5 * "A fair jaw-cracker dwarf-language must be." --Samwise Gamgee
8 /* This file contains functions for compiling a regular expression. See
9 * also regexec.c which funnily enough, contains functions for executing
10 * a regular expression.
12 * This file is also copied at build time to ext/re/re_comp.c, where
13 * it's built with -DPERL_EXT_RE_BUILD -DPERL_EXT_RE_DEBUG -DPERL_EXT.
14 * This causes the main functions to be compiled under new names and with
15 * debugging support added, which makes "use re 'debug'" work.
18 /* NOTE: this is derived from Henry Spencer's regexp code, and should not
19 * confused with the original package (see point 3 below). Thanks, Henry!
22 /* Additional note: this code is very heavily munged from Henry's version
23 * in places. In some spots I've traded clarity for efficiency, so don't
24 * blame Henry for some of the lack of readability.
27 /* The names of the functions have been changed from regcomp and
28 * regexec to pregcomp and pregexec in order to avoid conflicts
29 * with the POSIX routines of the same names.
32 #ifdef PERL_EXT_RE_BUILD
37 * pregcomp and pregexec -- regsub and regerror are not used in perl
39 * Copyright (c) 1986 by University of Toronto.
40 * Written by Henry Spencer. Not derived from licensed software.
42 * Permission is granted to anyone to use this software for any
43 * purpose on any computer system, and to redistribute it freely,
44 * subject to the following restrictions:
46 * 1. The author is not responsible for the consequences of use of
47 * this software, no matter how awful, even if they arise
50 * 2. The origin of this software must not be misrepresented, either
51 * by explicit claim or by omission.
53 * 3. Altered versions must be plainly marked as such, and must not
54 * be misrepresented as being the original software.
57 **** Alterations to Henry's code are...
59 **** Copyright (C) 1991, 1992, 1993, 1994, 1995, 1996, 1997, 1998, 1999,
60 **** 2000, 2001, 2002, 2003, 2004, 2005, 2006, by Larry Wall and others
62 **** You may distribute under the terms of either the GNU General Public
63 **** License or the Artistic License, as specified in the README file.
66 * Beware that some of this code is subtly aware of the way operator
67 * precedence is structured in regular expressions. Serious changes in
68 * regular-expression syntax might require a total rethink.
71 #define PERL_IN_REGCOMP_C
74 #ifndef PERL_IN_XSUB_RE
79 #ifdef PERL_IN_XSUB_RE
90 # if defined(BUGGY_MSC6)
91 /* MSC 6.00A breaks on op/regexp.t test 85 unless we turn this off */
92 # pragma optimize("a",off)
93 /* But MSC 6.00A is happy with 'w', for aliases only across function calls*/
94 # pragma optimize("w",on )
95 # endif /* BUGGY_MSC6 */
102 typedef struct RExC_state_t {
103 U32 flags; /* are we folding, multilining? */
104 char *precomp; /* uncompiled string. */
106 char *start; /* Start of input for compile */
107 char *end; /* End of input for compile */
108 char *parse; /* Input-scan pointer. */
109 I32 whilem_seen; /* number of WHILEM in this expr */
110 regnode *emit_start; /* Start of emitted-code area */
111 regnode *emit; /* Code-emit pointer; ®dummy = don't = compiling */
112 I32 naughty; /* How bad is this pattern? */
113 I32 sawback; /* Did we see \1, ...? */
115 I32 size; /* Code size. */
116 I32 npar; /* () count. */
117 I32 nestroot; /* root parens we are in - used by accept */
121 regnode **open_parens; /* pointers to open parens */
122 regnode **close_parens; /* pointers to close parens */
123 regnode *opend; /* END node in program */
125 HV *charnames; /* cache of named sequences */
126 HV *paren_names; /* Paren names */
127 regnode **recurse; /* Recurse regops */
128 I32 recurse_count; /* Number of recurse regops */
130 char *starttry; /* -Dr: where regtry was called. */
131 #define RExC_starttry (pRExC_state->starttry)
134 const char *lastparse;
136 #define RExC_lastparse (pRExC_state->lastparse)
137 #define RExC_lastnum (pRExC_state->lastnum)
141 #define RExC_flags (pRExC_state->flags)
142 #define RExC_precomp (pRExC_state->precomp)
143 #define RExC_rx (pRExC_state->rx)
144 #define RExC_start (pRExC_state->start)
145 #define RExC_end (pRExC_state->end)
146 #define RExC_parse (pRExC_state->parse)
147 #define RExC_whilem_seen (pRExC_state->whilem_seen)
148 #define RExC_offsets (pRExC_state->rx->offsets) /* I am not like the others */
149 #define RExC_emit (pRExC_state->emit)
150 #define RExC_emit_start (pRExC_state->emit_start)
151 #define RExC_naughty (pRExC_state->naughty)
152 #define RExC_sawback (pRExC_state->sawback)
153 #define RExC_seen (pRExC_state->seen)
154 #define RExC_size (pRExC_state->size)
155 #define RExC_npar (pRExC_state->npar)
156 #define RExC_nestroot (pRExC_state->nestroot)
157 #define RExC_extralen (pRExC_state->extralen)
158 #define RExC_seen_zerolen (pRExC_state->seen_zerolen)
159 #define RExC_seen_evals (pRExC_state->seen_evals)
160 #define RExC_utf8 (pRExC_state->utf8)
161 #define RExC_charnames (pRExC_state->charnames)
162 #define RExC_open_parens (pRExC_state->open_parens)
163 #define RExC_close_parens (pRExC_state->close_parens)
164 #define RExC_opend (pRExC_state->opend)
165 #define RExC_paren_names (pRExC_state->paren_names)
166 #define RExC_recurse (pRExC_state->recurse)
167 #define RExC_recurse_count (pRExC_state->recurse_count)
169 #define ISMULT1(c) ((c) == '*' || (c) == '+' || (c) == '?')
170 #define ISMULT2(s) ((*s) == '*' || (*s) == '+' || (*s) == '?' || \
171 ((*s) == '{' && regcurly(s)))
174 #undef SPSTART /* dratted cpp namespace... */
177 * Flags to be passed up and down.
179 #define WORST 0 /* Worst case. */
180 #define HASWIDTH 0x1 /* Known to match non-null strings. */
181 #define SIMPLE 0x2 /* Simple enough to be STAR/PLUS operand. */
182 #define SPSTART 0x4 /* Starts with * or +. */
183 #define TRYAGAIN 0x8 /* Weeded out a declaration. */
185 #define REG_NODE_NUM(x) ((x) ? (int)((x)-RExC_emit_start) : -1)
187 /* whether trie related optimizations are enabled */
188 #if PERL_ENABLE_EXTENDED_TRIE_OPTIMISATION
189 #define TRIE_STUDY_OPT
190 #define FULL_TRIE_STUDY
196 #define PBYTE(u8str,paren) ((U8*)(u8str))[(paren) >> 3]
197 #define PBITVAL(paren) (1 << ((paren) & 7))
198 #define PAREN_TEST(u8str,paren) ( PBYTE(u8str,paren) & PBITVAL(paren))
199 #define PAREN_SET(u8str,paren) PBYTE(u8str,paren) |= PBITVAL(paren)
200 #define PAREN_UNSET(u8str,paren) PBYTE(u8str,paren) &= (~PBITVAL(paren))
203 /* About scan_data_t.
205 During optimisation we recurse through the regexp program performing
206 various inplace (keyhole style) optimisations. In addition study_chunk
207 and scan_commit populate this data structure with information about
208 what strings MUST appear in the pattern. We look for the longest
209 string that must appear for at a fixed location, and we look for the
210 longest string that may appear at a floating location. So for instance
215 Both 'FOO' and 'A' are fixed strings. Both 'B' and 'BAR' are floating
216 strings (because they follow a .* construct). study_chunk will identify
217 both FOO and BAR as being the longest fixed and floating strings respectively.
219 The strings can be composites, for instance
223 will result in a composite fixed substring 'foo'.
225 For each string some basic information is maintained:
227 - offset or min_offset
228 This is the position the string must appear at, or not before.
229 It also implicitly (when combined with minlenp) tells us how many
230 character must match before the string we are searching.
231 Likewise when combined with minlenp and the length of the string
232 tells us how many characters must appear after the string we have
236 Only used for floating strings. This is the rightmost point that
237 the string can appear at. Ifset to I32 max it indicates that the
238 string can occur infinitely far to the right.
241 A pointer to the minimum length of the pattern that the string
242 was found inside. This is important as in the case of positive
243 lookahead or positive lookbehind we can have multiple patterns
248 The minimum length of the pattern overall is 3, the minimum length
249 of the lookahead part is 3, but the minimum length of the part that
250 will actually match is 1. So 'FOO's minimum length is 3, but the
251 minimum length for the F is 1. This is important as the minimum length
252 is used to determine offsets in front of and behind the string being
253 looked for. Since strings can be composites this is the length of the
254 pattern at the time it was commited with a scan_commit. Note that
255 the length is calculated by study_chunk, so that the minimum lengths
256 are not known until the full pattern has been compiled, thus the
257 pointer to the value.
261 In the case of lookbehind the string being searched for can be
262 offset past the start point of the final matching string.
263 If this value was just blithely removed from the min_offset it would
264 invalidate some of the calculations for how many chars must match
265 before or after (as they are derived from min_offset and minlen and
266 the length of the string being searched for).
267 When the final pattern is compiled and the data is moved from the
268 scan_data_t structure into the regexp structure the information
269 about lookbehind is factored in, with the information that would
270 have been lost precalculated in the end_shift field for the
273 The fields pos_min and pos_delta are used to store the minimum offset
274 and the delta to the maximum offset at the current point in the pattern.
278 typedef struct scan_data_t {
279 /*I32 len_min; unused */
280 /*I32 len_delta; unused */
284 I32 last_end; /* min value, <0 unless valid. */
287 SV **longest; /* Either &l_fixed, or &l_float. */
288 SV *longest_fixed; /* longest fixed string found in pattern */
289 I32 offset_fixed; /* offset where it starts */
290 I32 *minlen_fixed; /* pointer to the minlen relevent to the string */
291 I32 lookbehind_fixed; /* is the position of the string modfied by LB */
292 SV *longest_float; /* longest floating string found in pattern */
293 I32 offset_float_min; /* earliest point in string it can appear */
294 I32 offset_float_max; /* latest point in string it can appear */
295 I32 *minlen_float; /* pointer to the minlen relevent to the string */
296 I32 lookbehind_float; /* is the position of the string modified by LB */
300 struct regnode_charclass_class *start_class;
304 * Forward declarations for pregcomp()'s friends.
307 static const scan_data_t zero_scan_data =
308 { 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0 ,0};
310 #define SF_BEFORE_EOL (SF_BEFORE_SEOL|SF_BEFORE_MEOL)
311 #define SF_BEFORE_SEOL 0x0001
312 #define SF_BEFORE_MEOL 0x0002
313 #define SF_FIX_BEFORE_EOL (SF_FIX_BEFORE_SEOL|SF_FIX_BEFORE_MEOL)
314 #define SF_FL_BEFORE_EOL (SF_FL_BEFORE_SEOL|SF_FL_BEFORE_MEOL)
317 # define SF_FIX_SHIFT_EOL (0+2)
318 # define SF_FL_SHIFT_EOL (0+4)
320 # define SF_FIX_SHIFT_EOL (+2)
321 # define SF_FL_SHIFT_EOL (+4)
324 #define SF_FIX_BEFORE_SEOL (SF_BEFORE_SEOL << SF_FIX_SHIFT_EOL)
325 #define SF_FIX_BEFORE_MEOL (SF_BEFORE_MEOL << SF_FIX_SHIFT_EOL)
327 #define SF_FL_BEFORE_SEOL (SF_BEFORE_SEOL << SF_FL_SHIFT_EOL)
328 #define SF_FL_BEFORE_MEOL (SF_BEFORE_MEOL << SF_FL_SHIFT_EOL) /* 0x20 */
329 #define SF_IS_INF 0x0040
330 #define SF_HAS_PAR 0x0080
331 #define SF_IN_PAR 0x0100
332 #define SF_HAS_EVAL 0x0200
333 #define SCF_DO_SUBSTR 0x0400
334 #define SCF_DO_STCLASS_AND 0x0800
335 #define SCF_DO_STCLASS_OR 0x1000
336 #define SCF_DO_STCLASS (SCF_DO_STCLASS_AND|SCF_DO_STCLASS_OR)
337 #define SCF_WHILEM_VISITED_POS 0x2000
339 #define SCF_TRIE_RESTUDY 0x4000 /* Do restudy? */
340 #define SCF_SEEN_ACCEPT 0x8000
342 #define UTF (RExC_utf8 != 0)
343 #define LOC ((RExC_flags & PMf_LOCALE) != 0)
344 #define FOLD ((RExC_flags & PMf_FOLD) != 0)
346 #define OOB_UNICODE 12345678
347 #define OOB_NAMEDCLASS -1
349 #define CHR_SVLEN(sv) (UTF ? sv_len_utf8(sv) : SvCUR(sv))
350 #define CHR_DIST(a,b) (UTF ? utf8_distance(a,b) : a - b)
353 /* length of regex to show in messages that don't mark a position within */
354 #define RegexLengthToShowInErrorMessages 127
357 * If MARKER[12] are adjusted, be sure to adjust the constants at the top
358 * of t/op/regmesg.t, the tests in t/op/re_tests, and those in
359 * op/pragma/warn/regcomp.
361 #define MARKER1 "<-- HERE" /* marker as it appears in the description */
362 #define MARKER2 " <-- HERE " /* marker as it appears within the regex */
364 #define REPORT_LOCATION " in regex; marked by " MARKER1 " in m/%.*s" MARKER2 "%s/"
367 * Calls SAVEDESTRUCTOR_X if needed, then calls Perl_croak with the given
368 * arg. Show regex, up to a maximum length. If it's too long, chop and add
371 #define FAIL(msg) STMT_START { \
372 const char *ellipses = ""; \
373 IV len = RExC_end - RExC_precomp; \
376 SAVEDESTRUCTOR_X(clear_re,(void*)RExC_rx); \
377 if (len > RegexLengthToShowInErrorMessages) { \
378 /* chop 10 shorter than the max, to ensure meaning of "..." */ \
379 len = RegexLengthToShowInErrorMessages - 10; \
382 Perl_croak(aTHX_ "%s in regex m/%.*s%s/", \
383 msg, (int)len, RExC_precomp, ellipses); \
387 * Simple_vFAIL -- like FAIL, but marks the current location in the scan
389 #define Simple_vFAIL(m) STMT_START { \
390 const IV offset = RExC_parse - RExC_precomp; \
391 Perl_croak(aTHX_ "%s" REPORT_LOCATION, \
392 m, (int)offset, RExC_precomp, RExC_precomp + offset); \
396 * Calls SAVEDESTRUCTOR_X if needed, then Simple_vFAIL()
398 #define vFAIL(m) STMT_START { \
400 SAVEDESTRUCTOR_X(clear_re,(void*)RExC_rx); \
405 * Like Simple_vFAIL(), but accepts two arguments.
407 #define Simple_vFAIL2(m,a1) STMT_START { \
408 const IV offset = RExC_parse - RExC_precomp; \
409 S_re_croak2(aTHX_ m, REPORT_LOCATION, a1, \
410 (int)offset, RExC_precomp, RExC_precomp + offset); \
414 * Calls SAVEDESTRUCTOR_X if needed, then Simple_vFAIL2().
416 #define vFAIL2(m,a1) STMT_START { \
418 SAVEDESTRUCTOR_X(clear_re,(void*)RExC_rx); \
419 Simple_vFAIL2(m, a1); \
424 * Like Simple_vFAIL(), but accepts three arguments.
426 #define Simple_vFAIL3(m, a1, a2) STMT_START { \
427 const IV offset = RExC_parse - RExC_precomp; \
428 S_re_croak2(aTHX_ m, REPORT_LOCATION, a1, a2, \
429 (int)offset, RExC_precomp, RExC_precomp + offset); \
433 * Calls SAVEDESTRUCTOR_X if needed, then Simple_vFAIL3().
435 #define vFAIL3(m,a1,a2) STMT_START { \
437 SAVEDESTRUCTOR_X(clear_re,(void*)RExC_rx); \
438 Simple_vFAIL3(m, a1, a2); \
442 * Like Simple_vFAIL(), but accepts four arguments.
444 #define Simple_vFAIL4(m, a1, a2, a3) STMT_START { \
445 const IV offset = RExC_parse - RExC_precomp; \
446 S_re_croak2(aTHX_ m, REPORT_LOCATION, a1, a2, a3, \
447 (int)offset, RExC_precomp, RExC_precomp + offset); \
450 #define vWARN(loc,m) STMT_START { \
451 const IV offset = loc - RExC_precomp; \
452 Perl_warner(aTHX_ packWARN(WARN_REGEXP), "%s" REPORT_LOCATION, \
453 m, (int)offset, RExC_precomp, RExC_precomp + offset); \
456 #define vWARNdep(loc,m) STMT_START { \
457 const IV offset = loc - RExC_precomp; \
458 Perl_warner(aTHX_ packWARN2(WARN_DEPRECATED, WARN_REGEXP), \
459 "%s" REPORT_LOCATION, \
460 m, (int)offset, RExC_precomp, RExC_precomp + offset); \
464 #define vWARN2(loc, m, a1) STMT_START { \
465 const IV offset = loc - RExC_precomp; \
466 Perl_warner(aTHX_ packWARN(WARN_REGEXP), m REPORT_LOCATION, \
467 a1, (int)offset, RExC_precomp, RExC_precomp + offset); \
470 #define vWARN3(loc, m, a1, a2) STMT_START { \
471 const IV offset = loc - RExC_precomp; \
472 Perl_warner(aTHX_ packWARN(WARN_REGEXP), m REPORT_LOCATION, \
473 a1, a2, (int)offset, RExC_precomp, RExC_precomp + offset); \
476 #define vWARN4(loc, m, a1, a2, a3) STMT_START { \
477 const IV offset = loc - RExC_precomp; \
478 Perl_warner(aTHX_ packWARN(WARN_REGEXP), m REPORT_LOCATION, \
479 a1, a2, a3, (int)offset, RExC_precomp, RExC_precomp + offset); \
482 #define vWARN5(loc, m, a1, a2, a3, a4) STMT_START { \
483 const IV offset = loc - RExC_precomp; \
484 Perl_warner(aTHX_ packWARN(WARN_REGEXP), m REPORT_LOCATION, \
485 a1, a2, a3, a4, (int)offset, RExC_precomp, RExC_precomp + offset); \
489 /* Allow for side effects in s */
490 #define REGC(c,s) STMT_START { \
491 if (!SIZE_ONLY) *(s) = (c); else (void)(s); \
494 /* Macros for recording node offsets. 20001227 mjd@plover.com
495 * Nodes are numbered 1, 2, 3, 4. Node #n's position is recorded in
496 * element 2*n-1 of the array. Element #2n holds the byte length node #n.
497 * Element 0 holds the number n.
498 * Position is 1 indexed.
501 #define Set_Node_Offset_To_R(node,byte) STMT_START { \
503 MJD_OFFSET_DEBUG(("** (%d) offset of node %d is %d.\n", \
504 __LINE__, (int)(node), (int)(byte))); \
506 Perl_croak(aTHX_ "value of node is %d in Offset macro", (int)(node)); \
508 RExC_offsets[2*(node)-1] = (byte); \
513 #define Set_Node_Offset(node,byte) \
514 Set_Node_Offset_To_R((node)-RExC_emit_start, (byte)-RExC_start)
515 #define Set_Cur_Node_Offset Set_Node_Offset(RExC_emit, RExC_parse)
517 #define Set_Node_Length_To_R(node,len) STMT_START { \
519 MJD_OFFSET_DEBUG(("** (%d) size of node %d is %d.\n", \
520 __LINE__, (int)(node), (int)(len))); \
522 Perl_croak(aTHX_ "value of node is %d in Length macro", (int)(node)); \
524 RExC_offsets[2*(node)] = (len); \
529 #define Set_Node_Length(node,len) \
530 Set_Node_Length_To_R((node)-RExC_emit_start, len)
531 #define Set_Cur_Node_Length(len) Set_Node_Length(RExC_emit, len)
532 #define Set_Node_Cur_Length(node) \
533 Set_Node_Length(node, RExC_parse - parse_start)
535 /* Get offsets and lengths */
536 #define Node_Offset(n) (RExC_offsets[2*((n)-RExC_emit_start)-1])
537 #define Node_Length(n) (RExC_offsets[2*((n)-RExC_emit_start)])
539 #define Set_Node_Offset_Length(node,offset,len) STMT_START { \
540 Set_Node_Offset_To_R((node)-RExC_emit_start, (offset)); \
541 Set_Node_Length_To_R((node)-RExC_emit_start, (len)); \
545 #if PERL_ENABLE_EXPERIMENTAL_REGEX_OPTIMISATIONS
546 #define EXPERIMENTAL_INPLACESCAN
549 #define DEBUG_STUDYDATA(data,depth) \
550 DEBUG_OPTIMISE_MORE_r(if(data){ \
551 PerlIO_printf(Perl_debug_log, \
552 "%*s"/* Len:%"IVdf"/%"IVdf" */" Pos:%"IVdf"/%"IVdf \
553 " Flags: %"IVdf" Whilem_c: %"IVdf" Lcp: %"IVdf" ", \
554 (int)(depth)*2, "", \
555 (IV)((data)->pos_min), \
556 (IV)((data)->pos_delta), \
557 (IV)((data)->flags), \
558 (IV)((data)->whilem_c), \
559 (IV)((data)->last_closep ? *((data)->last_closep) : -1) \
561 if ((data)->last_found) \
562 PerlIO_printf(Perl_debug_log, \
563 "Last:'%s' %"IVdf":%"IVdf"/%"IVdf" %sFixed:'%s' @ %"IVdf \
564 " %sFloat: '%s' @ %"IVdf"/%"IVdf"", \
565 SvPVX_const((data)->last_found), \
566 (IV)((data)->last_end), \
567 (IV)((data)->last_start_min), \
568 (IV)((data)->last_start_max), \
569 ((data)->longest && \
570 (data)->longest==&((data)->longest_fixed)) ? "*" : "", \
571 SvPVX_const((data)->longest_fixed), \
572 (IV)((data)->offset_fixed), \
573 ((data)->longest && \
574 (data)->longest==&((data)->longest_float)) ? "*" : "", \
575 SvPVX_const((data)->longest_float), \
576 (IV)((data)->offset_float_min), \
577 (IV)((data)->offset_float_max) \
579 PerlIO_printf(Perl_debug_log,"\n"); \
582 static void clear_re(pTHX_ void *r);
584 /* Mark that we cannot extend a found fixed substring at this point.
