFF + depth*2, "", SvPVX_const(prop),(IV)c,(IV)max);
8589 #if !defined(PERL_IN_XSUB_RE) || defined(PLUGGABLE_RE_EXTENSION)
8591 - regclass_swash - prepare the utf8 swash. Wraps the shared core version to
8592 create a copy so that changes the caller makes won't change the shared one.
8593 If <altsvp> is non-null, will return NULL in it, for back-compat.
8596 Perl_regclass_swash(pTHX_ const regexp *prog, const regnode* node, bool doinit, SV** listsvp, SV **altsvp)
8598 PERL_ARGS_ASSERT_REGCLASS_SWASH;
8604 return newSVsv(_get_regclass_nonbitmap_data(prog, node, doinit, listsvp, NULL, NULL));
8607 #endif /* !defined(PERL_IN_XSUB_RE) || defined(PLUGGABLE_RE_EXTENSION) */
8610 - reginclass - determine if a character falls into a character class
8612 n is the ANYOF-type regnode
8613 p is the target string
8614 p_end points to one byte beyond the end of the target string
8615 utf8_target tells whether p is in UTF-8.
8617 Returns true if matched; false otherwise.
8619 Note that this can be a synthetic start class, a combination of various
8620 nodes, so things you think might be mutually exclusive, such as locale,
8621 aren't. It can match both locale and non-locale
8626 S_reginclass(pTHX_ regexp * const prog, const regnode * const n, const U8* const p, const U8* const p_end, const bool utf8_target)
8629 const char flags = ANYOF_FLAGS(n);
8633 PERL_ARGS_ASSERT_REGINCLASS;
8635 /* If c is not already the code point, get it. Note that
8636 * UTF8_IS_INVARIANT() works even if not in UTF-8 */
8637 if (! UTF8_IS_INVARIANT(c) && utf8_target) {
8639 c = utf8n_to_uvchr(p, p_end - p, &c_len,
8640 (UTF8_ALLOW_DEFAULT & UTF8_ALLOW_ANYUV)
8641 | UTF8_ALLOW_FFFF | UTF8_CHECK_ONLY);
8642 /* see [perl #37836] for UTF8_ALLOW_ANYUV; [perl #38293] for
8643 * UTF8_ALLOW_FFFF */
8644 if (c_len == (STRLEN)-1)
8645 Perl_croak(aTHX_ "Malformed UTF-8 character (fatal)");
8646 if (c > 255 && OP(n) == ANYOFL && ! (flags & ANYOF_LOC_REQ_UTF8)) {
8647 _CHECK_AND_OUTPUT_WIDE_LOCALE_CP_MSG(c);
8651 /* If this character is potentially in the bitmap, check it */
8652 if (c < NUM_ANYOF_CODE_POINTS) {
8653 if (ANYOF_BITMAP_TEST(n, c))
8656 & ANYOF_SHARED_d_MATCHES_ALL_NON_UTF8_NON_ASCII_non_d_WARN_SUPER)
8663 else if (flags & ANYOF_LOCALE_FLAGS) {
8664 if ((flags & ANYOF_LOC_FOLD)
8666 && ANYOF_BITMAP_TEST(n, PL_fold_locale[c]))
8670 else if (ANYOF_POSIXL_TEST_ANY_SET(n)
8674 /* The data structure is arranged so bits 0, 2, 4, ... are set
8675 * if the class includes the Posix character class given by
8676 * bit/2; and 1, 3, 5, ... are set if the class includes the
8677 * complemented Posix class given by int(bit/2). So we loop
8678 * through the bits, each time changing whether we complement
8679 * the result or not. Suppose for the sake of illustration
8680 * that bits 0-3 mean respectively, \w, \W, \s, \S. If bit 0
8681 * is set, it means there is a match for this ANYOF node if the
8682 * character is in the class given by the expression (0 / 2 = 0
8683 * = \w). If it is in that class, isFOO_lc() will return 1,
8684 * and since 'to_complement' is 0, the result will stay TRUE,
8685 * and we exit the loop. Suppose instead that bit 0 is 0, but
8686 * bit 1 is 1. That means there is a match if the character
8687 * matches \W. We won't bother to call isFOO_lc() on bit 0,
8688 * but will on bit 1. On the second iteration 'to_complement'
8689 * will be 1, so the exclusive or will reverse things, so we
8690 * are testing for \W. On the third iteration, 'to_complement'
8691 * will be 0, and we would be testing for \s; the fourth
8692 * iteration would test for \S, etc.