585 Update the longest found anchored substring and the longest found
586 floating substrings if needed. */
589 S_scan_commit(pTHX_ const RExC_state_t *pRExC_state, scan_data_t *data, I32 *minlenp)
591 const STRLEN l = CHR_SVLEN(data->last_found);
592 const STRLEN old_l = CHR_SVLEN(*data->longest);
593 GET_RE_DEBUG_FLAGS_DECL;
595 if ((l >= old_l) && ((l > old_l) || (data->flags & SF_BEFORE_EOL))) {
596 SvSetMagicSV(*data->longest, data->last_found);
597 if (*data->longest == data->longest_fixed) {
598 data->offset_fixed = l ? data->last_start_min : data->pos_min;
599 if (data->flags & SF_BEFORE_EOL)
601 |= ((data->flags & SF_BEFORE_EOL) << SF_FIX_SHIFT_EOL);
603 data->flags &= ~SF_FIX_BEFORE_EOL;
604 data->minlen_fixed=minlenp;
605 data->lookbehind_fixed=0;
608 data->offset_float_min = l ? data->last_start_min : data->pos_min;
609 data->offset_float_max = (l
610 ? data->last_start_max
611 : data->pos_min + data->pos_delta);
612 if ((U32)data->offset_float_max > (U32)I32_MAX)
613 data->offset_float_max = I32_MAX;
614 if (data->flags & SF_BEFORE_EOL)
616 |= ((data->flags & SF_BEFORE_EOL) << SF_FL_SHIFT_EOL);
618 data->flags &= ~SF_FL_BEFORE_EOL;
619 data->minlen_float=minlenp;
620 data->lookbehind_float=0;
623 SvCUR_set(data->last_found, 0);
625 SV * const sv = data->last_found;
626 if (SvUTF8(sv) && SvMAGICAL(sv)) {
627 MAGIC * const mg = mg_find(sv, PERL_MAGIC_utf8);
633 data->flags &= ~SF_BEFORE_EOL;
634 DEBUG_STUDYDATA(data,0);
637 /* Can match anything (initialization) */
639 S_cl_anything(const RExC_state_t *pRExC_state, struct regnode_charclass_class *cl)
641 ANYOF_CLASS_ZERO(cl);
642 ANYOF_BITMAP_SETALL(cl);
643 cl->flags = ANYOF_EOS|ANYOF_UNICODE_ALL;
645 cl->flags |= ANYOF_LOCALE;
648 /* Can match anything (initialization) */
650 S_cl_is_anything(const struct regnode_charclass_class *cl)
654 for (value = 0; value <= ANYOF_MAX; value += 2)
655 if (ANYOF_CLASS_TEST(cl, value) && ANYOF_CLASS_TEST(cl, value + 1))
657 if (!(cl->flags & ANYOF_UNICODE_ALL))
659 if (!ANYOF_BITMAP_TESTALLSET((const void*)cl))
664 /* Can match anything (initialization) */
666 S_cl_init(const RExC_state_t *pRExC_state, struct regnode_charclass_class *cl)
668 Zero(cl, 1, struct regnode_charclass_class);
670 cl_anything(pRExC_state, cl);
674 S_cl_init_zero(const RExC_state_t *pRExC_state, struct regnode_charclass_class *cl)
676 Zero(cl, 1, struct regnode_charclass_class);
678 cl_anything(pRExC_state, cl);
680 cl->flags |= ANYOF_LOCALE;
683 /* 'And' a given class with another one. Can create false positives */
684 /* We assume that cl is not inverted */
686 S_cl_and(struct regnode_charclass_class *cl,
687 const struct regnode_charclass_class *and_with)
690 assert(and_with->type == ANYOF);
691 if (!(and_with->flags & ANYOF_CLASS)
692 && !(cl->flags & ANYOF_CLASS)
693 && (and_with->flags & ANYOF_LOCALE) == (cl->flags & ANYOF_LOCALE)
694 && !(and_with->flags & ANYOF_FOLD)
695 && !(cl->flags & ANYOF_FOLD)) {
698 if (and_with->flags & ANYOF_INVERT)
699 for (i = 0; i < ANYOF_BITMAP_SIZE; i++)
700 cl->bitmap[i] &= ~and_with->bitmap[i];
702 for (i = 0; i < ANYOF_BITMAP_SIZE; i++)
703 cl->bitmap[i] &= and_with->bitmap[i];
704 } /* XXXX: logic is complicated otherwise, leave it along for a moment. */
705 if (!(and_with->flags & ANYOF_EOS))
706 cl->flags &= ~ANYOF_EOS;
708 if (cl->flags & ANYOF_UNICODE_ALL && and_with->flags & ANYOF_UNICODE &&
709 !(and_with->flags & ANYOF_INVERT)) {
710 cl->flags &= ~ANYOF_UNICODE_ALL;
711 cl->flags |= ANYOF_UNICODE;
712 ARG_SET(cl, ARG(and_with));
714 if (!(and_with->flags & ANYOF_UNICODE_ALL) &&
715 !(and_with->flags & ANYOF_INVERT))
716 cl->flags &= ~ANYOF_UNICODE_ALL;
717 if (!(and_with->flags & (ANYOF_UNICODE|ANYOF_UNICODE_ALL)) &&
718 !(and_with->flags & ANYOF_INVERT))
719 cl->flags &= ~ANYOF_UNICODE;
722 /* 'OR' a given class with another one. Can create false positives */
723 /* We assume that cl is not inverted */
725 S_cl_or(const RExC_state_t *pRExC_state, struct regnode_charclass_class *cl, const struct regnode_charclass_class *or_with)
727 if (or_with->flags & ANYOF_INVERT) {
729 * (B1 | CL1) | (!B2 & !CL2) = (B1 | !B2 & !CL2) | (CL1 | (!B2 & !CL2))
730 * <= (B1 | !B2) | (CL1 | !CL2)
731 * which is wasteful if CL2 is small, but we ignore CL2:
732 * (B1 | CL1) | (!B2 & !CL2) <= (B1 | CL1) | !B2 = (B1 | !B2) | CL1
733 * XXXX Can we handle case-fold? Unclear:
734 * (OK1(i) | OK1(i')) | !(OK1(i) | OK1(i')) =
735 * (OK1(i) | OK1(i')) | (!OK1(i) & !OK1(i'))
737 if ( (or_with->flags & ANYOF_LOCALE) == (cl->flags & ANYOF_LOCALE)
738 && !(or_with->flags & ANYOF_FOLD)
739 && !(cl->flags & ANYOF_FOLD) ) {
742 for (i = 0; i < ANYOF_BITMAP_SIZE; i++)
743 cl->bitmap[i] |= ~or_with->bitmap[i];
744 } /* XXXX: logic is complicated otherwise */
746 cl_anything(pRExC_state, cl);
749 /* (B1 | CL1) | (B2 | CL2) = (B1 | B2) | (CL1 | CL2)) */
750 if ( (or_with->flags & ANYOF_LOCALE) == (cl->flags & ANYOF_LOCALE)
751 && (!(or_with->flags & ANYOF_FOLD)
752 || (cl->flags & ANYOF_FOLD)) ) {
755 /* OR char bitmap and class bitmap separately */
756 for (i = 0; i < ANYOF_BITMAP_SIZE; i++)
757 cl->bitmap[i] |= or_with->bitmap[i];
758 if (or_with->flags & ANYOF_CLASS) {
759 for (i = 0; i < ANYOF_CLASSBITMAP_SIZE; i++)
760 cl->classflags[i] |= or_with->classflags[i];
761 cl->flags |= ANYOF_CLASS;
764 else { /* XXXX: logic is complicated, leave it along for a moment. */
765 cl_anything(pRExC_state, cl);
768 if (or_with->flags & ANYOF_EOS)
769 cl->flags |= ANYOF_EOS;
771 if (cl->flags & ANYOF_UNICODE && or_with->flags & ANYOF_UNICODE &&
772 ARG(cl) != ARG(or_with)) {
773 cl->flags |= ANYOF_UNICODE_ALL;
774 cl->flags &= ~ANYOF_UNICODE;
776 if (or_with->flags & ANYOF_UNICODE_ALL) {
777 cl->flags |= ANYOF_UNICODE_ALL;
778 cl->flags &= ~ANYOF_UNICODE;
782 #define TRIE_LIST_ITEM(state,idx) (trie->states[state].trans.list)[ idx ]
783 #define TRIE_LIST_CUR(state) ( TRIE_LIST_ITEM( state, 0 ).forid )
784 #define TRIE_LIST_LEN(state) ( TRIE_LIST_ITEM( state, 0 ).newstate )
785 #define TRIE_LIST_USED(idx) ( trie->states[state].trans.list ? (TRIE_LIST_CUR( idx ) - 1) : 0 )
791 dump_trie_interim_list(trie,next_alloc)
792 dump_trie_interim_table(trie,next_alloc)
794 These routines dump out a trie in a somewhat readable format.
795 The _interim_ variants are used for debugging the interim
796 tables that are used to generate the final compressed
797 representation which is what dump_trie expects.
799 Part of the reason for their existance is to provide a form
800 of documentation as to how the different representations function.
806 Dumps the final compressed table form of the trie to Perl_debug_log.
807 Used for debugging make_trie().
811 S_dump_trie(pTHX_ const struct _reg_trie_data *trie,U32 depth)
814 SV *sv=sv_newmortal();
815 int colwidth= trie->widecharmap ? 6 : 4;
816 GET_RE_DEBUG_FLAGS_DECL;
819 PerlIO_printf( Perl_debug_log, "%*sChar : %-6s%-6s%-4s ",
820 (int)depth * 2 + 2,"",
821 "Match","Base","Ofs" );
823 for( state = 0 ; state < trie->uniquecharcount ; state++ ) {
824 SV ** const tmp = av_fetch( trie->revcharmap, state, 0);
826 PerlIO_printf( Perl_debug_log, "%*s",
828 pv_pretty(sv, SvPV_nolen_const(*tmp), SvCUR(*tmp), colwidth,
829 PL_colors[0], PL_colors[1],
830 (SvUTF8(*tmp) ? PERL_PV_ESCAPE_UNI : 0) |
831 PERL_PV_ESCAPE_FIRSTCHAR
836 PerlIO_printf( Perl_debug_log, "\n%*sState|-----------------------",
837 (int)depth * 2 + 2,"");
839 for( state = 0 ; state < trie->uniquecharcount ; state++ )
840 PerlIO_printf( Perl_debug_log, "%.*s", colwidth, "--------");
841 PerlIO_printf( Perl_debug_log, "\n");
843 for( state = 1 ; state < trie->statecount ; state++ ) {
844 const U32 base = trie->states[ state ].trans.base;
846 PerlIO_printf( Perl_debug_log, "%*s#%4"UVXf"|", (int)depth * 2 + 2,"", (UV)state);
848 if ( trie->states[ state ].wordnum ) {
849 PerlIO_printf( Perl_debug_log, " W%4X", trie->states[ state ].wordnum );
851 PerlIO_printf( Perl_debug_log, "%6s", "" );
854 PerlIO_printf( Perl_debug_log, " @%4"UVXf" ", (UV)base );
859 while( ( base + ofs < trie->uniquecharcount ) ||
860 ( base + ofs - trie->uniquecharcount < trie->lasttrans
861 && trie->trans[ base + ofs - trie->uniquecharcount ].check != state))
864 PerlIO_printf( Perl_debug_log, "+%2"UVXf"[ ", (UV)ofs);
866 for ( ofs = 0 ; ofs < trie->uniquecharcount ; ofs++ ) {
867 if ( ( base + ofs >= trie->uniquecharcount ) &&
868 ( base + ofs - trie->uniquecharcount < trie->lasttrans ) &&
869 trie->trans[ base + ofs - trie->uniquecharcount ].check == state )
871 PerlIO_printf( Perl_debug_log, "%*"UVXf,
873 (UV)trie->trans[ base + ofs - trie->uniquecharcount ].next );
875 PerlIO_printf( Perl_debug_log, "%*s",colwidth," ." );
879 PerlIO_printf( Perl_debug_log, "]");
882 PerlIO_printf( Perl_debug_log, "\n" );
886 dump_trie_interim_list(trie,next_alloc)
887 Dumps a fully constructed but uncompressed trie in list form.
888 List tries normally only are used for construction when the number of
889 possible chars (trie->uniquecharcount) is very high.
890 Used for debugging make_trie().
893 S_dump_trie_interim_list(pTHX_ const struct _reg_trie_data *trie, U32 next_alloc,U32 depth)
896 SV *sv=sv_newmortal();
897 int colwidth= trie->widecharmap ? 6 : 4;
898 GET_RE_DEBUG_FLAGS_DECL;
899 /* print out the table precompression. */
900 PerlIO_printf( Perl_debug_log, "%*sState :Word | Transition Data\n%*s%s",
901 (int)depth * 2 + 2,"", (int)depth * 2 + 2,"",
902 "------:-----+-----------------\n" );
904 for( state=1 ; state < next_alloc ; state ++ ) {
907 PerlIO_printf( Perl_debug_log, "%*s %4"UVXf" :",
908 (int)depth * 2 + 2,"", (UV)state );
909 if ( ! trie->states[ state ].wordnum ) {
910 PerlIO_printf( Perl_debug_log, "%5s| ","");
912 PerlIO_printf( Perl_debug_log, "W%4x| ",
913 trie->states[ state ].wordnum
916 for( charid = 1 ; charid <= TRIE_LIST_USED( state ) ; charid++ ) {
917 SV ** const tmp = av_fetch( trie->revcharmap, TRIE_LIST_ITEM(state,charid).forid, 0);
919 PerlIO_printf( Perl_debug_log, "%*s:%3X=%4"UVXf" | ",
921 pv_pretty(sv, SvPV_nolen_const(*tmp), SvCUR(*tmp), colwidth,
922 PL_colors[0], PL_colors[1],
923 (SvUTF8(*tmp) ? PERL_PV_ESCAPE_UNI : 0) |
924 PERL_PV_ESCAPE_FIRSTCHAR
926 TRIE_LIST_ITEM(state,charid).forid,
927 (UV)TRIE_LIST_ITEM(state,charid).newstate
930 PerlIO_printf(Perl_debug_log, "\n%*s| ",
931 (int)((depth * 2) + 14), "");
934 PerlIO_printf( Perl_debug_log, "\n");
939 dump_trie_interim_table(trie,next_alloc)
940 Dumps a fully constructed but uncompressed trie in table form.
941 This is the normal DFA style state transition table, with a few
942 twists to facilitate compression later.
943 Used for debugging make_trie().
946 S_dump_trie_interim_table(pTHX_ const struct _reg_trie_data *trie, U32 next_alloc, U32 depth)
950 SV *sv=sv_newmortal();
951 int colwidth= trie->widecharmap ? 6 : 4;
952 GET_RE_DEBUG_FLAGS_DECL;
955 print out the table precompression so that we can do a visual check
956 that they are identical.
959 PerlIO_printf( Perl_debug_log, "%*sChar : ",(int)depth * 2 + 2,"" );
961 for( charid = 0 ; charid < trie->uniquecharcount ; charid++ ) {
962 SV ** const tmp = av_fetch( trie->revcharmap, charid, 0);
964 PerlIO_printf( Perl_debug_log, "%*s",
966 pv_pretty(sv, SvPV_nolen_const(*tmp), SvCUR(*tmp), colwidth,
967 PL_colors[0], PL_colors[1],
968 (SvUTF8(*tmp) ? PERL_PV_ESCAPE_UNI : 0) |
969 PERL_PV_ESCAPE_FIRSTCHAR
975 PerlIO_printf( Perl_debug_log, "\n%*sState+-",(int)depth * 2 + 2,"" );
977 for( charid=0 ; charid < trie->uniquecharcount ; charid++ ) {
978 PerlIO_printf( Perl_debug_log, "%.*s", colwidth,"--------");
981 PerlIO_printf( Perl_debug_log, "\n" );
983 for( state=1 ; state < next_alloc ; state += trie->uniquecharcount ) {
985 PerlIO_printf( Perl_debug_log, "%*s%4"UVXf" : ",
986 (int)depth * 2 + 2,"",
987 (UV)TRIE_NODENUM( state ) );
989 for( charid = 0 ; charid < trie->uniquecharcount ; charid++ ) {
990 UV v=(UV)SAFE_TRIE_NODENUM( trie->trans[ state + charid ].next );
992 PerlIO_printf( Perl_debug_log, "%*"UVXf, colwidth, v );
994 PerlIO_printf( Perl_debug_log, "%*s", colwidth, "." );
996 if ( ! trie->states[ TRIE_NODENUM( state ) ].wordnum ) {
997 PerlIO_printf( Perl_debug_log, " (%4"UVXf")\n", (UV)trie->trans[ state ].check );
999 PerlIO_printf( Perl_debug_log, " (%4"UVXf") W%4X\n", (UV)trie->trans[ state ].check,
1000 trie->states[ TRIE_NODENUM( state ) ].wordnum );
1007 /* make_trie(startbranch,first,last,tail,word_count,flags,depth)
1008 startbranch: the first branch in the whole branch sequence
1009 first : start branch of sequence of branch-exact nodes.
1010 May be the same as startbranch
1011 last : Thing following the last branch.
1012 May be the same as tail.
1013 tail : item following the branch sequence
1014 count : words in the sequence
1015 flags : currently the OP() type we will be building one of /EXACT(|F|Fl)/
1016 depth : indent depth
1018 Inplace optimizes a sequence of 2 or more Branch-Exact nodes into a TRIE node.
1020 A trie is an N'ary tree where the branches are determined by digital
1021 decomposition of the key. IE, at the root node you look up the 1st character and
1022 follow that branch repeat until you find the end of the branches. Nodes can be
1023 marked as "accepting" meaning they represent a complete word. Eg:
1027 would convert into the following structure. Numbers represent states, letters
1028 following numbers represent valid transitions on the letter from that state, if
1029 the number is in square brackets it represents an accepting state, otherwise it
1030 will be in parenthesis.
1032 +-h->+-e->[3]-+-r->(8)-+-s->[9]
1036 (1) +-i->(6)-+-s->[7]
1038 +-s->(3)-+-h->(4)-+-e->[5]
1040 Accept Word Mapping: 3=>1 (he),5=>2 (she), 7=>3 (his), 9=>4 (hers)
1042 This shows that when matching against the string 'hers' we will begin at state 1
1043 read 'h' and move to state 2, read 'e' and move to state 3 which is accepting,
1044 then read 'r' and go to state 8 followed by 's' which takes us to state 9 which
1045 is also accepting. Thus we know that we can match both 'he' and 'hers' with a
1046 single traverse. We store a mapping from accepting to state to which word was
1047 matched, and then when we have multiple possibilities we try to complete the
1048 rest of the regex in the order in which they occured in the alternation.
1050 The only prior NFA like behaviour that would be changed by the TRIE support is
1051 the silent ignoring of duplicate alternations which are of the form:
1053 / (DUPE|DUPE) X? (?{ ... }) Y /x
1055 Thus EVAL blocks follwing a trie may be called a different number of times with
1056 and without the optimisation. With the optimisations dupes will be silently
1057 ignored. This inconsistant behaviour of EVAL type nodes is well established as
1058 the following demonstrates:
1060 'words'=~/(word|word|word)(?{ print $1 })[xyz]/
1062 which prints out 'word' three times, but
1064 'words'=~/(word|word|word)(?{ print $1 })S/
1066 which doesnt print it out at all. This is due to other optimisations kicking in.
1068 Example of what happens on a structural level:
1070 The regexp /(ac|ad|ab)+/ will produce the folowing debug output:
1072 1: CURLYM[1] {1,32767}(18)
1083 This would be optimizable with startbranch=5, first=5, last=16, tail=16
1084 and should turn into:
1086 1: CURLYM[1] {1,32767}(18)
1088 [Words:3 Chars Stored:6 Unique Chars:4 States:5 NCP:1]
1096 Cases where tail != last would be like /(?foo|bar)baz/:
1106 which would be optimizable with startbranch=1, first=1, last=7, tail=8
1107 and would end up looking like:
1110 [Words:2 Chars Stored:6 Unique Chars:5 States:7 NCP:1]
1117 d = uvuni_to_utf8_flags(d, uv, 0);
1119 is the recommended Unicode-aware way of saying
1124 #define TRIE_STORE_REVCHAR \
1126 SV *tmp = newSVpvs(""); \
1127 if (UTF) SvUTF8_on(tmp); \
1128 Perl_sv_catpvf( aTHX_ tmp, "%c", (int)uvc ); \
1129 av_push( TRIE_REVCHARMAP(trie), tmp ); \
1132 #define TRIE_READ_CHAR STMT_START { \
1136 if ( foldlen > 0 ) { \
1137 uvc = utf8n_to_uvuni( scan, UTF8_MAXLEN, &len, uniflags ); \
1142 uvc = utf8n_to_uvuni( (const U8*)uc, UTF8_MAXLEN, &len, uniflags);\
1143 uvc = to_uni_fold( uvc, foldbuf, &foldlen ); \
1144 foldlen -= UNISKIP( uvc ); \
1145 scan = foldbuf + UNISKIP( uvc ); \
1148 uvc = utf8n_to_uvuni( (const U8*)uc, UTF8_MAXLEN, &len, uniflags);\
1158 #define TRIE_LIST_PUSH(state,fid,ns) STMT_START { \
1159 if ( TRIE_LIST_CUR( state ) >=TRIE_LIST_LEN( state ) ) { \
1160 U32 ging = TRIE_LIST_LEN( state ) *= 2; \
1161 Renew( trie->states[ state ].trans.list, ging, reg_trie_trans_le ); \
1163 TRIE_LIST_ITEM( state, TRIE_LIST_CUR( state ) ).forid = fid; \
1164 TRIE_LIST_ITEM( state, TRIE_LIST_CUR( state ) ).newstate = ns; \
1165 TRIE_LIST_CUR( state )++; \
1168 #define TRIE_LIST_NEW(state) STMT_START { \
1169 Newxz( trie->states[ state ].trans.list, \
1170 4, reg_trie_trans_le ); \
1171 TRIE_LIST_CUR( state ) = 1; \
1172 TRIE_LIST_LEN( state ) = 4; \
1175 #define TRIE_HANDLE_WORD(state) STMT_START { \
1176 U16 dupe= trie->states[ state ].wordnum; \
1177 regnode * const noper_next = regnext( noper ); \
1179 if (trie->wordlen) \
1180 trie->wordlen[ curword ] = wordlen; \
1182 /* store the word for dumping */ \
1184 if (OP(noper) != NOTHING) \
1185 tmp = newSVpvn(STRING(noper), STR_LEN(noper)); \
1187 tmp = newSVpvn( "", 0 ); \
1188 if ( UTF ) SvUTF8_on( tmp ); \
1189 av_push( trie->words, tmp ); \
1194 if ( noper_next < tail ) { \
1196 Newxz( trie->jump, word_count + 1, U16); \
1197 trie->jump[curword] = (U16)(noper_next - convert); \
1199 jumper = noper_next; \
1201 nextbranch= regnext(cur); \
1205 /* So it's a dupe. This means we need to maintain a */\
1206 /* linked-list from the first to the next. */\
1207 /* we only allocate the nextword buffer when there */\
1208 /* a dupe, so first time we have to do the allocation */\
1209 if (!trie->nextword) \
1210 Newxz( trie->nextword, word_count + 1, U16); \
1211 while ( trie->nextword[dupe] ) \
1212 dupe= trie->nextword[dupe]; \
1213 trie->nextword[dupe]= curword; \
1215 /* we haven't inserted this word yet. */ \
1216 trie->states[ state ].wordnum = curword; \
1221 #define TRIE_TRANS_STATE(state,base,ucharcount,charid,special) \
1222 ( ( base + charid >= ucharcount \
1223 && base + charid < ubound \
1224 && state == trie->trans[ base - ucharcount + charid ].check \
1225 && trie->trans[ base - ucharcount + charid ].next ) \
1226 ? trie->trans[ base - ucharcount + charid ].next \
1227 : ( state==1 ? special : 0 ) \
1231 #define MADE_JUMP_TRIE 2
1232 #define MADE_EXACT_TRIE 4
1235 S_make_trie(pTHX_ RExC_state_t *pRExC_state, regnode *startbranch, regnode *first, regnode *last, regnode *tail, U32 word_count, U32 flags, U32 depth)
1238 /* first pass, loop through and scan words */
1239 reg_trie_data *trie;
1241 const U32 uniflags = UTF8_ALLOW_DEFAULT;
1246 regnode *jumper = NULL;
1247 regnode *nextbranch = NULL;
1248 regnode *convert = NULL;
1249 /* we just use folder as a flag in utf8 */
1250 const U8 * const folder = ( flags == EXACTF
1252 : ( flags == EXACTFL
1258 const U32 data_slot = add_data( pRExC_state, 1, "t" );
1259 SV *re_trie_maxbuff;
1261 /* these are only used during construction but are useful during
1262 * debugging so we store them in the struct when debugging.
1264 STRLEN trie_charcount=0;
1265 AV *trie_revcharmap;
1267 GET_RE_DEBUG_FLAGS_DECL;
1269 PERL_UNUSED_ARG(depth);
1272 Newxz( trie, 1, reg_trie_data );
1274 trie->startstate = 1;
1275 trie->wordcount = word_count;
1276 RExC_rx->data->data[ data_slot ] = (void*)trie;
1277 Newxz( trie->charmap, 256, U16 );
1278 if (!(UTF && folder))
1279 Newxz( trie->bitmap, ANYOF_BITMAP_SIZE, char );
1281 trie->words = newAV();
1283 TRIE_REVCHARMAP(trie) = newAV();
1285 re_trie_maxbuff = get_sv(RE_TRIE_MAXBUF_NAME, 1);
1286 if (!SvIOK(re_trie_maxbuff)) {
1287 sv_setiv(re_trie_maxbuff, RE_TRIE_MAXBUF_INIT);
1290 PerlIO_printf( Perl_debug_log,
1291 "%*smake_trie start==%d, first==%d, last==%d, tail==%d depth=%d\n",
1292 (int)depth * 2 + 2, "",
1293 REG_NODE_NUM(startbranch),REG_NODE_NUM(first),
1294 REG_NODE_NUM(last), REG_NODE_NUM(tail),
1298 /* Find the node we are going to overwrite */
1299 if ( first == startbranch && OP( last ) != BRANCH ) {
1300 /* whole branch chain */
1303 /* branch sub-chain */
1304 convert = NEXTOPER( first );
1307 /* -- First loop and Setup --
1309 We first traverse the branches and scan each word to determine if it
1310 contains widechars, and how many unique chars there are, this is
1311 important as we have to build a table with at least as many columns as we
1314 We use an array of integers to represent the character codes 0..255
1315 (trie->charmap) and we use a an HV* to store unicode characters. We use the
1316 native representation of the character value as the key and IV's for the
1319 *TODO* If we keep track of how many times each character is used we can
1320 remap the columns so that the table compression later on is more
1321 efficient in terms of memory by ensuring most common value is in the
1322 middle and the least common are on the outside. IMO this would be better
1323 than a most to least common mapping as theres a decent chance the most
1324 common letter will share a node with the least common, meaning the node
1325 will not be compressable. With a middle is most common approach the worst
1326 case is when we have the least common nodes twice.
1330 for ( cur = first ; cur < last ; cur = regnext( cur ) ) {
1331 regnode * const noper = NEXTOPER( cur );
1332 const U8 *uc = (U8*)STRING( noper );
1333 const U8 * const e = uc + STR_LEN( noper );
1335 U8 foldbuf[ UTF8_MAXBYTES_CASE + 1 ];
1336 const U8 *scan = (U8*)NULL;
1337 U32 wordlen = 0; /* required init */
1340 if (OP(noper) == NOTHING) {
1345 TRIE_BITMAP_SET(trie,*uc);
1346 if ( folder ) TRIE_BITMAP_SET(trie,folder[ *uc ]);
1348 for ( ; uc < e ; uc += len ) {
1349 TRIE_CHARCOUNT(trie)++;
1353 if ( !trie->charmap[ uvc ] ) {
1354 trie->charmap[ uvc ]=( ++trie->uniquecharcount );
1356 trie->charmap[ folder[ uvc ] ] = trie->charmap[ uvc ];
1361 if ( !trie->widecharmap )
1362 trie->widecharmap = newHV();
1364 svpp = hv_fetch( trie->widecharmap, (char*)&uvc, sizeof( UV ), 1 );
1367 Perl_croak( aTHX_ "error creating/fetching widecharmap entry for 0x%"UVXf, uvc );
1369 if ( !SvTRUE( *svpp ) ) {
1370 sv_setiv( *svpp, ++trie->uniquecharcount );
1375 if( cur == first ) {
1378 } else if (chars < trie->minlen) {
1380 } else if (chars > trie->maxlen) {
1384 } /* end first pass */
1385 DEBUG_TRIE_COMPILE_r(
1386 PerlIO_printf( Perl_debug_log, "%*sTRIE(%s): W:%d C:%d Uq:%d Min:%d Max:%d\n",
1387 (int)depth * 2 + 2,"",
1388 ( trie->widecharmap ? "UTF8" : "NATIVE" ), (int)word_count,
1389 (int)TRIE_CHARCOUNT(trie), trie->uniquecharcount,
1390 (int)trie->minlen, (int)trie->maxlen )
1392 Newxz( trie->wordlen, word_count, U32 );
1395 We now know what we are dealing with in terms of unique chars and
1396 string sizes so we can calculate how much memory a naive
1397 representation using a flat table will take. If it's over a reasonable
1398 limit (as specified by ${^RE_TRIE_MAXBUF}) we use a more memory
1399 conservative but potentially much slower representation using an array
1402 At the end we convert both representations into the same compressed
1403 form that will be used in regexec.c for matching with. The latter
1404 is a form that cannot be used to construct with but has memory
1405 properties similar to the list form and access properties similar
1406 to the table form making it both suitable for fast searches and
1407 small enough that its feasable to store for the duration of a program.
1409 See the comment in the code where the compressed table is produced
1410 inplace from the flat tabe representation for an explanation of how
1411 the compression works.
1416 if ( (IV)( ( TRIE_CHARCOUNT(trie) + 1 ) * trie->uniquecharcount + 1) > SvIV(re_trie_maxbuff) ) {
1418 Second Pass -- Array Of Lists Representation
1420 Each state will be represented by a list of charid:state records
1421 (reg_trie_trans_le) the first such element holds the CUR and LEN
1422 points of the allocated array. (See defines above).
1424 We build the initial structure using the lists, and then convert
1425 it into the compressed table form which allows faster lookups
1426 (but cant be modified once converted).
1429 STRLEN transcount = 1;
1431 DEBUG_TRIE_COMPILE_MORE_r( PerlIO_printf( Perl_debug_log,
1432 "%*sCompiling trie using list compiler\n",
1433 (int)depth * 2 + 2, ""));
1435 Newxz( trie->states, TRIE_CHARCOUNT(trie) + 2, reg_trie_state );
1439 for ( cur = first ; cur < last ; cur = regnext( cur ) ) {
1441 regnode * const noper = NEXTOPER( cur );
1442 U8 *uc = (U8*)STRING( noper );
1443 const U8 * const e = uc + STR_LEN( noper );
1444 U32 state = 1; /* required init */
1445 U16 charid = 0; /* sanity init */
1446 U8 *scan = (U8*)NULL; /* sanity init */
1447 STRLEN foldlen = 0; /* required init */
1448 U32 wordlen = 0; /* required init */
1449 U8 foldbuf[ UTF8_MAXBYTES_CASE + 1 ];
1451 if (OP(noper) != NOTHING) {
1452 for ( ; uc < e ; uc += len ) {
1457 charid = trie->charmap[ uvc ];
1459 SV** const svpp = hv_fetch( trie->widecharmap, (char*)&uvc, sizeof( UV ), 0);
1463 charid=(U16)SvIV( *svpp );
1466 /* charid is now 0 if we dont know the char read, or nonzero if we do */
1473 if ( !trie->states[ state ].trans.list ) {
1474 TRIE_LIST_NEW( state );
1476 for ( check = 1; check <= TRIE_LIST_USED( state ); check++ ) {
1477 if ( TRIE_LIST_ITEM( state, check ).forid == charid ) {
1478 newstate = TRIE_LIST_ITEM( state, check ).newstate;
1483 newstate = next_alloc++;
1484 TRIE_LIST_PUSH( state, charid, newstate );
1489 Perl_croak( aTHX_ "panic! In trie construction, no char mapping for %"IVdf, uvc );
1493 TRIE_HANDLE_WORD(state);
1495 } /* end second pass */
1497 /* next alloc is the NEXT state to be allocated */
1498 trie->statecount = next_alloc;
1499 Renew( trie->states, next_alloc, reg_trie_state );
1501 /* and now dump it out before we compress it */
1502 DEBUG_TRIE_COMPILE_MORE_r(
1503 dump_trie_interim_list(trie,next_alloc,depth+1)
1506 Newxz( trie->trans, transcount ,reg_trie_trans );
1513 for( state=1 ; state < next_alloc ; state ++ ) {
1517 DEBUG_TRIE_COMPILE_MORE_r(
1518 PerlIO_printf( Perl_debug_log, "tp: %d zp: %d ",tp,zp)
1522 if (trie->states[state].trans.list) {
1523 U16 minid=TRIE_LIST_ITEM( state, 1).forid;
1527 for( idx = 2 ; idx <= TRIE_LIST_USED( state ) ; idx++ ) {
1528 const U16 forid = TRIE_LIST_ITEM( state, idx).forid;
1529 if ( forid < minid ) {
1531 } else if ( forid > maxid ) {
1535 if ( transcount < tp + maxid - minid + 1) {
1537 Renew( trie->trans, transcount, reg_trie_trans );
1538 Zero( trie->trans + (transcount / 2), transcount / 2 , reg_trie_trans );
1540 base = trie->uniquecharcount + tp - minid;
1541 if ( maxid == minid ) {
1543 for ( ; zp < tp ; zp++ ) {
1544 if ( ! trie->trans[ zp ].next ) {
1545 base = trie->uniquecharcount + zp - minid;
1546 trie->trans[ zp ].next = TRIE_LIST_ITEM( state, 1).newstate;
1547 trie->trans[ zp ].check = state;
1553 trie->trans[ tp ].next = TRIE_LIST_ITEM( state, 1).newstate;
1554 trie->trans[ tp ].check = state;
1559 for ( idx=1; idx <= TRIE_LIST_USED( state ) ; idx++ ) {
1560 const U32 tid = base - trie->uniquecharcount + TRIE_LIST_ITEM( state, idx ).forid;
1561 trie->trans[ tid ].next = TRIE_LIST_ITEM( state, idx ).newstate;
1562 trie->trans[ tid ].check = state;
1564 tp += ( maxid - minid + 1 );
1566 Safefree(trie->states[ state ].trans.list);
1569 DEBUG_TRIE_COMPILE_MORE_r(
1570 PerlIO_printf( Perl_debug_log, " base: %d\n",base);
1573 trie->states[ state ].trans.base=base;
1575 trie->lasttrans = tp + 1;
1579 Second Pass -- Flat Table Representation.
1581 we dont use the 0 slot of either trans[] or states[] so we add 1 to each.
1582 We know that we will need Charcount+1 trans at most to store the data
1583 (one row per char at worst case) So we preallocate both structures
1584 assuming worst case.
1586 We then construct the trie using only the .next slots of the entry
1589 We use the .check field of the first entry of the node temporarily to
1590 make compression both faster and easier by keeping track of how many non
1591 zero fields are in the node.
1593 Since trans are numbered from 1 any 0 pointer in the table is a FAIL
1596 There are two terms at use here: state as a TRIE_NODEIDX() which is a
1597 number representing the first entry of the node, and state as a
1598 TRIE_NODENUM() which is the trans number. state 1 is TRIE_NODEIDX(1) and
1599 TRIE_NODENUM(1), state 2 is TRIE_NODEIDX(2) and TRIE_NODENUM(3) if there
1600 are 2 entrys per node. eg:
1608 The table is internally in the right hand, idx form. However as we also
1609 have to deal with the states array which is indexed by nodenum we have to
1610 use TRIE_NODENUM() to convert.
1613 DEBUG_TRIE_COMPILE_MORE_r( PerlIO_printf( Perl_debug_log,
1614 "%*sCompiling trie using table compiler\n",
1615 (int)depth * 2 + 2, ""));
1617 Newxz( trie->trans, ( TRIE_CHARCOUNT(trie) + 1 ) * trie->uniquecharcount + 1,
1619 Newxz( trie->states, TRIE_CHARCOUNT(trie) + 2, reg_trie_state );
1620 next_alloc = trie->uniquecharcount + 1;
1623 for ( cur = first ; cur < last ; cur = regnext( cur ) ) {
1625 regnode * const noper = NEXTOPER( cur );
1626 const U8 *uc = (U8*)STRING( noper );
1627 const U8 * const e = uc + STR_LEN( noper );
1629 U32 state = 1; /* required init */
1631 U16 charid = 0; /* sanity init */
1632 U32 accept_state = 0; /* sanity init */
1633 U8 *scan = (U8*)NULL; /* sanity init */
1635 STRLEN foldlen = 0; /* required init */
1636 U32 wordlen = 0; /* required init */
1637 U8 foldbuf[ UTF8_MAXBYTES_CASE + 1 ];
1639 if ( OP(noper) != NOTHING ) {
1640 for ( ; uc < e ; uc += len ) {
1645 charid = trie->charmap[ uvc ];
1647 SV* const * const svpp = hv_fetch( trie->widecharmap, (char*)&uvc, sizeof( UV ), 0);
1648 charid = svpp ? (U16)SvIV(*svpp) : 0;
1652 if ( !trie->trans[ state + charid ].next ) {
1653 trie->trans[ state + charid ].next = next_alloc;
1654 trie->trans[ state ].check++;
1655 next_alloc += trie->uniquecharcount;
1657 state = trie->trans[ state + charid ].next;
1659 Perl_croak( aTHX_ "panic! In trie construction, no char mapping for %"IVdf, uvc );
1661 /* charid is now 0 if we dont know the char read, or nonzero if we do */
1664 accept_state = TRIE_NODENUM( state );
1665 TRIE_HANDLE_WORD(accept_state);
1667 } /* end second pass */
1669 /* and now dump it out before we compress it */
1670 DEBUG_TRIE_COMPILE_MORE_r(
1671 dump_trie_interim_table(trie,next_alloc,depth+1)
1676 * Inplace compress the table.*
1678 For sparse data sets the table constructed by the trie algorithm will
1679 be mostly 0/FAIL transitions or to put it another way mostly empty.
1680 (Note that leaf nodes will not contain any transitions.)
1682 This algorithm compresses the tables by eliminating most such
1683 transitions, at the cost of a modest bit of extra work during lookup:
1685 - Each states[] entry contains a .base field which indicates the
1686 index in the state[] array wheres its transition data is stored.
1688 - If .base is 0 there are no valid transitions from that node.
1690 - If .base is nonzero then charid is added to it to find an entry in
1693 -If trans[states[state].base+charid].check!=state then the
1694 transition is taken to be a 0/Fail transition. Thus if there are fail
1695 transitions at the front of the node then the .base offset will point
1696 somewhere inside the previous nodes data (or maybe even into a node
1697 even earlier), but the .check field determines if the transition is
1701 The following process inplace converts the table to the compressed
1702 table: We first do not compress the root node 1,and mark its all its
1703 .check pointers as 1 and set its .base pointer as 1 as well. This
1704 allows to do a DFA construction from the compressed table later, and
1705 ensures that any .base pointers we calculate later are greater than
1708 - We set 'pos' to indicate the first entry of the second node.
1710 - We then iterate over the columns of the node, finding the first and
1711 last used entry at l and m. We then copy l..m into pos..(pos+m-l),
1712 and set the .check pointers accordingly, and advance pos
1713 appropriately and repreat for the next node. Note that when we copy
1714 the next pointers we have to convert them from the original
1715 NODEIDX form to NODENUM form as the former is not valid post
1718 - If a node has no transitions used we mark its base as 0 and do not
1719 advance the pos pointer.
1721 - If a node only has one transition we use a second pointer into the
1722 structure to fill in allocated fail transitions from other states.
1723 This pointer is independent of the main pointer and scans forward
1724 looking for null transitions that are allocated to a state. When it
1725 finds one it writes the single transition into the "hole". If the
1726 pointer doesnt find one the single transition is appended as normal.
1728 - Once compressed we can Renew/realloc the structures to release the
1731 See "Table-Compression Methods" in sec 3.9 of the Red Dragon,
1732 specifically Fig 3.47 and the associated pseudocode.
1736 const U32 laststate = TRIE_NODENUM( next_alloc );
1739 trie->statecount = laststate;
1741 for ( state = 1 ; state < laststate ; state++ ) {
1743 const U32 stateidx = TRIE_NODEIDX( state );
1744 const U32 o_used = trie->trans[ stateidx ].check;
1745 U32 used = trie->trans[ stateidx ].check;
1746 trie->trans[ stateidx ].check = 0;
1748 for ( charid = 0 ; used && charid < trie->uniquecharcount ; charid++ ) {
1749 if ( flag || trie->trans[ stateidx + charid ].next ) {
1750 if ( trie->trans[ stateidx + charid ].next ) {
1752 for ( ; zp < pos ; zp++ ) {
1753 if ( ! trie->trans[ zp ].next ) {
1757 trie->states[ state ].trans.base = zp + trie->uniquecharcount - charid ;
1758 trie->trans[ zp ].next = SAFE_TRIE_NODENUM( trie->trans[ stateidx + charid ].next );
1759 trie->trans[ zp ].check = state;
1760 if ( ++zp > pos ) pos = zp;
1767 trie->states[ state ].trans.base = pos + trie->uniquecharcount - charid ;
1769 trie->trans[ pos ].next = SAFE_TRIE_NODENUM( trie->trans[ stateidx + charid ].next );
1770 trie->trans[ pos ].check = state;
1775 trie->lasttrans = pos + 1;
1776 Renew( trie->states, laststate, reg_trie_state);
1777 DEBUG_TRIE_COMPILE_MORE_r(
1778 PerlIO_printf( Perl_debug_log,
1779 "%*sAlloc: %d Orig: %"IVdf" elements, Final:%"IVdf". Savings of %%%5.2f\n",
1780 (int)depth * 2 + 2,"",
1781 (int)( ( TRIE_CHARCOUNT(trie) + 1 ) * trie->uniquecharcount + 1 ),
1784 ( ( next_alloc - pos ) * 100 ) / (double)next_alloc );
1787 } /* end table compress */
1789 DEBUG_TRIE_COMPILE_MORE_r(
1790 PerlIO_printf(Perl_debug_log, "%*sStatecount:%"UVxf" Lasttrans:%"UVxf"\n",
1791 (int)depth * 2 + 2, "",
1792 (UV)trie->statecount,
1793 (UV)trie->lasttrans)
1795 /* resize the trans array to remove unused space */
1796 Renew( trie->trans, trie->lasttrans, reg_trie_trans);
1798 /* and now dump out the compressed format */
1799 DEBUG_TRIE_COMPILE_r(
1800 dump_trie(trie,depth+1)
1803 { /* Modify the program and insert the new TRIE node*/
1804 U8 nodetype =(U8)(flags & 0xFF);
1808 regnode *optimize = NULL;
1810 U32 mjd_nodelen = 0;
1813 This means we convert either the first branch or the first Exact,
1814 depending on whether the thing following (in 'last') is a branch
1815 or not and whther first is the startbranch (ie is it a sub part of
1816 the alternation or is it the whole thing.)
1817 Assuming its a sub part we conver the EXACT otherwise we convert
1818 the whole branch sequence, including the first.