8694 * Note that this code assumes that all the classes are closed
8695 * under folding. For example, if a character matches \w, then
8696 * its fold does too; and vice versa. This should be true for
8697 * any well-behaved locale for all the currently defined Posix
8698 * classes, except for :lower: and :upper:, which are handled
8699 * by the pseudo-class :cased: which matches if either of the
8700 * other two does. To get rid of this assumption, an outer
8701 * loop could be used below to iterate over both the source
8702 * character, and its fold (if different) */
8705 int to_complement = 0;
8707 while (count < ANYOF_MAX) {
8708 if (ANYOF_POSIXL_TEST(n, count)
8709 && to_complement ^ cBOOL(isFOO_lc(count/2, (U8) c)))
8722 /* If the bitmap didn't (or couldn't) match, and something outside the
8723 * bitmap could match, try that. */
8725 if (c >= NUM_ANYOF_CODE_POINTS
8726 && (flags & ANYOF_MATCHES_ALL_ABOVE_BITMAP))
8728 match = TRUE; /* Everything above the bitmap matches */
8730 else if ((flags & ANYOF_HAS_NONBITMAP_NON_UTF8_MATCHES)
8731 || (utf8_target && (flags & ANYOF_HAS_UTF8_NONBITMAP_MATCHES))
8732 || ((flags & ANYOF_LOC_FOLD)
8733 && IN_UTF8_CTYPE_LOCALE
8734 && ARG(n) != ANYOF_ONLY_HAS_BITMAP))
8736 SV* only_utf8_locale = NULL;
8737 SV * const sw = _get_regclass_nonbitmap_data(prog, n, TRUE, 0,
8738 &only_utf8_locale, NULL);
8744 } else { /* Convert to utf8 */
8745 utf8_p = utf8_buffer;
8746 append_utf8_from_native_byte(*p, &utf8_p);
8747 utf8_p = utf8_buffer;
8750 if (swash_fetch(sw, utf8_p, TRUE)) {
8754 if (! match && only_utf8_locale && IN_UTF8_CTYPE_LOCALE) {
8755 match = _invlist_contains_cp(only_utf8_locale, c);
8759 if (UNICODE_IS_SUPER(c)
8761 & ANYOF_SHARED_d_MATCHES_ALL_NON_UTF8_NON_ASCII_non_d_WARN_SUPER)
8763 && ckWARN_d(WARN_NON_UNICODE))
8765 Perl_warner(aTHX_ packWARN(WARN_NON_UNICODE),
8766 "Matched non-Unicode code point 0x%04"UVXf" against Unicode property; may not be portable", c);
8770 #if ANYOF_INVERT != 1
8771 /* Depending on compiler optimization cBOOL takes time, so if don't have to
8773 # error ANYOF_INVERT needs to be set to 1, or guarded with cBOOL below,
8776 /* The xor complements the return if to invert: 1^1 = 0, 1^0 = 1 */
8777 return (flags & ANYOF_INVERT) ^ match;
8781 S_reghop3(U8 *s, SSize_t off, const U8* lim)
8783 /* return the position 'off' UTF-8 characters away from 's', forward if
8784 * 'off' >= 0, backwards if negative. But don't go outside of position
8785 * 'lim', which better be < s if off < 0 */
8787 PERL_ARGS_ASSERT_REGHOP3;
8790 while (off-- && s < lim) {
8791 /* XXX could check well-formedness here */
8796 while (off++ && s > lim) {
8798 if (UTF8_IS_CONTINUED(*s)) {
8799 while (s > lim && UTF8_IS_CONTINUATION(*s))
8801 if (! UTF8_IS_START(*s)) {
8802 Perl_croak_nocontext("Malformed UTF-8 character (fatal)");
8805 /* XXX could check well-formedness here */
8812 S_reghop4(U8 *s, SSize_t off, const U8* llim, const U8* rlim)
8814 PERL_ARGS_ASSERT_REGHOP4;
8817 while (off-- && s < rlim) {
8818 /* XXX could check well-formedness here */
8823 while (off++ && s > llim) {
8825 if (UTF8_IS_CONTINUED(*s)) {