1820 /* Find the node we are going to overwrite */
1821 if ( first != startbranch || OP( last ) == BRANCH ) {
1822 /* branch sub-chain */
1823 NEXT_OFF( first ) = (U16)(last - first);
1825 mjd_offset= Node_Offset((convert));
1826 mjd_nodelen= Node_Length((convert));
1828 /* whole branch chain */
1831 const regnode *nop = NEXTOPER( convert );
1832 mjd_offset= Node_Offset((nop));
1833 mjd_nodelen= Node_Length((nop));
1838 PerlIO_printf(Perl_debug_log, "%*sMJD offset:%"UVuf" MJD length:%"UVuf"\n",
1839 (int)depth * 2 + 2, "",
1840 (UV)mjd_offset, (UV)mjd_nodelen)
1843 /* But first we check to see if there is a common prefix we can
1844 split out as an EXACT and put in front of the TRIE node. */
1845 trie->startstate= 1;
1846 if ( trie->bitmap && !trie->widecharmap && !trie->jump ) {
1848 for ( state = 1 ; state < trie->statecount-1 ; state++ ) {
1852 const U32 base = trie->states[ state ].trans.base;
1854 if ( trie->states[state].wordnum )
1857 for ( ofs = 0 ; ofs < trie->uniquecharcount ; ofs++ ) {
1858 if ( ( base + ofs >= trie->uniquecharcount ) &&
1859 ( base + ofs - trie->uniquecharcount < trie->lasttrans ) &&
1860 trie->trans[ base + ofs - trie->uniquecharcount ].check == state )
1862 if ( ++count > 1 ) {
1863 SV **tmp = av_fetch( TRIE_REVCHARMAP(trie), ofs, 0);
1864 const U8 *ch = (U8*)SvPV_nolen_const( *tmp );
1865 if ( state == 1 ) break;
1867 Zero(trie->bitmap, ANYOF_BITMAP_SIZE, char);
1869 PerlIO_printf(Perl_debug_log,
1870 "%*sNew Start State=%"UVuf" Class: [",
1871 (int)depth * 2 + 2, "",
1874 SV ** const tmp = av_fetch( TRIE_REVCHARMAP(trie), idx, 0);
1875 const U8 * const ch = (U8*)SvPV_nolen_const( *tmp );
1877 TRIE_BITMAP_SET(trie,*ch);
1879 TRIE_BITMAP_SET(trie, folder[ *ch ]);
1881 PerlIO_printf(Perl_debug_log, (char*)ch)
1885 TRIE_BITMAP_SET(trie,*ch);
1887 TRIE_BITMAP_SET(trie,folder[ *ch ]);
1888 DEBUG_OPTIMISE_r(PerlIO_printf( Perl_debug_log,"%s", ch));
1894 SV **tmp = av_fetch( TRIE_REVCHARMAP(trie), idx, 0);
1895 const char *ch = SvPV_nolen_const( *tmp );
1897 PerlIO_printf( Perl_debug_log,
1898 "%*sPrefix State: %"UVuf" Idx:%"UVuf" Char='%s'\n",
1899 (int)depth * 2 + 2, "",
1900 (UV)state, (UV)idx, ch)
1903 OP( convert ) = nodetype;
1904 str=STRING(convert);
1913 DEBUG_OPTIMISE_r(PerlIO_printf( Perl_debug_log,"]\n"));
1919 regnode *n = convert+NODE_SZ_STR(convert);
1920 NEXT_OFF(convert) = NODE_SZ_STR(convert);
1921 trie->startstate = state;
1922 trie->minlen -= (state - 1);
1923 trie->maxlen -= (state - 1);
1925 regnode *fix = convert;
1927 Set_Node_Offset_Length(convert, mjd_offset, state - 1);
1928 while( ++fix < n ) {
1929 Set_Node_Offset_Length(fix, 0, 0);
1935 NEXT_OFF(convert) = (U16)(tail - convert);
1936 DEBUG_r(optimize= n);
1942 if ( trie->maxlen ) {
1943 NEXT_OFF( convert ) = (U16)(tail - convert);
1944 ARG_SET( convert, data_slot );
1945 /* Store the offset to the first unabsorbed branch in
1946 jump[0], which is otherwise unused by the jump logic.
1947 We use this when dumping a trie and during optimisation. */
1949 trie->jump[0] = (U16)(nextbranch - convert);
1952 if ( !trie->states[trie->startstate].wordnum && trie->bitmap &&
1953 ( (char *)jumper - (char *)convert) >= (int)sizeof(struct regnode_charclass) )
1955 OP( convert ) = TRIEC;
1956 Copy(trie->bitmap, ((struct regnode_charclass *)convert)->bitmap, ANYOF_BITMAP_SIZE, char);
1957 Safefree(trie->bitmap);
1960 OP( convert ) = TRIE;
1962 /* store the type in the flags */
1963 convert->flags = nodetype;
1967 + regarglen[ OP( convert ) ];
1969 /* XXX We really should free up the resource in trie now,
1970 as we won't use them - (which resources?) dmq */
1972 /* needed for dumping*/
1973 DEBUG_r(if (optimize) {
1974 regnode *opt = convert;
1975 while ( ++opt < optimize) {
1976 Set_Node_Offset_Length(opt,0,0);
1979 Try to clean up some of the debris left after the
1982 while( optimize < jumper ) {
1983 mjd_nodelen += Node_Length((optimize));
1984 OP( optimize ) = OPTIMIZED;
1985 Set_Node_Offset_Length(optimize,0,0);
1988 Set_Node_Offset_Length(convert,mjd_offset,mjd_nodelen);
1990 } /* end node insert */
1992 SvREFCNT_dec(TRIE_REVCHARMAP(trie));
1996 : trie->startstate>1
2002 S_make_trie_failtable(pTHX_ RExC_state_t *pRExC_state, regnode *source, regnode *stclass, U32 depth)
2004 /* The Trie is constructed and compressed now so we can build a fail array now if its needed
2006 This is basically the Aho-Corasick algorithm. Its from exercise 3.31 and 3.32 in the
2007 "Red Dragon" -- Compilers, principles, techniques, and tools. Aho, Sethi, Ullman 1985/88
2010 We find the fail state for each state in the trie, this state is the longest proper
2011 suffix of the current states 'word' that is also a proper prefix of another word in our
2012 trie. State 1 represents the word '' and is the thus the default fail state. This allows
2013 the DFA not to have to restart after its tried and failed a word at a given point, it
2014 simply continues as though it had been matching the other word in the first place.
2016 'abcdgu'=~/abcdefg|cdgu/
2017 When we get to 'd' we are still matching the first word, we would encounter 'g' which would
2018 fail, which would bring use to the state representing 'd' in the second word where we would
2019 try 'g' and succeed, prodceding to match 'cdgu'.
2021 /* add a fail transition */
2022 reg_trie_data *trie=(reg_trie_data *)RExC_rx->data->data[ARG(source)];
2024 const U32 ucharcount = trie->uniquecharcount;
2025 const U32 numstates = trie->statecount;
2026 const U32 ubound = trie->lasttrans + ucharcount;
2030 U32 base = trie->states[ 1 ].trans.base;
2033 const U32 data_slot = add_data( pRExC_state, 1, "T" );
2034 GET_RE_DEBUG_FLAGS_DECL;
2036 PERL_UNUSED_ARG(depth);
2040 ARG_SET( stclass, data_slot );
2041 Newxz( aho, 1, reg_ac_data );
2042 RExC_rx->data->data[ data_slot ] = (void*)aho;
2044 aho->states=(reg_trie_state *)savepvn((const char*)trie->states,
2045 numstates * sizeof(reg_trie_state));
2046 Newxz( q, numstates, U32);
2047 Newxz( aho->fail, numstates, U32 );
2050 /* initialize fail[0..1] to be 1 so that we always have
2051 a valid final fail state */
2052 fail[ 0 ] = fail[ 1 ] = 1;
2054 for ( charid = 0; charid < ucharcount ; charid++ ) {
2055 const U32 newstate = TRIE_TRANS_STATE( 1, base, ucharcount, charid, 0 );
2057 q[ q_write ] = newstate;
2058 /* set to point at the root */
2059 fail[ q[ q_write++ ] ]=1;
2062 while ( q_read < q_write) {
2063 const U32 cur = q[ q_read++ % numstates ];
2064 base = trie->states[ cur ].trans.base;
2066 for ( charid = 0 ; charid < ucharcount ; charid++ ) {
2067 const U32 ch_state = TRIE_TRANS_STATE( cur, base, ucharcount, charid, 1 );
2069 U32 fail_state = cur;
2072 fail_state = fail[ fail_state ];
2073 fail_base = aho->states[ fail_state ].trans.base;
2074 } while ( !TRIE_TRANS_STATE( fail_state, fail_base, ucharcount, charid, 1 ) );
2076 fail_state = TRIE_TRANS_STATE( fail_state, fail_base, ucharcount, charid, 1 );
2077 fail[ ch_state ] = fail_state;
2078 if ( !aho->states[ ch_state ].wordnum && aho->states[ fail_state ].wordnum )
2080 aho->states[ ch_state ].wordnum = aho->states[ fail_state ].wordnum;
2082 q[ q_write++ % numstates] = ch_state;
2086 /* restore fail[0..1] to 0 so that we "fall out" of the AC loop
2087 when we fail in state 1, this allows us to use the
2088 charclass scan to find a valid start char. This is based on the principle
2089 that theres a good chance the string being searched contains lots of stuff
2090 that cant be a start char.
2092 fail[ 0 ] = fail[ 1 ] = 0;
2093 DEBUG_TRIE_COMPILE_r({
2094 PerlIO_printf(Perl_debug_log,
2095 "%*sStclass Failtable (%"UVuf" states): 0",
2096 (int)(depth * 2), "", (UV)numstates
2098 for( q_read=1; q_read<numstates; q_read++ ) {
2099 PerlIO_printf(Perl_debug_log, ", %"UVuf, (UV)fail[q_read]);
2101 PerlIO_printf(Perl_debug_log, "\n");
2104 /*RExC_seen |= REG_SEEN_TRIEDFA;*/
2109 * There are strange code-generation bugs caused on sparc64 by gcc-2.95.2.
2110 * These need to be revisited when a newer toolchain becomes available.
2112 #if defined(__sparc64__) && defined(__GNUC__)
2113 # if __GNUC__ < 2 || (__GNUC__ == 2 && __GNUC_MINOR__ < 96)
2114 # undef SPARC64_GCC_WORKAROUND
2115 # define SPARC64_GCC_WORKAROUND 1
2119 #define DEBUG_PEEP(str,scan,depth) \
2120 DEBUG_OPTIMISE_r({ \
2121 SV * const mysv=sv_newmortal(); \
2122 regnode *Next = regnext(scan); \
2123 regprop(RExC_rx, mysv, scan); \
2124 PerlIO_printf(Perl_debug_log, "%*s" str ">%3d: %s (%d)\n", \
2125 (int)depth*2, "", REG_NODE_NUM(scan), SvPV_nolen_const(mysv),\
2126 Next ? (REG_NODE_NUM(Next)) : 0 ); \
2133 #define JOIN_EXACT(scan,min,flags) \
2134 if (PL_regkind[OP(scan)] == EXACT) \
2135 join_exact(pRExC_state,(scan),(min),(flags),NULL,depth+1)
2138 S_join_exact(pTHX_ RExC_state_t *pRExC_state, regnode *scan, I32 *min, U32 flags,regnode *val, U32 depth) {
2139 /* Merge several consecutive EXACTish nodes into one. */
2140 regnode *n = regnext(scan);
2142 regnode *next = scan + NODE_SZ_STR(scan);
2146 regnode *stop = scan;
2147 GET_RE_DEBUG_FLAGS_DECL;
2149 PERL_UNUSED_ARG(depth);
2151 #ifndef EXPERIMENTAL_INPLACESCAN
2152 PERL_UNUSED_ARG(flags);
2153 PERL_UNUSED_ARG(val);
2155 DEBUG_PEEP("join",scan,depth);
2157 /* Skip NOTHING, merge EXACT*. */
2159 ( PL_regkind[OP(n)] == NOTHING ||
2160 (stringok && (OP(n) == OP(scan))))
2162 && NEXT_OFF(scan) + NEXT_OFF(n) < I16_MAX) {
2164 if (OP(n) == TAIL || n > next)
2166 if (PL_regkind[OP(n)] == NOTHING) {
2167 DEBUG_PEEP("skip:",n,depth);
2168 NEXT_OFF(scan) += NEXT_OFF(n);
2169 next = n + NODE_STEP_REGNODE;
2176 else if (stringok) {
2177 const unsigned int oldl = STR_LEN(scan);
2178 regnode * const nnext = regnext(n);
2180 DEBUG_PEEP("merg",n,depth);
2183 if (oldl + STR_LEN(n) > U8_MAX)
2185 NEXT_OFF(scan) += NEXT_OFF(n);
2186 STR_LEN(scan) += STR_LEN(n);
2187 next = n + NODE_SZ_STR(n);
2188 /* Now we can overwrite *n : */
2189 Move(STRING(n), STRING(scan) + oldl, STR_LEN(n), char);
2197 #ifdef EXPERIMENTAL_INPLACESCAN
2198 if (flags && !NEXT_OFF(n)) {
2199 DEBUG_PEEP("atch", val, depth);
2200 if (reg_off_by_arg[OP(n)]) {
2201 ARG_SET(n, val - n);
2204 NEXT_OFF(n) = val - n;
2211 if (UTF && ( OP(scan) == EXACTF ) && ( STR_LEN(scan) >= 6 ) ) {
2213 Two problematic code points in Unicode casefolding of EXACT nodes:
2215 U+0390 - GREEK SMALL LETTER IOTA WITH DIALYTIKA AND TONOS
2216 U+03B0 - GREEK SMALL LETTER UPSILON WITH DIALYTIKA AND TONOS
2222 U+03B9 U+0308 U+0301 0xCE 0xB9 0xCC 0x88 0xCC 0x81
2223 U+03C5 U+0308 U+0301 0xCF 0x85 0xCC 0x88 0xCC 0x81
2225 This means that in case-insensitive matching (or "loose matching",
2226 as Unicode calls it), an EXACTF of length six (the UTF-8 encoded byte
2227 length of the above casefolded versions) can match a target string
2228 of length two (the byte length of UTF-8 encoded U+0390 or U+03B0).
2229 This would rather mess up the minimum length computation.
2231 What we'll do is to look for the tail four bytes, and then peek
2232 at the preceding two bytes to see whether we need to decrease
2233 the minimum length by four (six minus two).
2235 Thanks to the design of UTF-8, there cannot be false matches:
2236 A sequence of valid UTF-8 bytes cannot be a subsequence of
2237 another valid sequence of UTF-8 bytes.
2240 char * const s0 = STRING(scan), *s, *t;
2241 char * const s1 = s0 + STR_LEN(scan) - 1;
2242 char * const s2 = s1 - 4;
2243 #ifdef EBCDIC /* RD tunifold greek 0390 and 03B0 */
2244 const char t0[] = "\xaf\x49\xaf\x42";
2246 const char t0[] = "\xcc\x88\xcc\x81";
2248 const char * const t1 = t0 + 3;
2251 s < s2 && (t = ninstr(s, s1, t0, t1));
2254 if (((U8)t[-1] == 0x68 && (U8)t[-2] == 0xB4) ||
2255 ((U8)t[-1] == 0x46 && (U8)t[-2] == 0xB5))
2257 if (((U8)t[-1] == 0xB9 && (U8)t[-2] == 0xCE) ||
2258 ((U8)t[-1] == 0x85 && (U8)t[-2] == 0xCF))
2266 n = scan + NODE_SZ_STR(scan);
2268 if (PL_regkind[OP(n)] != NOTHING || OP(n) == NOTHING) {
2275 DEBUG_OPTIMISE_r(if (merged){DEBUG_PEEP("finl",scan,depth)});
2279 /* REx optimizer. Converts nodes into quickier variants "in place".
2280 Finds fixed substrings. */
2282 /* Stops at toplevel WHILEM as well as at "last". At end *scanp is set
2283 to the position after last scanned or to NULL. */
2285 #define INIT_AND_WITHP \
2286 assert(!and_withp); \
2287 Newx(and_withp,1,struct regnode_charclass_class); \
2288 SAVEFREEPV(and_withp)
2291 S_study_chunk(pTHX_ RExC_state_t *pRExC_state, regnode **scanp,
2292 I32 *minlenp, I32 *deltap,
2297 struct regnode_charclass_class *and_withp,
2298 U32 flags, U32 depth)
2299 /* scanp: Start here (read-write). */
2300 /* deltap: Write maxlen-minlen here. */
2301 /* last: Stop before this one. */
2302 /* data: string data about the pattern */
2303 /* stopparen: treat close N as END */
2304 /* recursed: which subroutines have we recursed into */
2305 /* and_withp: Valid if flags & SCF_DO_STCLASS_OR */
2308 I32 min = 0, pars = 0, code;
2309 regnode *scan = *scanp, *next;
2311 int is_inf = (flags & SCF_DO_SUBSTR) && (data->flags & SF_IS_INF);
2312 int is_inf_internal = 0; /* The studied chunk is infinite */
2313 I32 is_par = OP(scan) == OPEN ? ARG(scan) : 0;
2314 scan_data_t data_fake;
2315 SV *re_trie_maxbuff = NULL;
2316 regnode *first_non_open = scan;
2317 I32 stopmin = I32_MAX;
2318 GET_RE_DEBUG_FLAGS_DECL;
2320 StructCopy(&zero_scan_data, &data_fake, scan_data_t);
2324 while (first_non_open && OP(first_non_open) == OPEN)
2325 first_non_open=regnext(first_non_open);
2329 while (scan && OP(scan) != END && scan < last) {
2330 /* Peephole optimizer: */
2331 DEBUG_STUDYDATA(data,depth);
2332 DEBUG_PEEP("Peep",scan,depth);
2333 JOIN_EXACT(scan,&min,0);
2335 /* Follow the next-chain of the current node and optimize
2336 away all the NOTHINGs from it. */
2337 if (OP(scan) != CURLYX) {
2338 const int max = (reg_off_by_arg[OP(scan)]
2340 /* I32 may be smaller than U16 on CRAYs! */
2341 : (I32_MAX < U16_MAX ? I32_MAX : U16_MAX));
2342 int off = (reg_off_by_arg[OP(scan)] ? ARG(scan) : NEXT_OFF(scan));
2346 /* Skip NOTHING and LONGJMP. */
2347 while ((n = regnext(n))
2348 && ((PL_regkind[OP(n)] == NOTHING && (noff = NEXT_OFF(n)))
2349 || ((OP(n) == LONGJMP) && (noff = ARG(n))))
2350 && off + noff < max)
2352 if (reg_off_by_arg[OP(scan)])
2355 NEXT_OFF(scan) = off;
2360 /* The principal pseudo-switch. Cannot be a switch, since we
2361 look into several different things. */
2362 if (OP(scan) == BRANCH || OP(scan) == BRANCHJ
2363 || OP(scan) == IFTHEN || OP(scan) == SUSPEND) {
2364 next = regnext(scan);
2366 /* demq: the op(next)==code check is to see if we have "branch-branch" AFAICT */
2368 if (OP(next) == code || code == IFTHEN || code == SUSPEND) {
2369 /* NOTE - There is similar code to this block below for handling
2370 TRIE nodes on a re-study. If you change stuff here check there
2372 I32 max1 = 0, min1 = I32_MAX, num = 0;
2373 struct regnode_charclass_class accum;
2374 regnode * const startbranch=scan;
2376 if (flags & SCF_DO_SUBSTR) /* XXXX Add !SUSPEND? */
2377 scan_commit(pRExC_state, data, minlenp); /* Cannot merge strings after this. */
2378 if (flags & SCF_DO_STCLASS)
2379 cl_init_zero(pRExC_state, &accum);
2381 while (OP(scan) == code) {
2382 I32 deltanext, minnext, f = 0, fake;
2383 struct regnode_charclass_class this_class;
2386 data_fake.flags = 0;
2388 data_fake.whilem_c = data->whilem_c;
2389 data_fake.last_closep = data->last_closep;
2392 data_fake.last_closep = &fake;
2393 next = regnext(scan);
2394 scan = NEXTOPER(scan);
2396 scan = NEXTOPER(scan);
2397 if (flags & SCF_DO_STCLASS) {
2398 cl_init(pRExC_state, &this_class);
2399 data_fake.start_class = &this_class;
2400 f = SCF_DO_STCLASS_AND;
2402 if (flags & SCF_WHILEM_VISITED_POS)
2403 f |= SCF_WHILEM_VISITED_POS;
2405 /* we suppose the run is continuous, last=next...*/
2406 minnext = study_chunk(pRExC_state, &scan, minlenp, &deltanext,
2408 stopparen, recursed, NULL, f,depth+1);
2411 if (max1 < minnext + deltanext)
2412 max1 = minnext + deltanext;
2413 if (deltanext == I32_MAX)
2414 is_inf = is_inf_internal = 1;
2416 if (data_fake.flags & (SF_HAS_PAR|SF_IN_PAR))
2418 if (data_fake.flags & SCF_SEEN_ACCEPT) {
2419 if ( stopmin > minnext)
2420 stopmin = min + min1;
2421 flags &= ~SCF_DO_SUBSTR;
2423 data->flags |= SCF_SEEN_ACCEPT;
2426 if (data_fake.flags & SF_HAS_EVAL)
2427 data->flags |= SF_HAS_EVAL;
2428 data->whilem_c = data_fake.whilem_c;
2430 if (flags & SCF_DO_STCLASS)
2431 cl_or(pRExC_state, &accum, &this_class);
2432 if (code == SUSPEND)
2435 if (code == IFTHEN && num < 2) /* Empty ELSE branch */
2437 if (flags & SCF_DO_SUBSTR) {
2438 data->pos_min += min1;
2439 data->pos_delta += max1 - min1;
2440 if (max1 != min1 || is_inf)
2441 data->longest = &(data->longest_float);
2444 delta += max1 - min1;
2445 if (flags & SCF_DO_STCLASS_OR) {
2446 cl_or(pRExC_state, data->start_class, &accum);
2448 cl_and(data->start_class, and_withp);
2449 flags &= ~SCF_DO_STCLASS;
2452 else if (flags & SCF_DO_STCLASS_AND) {
2454 cl_and(data->start_class, &accum);
2455 flags &= ~SCF_DO_STCLASS;
2458 /* Switch to OR mode: cache the old value of
2459 * data->start_class */
2461 StructCopy(data->start_class, and_withp,
2462 struct regnode_charclass_class);
2463 flags &= ~SCF_DO_STCLASS_AND;
2464 StructCopy(&accum, data->start_class,
2465 struct regnode_charclass_class);
2466 flags |= SCF_DO_STCLASS_OR;
2467 data->start_class->flags |= ANYOF_EOS;
2471 if (PERL_ENABLE_TRIE_OPTIMISATION && OP( startbranch ) == BRANCH ) {
2474 Assuming this was/is a branch we are dealing with: 'scan' now
2475 points at the item that follows the branch sequence, whatever
2476 it is. We now start at the beginning of the sequence and look
2483 which would be constructed from a pattern like /A|LIST|OF|WORDS/
2485 If we can find such a subseqence we need to turn the first
2486 element into a trie and then add the subsequent branch exact
2487 strings to the trie.