8826 while (s > llim && UTF8_IS_CONTINUATION(*s))
8828 if (! UTF8_IS_START(*s)) {
8829 Perl_croak_nocontext("Malformed UTF-8 character (fatal)");
8832 /* XXX could check well-formedness here */
8838 /* like reghop3, but returns NULL on overrun, rather than returning last
8842 S_reghopmaybe3(U8* s, SSize_t off, const U8* lim)
8844 PERL_ARGS_ASSERT_REGHOPMAYBE3;
8847 while (off-- && s < lim) {
8848 /* XXX could check well-formedness here */
8855 while (off++ && s > lim) {
8857 if (UTF8_IS_CONTINUED(*s)) {
8858 while (s > lim && UTF8_IS_CONTINUATION(*s))
8860 if (! UTF8_IS_START(*s)) {
8861 Perl_croak_nocontext("Malformed UTF-8 character (fatal)");
8864 /* XXX could check well-formedness here */
8873 /* when executing a regex that may have (?{}), extra stuff needs setting
8874 up that will be visible to the called code, even before the current
8875 match has finished. In particular:
8877 * $_ is localised to the SV currently being matched;
8878 * pos($_) is created if necessary, ready to be updated on each call-out
8880 * a fake PMOP is created that can be set to PL_curpm (normally PL_curpm
8881 isn't set until the current pattern is successfully finished), so that
8882 $1 etc of the match-so-far can be seen;
8883 * save the old values of subbeg etc of the current regex, and set then
8884 to the current string (again, this is normally only done at the end
8889 S_setup_eval_state(pTHX_ regmatch_info *const reginfo)
8892 regexp *const rex = ReANY(reginfo->prog);
8893 regmatch_info_aux_eval *eval_state = reginfo->info_aux_eval;
8895 eval_state->rex = rex;
8898 /* Make $_ available to executed code. */
8899 if (reginfo->sv != DEFSV) {
8901 DEFSV_set(reginfo->sv);
8904 if (!(mg = mg_find_mglob(reginfo->sv))) {
8905 /* prepare for quick setting of pos */
8906 mg = sv_magicext_mglob(reginfo->sv);
8909 eval_state->pos_magic = mg;
8910 eval_state->pos = mg->mg_len;
8911 eval_state->pos_flags = mg->mg_flags;
8914 eval_state->pos_magic = NULL;
8916 if (!PL_reg_curpm) {
8917 /* PL_reg_curpm is a fake PMOP that we can attach the current
8918 * regex to and point PL_curpm at, so that $1 et al are visible
8919 * within a /(?{})/. It's just allocated once per interpreter the
8920 * first time its needed */
8921 Newxz(PL_reg_curpm, 1, PMOP);
8924 SV* const repointer = &PL_sv_undef;
8925 /* this regexp is also owned by the new PL_reg_curpm, which
8926 will try to free it. */
8927 av_push(PL_regex_padav, repointer);
8928 PL_reg_curpm->op_pmoffset = av_tindex(PL_regex_padav);
8929 PL_regex_pad = AvARRAY(PL_regex_padav);
8933 SET_reg_curpm(reginfo->prog);
8934 eval_state->curpm = PL_curpm;
8935 PL_curpm = PL_reg_curpm;
8936 if (RXp_MATCH_COPIED(rex)) {
8937 /* Here is a serious problem: we cannot rewrite subbeg,
8938 since it may be needed if this match fails. Thus
8939 $` inside (?{}) could fail... */
8940 eval_state->subbeg = rex->subbeg;
8941 eval_state->sublen = rex->sublen;
8942 eval_state->suboffset = rex->suboffset;
8943 eval_state->subcoffset = rex->subcoffset;
8945 eval_state->saved_copy = rex->saved_copy;
8947 RXp_MATCH_COPIED_off(rex);
8950 eval_state->subbeg = NULL;
8951 rex->subbeg = (char *)reginfo->strbeg;