2491 1. patterns where the whole set of branch can be converted.
2493 2. patterns where only a subset can be converted.
2495 In case 1 we can replace the whole set with a single regop
2496 for the trie. In case 2 we need to keep the start and end
2499 'BRANCH EXACT; BRANCH EXACT; BRANCH X'
2500 becomes BRANCH TRIE; BRANCH X;
2502 There is an additional case, that being where there is a
2503 common prefix, which gets split out into an EXACT like node
2504 preceding the TRIE node.
2506 If x(1..n)==tail then we can do a simple trie, if not we make
2507 a "jump" trie, such that when we match the appropriate word
2508 we "jump" to the appopriate tail node. Essentailly we turn
2509 a nested if into a case structure of sorts.
2514 if (!re_trie_maxbuff) {
2515 re_trie_maxbuff = get_sv(RE_TRIE_MAXBUF_NAME, 1);
2516 if (!SvIOK(re_trie_maxbuff))
2517 sv_setiv(re_trie_maxbuff, RE_TRIE_MAXBUF_INIT);
2519 if ( SvIV(re_trie_maxbuff)>=0 ) {
2521 regnode *first = (regnode *)NULL;
2522 regnode *last = (regnode *)NULL;
2523 regnode *tail = scan;
2528 SV * const mysv = sv_newmortal(); /* for dumping */
2530 /* var tail is used because there may be a TAIL
2531 regop in the way. Ie, the exacts will point to the
2532 thing following the TAIL, but the last branch will
2533 point at the TAIL. So we advance tail. If we
2534 have nested (?:) we may have to move through several
2538 while ( OP( tail ) == TAIL ) {
2539 /* this is the TAIL generated by (?:) */
2540 tail = regnext( tail );
2545 regprop(RExC_rx, mysv, tail );
2546 PerlIO_printf( Perl_debug_log, "%*s%s%s\n",
2547 (int)depth * 2 + 2, "",
2548 "Looking for TRIE'able sequences. Tail node is: ",
2549 SvPV_nolen_const( mysv )
2555 step through the branches, cur represents each
2556 branch, noper is the first thing to be matched
2557 as part of that branch and noper_next is the
2558 regnext() of that node. if noper is an EXACT
2559 and noper_next is the same as scan (our current
2560 position in the regex) then the EXACT branch is
2561 a possible optimization target. Once we have
2562 two or more consequetive such branches we can
2563 create a trie of the EXACT's contents and stich
2564 it in place. If the sequence represents all of
2565 the branches we eliminate the whole thing and
2566 replace it with a single TRIE. If it is a
2567 subsequence then we need to stitch it in. This
2568 means the first branch has to remain, and needs
2569 to be repointed at the item on the branch chain
2570 following the last branch optimized. This could
2571 be either a BRANCH, in which case the
2572 subsequence is internal, or it could be the
2573 item following the branch sequence in which
2574 case the subsequence is at the end.
2578 /* dont use tail as the end marker for this traverse */
2579 for ( cur = startbranch ; cur != scan ; cur = regnext( cur ) ) {
2580 regnode * const noper = NEXTOPER( cur );
2581 #if defined(DEBUGGING) || defined(NOJUMPTRIE)
2582 regnode * const noper_next = regnext( noper );
2586 regprop(RExC_rx, mysv, cur);
2587 PerlIO_printf( Perl_debug_log, "%*s- %s (%d)",
2588 (int)depth * 2 + 2,"", SvPV_nolen_const( mysv ), REG_NODE_NUM(cur) );
2590 regprop(RExC_rx, mysv, noper);
2591 PerlIO_printf( Perl_debug_log, " -> %s",
2592 SvPV_nolen_const(mysv));
2595 regprop(RExC_rx, mysv, noper_next );
2596 PerlIO_printf( Perl_debug_log,"\t=> %s\t",
2597 SvPV_nolen_const(mysv));
2599 PerlIO_printf( Perl_debug_log, "(First==%d,Last==%d,Cur==%d)\n",
2600 REG_NODE_NUM(first), REG_NODE_NUM(last), REG_NODE_NUM(cur) );
2602 if ( (((first && optype!=NOTHING) ? OP( noper ) == optype
2603 : PL_regkind[ OP( noper ) ] == EXACT )
2604 || OP(noper) == NOTHING )
2606 && noper_next == tail
2611 if ( !first || optype == NOTHING ) {
2612 if (!first) first = cur;
2613 optype = OP( noper );
2619 make_trie( pRExC_state,
2620 startbranch, first, cur, tail, count,
2623 if ( PL_regkind[ OP( noper ) ] == EXACT
2625 && noper_next == tail
2630 optype = OP( noper );
2640 regprop(RExC_rx, mysv, cur);
2641 PerlIO_printf( Perl_debug_log,
2642 "%*s- %s (%d) <SCAN FINISHED>\n", (int)depth * 2 + 2,
2643 "", SvPV_nolen_const( mysv ),REG_NODE_NUM(cur));
2647 made= make_trie( pRExC_state, startbranch, first, scan, tail, count, optype, depth+1 );
2648 #ifdef TRIE_STUDY_OPT
2649 if ( ((made == MADE_EXACT_TRIE &&
2650 startbranch == first)
2651 || ( first_non_open == first )) &&
2653 flags |= SCF_TRIE_RESTUDY;
2654 if ( startbranch == first
2657 RExC_seen &=~REG_TOP_LEVEL_BRANCHES;
2667 else if ( code == BRANCHJ ) { /* single branch is optimized. */
2668 scan = NEXTOPER(NEXTOPER(scan));
2669 } else /* single branch is optimized. */
2670 scan = NEXTOPER(scan);
2673 else if (OP(scan) == EXACT) {
2674 I32 l = STR_LEN(scan);
2677 const U8 * const s = (U8*)STRING(scan);
2678 l = utf8_length(s, s + l);
2679 uc = utf8_to_uvchr(s, NULL);
2681 uc = *((U8*)STRING(scan));
2684 if (flags & SCF_DO_SUBSTR) { /* Update longest substr. */
2685 /* The code below prefers earlier match for fixed
2686 offset, later match for variable offset. */
2687 if (data->last_end == -1) { /* Update the start info. */
2688 data->last_start_min = data->pos_min;
2689 data->last_start_max = is_inf
2690 ? I32_MAX : data->pos_min + data->pos_delta;
2692 sv_catpvn(data->last_found, STRING(scan), STR_LEN(scan));
2694 SvUTF8_on(data->last_found);
2696 SV * const sv = data->last_found;
2697 MAGIC * const mg = SvUTF8(sv) && SvMAGICAL(sv) ?
2698 mg_find(sv, PERL_MAGIC_utf8) : NULL;
2699 if (mg && mg->mg_len >= 0)
2700 mg->mg_len += utf8_length((U8*)STRING(scan),
2701 (U8*)STRING(scan)+STR_LEN(scan));
2703 data->last_end = data->pos_min + l;
2704 data->pos_min += l; /* As in the first entry. */
2705 data->flags &= ~SF_BEFORE_EOL;
2707 if (flags & SCF_DO_STCLASS_AND) {
2708 /* Check whether it is compatible with what we know already! */
2712 (!(data->start_class->flags & (ANYOF_CLASS | ANYOF_LOCALE))
2713 && !ANYOF_BITMAP_TEST(data->start_class, uc)
2714 && (!(data->start_class->flags & ANYOF_FOLD)
2715 || !ANYOF_BITMAP_TEST(data->start_class, PL_fold[uc])))
2718 ANYOF_CLASS_ZERO(data->start_class);
2719 ANYOF_BITMAP_ZERO(data->start_class);
2721 ANYOF_BITMAP_SET(data->start_class, uc);
2722 data->start_class->flags &= ~ANYOF_EOS;
2724 data->start_class->flags &= ~ANYOF_UNICODE_ALL;
2726 else if (flags & SCF_DO_STCLASS_OR) {
2727 /* false positive possible if the class is case-folded */
2729 ANYOF_BITMAP_SET(data->start_class, uc);
2731 data->start_class->flags |= ANYOF_UNICODE_ALL;
2732 data->start_class->flags &= ~ANYOF_EOS;
2733 cl_and(data->start_class, and_withp);
2735 flags &= ~SCF_DO_STCLASS;
2737 else if (PL_regkind[OP(scan)] == EXACT) { /* But OP != EXACT! */
2738 I32 l = STR_LEN(scan);
2739 UV uc = *((U8*)STRING(scan));
2741 /* Search for fixed substrings supports EXACT only. */
2742 if (flags & SCF_DO_SUBSTR) {
2744 scan_commit(pRExC_state, data, minlenp);
2747 const U8 * const s = (U8 *)STRING(scan);
2748 l = utf8_length(s, s + l);
2749 uc = utf8_to_uvchr(s, NULL);
2752 if (flags & SCF_DO_SUBSTR)
2754 if (flags & SCF_DO_STCLASS_AND) {
2755 /* Check whether it is compatible with what we know already! */
2759 (!(data->start_class->flags & (ANYOF_CLASS | ANYOF_LOCALE))
2760 && !ANYOF_BITMAP_TEST(data->start_class, uc)
2761 && !ANYOF_BITMAP_TEST(data->start_class, PL_fold[uc])))
2763 ANYOF_CLASS_ZERO(data->start_class);
2764 ANYOF_BITMAP_ZERO(data->start_class);
2766 ANYOF_BITMAP_SET(data->start_class, uc);
2767 data->start_class->flags &= ~ANYOF_EOS;
2768 data->start_class->flags |= ANYOF_FOLD;
2769 if (OP(scan) == EXACTFL)
2770 data->start_class->flags |= ANYOF_LOCALE;
2773 else if (flags & SCF_DO_STCLASS_OR) {
2774 if (data->start_class->flags & ANYOF_FOLD) {
2775 /* false positive possible if the class is case-folded.
2776 Assume that the locale settings are the same... */
2778 ANYOF_BITMAP_SET(data->start_class, uc);
2779 data->start_class->flags &= ~ANYOF_EOS;
2781 cl_and(data->start_class, and_withp);
2783 flags &= ~SCF_DO_STCLASS;
2785 else if (strchr((const char*)PL_varies,OP(scan))) {
2786 I32 mincount, maxcount, minnext, deltanext, fl = 0;
2787 I32 f = flags, pos_before = 0;
2788 regnode * const oscan = scan;
2789 struct regnode_charclass_class this_class;
2790 struct regnode_charclass_class *oclass = NULL;
2791 I32 next_is_eval = 0;
2793 switch (PL_regkind[OP(scan)]) {
2794 case WHILEM: /* End of (?:...)* . */
2795 scan = NEXTOPER(scan);
2798 if (flags & (SCF_DO_SUBSTR | SCF_DO_STCLASS)) {
2799 next = NEXTOPER(scan);
2800 if (OP(next) == EXACT || (flags & SCF_DO_STCLASS)) {
2802 maxcount = REG_INFTY;
2803 next = regnext(scan);
2804 scan = NEXTOPER(scan);
2808 if (flags & SCF_DO_SUBSTR)
2813 if (flags & SCF_DO_STCLASS) {
2815 maxcount = REG_INFTY;
2816 next = regnext(scan);
2817 scan = NEXTOPER(scan);
2820 is_inf = is_inf_internal = 1;
2821 scan = regnext(scan);
2822 if (flags & SCF_DO_SUBSTR) {
2823 scan_commit(pRExC_state, data, minlenp); /* Cannot extend fixed substrings */
2824 data->longest = &(data->longest_float);
2826 goto optimize_curly_tail;
2828 if (stopparen>0 && (OP(scan)==CURLYN || OP(scan)==CURLYM)
2829 && (scan->flags == stopparen))
2834 mincount = ARG1(scan);
2835 maxcount = ARG2(scan);
2837 next = regnext(scan);
2838 if (OP(scan) == CURLYX) {
2839 I32 lp = (data ? *(data->last_closep) : 0);
2840 scan->flags = ((lp <= (I32)U8_MAX) ? (U8)lp : U8_MAX);
2842 scan = NEXTOPER(scan) + EXTRA_STEP_2ARGS;
2843 next_is_eval = (OP(scan) == EVAL);
2845 if (flags & SCF_DO_SUBSTR) {
2846 if (mincount == 0) scan_commit(pRExC_state,data,minlenp); /* Cannot extend fixed substrings */
2847 pos_before = data->pos_min;
2851 data->flags &= ~(SF_HAS_PAR|SF_IN_PAR|SF_HAS_EVAL);
2853 data->flags |= SF_IS_INF;
2855 if (flags & SCF_DO_STCLASS) {
2856 cl_init(pRExC_state, &this_class);
2857 oclass = data->start_class;
2858 data->start_class = &this_class;
2859 f |= SCF_DO_STCLASS_AND;
2860 f &= ~SCF_DO_STCLASS_OR;
2862 /* These are the cases when once a subexpression
2863 fails at a particular position, it cannot succeed
2864 even after backtracking at the enclosing scope.
2866 XXXX what if minimal match and we are at the
2867 initial run of {n,m}? */
2868 if ((mincount != maxcount - 1) && (maxcount != REG_INFTY))
2869 f &= ~SCF_WHILEM_VISITED_POS;
2871 /* This will finish on WHILEM, setting scan, or on NULL: */
2872 minnext = study_chunk(pRExC_state, &scan, minlenp, &deltanext,
2873 last, data, stopparen, recursed, NULL,
2875 ? (f & ~SCF_DO_SUBSTR) : f),depth+1);
2877 if (flags & SCF_DO_STCLASS)
2878 data->start_class = oclass;
2879 if (mincount == 0 || minnext == 0) {
2880 if (flags & SCF_DO_STCLASS_OR) {
2881 cl_or(pRExC_state, data->start_class, &this_class);
2883 else if (flags & SCF_DO_STCLASS_AND) {
2884 /* Switch to OR mode: cache the old value of
2885 * data->start_class */
2887 StructCopy(data->start_class, and_withp,
2888 struct regnode_charclass_class);
2889 flags &= ~SCF_DO_STCLASS_AND;
2890 StructCopy(&this_class, data->start_class,
2891 struct regnode_charclass_class);
2892 flags |= SCF_DO_STCLASS_OR;
2893 data->start_class->flags |= ANYOF_EOS;
2895 } else { /* Non-zero len */
2896 if (flags & SCF_DO_STCLASS_OR) {
2897 cl_or(pRExC_state, data->start_class, &this_class);
2898 cl_and(data->start_class, and_withp);
2900 else if (flags & SCF_DO_STCLASS_AND)
2901 cl_and(data->start_class, &this_class);
2902 flags &= ~SCF_DO_STCLASS;
2904 if (!scan) /* It was not CURLYX, but CURLY. */
2906 if ( /* ? quantifier ok, except for (?{ ... }) */
2907 (next_is_eval || !(mincount == 0 && maxcount == 1))
2908 && (minnext == 0) && (deltanext == 0)
2909 && data && !(data->flags & (SF_HAS_PAR|SF_IN_PAR))
2910 && maxcount <= REG_INFTY/3 /* Complement check for big count */
2911 && ckWARN(WARN_REGEXP))
2914 "Quantifier unexpected on zero-length expression");
2917 min += minnext * mincount;
2918 is_inf_internal |= ((maxcount == REG_INFTY
2919 && (minnext + deltanext) > 0)
2920 || deltanext == I32_MAX);
2921 is_inf |= is_inf_internal;
2922 delta += (minnext + deltanext) * maxcount - minnext * mincount;
2924 /* Try powerful optimization CURLYX => CURLYN. */
2925 if ( OP(oscan) == CURLYX && data
2926 && data->flags & SF_IN_PAR
2927 && !(data->flags & SF_HAS_EVAL)
2928 && !deltanext && minnext == 1 ) {
2929 /* Try to optimize to CURLYN. */
2930 regnode *nxt = NEXTOPER(oscan) + EXTRA_STEP_2ARGS;
2931 regnode * const nxt1 = nxt;
2938 if (!strchr((const char*)PL_simple,OP(nxt))
2939 && !(PL_regkind[OP(nxt)] == EXACT
2940 && STR_LEN(nxt) == 1))
2946 if (OP(nxt) != CLOSE)
2948 if (RExC_open_parens) {
2949 RExC_open_parens[ARG(nxt1)-1]=oscan; /*open->CURLYM*/
2950 RExC_close_parens[ARG(nxt1)-1]=nxt+2; /*close->while*/
2952 /* Now we know that nxt2 is the only contents: */
2953 oscan->flags = (U8)ARG(nxt);
2955 OP(nxt1) = NOTHING; /* was OPEN. */
2958 OP(nxt1 + 1) = OPTIMIZED; /* was count. */
2959 NEXT_OFF(nxt1+ 1) = 0; /* just for consistancy. */
2960 NEXT_OFF(nxt2) = 0; /* just for consistancy with CURLY. */
2961 OP(nxt) = OPTIMIZED; /* was CLOSE. */
2962 OP(nxt + 1) = OPTIMIZED; /* was count. */
2963 NEXT_OFF(nxt+ 1) = 0; /* just for consistancy. */
2968 /* Try optimization CURLYX => CURLYM. */
2969 if ( OP(oscan) == CURLYX && data
2970 && !(data->flags & SF_HAS_PAR)
2971 && !(data->flags & SF_HAS_EVAL)
2972 && !deltanext /* atom is fixed width */
2973 && minnext != 0 /* CURLYM can't handle zero width */
2975 /* XXXX How to optimize if data == 0? */
2976 /* Optimize to a simpler form. */
2977 regnode *nxt = NEXTOPER(oscan) + EXTRA_STEP_2ARGS; /* OPEN */
2981 while ( (nxt2 = regnext(nxt)) /* skip over embedded stuff*/
2982 && (OP(nxt2) != WHILEM))
2984 OP(nxt2) = SUCCEED; /* Whas WHILEM */
2985 /* Need to optimize away parenths. */
2986 if (data->flags & SF_IN_PAR) {
2987 /* Set the parenth number. */
2988 regnode *nxt1 = NEXTOPER(oscan) + EXTRA_STEP_2ARGS; /* OPEN*/
2990 if (OP(nxt) != CLOSE)
2991 FAIL("Panic opt close");
2992 oscan->flags = (U8)ARG(nxt);
2993 if (RExC_open_parens) {
2994 RExC_open_parens[ARG(nxt1)-1]=oscan; /*open->CURLYM*/
2995 RExC_close_parens[ARG(nxt1)-1]=nxt2+1; /*close->NOTHING*/
2997 OP(nxt1) = OPTIMIZED; /* was OPEN. */
2998 OP(nxt) = OPTIMIZED; /* was CLOSE. */
3001 OP(nxt1 + 1) = OPTIMIZED; /* was count. */
3002 OP(nxt + 1) = OPTIMIZED; /* was count. */
3003 NEXT_OFF(nxt1 + 1) = 0; /* just for consistancy. */
3004 NEXT_OFF(nxt + 1) = 0; /* just for consistancy. */
3007 while ( nxt1 && (OP(nxt1) != WHILEM)) {
3008 regnode *nnxt = regnext(nxt1);
3011 if (reg_off_by_arg[OP(nxt1)])
3012 ARG_SET(nxt1, nxt2 - nxt1);
3013 else if (nxt2 - nxt1 < U16_MAX)
3014 NEXT_OFF(nxt1) = nxt2 - nxt1;
3016 OP(nxt) = NOTHING; /* Cannot beautify */
3021 /* Optimize again: */
3022 study_chunk(pRExC_state, &nxt1, minlenp, &deltanext, nxt,
3023 NULL, stopparen, recursed, NULL, 0,depth+1);
3028 else if ((OP(oscan) == CURLYX)
3029 && (flags & SCF_WHILEM_VISITED_POS)
3030 /* See the comment on a similar expression above.
3031 However, this time it not a subexpression
3032 we care about, but the expression itself. */
3033 && (maxcount == REG_INFTY)
3034 && data && ++data->whilem_c < 16) {
3035 /* This stays as CURLYX, we can put the count/of pair. */
3036 /* Find WHILEM (as in regexec.c) */
3037 regnode *nxt = oscan + NEXT_OFF(oscan);
3039 if (OP(PREVOPER(nxt)) == NOTHING) /* LONGJMP */
3041 PREVOPER(nxt)->flags = (U8)(data->whilem_c
3042 | (RExC_whilem_seen << 4)); /* On WHILEM */
3044 if (data && fl & (SF_HAS_PAR|SF_IN_PAR))
3046 if (flags & SCF_DO_SUBSTR) {
3047 SV *last_str = NULL;
3048 int counted = mincount != 0;
3050 if (data->last_end > 0 && mincount != 0) { /* Ends with a string. */
3051 #if defined(SPARC64_GCC_WORKAROUND)
3054 const char *s = NULL;
3057 if (pos_before >= data->last_start_min)
3060 b = data->last_start_min;
3063 s = SvPV_const(data->last_found, l);
3064 old = b - data->last_start_min;
3067 I32 b = pos_before >= data->last_start_min
3068 ? pos_before : data->last_start_min;
3070 const char * const s = SvPV_const(data->last_found, l);
3071 I32 old = b - data->last_start_min;
3075 old = utf8_hop((U8*)s, old) - (U8*)s;
3078 /* Get the added string: */
3079 last_str = newSVpvn(s + old, l);
3081 SvUTF8_on(last_str);
3082 if (deltanext == 0 && pos_before == b) {
3083 /* What was added is a constant string */
3085 SvGROW(last_str, (mincount * l) + 1);
3086 repeatcpy(SvPVX(last_str) + l,
3087 SvPVX_const(last_str), l, mincount - 1);
3088 SvCUR_set(last_str, SvCUR(last_str) * mincount);
3089 /* Add additional parts. */
3090 SvCUR_set(data->last_found,
3091 SvCUR(data->last_found) - l);
3092 sv_catsv(data->last_found, last_str);
3094 SV * sv = data->last_found;
3096 SvUTF8(sv) && SvMAGICAL(sv) ?
3097 mg_find(sv, PERL_MAGIC_utf8) : NULL;
3098 if (mg && mg->mg_len >= 0)
3099 mg->mg_len += CHR_SVLEN(last_str);
3101 data->last_end += l * (mincount - 1);
3104 /* start offset must point into the last copy */
3105 data->last_start_min += minnext * (mincount - 1);
3106 data->last_start_max += is_inf ? I32_MAX
3107 : (maxcount - 1) * (minnext + data->pos_delta);
3110 /* It is counted once already... */
3111 data->pos_min += minnext * (mincount - counted);
3112 data->pos_delta += - counted * deltanext +
3113 (minnext + deltanext) * maxcount - minnext * mincount;
3114 if (mincount != maxcount) {
3115 /* Cannot extend fixed substrings found inside
3117 scan_commit(pRExC_state,data,minlenp);
3118 if (mincount && last_str) {
3119 SV * const sv = data->last_found;
3120 MAGIC * const mg = SvUTF8(sv) && SvMAGICAL(sv) ?
3121 mg_find(sv, PERL_MAGIC_utf8) : NULL;
3125 sv_setsv(sv, last_str);
3126 data->last_end = data->pos_min;
3127 data->last_start_min =
3128 data->pos_min - CHR_SVLEN(last_str);
3129 data->last_start_max = is_inf
3131 : data->pos_min + data->pos_delta
3132 - CHR_SVLEN(last_str);
3134 data->longest = &(data->longest_float);
3136 SvREFCNT_dec(last_str);
3138 if (data && (fl & SF_HAS_EVAL))
3139 data->flags |= SF_HAS_EVAL;
3140 optimize_curly_tail:
3141 if (OP(oscan) != CURLYX) {
3142 while (PL_regkind[OP(next = regnext(oscan))] == NOTHING
3144 NEXT_OFF(oscan) += NEXT_OFF(next);
3147 default: /* REF and CLUMP only? */
3148 if (flags & SCF_DO_SUBSTR) {
3149 scan_commit(pRExC_state,data,minlenp); /* Cannot expect anything... */
3150 data->longest = &(data->longest_float);
3152 is_inf = is_inf_internal = 1;
3153 if (flags & SCF_DO_STCLASS_OR)
3154 cl_anything(pRExC_state, data->start_class);
3155 flags &= ~SCF_DO_STCLASS;
3159 else if (strchr((const char*)PL_simple,OP(scan))) {
3162 if (flags & SCF_DO_SUBSTR) {
3163 scan_commit(pRExC_state,data,minlenp);
3167 if (flags & SCF_DO_STCLASS) {
3168 data->start_class->flags &= ~ANYOF_EOS; /* No match on empty */
3170 /* Some of the logic below assumes that switching
3171 locale on will only add false positives. */
3172 switch (PL_regkind[OP(scan)]) {
3176 /* Perl_croak(aTHX_ "panic: unexpected simple REx opcode %d", OP(scan)); */
3177 if (flags & SCF_DO_STCLASS_OR) /* Allow everything */
3178 cl_anything(pRExC_state, data->start_class);
3181 if (OP(scan) == SANY)
3183 if (flags & SCF_DO_STCLASS_OR) { /* Everything but \n */
3184 value = (ANYOF_BITMAP_TEST(data->start_class,'\n')
3185 || (data->start_class->flags & ANYOF_CLASS));
3186 cl_anything(pRExC_state, data->start_class);
3188 if (flags & SCF_DO_STCLASS_AND || !value)
3189 ANYOF_BITMAP_CLEAR(data->start_class,'\n');
3192 if (flags & SCF_DO_STCLASS_AND)
3193 cl_and(data->start_class,
3194 (struct regnode_charclass_class*)scan);
3196 cl_or(pRExC_state, data->start_class,
3197 (struct regnode_charclass_class*)scan);
3200 if (flags & SCF_DO_STCLASS_AND) {
3201 if (!(data->start_class->flags & ANYOF_LOCALE)) {
3202 ANYOF_CLASS_CLEAR(data->start_class,ANYOF_NALNUM);
3203 for (value = 0; value < 256; value++)
3204 if (!isALNUM(value))
3205 ANYOF_BITMAP_CLEAR(data->start_class, value);
3209 if (data->start_class->flags & ANYOF_LOCALE)
3210 ANYOF_CLASS_SET(data->start_class,ANYOF_ALNUM);
3212 for (value = 0; value < 256; value++)
3214 ANYOF_BITMAP_SET(data->start_class, value);
3219 if (flags & SCF_DO_STCLASS_AND) {
3220 if (data->start_class->flags & ANYOF_LOCALE)
3221 ANYOF_CLASS_CLEAR(data->start_class,ANYOF_NALNUM);
3224 ANYOF_CLASS_SET(data->start_class,ANYOF_ALNUM);
3225 data->start_class->flags |= ANYOF_LOCALE;
3229 if (flags & SCF_DO_STCLASS_AND) {
3230 if (!(data->start_class->flags & ANYOF_LOCALE)) {
3231 ANYOF_CLASS_CLEAR(data->start_class,ANYOF_ALNUM);
3232 for (value = 0; value < 256; value++)
3234 ANYOF_BITMAP_CLEAR(data->start_class, value);
3238 if (data->start_class->flags & ANYOF_LOCALE)
3239 ANYOF_CLASS_SET(data->start_class,ANYOF_NALNUM);
3241 for (value = 0; value < 256; value++)
3242 if (!isALNUM(value))
3243 ANYOF_BITMAP_SET(data->start_class, value);
3248 if (flags & SCF_DO_STCLASS_AND) {
3249 if (data->start_class->flags & ANYOF_LOCALE)
3250 ANYOF_CLASS_CLEAR(data->start_class,ANYOF_ALNUM);
3253 data->start_class->flags |= ANYOF_LOCALE;
3254 ANYOF_CLASS_SET(data->start_class,ANYOF_NALNUM);
3258 if (flags & SCF_DO_STCLASS_AND) {
3259 if (!(data->start_class->flags & ANYOF_LOCALE)) {
3260 ANYOF_CLASS_CLEAR(data->start_class,ANYOF_NSPACE);
3261 for (value = 0; value < 256; value++)
3262 if (!isSPACE(value))
3263 ANYOF_BITMAP_CLEAR(data->start_class, value);
3267 if (data->start_class->flags & ANYOF_LOCALE)
3268 ANYOF_CLASS_SET(data->start_class,ANYOF_SPACE);
3270 for (value = 0; value < 256; value++)
3272 ANYOF_BITMAP_SET(data->start_class, value);
3277 if (flags & SCF_DO_STCLASS_AND) {
3278 if (data->start_class->flags & ANYOF_LOCALE)
3279 ANYOF_CLASS_CLEAR(data->start_class,ANYOF_NSPACE);
3282 data->start_class->flags |= ANYOF_LOCALE;
3283 ANYOF_CLASS_SET(data->start_class,ANYOF_SPACE);
3287 if (flags & SCF_DO_STCLASS_AND) {
3288 if (!(data->start_class->flags & ANYOF_LOCALE)) {
3289 ANYOF_CLASS_CLEAR(data->start_class,ANYOF_SPACE);
3290 for (value = 0; value < 256; value++)
3292 ANYOF_BITMAP_CLEAR(data->start_class, value);
3296 if (data->start_class->flags & ANYOF_LOCALE)
3297 ANYOF_CLASS_SET(data->start_class,ANYOF_NSPACE);
3299 for (value = 0; value < 256; value++)
3300 if (!isSPACE(value))
3301 ANYOF_BITMAP_SET(data->start_class, value);
3306 if (flags & SCF_DO_STCLASS_AND) {
3307 if (data->start_class->flags & ANYOF_LOCALE) {
3308 ANYOF_CLASS_CLEAR(data->start_class,ANYOF_SPACE);
3309 for (value = 0; value < 256; value++)
3310 if (!isSPACE(value))
3311 ANYOF_BITMAP_CLEAR(data->start_class, value);
3315 data->start_class->flags |= ANYOF_LOCALE;
3316 ANYOF_CLASS_SET(data->start_class,ANYOF_NSPACE);
3320 if (flags & SCF_DO_STCLASS_AND) {
3321 ANYOF_CLASS_CLEAR(data->start_class,ANYOF_NDIGIT);
3322 for (value = 0; value < 256; value++)
3323 if (!isDIGIT(value))
3324 ANYOF_BITMAP_CLEAR(data->start_class, value);
3327 if (data->start_class->flags & ANYOF_LOCALE)
3328 ANYOF_CLASS_SET(data->start_class,ANYOF_DIGIT);
3330 for (value = 0; value < 256; value++)
3332 ANYOF_BITMAP_SET(data->start_class, value);
3337 if (flags & SCF_DO_STCLASS_AND) {
3338 ANYOF_CLASS_CLEAR(data->start_class,ANYOF_DIGIT);
3339 for (value = 0; value < 256; value++)
3341 ANYOF_BITMAP_CLEAR(data->start_class, value);
3344 if (data->start_class->flags & ANYOF_LOCALE)
3345 ANYOF_CLASS_SET(data->start_class,ANYOF_NDIGIT);
3347 for (value = 0; value < 256; value++)
3348 if (!isDIGIT(value))
3349 ANYOF_BITMAP_SET(data->start_class, value);
3354 if (flags & SCF_DO_STCLASS_OR)
3355 cl_and(data->start_class, and_withp);
3356 flags &= ~SCF_DO_STCLASS;
3359 else if (PL_regkind[OP(scan)] == EOL && flags & SCF_DO_SUBSTR) {
3360 data->flags |= (OP(scan) == MEOL
3364 else if ( PL_regkind[OP(scan)] == BRANCHJ
3365 /* Lookbehind, or need to calculate parens/evals/stclass: */
3366 && (scan->flags || data || (flags & SCF_DO_STCLASS))
3367 && (OP(scan) == IFMATCH || OP(scan) == UNLESSM)) {
3368 if ( !PERL_ENABLE_POSITIVE_ASSERTION_STUDY
3369 || OP(scan) == UNLESSM )
3371 /* Negative Lookahead/lookbehind
3372 In this case we can't do fixed string optimisation.
3375 I32 deltanext, minnext, fake = 0;
3377 struct regnode_charclass_class intrnl;
3380 data_fake.flags = 0;
3382 data_fake.whilem_c = data->whilem_c;
3383 data_fake.last_closep = data->last_closep;
3386 data_fake.last_closep = &fake;
3387 if ( flags & SCF_DO_STCLASS && !scan->flags
3388 && OP(scan) == IFMATCH ) { /* Lookahead */
3389 cl_init(pRExC_state, &intrnl);
3390 data_fake.start_class = &intrnl;
3391 f |= SCF_DO_STCLASS_AND;
3393 if (flags & SCF_WHILEM_VISITED_POS)
3394 f |= SCF_WHILEM_VISITED_POS;
3395 next = regnext(scan);
3396 nscan = NEXTOPER(NEXTOPER(scan));
3397 minnext = study_chunk(pRExC_state, &nscan, minlenp, &deltanext,
3398 last, &data_fake, stopparen, recursed, NULL, f, depth+1);
3401 vFAIL("Variable length lookbehind not implemented");
3403 else if (minnext > (I32)U8_MAX) {
3404 vFAIL2("Lookbehind longer than %"UVuf" not implemented", (UV)U8_MAX);
3406 scan->flags = (U8)minnext;
3409 if (data_fake.flags & (SF_HAS_PAR|SF_IN_PAR))
3411 if (data_fake.flags & SF_HAS_EVAL)
3412 data->flags |= SF_HAS_EVAL;
3413 data->whilem_c = data_fake.whilem_c;
3415 if (f & SCF_DO_STCLASS_AND) {
3416 const int was = (data->start_class->flags & ANYOF_EOS);
3418 cl_and(data->start_class, &intrnl);
3420 data->start_class->flags |= ANYOF_EOS;
3423 #if PERL_ENABLE_POSITIVE_ASSERTION_STUDY
3425 /* Positive Lookahead/lookbehind
3426 In this case we can do fixed string optimisation,
3427 but we must be careful about it. Note in the case of
3428 lookbehind the positions will be offset by the minimum
3429 length of the pattern, something we won't know about
3430 until after the recurse.
3432 I32 deltanext, fake = 0;
3434 struct regnode_charclass_class intrnl;
3436 /* We use SAVEFREEPV so that when the full compile
3437 is finished perl will clean up the allocated
3438 minlens when its all done. This was we don't
3439 have to worry about freeing them when we know
3440 they wont be used, which would be a pain.
3443 Newx( minnextp, 1, I32 );
3444 SAVEFREEPV(minnextp);
3447 StructCopy(data, &data_fake, scan_data_t);
3448 if ((flags & SCF_DO_SUBSTR) && data->last_found) {
3451 scan_commit(pRExC_state, &data_fake,minlenp);
3452 data_fake.last_found=newSVsv(data->last_found);
3456 data_fake.last_closep = &fake;
3457 data_fake.flags = 0;
3459 data_fake.flags |= SF_IS_INF;
3460 if ( flags & SCF_DO_STCLASS && !scan->flags
3461 && OP(scan) == IFMATCH ) { /* Lookahead */
3462 cl_init(pRExC_state, &intrnl);
3463 data_fake.start_class = &intrnl;
3464 f |= SCF_DO_STCLASS_AND;
3466 if (flags & SCF_WHILEM_VISITED_POS)
3467 f |= SCF_WHILEM_VISITED_POS;
3468 next = regnext(scan);
3469 nscan = NEXTOPER(NEXTOPER(scan));
3471 *minnextp = study_chunk(pRExC_state, &nscan, minnextp, &deltanext,
3472 last, &data_fake, stopparen, recursed, NULL, f,depth+1);
3475 vFAIL("Variable length lookbehind not implemented");
3477 else if (*minnextp > (I32)U8_MAX) {
3478 vFAIL2("Lookbehind longer than %"UVuf" not implemented", (UV)U8_MAX);
3480 scan->flags = (U8)*minnextp;
3485 if (f & SCF_DO_STCLASS_AND) {
3486 const int was = (data->start_class->flags & ANYOF_EOS);
3488 cl_and(data->start_class, &intrnl);
3490 data->start_class->flags |= ANYOF_EOS;
3493 if (data_fake.flags & (SF_HAS_PAR|SF_IN_PAR))
3495 if (data_fake.flags & SF_HAS_EVAL)
3496 data->flags |= SF_HAS_EVAL;
3497 data->whilem_c = data_fake.whilem_c;
3498 if ((flags & SCF_DO_SUBSTR) && data_fake.last_found) {
3499 if (RExC_rx->minlen<*minnextp)
3500 RExC_rx->minlen=*minnextp;
3501 scan_commit(pRExC_state, &data_fake, minnextp);
3502 SvREFCNT_dec(data_fake.last_found);
3504 if ( data_fake.minlen_fixed != minlenp )
3506 data->offset_fixed= data_fake.offset_fixed;
3507 data->minlen_fixed= data_fake.minlen_fixed;
3508 data->lookbehind_fixed+= scan->flags;
3510 if ( data_fake.minlen_float != minlenp )
3512 data->minlen_float= data_fake.minlen_float;
3513 data->offset_float_min=data_fake.offset_float_min;
3514 data->offset_float_max=data_fake.offset_float_max;
3515 data->lookbehind_float+= scan->flags;
3524 else if (OP(scan) == OPEN) {
3525 if (stopparen != (I32)ARG(scan))
3528 else if (OP(scan) == CLOSE) {
3529 if (stopparen == (I32)ARG(scan)) {
3532 if ((I32)ARG(scan) == is_par) {
3533 next = regnext(scan);
3535 if ( next && (OP(next) != WHILEM) && next < last)
3536 is_par = 0; /* Disable optimization */
3539 *(data->last_closep) = ARG(scan);
3541 else if (OP(scan) == GOSUB || OP(scan) == GOSTART) {
3542 /* set the pointer */
3546 if (OP(scan) == GOSUB) {
3548 RExC_recurse[ARG2L(scan)] = scan;
3549 start = RExC_open_parens[paren-1];
3550 end = RExC_close_parens[paren-1];
3553 start = RExC_rx->program + 1;
3559 Newxz(recursed, (((RExC_npar)>>3) +1), U8);
3560 SAVEFREEPV(recursed);
3562 if (!PAREN_TEST(recursed,paren+1)) {
3564 PAREN_SET(recursed,paren+1);
3566 DEBUG_PEEP("goto",start,depth);
3579 if (deltanext == I32_MAX) {
3580 is_inf = is_inf_internal = 1;
3583 DEBUG_PEEP("rtrn",end,depth);
3584 PAREN_UNSET(recursed,paren+1);
3586 if (flags & SCF_DO_SUBSTR) {
3587 scan_commit(pRExC_state,data,minlenp);
3588 data->longest = &(data->longest_float);
3590 is_inf = is_inf_internal = 1;
3591 if (flags & SCF_DO_STCLASS_OR) /* Allow everything */
3592 cl_anything(pRExC_state, data->start_class);
3593 flags &= ~SCF_DO_STCLASS;
3596 else if (OP(scan) == EVAL) {
3598 data->flags |= SF_HAS_EVAL;
3600 else if ( PL_regkind[OP(scan)] == ENDLIKE ) {
3601 if (flags & SCF_DO_SUBSTR) {
3602 scan_commit(pRExC_state,data,minlenp);
3603 flags &= ~SCF_DO_SUBSTR;
3605 if (data && OP(scan)==ACCEPT) {
3606 data->flags |= SCF_SEEN_ACCEPT;
3611 else if (OP(scan) == LOGICAL && scan->flags == 2) /* Embedded follows */
3613 if (flags & SCF_DO_SUBSTR) {
3614 scan_commit(pRExC_state,data,minlenp);
3615 data->longest = &(data->longest_float);
3617 is_inf = is_inf_internal = 1;
3618 if (flags & SCF_DO_STCLASS_OR) /* Allow everything */
3619 cl_anything(pRExC_state, data->start_class);
3620 flags &= ~SCF_DO_STCLASS;
3622 #ifdef TRIE_STUDY_OPT
3623 #ifdef FULL_TRIE_STUDY
3624 else if (PL_regkind[OP(scan)] == TRIE) {
3625 /* NOTE - There is similar code to this block above for handling
3626 BRANCH nodes on the initial study. If you change stuff here
3628 regnode *trie_node= scan;
3629 regnode *tail= regnext(scan);
3630 reg_trie_data *trie = (reg_trie_data*)RExC_rx->data->data[ ARG(scan) ];
3631 I32 max1 = 0, min1 = I32_MAX;
3632 struct regnode_charclass_class accum;
3634 if (flags & SCF_DO_SUBSTR) /* XXXX Add !SUSPEND? */
3635 scan_commit(pRExC_state, data,minlenp); /* Cannot merge strings after this. */
3636 if (flags & SCF_DO_STCLASS)
3637 cl_init_zero(pRExC_state, &accum);
3643 const regnode *nextbranch= NULL;
3646 for ( word=1 ; word <= trie->wordcount ; word++)
3648 I32 deltanext=0, minnext=0, f = 0, fake;
3649 struct regnode_charclass_class this_class;
3651 data_fake.flags = 0;
3653 data_fake.whilem_c = data->whilem_c;
3654 data_fake.last_closep = data->last_closep;
3657 data_fake.last_closep = &fake;
3659 if (flags & SCF_DO_STCLASS) {
3660 cl_init(pRExC_state, &this_class);
3661 data_fake.start_class = &this_class;
3662 f = SCF_DO_STCLASS_AND;
3664 if (flags & SCF_WHILEM_VISITED_POS)
3665 f |= SCF_WHILEM_VISITED_POS;
3667 if (trie->jump[word]) {
3669 nextbranch = trie_node + trie->jump[0];
3670 scan= trie_node + trie->jump[word];
3671 /* We go from the jump point to the branch that follows
3672 it. Note this means we need the vestigal unused branches
3673 even though they arent otherwise used.