8953 rex->subcoffset = 0;
8954 rex->sublen = reginfo->strend - reginfo->strbeg;
8958 /* destructor to clear up regmatch_info_aux and regmatch_info_aux_eval */
8961 S_cleanup_regmatch_info_aux(pTHX_ void *arg)
8963 regmatch_info_aux *aux = (regmatch_info_aux *) arg;
8964 regmatch_info_aux_eval *eval_state = aux->info_aux_eval;
8967 Safefree(aux->poscache);
8971 /* undo the effects of S_setup_eval_state() */
8973 if (eval_state->subbeg) {
8974 regexp * const rex = eval_state->rex;
8975 rex->subbeg = eval_state->subbeg;
8976 rex->sublen = eval_state->sublen;
8977 rex->suboffset = eval_state->suboffset;
8978 rex->subcoffset = eval_state->subcoffset;
8980 rex->saved_copy = eval_state->saved_copy;
8982 RXp_MATCH_COPIED_on(rex);
8984 if (eval_state->pos_magic)
8986 eval_state->pos_magic->mg_len = eval_state->pos;
8987 eval_state->pos_magic->mg_flags =
8988 (eval_state->pos_magic->mg_flags & ~MGf_BYTES)
8989 | (eval_state->pos_flags & MGf_BYTES);
8992 PL_curpm = eval_state->curpm;
8995 PL_regmatch_state = aux->old_regmatch_state;
8996 PL_regmatch_slab = aux->old_regmatch_slab;
8998 /* free all slabs above current one - this must be the last action
8999 * of this function, as aux and eval_state are allocated within
9000 * slabs and may be freed here */
9002 s = PL_regmatch_slab->next;
9004 PL_regmatch_slab->next = NULL;
9006 regmatch_slab * const osl = s;
9015 S_to_utf8_substr(pTHX_ regexp *prog)
9017 /* Converts substr fields in prog from bytes to UTF-8, calling fbm_compile
9018 * on the converted value */
9022 PERL_ARGS_ASSERT_TO_UTF8_SUBSTR;
9025 if (prog->substrs->data[i].substr
9026 && !prog->substrs->data[i].utf8_substr) {
9027 SV* const sv = newSVsv(prog->substrs->data[i].substr);
9028 prog->substrs->data[i].utf8_substr = sv;
9029 sv_utf8_upgrade(sv);
9030 if (SvVALID(prog->substrs->data[i].substr)) {
9031 if (SvTAIL(prog->substrs->data[i].substr)) {
9032 /* Trim the trailing \n that fbm_compile added last
9034 SvCUR_set(sv, SvCUR(sv) - 1);
9035 /* Whilst this makes the SV technically "invalid" (as its
9036 buffer is no longer followed by "\0") when fbm_compile()
9037 adds the "\n" back, a "\0" is restored. */
9038 fbm_compile(sv, FBMcf_TAIL);
9042 if (prog->substrs->data[i].substr == prog->check_substr)
9043 prog->check_utf8 = sv;
9049 S_to_byte_substr(pTHX_ regexp *prog)
9051 /* Converts substr fields in prog from UTF-8 to bytes, calling fbm_compile
9052 * on the converted value; returns FALSE if can't be converted. */
9056 PERL_ARGS_ASSERT_TO_BYTE_SUBSTR;
9059 if (prog->substrs->data[i].utf8_substr
9060 && !prog->substrs->data[i].substr) {
9061 SV* sv = newSVsv(prog->substrs->data[i].utf8_substr);
9062 if (! sv_utf8_downgrade(sv, TRUE)) {
9065 if (SvVALID(prog->substrs->data[i].utf8_substr)) {
9066 if (SvTAIL(prog->substrs->data[i].utf8_substr)) {
9067 /* Trim the trailing \n that fbm_compile added last
9069 SvCUR_set(sv, SvCUR(sv) - 1);
9070 fbm_compile(sv, FBMcf_TAIL);
9074 prog->substrs->data[i].substr = sv;
9075 if (prog->substrs->data[i].utf8_substr == prog->check_utf8)
9076 prog->check_substr = sv;
9084 * ex: set ts=8 sts=4 sw=4 et:
https://perl5.git.perl.org / perl5.git / blob