3675 minnext = study_chunk(pRExC_state, &scan, minlenp,
3676 &deltanext, (regnode *)nextbranch, &data_fake,
3677 stopparen, recursed, NULL, f,depth+1);
3679 if (nextbranch && PL_regkind[OP(nextbranch)]==BRANCH)
3680 nextbranch= regnext((regnode*)nextbranch);
3682 if (min1 > (I32)(minnext + trie->minlen))
3683 min1 = minnext + trie->minlen;
3684 if (max1 < (I32)(minnext + deltanext + trie->maxlen))
3685 max1 = minnext + deltanext + trie->maxlen;
3686 if (deltanext == I32_MAX)
3687 is_inf = is_inf_internal = 1;
3689 if (data_fake.flags & (SF_HAS_PAR|SF_IN_PAR))
3691 if (data_fake.flags & SCF_SEEN_ACCEPT) {
3692 if ( stopmin > min + min1)
3693 stopmin = min + min1;
3694 flags &= ~SCF_DO_SUBSTR;
3696 data->flags |= SCF_SEEN_ACCEPT;
3699 if (data_fake.flags & SF_HAS_EVAL)
3700 data->flags |= SF_HAS_EVAL;
3701 data->whilem_c = data_fake.whilem_c;
3703 if (flags & SCF_DO_STCLASS)
3704 cl_or(pRExC_state, &accum, &this_class);
3707 if (flags & SCF_DO_SUBSTR) {
3708 data->pos_min += min1;
3709 data->pos_delta += max1 - min1;
3710 if (max1 != min1 || is_inf)
3711 data->longest = &(data->longest_float);
3714 delta += max1 - min1;
3715 if (flags & SCF_DO_STCLASS_OR) {
3716 cl_or(pRExC_state, data->start_class, &accum);
3718 cl_and(data->start_class, and_withp);
3719 flags &= ~SCF_DO_STCLASS;
3722 else if (flags & SCF_DO_STCLASS_AND) {
3724 cl_and(data->start_class, &accum);
3725 flags &= ~SCF_DO_STCLASS;
3728 /* Switch to OR mode: cache the old value of
3729 * data->start_class */
3731 StructCopy(data->start_class, and_withp,
3732 struct regnode_charclass_class);
3733 flags &= ~SCF_DO_STCLASS_AND;
3734 StructCopy(&accum, data->start_class,
3735 struct regnode_charclass_class);
3736 flags |= SCF_DO_STCLASS_OR;
3737 data->start_class->flags |= ANYOF_EOS;
3744 else if (PL_regkind[OP(scan)] == TRIE) {
3745 reg_trie_data *trie = (reg_trie_data*)RExC_rx->data->data[ ARG(scan) ];
3748 min += trie->minlen;
3749 delta += (trie->maxlen - trie->minlen);
3750 flags &= ~SCF_DO_STCLASS; /* xxx */
3751 if (flags & SCF_DO_SUBSTR) {
3752 scan_commit(pRExC_state,data,minlenp); /* Cannot expect anything... */
3753 data->pos_min += trie->minlen;
3754 data->pos_delta += (trie->maxlen - trie->minlen);
3755 if (trie->maxlen != trie->minlen)
3756 data->longest = &(data->longest_float);
3758 if (trie->jump) /* no more substrings -- for now /grr*/
3759 flags &= ~SCF_DO_SUBSTR;
3761 #endif /* old or new */
3762 #endif /* TRIE_STUDY_OPT */
3763 /* Else: zero-length, ignore. */
3764 scan = regnext(scan);
3769 *deltap = is_inf_internal ? I32_MAX : delta;
3770 if (flags & SCF_DO_SUBSTR && is_inf)
3771 data->pos_delta = I32_MAX - data->pos_min;
3772 if (is_par > (I32)U8_MAX)
3774 if (is_par && pars==1 && data) {
3775 data->flags |= SF_IN_PAR;
3776 data->flags &= ~SF_HAS_PAR;
3778 else if (pars && data) {
3779 data->flags |= SF_HAS_PAR;
3780 data->flags &= ~SF_IN_PAR;
3782 if (flags & SCF_DO_STCLASS_OR)
3783 cl_and(data->start_class, and_withp);
3784 if (flags & SCF_TRIE_RESTUDY)
3785 data->flags |= SCF_TRIE_RESTUDY;
3787 DEBUG_STUDYDATA(data,depth);
3789 return min < stopmin ? min : stopmin;
3793 S_add_data(RExC_state_t *pRExC_state, I32 n, const char *s)
3795 if (RExC_rx->data) {
3796 const U32 count = RExC_rx->data->count;
3797 Renewc(RExC_rx->data,
3798 sizeof(*RExC_rx->data) + sizeof(void*) * (count + n - 1),
3799 char, struct reg_data);
3800 Renew(RExC_rx->data->what, count + n, U8);
3801 RExC_rx->data->count += n;
3804 Newxc(RExC_rx->data, sizeof(*RExC_rx->data) + sizeof(void*) * (n - 1),
3805 char, struct reg_data);
3806 Newx(RExC_rx->data->what, n, U8);
3807 RExC_rx->data->count = n;
3809 Copy(s, RExC_rx->data->what + RExC_rx->data->count - n, n, U8);
3810 return RExC_rx->data->count - n;
3813 #ifndef PERL_IN_XSUB_RE
3815 Perl_reginitcolors(pTHX)
3818 const char * const s = PerlEnv_getenv("PERL_RE_COLORS");
3820 char *t = savepv(s);
3824 t = strchr(t, '\t');
3830 PL_colors[i] = t = (char *)"";
3835 PL_colors[i++] = (char *)"";
3842 #ifdef TRIE_STUDY_OPT
3843 #define CHECK_RESTUDY_GOTO \
3845 (data.flags & SCF_TRIE_RESTUDY) \
3849 #define CHECK_RESTUDY_GOTO
3853 - pregcomp - compile a regular expression into internal code
3855 * We can't allocate space until we know how big the compiled form will be,
3856 * but we can't compile it (and thus know how big it is) until we've got a
3857 * place to put the code. So we cheat: we compile it twice, once with code
3858 * generation turned off and size counting turned on, and once "for real".
3859 * This also means that we don't allocate space until we are sure that the
3860 * thing really will compile successfully, and we never have to move the
3861 * code and thus invalidate pointers into it. (Note that it has to be in
3862 * one piece because free() must be able to free it all.) [NB: not true in perl]
3864 * Beware that the optimization-preparation code in here knows about some
3865 * of the structure of the compiled regexp. [I'll say.]
3870 #ifndef PERL_IN_XSUB_RE
3871 #define RE_ENGINE_PTR &PL_core_reg_engine
3873 extern const struct regexp_engine my_reg_engine;
3874 #define RE_ENGINE_PTR &my_reg_engine
3876 /* these make a few things look better, to avoid indentation */
3877 #define BEGIN_BLOCK {
3881 Perl_pregcomp(pTHX_ char *exp, char *xend, PMOP *pm)
3884 GET_RE_DEBUG_FLAGS_DECL;
3885 DEBUG_r(if (!PL_colorset) reginitcolors());
3886 #ifndef PERL_IN_XSUB_RE
3888 /* Dispatch a request to compile a regexp to correct
3890 HV * const table = GvHV(PL_hintgv);
3892 SV **ptr= hv_fetchs(table, "regcomp", FALSE);
3893 if (ptr && SvIOK(*ptr) && SvIV(*ptr)) {
3894 const regexp_engine *eng=INT2PTR(regexp_engine*,SvIV(*ptr));
3896 PerlIO_printf(Perl_debug_log, "Using engine %"UVxf"\n",
3899 return CALLREGCOMP_ENG(eng, exp, xend, pm);
3913 RExC_state_t RExC_state;
3914 RExC_state_t * const pRExC_state = &RExC_state;
3915 #ifdef TRIE_STUDY_OPT
3917 RExC_state_t copyRExC_state;
3920 FAIL("NULL regexp argument");
3922 RExC_utf8 = pm->op_pmdynflags & PMdf_CMP_UTF8;
3926 SV *dsv= sv_newmortal();
3927 RE_PV_QUOTED_DECL(s, RExC_utf8,
3928 dsv, RExC_precomp, (xend - exp), 60);
3929 PerlIO_printf(Perl_debug_log, "%sCompiling REx%s %s\n",
3930 PL_colors[4],PL_colors[5],s);
3932 RExC_flags = pm->op_pmflags;
3936 RExC_seen_zerolen = *exp == '^' ? -1 : 0;
3937 RExC_seen_evals = 0;
3940 /* First pass: determine size, legality. */
3948 RExC_emit = &PL_regdummy;
3949 RExC_whilem_seen = 0;
3950 RExC_charnames = NULL;
3951 RExC_open_parens = NULL;
3952 RExC_close_parens = NULL;
3954 RExC_paren_names = NULL;
3955 RExC_recurse = NULL;
3956 RExC_recurse_count = 0;
3958 #if 0 /* REGC() is (currently) a NOP at the first pass.
3959 * Clever compilers notice this and complain. --jhi */
3960 REGC((U8)REG_MAGIC, (char*)RExC_emit);
3962 DEBUG_PARSE_r(PerlIO_printf(Perl_debug_log, "Starting first pass (sizing)\n"));
3963 if (reg(pRExC_state, 0, &flags,1) == NULL) {
3964 RExC_precomp = NULL;
3968 PerlIO_printf(Perl_debug_log,
3969 "Required size %"IVdf" nodes\n"
3970 "Starting second pass (creation)\n",
3973 RExC_lastparse=NULL;
3975 /* Small enough for pointer-storage convention?
3976 If extralen==0, this means that we will not need long jumps. */
3977 if (RExC_size >= 0x10000L && RExC_extralen)
3978 RExC_size += RExC_extralen;
3981 if (RExC_whilem_seen > 15)
3982 RExC_whilem_seen = 15;
3985 /* Make room for a sentinel value at the end of the program */
3989 /* Allocate space and zero-initialize. Note, the two step process
3990 of zeroing when in debug mode, thus anything assigned has to
3991 happen after that */
3992 Newxc(r, sizeof(regexp) + (unsigned)RExC_size * sizeof(regnode),
3995 FAIL("Regexp out of space");
3997 /* avoid reading uninitialized memory in DEBUGGING code in study_chunk() */
3998 Zero(r, sizeof(regexp) + (unsigned)RExC_size * sizeof(regnode), char);
4000 /* initialization begins here */
4001 r->engine= RE_ENGINE_PTR;
4003 r->prelen = xend - exp;
4004 r->precomp = savepvn(RExC_precomp, r->prelen);
4006 #ifdef PERL_OLD_COPY_ON_WRITE
4007 r->saved_copy = NULL;
4009 r->reganch = pm->op_pmflags & PMf_COMPILETIME;
4010 r->nparens = RExC_npar - 1; /* set early to validate backrefs */
4011 r->lastparen = 0; /* mg.c reads this. */
4013 r->substrs = 0; /* Useful during FAIL. */
4014 r->startp = 0; /* Useful during FAIL. */
4018 if (RExC_seen & REG_SEEN_RECURSE) {
4019 Newxz(RExC_open_parens, RExC_npar,regnode *);
4020 SAVEFREEPV(RExC_open_parens);
4021 Newxz(RExC_close_parens,RExC_npar,regnode *);
4022 SAVEFREEPV(RExC_close_parens);
4025 /* Useful during FAIL. */
4026 Newxz(r->offsets, 2*RExC_size+1, U32); /* MJD 20001228 */
4028 r->offsets[0] = RExC_size;
4030 DEBUG_OFFSETS_r(PerlIO_printf(Perl_debug_log,
4031 "%s %"UVuf" bytes for offset annotations.\n",
4032 r->offsets ? "Got" : "Couldn't get",
4033 (UV)((2*RExC_size+1) * sizeof(U32))));
4037 /* Second pass: emit code. */
4038 RExC_flags = pm->op_pmflags; /* don't let top level (?i) bleed */
4043 RExC_emit_start = r->program;
4044 RExC_emit = r->program;
4046 /* put a sentinal on the end of the program so we can check for
4048 r->program[RExC_size].type = 255;
4050 /* Store the count of eval-groups for security checks: */
4051 RExC_emit->next_off = (RExC_seen_evals > (I32)U16_MAX) ? U16_MAX : (U16)RExC_seen_evals;
4052 REGC((U8)REG_MAGIC, (char*) RExC_emit++);
4054 if (reg(pRExC_state, 0, &flags,1) == NULL)
4057 /* XXXX To minimize changes to RE engine we always allocate
4058 3-units-long substrs field. */
4059 Newx(r->substrs, 1, struct reg_substr_data);
4060 if (RExC_recurse_count) {
4061 Newxz(RExC_recurse,RExC_recurse_count,regnode *);
4062 SAVEFREEPV(RExC_recurse);
4066 r->minlen = minlen = sawplus = sawopen = 0;
4067 Zero(r->substrs, 1, struct reg_substr_data);
4069 #ifdef TRIE_STUDY_OPT
4072 DEBUG_OPTIMISE_r(PerlIO_printf(Perl_debug_log,"Restudying\n"));
4074 RExC_state = copyRExC_state;
4075 if (seen & REG_TOP_LEVEL_BRANCHES)
4076 RExC_seen |= REG_TOP_LEVEL_BRANCHES;
4078 RExC_seen &= ~REG_TOP_LEVEL_BRANCHES;
4079 if (data.last_found) {
4080 SvREFCNT_dec(data.longest_fixed);
4081 SvREFCNT_dec(data.longest_float);
4082 SvREFCNT_dec(data.last_found);
4084 StructCopy(&zero_scan_data, &data, scan_data_t);
4086 StructCopy(&zero_scan_data, &data, scan_data_t);
4087 copyRExC_state = RExC_state;
4090 StructCopy(&zero_scan_data, &data, scan_data_t);
4093 /* Dig out information for optimizations. */
4094 r->reganch = pm->op_pmflags & PMf_COMPILETIME; /* Again? */
4095 pm->op_pmflags = RExC_flags;
4097 r->reganch |= ROPT_UTF8; /* Unicode in it? */
4098 r->regstclass = NULL;
4099 if (RExC_naughty >= 10) /* Probably an expensive pattern. */
4100 r->reganch |= ROPT_NAUGHTY;
4101 scan = r->program + 1; /* First BRANCH. */
4103 /* testing for BRANCH here tells us whether there is "must appear"
4104 data in the pattern. If there is then we can use it for optimisations */
4105 if (!(RExC_seen & REG_TOP_LEVEL_BRANCHES)) { /* Only one top-level choice. */
4107 STRLEN longest_float_length, longest_fixed_length;
4108 struct regnode_charclass_class ch_class; /* pointed to by data */
4110 I32 last_close = 0; /* pointed to by data */
4113 /* Skip introductions and multiplicators >= 1. */
4114 while ((OP(first) == OPEN && (sawopen = 1)) ||
4115 /* An OR of *one* alternative - should not happen now. */
4116 (OP(first) == BRANCH && OP(regnext(first)) != BRANCH) ||
4117 /* for now we can't handle lookbehind IFMATCH*/
4118 (OP(first) == IFMATCH && !first->flags) ||
4119 (OP(first) == PLUS) ||
4120 (OP(first) == MINMOD) ||
4121 /* An {n,m} with n>0 */
4122 (PL_regkind[OP(first)] == CURLY && ARG1(first) > 0) )
4125 if (OP(first) == PLUS)
4128 first += regarglen[OP(first)];
4129 if (OP(first) == IFMATCH) {
4130 first = NEXTOPER(first);
4131 first += EXTRA_STEP_2ARGS;
4132 } else /* XXX possible optimisation for /(?=)/ */
4133 first = NEXTOPER(first);
4136 /* Starting-point info. */
4138 DEBUG_PEEP("first:",first,0);
4139 /* Ignore EXACT as we deal with it later. */
4140 if (PL_regkind[OP(first)] == EXACT) {
4141 if (OP(first) == EXACT)
4142 NOOP; /* Empty, get anchored substr later. */
4143 else if ((OP(first) == EXACTF || OP(first) == EXACTFL))
4144 r->regstclass = first;
4147 else if (PL_regkind[OP(first)] == TRIE &&
4148 ((reg_trie_data *)r->data->data[ ARG(first) ])->minlen>0)
4151 /* this can happen only on restudy */
4152 if ( OP(first) == TRIE ) {
4153 struct regnode_1 *trieop;
4154 Newxz(trieop,1,struct regnode_1);
4155 StructCopy(first,trieop,struct regnode_1);
4156 trie_op=(regnode *)trieop;
4158 struct regnode_charclass *trieop;
4159 Newxz(trieop,1,struct regnode_charclass);
4160 StructCopy(first,trieop,struct regnode_charclass);
4161 trie_op=(regnode *)trieop;
4164 make_trie_failtable(pRExC_state, (regnode *)first, trie_op, 0);
4165 r->regstclass = trie_op;
4168 else if (strchr((const char*)PL_simple,OP(first)))
4169 r->regstclass = first;
4170 else if (PL_regkind[OP(first)] == BOUND ||
4171 PL_regkind[OP(first)] == NBOUND)
4172 r->regstclass = first;
4173 else if (PL_regkind[OP(first)] == BOL) {
4174 r->reganch |= (OP(first) == MBOL
4176 : (OP(first) == SBOL
4179 first = NEXTOPER(first);
4182 else if (OP(first) == GPOS) {
4183 r->reganch |= ROPT_ANCH_GPOS;
4184 first = NEXTOPER(first);
4187 else if (!sawopen && (OP(first) == STAR &&
4188 PL_regkind[OP(NEXTOPER(first))] == REG_ANY) &&
4189 !(r->reganch & ROPT_ANCH) )
4191 /* turn .* into ^.* with an implied $*=1 */
4193 (OP(NEXTOPER(first)) == REG_ANY)
4196 r->reganch |= type | ROPT_IMPLICIT;
4197 first = NEXTOPER(first);
4200 if (sawplus && (!sawopen || !RExC_sawback)
4201 && !(RExC_seen & REG_SEEN_EVAL)) /* May examine pos and $& */
4202 /* x+ must match at the 1st pos of run of x's */
4203 r->reganch |= ROPT_SKIP;
4205 /* Scan is after the zeroth branch, first is atomic matcher. */
4206 #ifdef TRIE_STUDY_OPT
4209 PerlIO_printf(Perl_debug_log, "first at %"IVdf"\n",
4210 (IV)(first - scan + 1))
4214 PerlIO_printf(Perl_debug_log, "first at %"IVdf"\n",
4215 (IV)(first - scan + 1))
4221 * If there's something expensive in the r.e., find the
4222 * longest literal string that must appear and make it the
4223 * regmust. Resolve ties in favor of later strings, since
4224 * the regstart check works with the beginning of the r.e.
4225 * and avoiding duplication strengthens checking. Not a
4226 * strong reason, but sufficient in the absence of others.
4227 * [Now we resolve ties in favor of the earlier string if
4228 * it happens that c_offset_min has been invalidated, since the
4229 * earlier string may buy us something the later one won't.]
4232 data.longest_fixed = newSVpvs("");
4233 data.longest_float = newSVpvs("");
4234 data.last_found = newSVpvs("");
4235 data.longest = &(data.longest_fixed);
4237 if (!r->regstclass) {
4238 cl_init(pRExC_state, &ch_class);
4239 data.start_class = &ch_class;
4240 stclass_flag = SCF_DO_STCLASS_AND;
4241 } else /* XXXX Check for BOUND? */
4243 data.last_closep = &last_close;
4245 minlen = study_chunk(pRExC_state, &first, &minlen, &fake, scan + RExC_size, /* Up to end */
4246 &data, -1, NULL, NULL,
4247 SCF_DO_SUBSTR | SCF_WHILEM_VISITED_POS | stclass_flag,0);
4253 if ( RExC_npar == 1 && data.longest == &(data.longest_fixed)
4254 && data.last_start_min == 0 && data.last_end > 0
4255 && !RExC_seen_zerolen
4256 && (!(RExC_seen & REG_SEEN_GPOS) || (r->reganch & ROPT_ANCH_GPOS)))
4257 r->reganch |= ROPT_CHECK_ALL;
4258 scan_commit(pRExC_state, &data,&minlen);
4259 SvREFCNT_dec(data.last_found);
4261 /* Note that code very similar to this but for anchored string
4262 follows immediately below, changes may need to be made to both.
4265 longest_float_length = CHR_SVLEN(data.longest_float);
4266 if (longest_float_length
4267 || (data.flags & SF_FL_BEFORE_EOL
4268 && (!(data.flags & SF_FL_BEFORE_MEOL)
4269 || (RExC_flags & PMf_MULTILINE))))
4273 if (SvCUR(data.longest_fixed) /* ok to leave SvCUR */
4274 && data.offset_fixed == data.offset_float_min
4275 && SvCUR(data.longest_fixed) == SvCUR(data.longest_float))
4276 goto remove_float; /* As in (a)+. */
4278 /* copy the information about the longest float from the reg_scan_data
4279 over to the program. */
4280 if (SvUTF8(data.longest_float)) {
4281 r->float_utf8 = data.longest_float;
4282 r->float_substr = NULL;
4284 r->float_substr = data.longest_float;
4285 r->float_utf8 = NULL;
4287 /* float_end_shift is how many chars that must be matched that
4288 follow this item. We calculate it ahead of time as once the
4289 lookbehind offset is added in we lose the ability to correctly
4291 ml = data.minlen_float ? *(data.minlen_float)
4292 : (I32)longest_float_length;
4293 r->float_end_shift = ml - data.offset_float_min
4294 - longest_float_length + (SvTAIL(data.longest_float) != 0)
4295 + data.lookbehind_float;
4296 r->float_min_offset = data.offset_float_min - data.lookbehind_float;
4297 r->float_max_offset = data.offset_float_max;
4298 if (data.offset_float_max < I32_MAX) /* Don't offset infinity */
4299 r->float_max_offset -= data.lookbehind_float;
4301 t = (data.flags & SF_FL_BEFORE_EOL /* Can't have SEOL and MULTI */
4302 && (!(data.flags & SF_FL_BEFORE_MEOL)
4303 || (RExC_flags & PMf_MULTILINE)));
4304 fbm_compile(data.longest_float, t ? FBMcf_TAIL : 0);
4308 r->float_substr = r->float_utf8 = NULL;
4309 SvREFCNT_dec(data.longest_float);
4310 longest_float_length = 0;
4313 /* Note that code very similar to this but for floating string
4314 is immediately above, changes may need to be made to both.
4317 longest_fixed_length = CHR_SVLEN(data.longest_fixed);
4318 if (longest_fixed_length
4319 || (data.flags & SF_FIX_BEFORE_EOL /* Cannot have SEOL and MULTI */
4320 && (!(data.flags & SF_FIX_BEFORE_MEOL)
4321 || (RExC_flags & PMf_MULTILINE))))
4325 /* copy the information about the longest fixed
4326 from the reg_scan_data over to the program. */
4327 if (SvUTF8(data.longest_fixed)) {
4328 r->anchored_utf8 = data.longest_fixed;
4329 r->anchored_substr = NULL;
4331 r->anchored_substr = data.longest_fixed;
4332 r->anchored_utf8 = NULL;
4334 /* fixed_end_shift is how many chars that must be matched that
4335 follow this item. We calculate it ahead of time as once the
4336 lookbehind offset is added in we lose the ability to correctly
4338 ml = data.minlen_fixed ? *(data.minlen_fixed)
4339 : (I32)longest_fixed_length;
4340 r->anchored_end_shift = ml - data.offset_fixed
4341 - longest_fixed_length + (SvTAIL(data.longest_fixed) != 0)
4342 + data.lookbehind_fixed;
4343 r->anchored_offset = data.offset_fixed - data.lookbehind_fixed;
4345 t = (data.flags & SF_FIX_BEFORE_EOL /* Can't have SEOL and MULTI */
4346 && (!(data.flags & SF_FIX_BEFORE_MEOL)
4347 || (RExC_flags & PMf_MULTILINE)));
4348 fbm_compile(data.longest_fixed, t ? FBMcf_TAIL : 0);
4351 r->anchored_substr = r->anchored_utf8 = NULL;
4352 SvREFCNT_dec(data.longest_fixed);
4353 longest_fixed_length = 0;
4356 && (OP(r->regstclass) == REG_ANY || OP(r->regstclass) == SANY))
4357 r->regstclass = NULL;
4358 if ((!(r->anchored_substr || r->anchored_utf8) || r->anchored_offset)
4360 && !(data.start_class->flags & ANYOF_EOS)
4361 && !cl_is_anything(data.start_class))
4363 const I32 n = add_data(pRExC_state, 1, "f");
4365 Newx(RExC_rx->data->data[n], 1,
4366 struct regnode_charclass_class);
4367 StructCopy(data.start_class,
4368 (struct regnode_charclass_class*)RExC_rx->data->data[n],
4369 struct regnode_charclass_class);
4370 r->regstclass = (regnode*)RExC_rx->data->data[n];
4371 r->reganch &= ~ROPT_SKIP; /* Used in find_byclass(). */
4372 DEBUG_COMPILE_r({ SV *sv = sv_newmortal();
4373 regprop(r, sv, (regnode*)data.start_class);
4374 PerlIO_printf(Perl_debug_log,
4375 "synthetic stclass \"%s\".\n",
4376 SvPVX_const(sv));});
4379 /* A temporary algorithm prefers floated substr to fixed one to dig more info. */
4380 if (longest_fixed_length > longest_float_length) {
4381 r->check_end_shift = r->anchored_end_shift;
4382 r->check_substr = r->anchored_substr;
4383 r->check_utf8 = r->anchored_utf8;
4384 r->check_offset_min = r->check_offset_max = r->anchored_offset;
4385 if (r->reganch & ROPT_ANCH_SINGLE)
4386 r->reganch |= ROPT_NOSCAN;
4389 r->check_end_shift = r->float_end_shift;
4390 r->check_substr = r->float_substr;
4391 r->check_utf8 = r->float_utf8;
4392 r->check_offset_min = r->float_min_offset;
4393 r->check_offset_max = r->float_max_offset;
4395 /* XXXX Currently intuiting is not compatible with ANCH_GPOS.
4396 This should be changed ASAP! */
4397 if ((r->check_substr || r->check_utf8) && !(r->reganch & ROPT_ANCH_GPOS)) {
4398 r->reganch |= RE_USE_INTUIT;
4399 if (SvTAIL(r->check_substr ? r->check_substr : r->check_utf8))
4400 r->reganch |= RE_INTUIT_TAIL;
4402 /* XXX Unneeded? dmq (shouldn't as this is handled elsewhere)
4403 if ( (STRLEN)minlen < longest_float_length )
4404 minlen= longest_float_length;
4405 if ( (STRLEN)minlen < longest_fixed_length )
4406 minlen= longest_fixed_length;
4410 /* Several toplevels. Best we can is to set minlen. */
4412 struct regnode_charclass_class ch_class;
4415 DEBUG_PARSE_r(PerlIO_printf(Perl_debug_log, "\nMulti Top Level\n"));
4417 scan = r->program + 1;
4418 cl_init(pRExC_state, &ch_class);
4419 data.start_class = &ch_class;
4420 data.last_closep = &last_close;
4423 minlen = study_chunk(pRExC_state, &scan, &minlen, &fake, scan + RExC_size,
4424 &data, -1, NULL, NULL, SCF_DO_STCLASS_AND|SCF_WHILEM_VISITED_POS,0);
4428 r->check_substr = r->check_utf8 = r->anchored_substr = r->anchored_utf8
4429 = r->float_substr = r->float_utf8 = NULL;
4430 if (!(data.start_class->flags & ANYOF_EOS)
4431 && !cl_is_anything(data.start_class))
4433 const I32 n = add_data(pRExC_state, 1, "f");
4435 Newx(RExC_rx->data->data[n], 1,
4436 struct regnode_charclass_class);
4437 StructCopy(data.start_class,
4438 (struct regnode_charclass_class*)RExC_rx->data->data[n],
4439 struct regnode_charclass_class);
4440 r->regstclass = (regnode*)RExC_rx->data->data[n];
4441 r->reganch &= ~ROPT_SKIP; /* Used in find_byclass(). */
4442 DEBUG_COMPILE_r({ SV* sv = sv_newmortal();
4443 regprop(r, sv, (regnode*)data.start_class);
4444 PerlIO_printf(Perl_debug_log,
4445 "synthetic stclass \"%s\".\n",
4446 SvPVX_const(sv));});
4450 /* Guard against an embedded (?=) or (?<=) with a longer minlen than
4451 the "real" pattern. */
4453 PerlIO_printf(Perl_debug_log,"minlen: %"IVdf" r->minlen:%"IVdf"\n",
4456 r->minlenret = minlen;
4457 if (r->minlen < minlen)
4460 if (RExC_seen & REG_SEEN_GPOS)
4461 r->reganch |= ROPT_GPOS_SEEN;
4462 if (RExC_seen & REG_SEEN_LOOKBEHIND)
4463 r->reganch |= ROPT_LOOKBEHIND_SEEN;
4464 if (RExC_seen & REG_SEEN_EVAL)
4465 r->reganch |= ROPT_EVAL_SEEN;
4466 if (RExC_seen & REG_SEEN_CANY)
4467 r->reganch |= ROPT_CANY_SEEN;
4468 if (RExC_seen & REG_SEEN_VERBARG)
4469 r->reganch |= ROPT_VERBARG_SEEN;
4470 if (RExC_seen & REG_SEEN_CUTGROUP)
4471 r->reganch |= ROPT_CUTGROUP_SEEN;
4472 if (RExC_paren_names)
4473 r->paren_names = (HV*)SvREFCNT_inc(RExC_paren_names);
4475 r->paren_names = NULL;
4477 if (RExC_recurse_count) {
4478 for ( ; RExC_recurse_count ; RExC_recurse_count-- ) {
4479 const regnode *scan = RExC_recurse[RExC_recurse_count-1];
4480 ARG2L_SET( scan, RExC_open_parens[ARG(scan)-1] - scan );
4483 Newxz(r->startp, RExC_npar, I32);
4484 Newxz(r->endp, RExC_npar, I32);
4487 PerlIO_printf(Perl_debug_log,"Final program:\n");
4490 DEBUG_OFFSETS_r(if (r->offsets) {
4491 const U32 len = r->offsets[0];
4493 GET_RE_DEBUG_FLAGS_DECL;
4494 PerlIO_printf(Perl_debug_log, "Offsets: [%"UVuf"]\n\t", (UV)r->offsets[0]);
4495 for (i = 1; i <= len; i++) {
4496 if (r->offsets[i*2-1] || r->offsets[i*2])
4497 PerlIO_printf(Perl_debug_log, "%"UVuf":%"UVuf"[%"UVuf"] ",
4498 (UV)i, (UV)r->offsets[i*2-1], (UV)r->offsets[i*2]);
4500 PerlIO_printf(Perl_debug_log, "\n");
4506 #undef CORE_ONLY_BLOCK
4508 #undef RE_ENGINE_PTR
4510 #ifndef PERL_IN_XSUB_RE
4512 Perl_reg_named_buff_sv(pTHX_ SV* namesv)
4514 I32 parno = 0; /* no match */
4516 const REGEXP * const rx = PM_GETRE(PL_curpm);
4517 if (rx && rx->paren_names) {
4518 HE *he_str = hv_fetch_ent( rx->paren_names, namesv, 0, 0 );
4521 SV* sv_dat=HeVAL(he_str);
4522 I32 *nums=(I32*)SvPVX(sv_dat);
4523 for ( i=0; i<SvIVX(sv_dat); i++ ) {
4524 if ((I32)(rx->lastparen) >= nums[i] &&
4525 rx->endp[nums[i]] != -1)
4538 SV *sv= sv_newmortal();
4539 Perl_sv_setpvf(aTHX_ sv, "%"IVdf,(IV)parno);
4540 gv_paren= Perl_gv_fetchsv(aTHX_ sv, GV_ADD, SVt_PVGV);
4541 return GvSVn(gv_paren);
4546 /* Scans the name of a named buffer from the pattern.
4547 * If flags is REG_RSN_RETURN_NULL returns null.
4548 * If flags is REG_RSN_RETURN_NAME returns an SV* containing the name
4549 * If flags is REG_RSN_RETURN_DATA returns the data SV* corresponding
4550 * to the parsed name as looked up in the RExC_paren_names hash.
4551 * If there is an error throws a vFAIL().. type exception.
4554 #define REG_RSN_RETURN_NULL 0
4555 #define REG_RSN_RETURN_NAME 1
4556 #define REG_RSN_RETURN_DATA 2
4559 S_reg_scan_name(pTHX_ RExC_state_t *pRExC_state, U32 flags) {
4560 char *name_start = RExC_parse;
4563 while( isIDFIRST_uni(utf8n_to_uvchr((U8*)RExC_parse,
4564 RExC_end - RExC_parse, &numlen, UTF8_ALLOW_DEFAULT)))
4566 RExC_parse += numlen;
4569 while( isIDFIRST(*RExC_parse) )
4573 SV* sv_name = sv_2mortal(Perl_newSVpvn(aTHX_ name_start,
4574 (int)(RExC_parse - name_start)));
4577 if ( flags == REG_RSN_RETURN_NAME)
4579 else if (flags==REG_RSN_RETURN_DATA) {
4582 if ( ! sv_name ) /* should not happen*/
4583 Perl_croak(aTHX_ "panic: no svname in reg_scan_name");
4584 if (RExC_paren_names)
4585 he_str = hv_fetch_ent( RExC_paren_names, sv_name, 0, 0 );
4587 sv_dat = HeVAL(he_str);
4589 vFAIL("Reference to nonexistent named group");
4593 Perl_croak(aTHX_ "panic: bad flag in reg_scan_name");
4600 #define DEBUG_PARSE_MSG(funcname) DEBUG_PARSE_r({ \
4601 int rem=(int)(RExC_end - RExC_parse); \
4610 if (RExC_lastparse!=RExC_parse) \
4611 PerlIO_printf(Perl_debug_log," >%.*s%-*s", \
4614 iscut ? "..." : "<" \
4617 PerlIO_printf(Perl_debug_log,"%16s",""); \
4622 num=REG_NODE_NUM(RExC_emit); \
4623 if (RExC_lastnum!=num) \
4624 PerlIO_printf(Perl_debug_log,"|%4d",num); \
4626 PerlIO_printf(Perl_debug_log,"|%4s",""); \
4627 PerlIO_printf(Perl_debug_log,"|%*s%-4s", \
4628 (int)((depth*2)), "", \
4632 RExC_lastparse=RExC_parse; \
4637 #define DEBUG_PARSE(funcname) DEBUG_PARSE_r({ \
4638 DEBUG_PARSE_MSG((funcname)); \
4639 PerlIO_printf(Perl_debug_log,"%4s","\n"); \
4641 #define DEBUG_PARSE_FMT(funcname,fmt,args) DEBUG_PARSE_r({ \
4642 DEBUG_PARSE_MSG((funcname)); \
4643 PerlIO_printf(Perl_debug_log,fmt "\n",args); \
4646 - reg - regular expression, i.e. main body or parenthesized thing
4648 * Caller must absorb opening parenthesis.
4650 * Combining parenthesis handling with the base level of regular expression
4651 * is a trifle forced, but the need to tie the tails of the branches to what
4652 * follows makes it hard to avoid.
4654 #define REGTAIL(x,y,z) regtail((x),(y),(z),depth+1)
4656 #define REGTAIL_STUDY(x,y,z) regtail_study((x),(y),(z),depth+1)
4658 #define REGTAIL_STUDY(x,y,z) regtail((x),(y),(z),depth+1)
4661 /* this idea is borrowed from STR_WITH_LEN in handy.h */
4662 #define CHECK_WORD(s,v,l) \
4663 (((sizeof(s)-1)==(l)) && (strnEQ(start_verb, (s ""), (sizeof(s)-1))))
4666 S_reg(pTHX_ RExC_state_t *pRExC_state, I32 paren, I32 *flagp,U32 depth)
4667 /* paren: Parenthesized? 0=top, 1=(, inside: changed to letter. */
4670 register regnode *ret; /* Will be the head of the group. */
4671 register regnode *br;
4672 register regnode *lastbr;
4673 register regnode *ender = NULL;
4674 register I32 parno = 0;
4676 const I32 oregflags = RExC_flags;
4677 bool have_branch = 0;
4680 /* for (?g), (?gc), and (?o) warnings; warning
4681 about (?c) will warn about (?g) -- japhy */
4683 #define WASTED_O 0x01
4684 #define WASTED_G 0x02
4685 #define WASTED_C 0x04
4686 #define WASTED_GC (0x02|0x04)
4687 I32 wastedflags = 0x00;
4689 char * parse_start = RExC_parse; /* MJD */
4690 char * const oregcomp_parse = RExC_parse;
4692 GET_RE_DEBUG_FLAGS_DECL;
4693 DEBUG_PARSE("reg ");
4696 *flagp = 0; /* Tentatively. */
4699 /* Make an OPEN node, if parenthesized. */
4701 if ( *RExC_parse == '*') { /* (*VERB:ARG) */
4702 char *start_verb = RExC_parse;
4703 STRLEN verb_len = 0;
4704 char *start_arg = NULL;
4705 unsigned char op = 0;
4707 int internal_argval = 0; /* internal_argval is only useful if !argok */
4708 while ( *RExC_parse && *RExC_parse != ')' ) {
4709 if ( *RExC_parse == ':' ) {
4710 start_arg = RExC_parse + 1;
4716 verb_len = RExC_parse - start_verb;
4719 while ( *RExC_parse && *RExC_parse != ')' )
4721 if ( *RExC_parse != ')' )
4722 vFAIL("Unterminated verb pattern argument");
4723 if ( RExC_parse == start_arg )
4726 if ( *RExC_parse != ')' )
4727 vFAIL("Unterminated verb pattern");
4730 switch ( *start_verb ) {
4731 case 'A': /* (*ACCEPT) */
4732 if ( CHECK_WORD("ACCEPT",start_verb,verb_len) ) {
4734 internal_argval = RExC_nestroot;
4737 case 'C': /* (*COMMIT) */
4738 if ( CHECK_WORD("COMMIT",start_verb,verb_len) )
4741 case 'F': /* (*FAIL) */
4742 if ( verb_len==1 || CHECK_WORD("FAIL",start_verb,verb_len) ) {
4747 case ':': /* (*:NAME) */
4748 case 'M': /* (*MARK:NAME) */
4749 if ( verb_len==0 || CHECK_WORD("MARK",start_verb,verb_len) ) {
4754 case 'P': /* (*PRUNE) */
4755 if ( CHECK_WORD("PRUNE",start_verb,verb_len) )
4758 case 'S': /* (*SKIP) */
4759 if ( CHECK_WORD("SKIP",start_verb,verb_len) )
4762 case 'T': /* (*THEN) */
4763 /* [19:06] <TimToady> :: is then */
4764 if ( CHECK_WORD("THEN",start_verb,verb_len) ) {
4766 RExC_seen |= REG_SEEN_CUTGROUP;
4772 vFAIL3("Unknown verb pattern '%.*s'",
4773 verb_len, start_verb);
4776 if ( start_arg && internal_argval ) {
4777 vFAIL3("Verb pattern '%.*s' may not have an argument",
4778 verb_len, start_verb);
4779 } else if ( argok < 0 && !start_arg ) {
4780 vFAIL3("Verb pattern '%.*s' has a mandatory argument",
4781 verb_len, start_verb);
4783 ret = reganode(pRExC_state, op, internal_argval);
4784 if ( ! internal_argval && ! SIZE_ONLY ) {
4786 SV *sv = newSVpvn( start_arg, RExC_parse - start_arg);
4787 ARG(ret) = add_data( pRExC_state, 1, "S" );
4788 RExC_rx->data->data[ARG(ret)]=(void*)sv;
4795 if (!internal_argval)
4796 RExC_seen |= REG_SEEN_VERBARG;
4797 } else if ( start_arg ) {
4798 vFAIL3("Verb pattern '%.*s' may not have an argument",
4799 verb_len, start_verb);
4801 ret = reg_node(pRExC_state, op);
4803 nextchar(pRExC_state);
4806 if (*RExC_parse == '?') { /* (?...) */
4807 U32 posflags = 0, negflags = 0;
4808 U32 *flagsp = &posflags;
4809 bool is_logical = 0;
4810 const char * const seqstart = RExC_parse;
4813 paren = *RExC_parse++;
4814 ret = NULL; /* For look-ahead/behind. */
4817 case '<': /* (?<...) */
4818 if (*RExC_parse == '!')
4820 else if (*RExC_parse != '=')
4825 case '\'': /* (?'...') */
4826 name_start= RExC_parse;
4827 svname = reg_scan_name(pRExC_state,
4828 SIZE_ONLY ? /* reverse test from the others */
4829 REG_RSN_RETURN_NAME :
4830 REG_RSN_RETURN_NULL);
4831 if (RExC_parse == name_start)
4833 if (*RExC_parse != paren)
4834 vFAIL2("Sequence (?%c... not terminated",
4835 paren=='>' ? '<' : paren);
4839 if (!svname) /* shouldnt happen */
4841 "panic: reg_scan_name returned NULL");
4842 if (!RExC_paren_names) {
4843 RExC_paren_names= newHV();
4844 sv_2mortal((SV*)RExC_paren_names);
4846 he_str = hv_fetch_ent( RExC_paren_names, svname, 1, 0 );
4848 sv_dat = HeVAL(he_str);
4850 /* croak baby croak */
4852 "panic: paren_name hash element allocation failed");
4853 } else if ( SvPOK(sv_dat) ) {
4854 IV count=SvIV(sv_dat);
4855 I32 *pv=(I32*)SvGROW(sv_dat,SvCUR(sv_dat)+sizeof(I32)+1);
4856 SvCUR_set(sv_dat,SvCUR(sv_dat)+sizeof(I32));
4857 pv[count]=RExC_npar;
4860 (void)SvUPGRADE(sv_dat,SVt_PVNV);
4861 sv_setpvn(sv_dat, (char *)&(RExC_npar), sizeof(I32));
4866 /*sv_dump(sv_dat);*/
4868 nextchar(pRExC_state);
4870 goto capturing_parens;
4872 RExC_seen |= REG_SEEN_LOOKBEHIND;
4874 case '=': /* (?=...) */
4875 case '!': /* (?!...) */
4876 RExC_seen_zerolen++;
4877 if (*RExC_parse == ')') {
4878 ret=reg_node(pRExC_state, OPFAIL);
4879 nextchar(pRExC_state);
4882 case ':': /* (?:...) */
4883 case '>': /* (?>...) */
4885 case '$': /* (?$...) */
4886 case '@': /* (?@...) */
4887 vFAIL2("Sequence (?%c...) not implemented", (int)paren);
4889 case '#': /* (?#...) */
4890 while (*RExC_parse && *RExC_parse != ')')
4892 if (*RExC_parse != ')')
4893 FAIL("Sequence (?#... not terminated");
4894 nextchar(pRExC_state);
4897 case '0' : /* (?0) */
4898 case 'R' : /* (?R) */
4899 if (*RExC_parse != ')')
4900 FAIL("Sequence (?R) not terminated");
4901 ret = reg_node(pRExC_state, GOSTART);
4902 nextchar(pRExC_state);
4905 { /* named and numeric backreferences */
4908 case '&': /* (?&NAME) */
4909 parse_start = RExC_parse - 1;
4911 SV *sv_dat = reg_scan_name(pRExC_state,
4912 SIZE_ONLY ? REG_RSN_RETURN_NULL : REG_RSN_RETURN_DATA);
4913 num = sv_dat ? *((I32 *)SvPVX(sv_dat)) : 0;
4915 goto gen_recurse_regop;
4918 if (!(RExC_parse[0] >= '1' && RExC_parse[0] <= '9')) {
4920 vFAIL("Illegal pattern");
4922 goto parse_recursion;
4924 case '-': /* (?-1) */
4925 if (!(RExC_parse[0] >= '1' && RExC_parse[0] <= '9')) {
4926 RExC_parse--; /* rewind to let it be handled later */
4930 case '1': case '2': case '3': case '4': /* (?1) */
4931 case '5': case '6': case '7': case '8': case '9':
4934 num = atoi(RExC_parse);
4935 parse_start = RExC_parse - 1; /* MJD */
4936 if (*RExC_parse == '-')
4938 while (isDIGIT(*RExC_parse))
4940 if (*RExC_parse!=')')
4941 vFAIL("Expecting close bracket");
4944 if ( paren == '-' ) {
4946 Diagram of capture buffer numbering.
4947 Top line is the normal capture buffer numbers
4948 Botton line is the negative indexing as from
4952 /(a(x)y)(a(b(c(?-2)d)e)f)(g(h))/
4956 num = RExC_npar + num;
4959 vFAIL("Reference to nonexistent group");
4961 } else if ( paren == '+' ) {
4962 num = RExC_npar + num - 1;
4965 ret = reganode(pRExC_state, GOSUB, num);
4967 if (num > (I32)RExC_rx->nparens) {
4969 vFAIL("Reference to nonexistent group");
4971 ARG2L_SET( ret, RExC_recurse_count++);
4973 DEBUG_OPTIMISE_MORE_r(PerlIO_printf(Perl_debug_log,
4974 "Recurse #%"UVuf" to %"IVdf"\n", (UV)ARG(ret), (IV)ARG2L(ret)));
4978 RExC_seen |= REG_SEEN_RECURSE;
4979 Set_Node_Length(ret, 1 + regarglen[OP(ret)]); /* MJD */
4980 Set_Node_Offset(ret, parse_start); /* MJD */
4982 nextchar(pRExC_state);
4984 } /* named and numeric backreferences */
4987 case 'p': /* (?p...) */
4988 if (SIZE_ONLY && ckWARN2(WARN_DEPRECATED, WARN_REGEXP))
4989 vWARNdep(RExC_parse, "(?p{}) is deprecated - use (??{})");
4991 case '?': /* (??...) */
4993 if (*RExC_parse != '{')
4995 paren = *RExC_parse++;
4997 case '{': /* (?{...}) */
4999 I32 count = 1, n = 0;
5001 char *s = RExC_parse;
5003 RExC_seen_zerolen++;
5004 RExC_seen |= REG_SEEN_EVAL;
5005 while (count && (c = *RExC_parse)) {
5016 if (*RExC_parse != ')') {
5018 vFAIL("Sequence (?{...}) not terminated or not {}-balanced");
5022 OP_4tree *sop, *rop;
5023 SV * const sv = newSVpvn(s, RExC_parse - 1 - s);
5026 Perl_save_re_context(aTHX);
5027 rop = sv_compile_2op(sv, &sop, "re", &pad);
5028 sop->op_private |= OPpREFCOUNTED;
5029 /* re_dup will OpREFCNT_inc */
5030 OpREFCNT_set(sop, 1);
5033 n = add_data(pRExC_state, 3, "nop");
5034 RExC_rx->data->data[n] = (void*)rop;
5035 RExC_rx->data->data[n+1] = (void*)sop;
5036 RExC_rx->data->data[n+2] = (void*)pad;
5039 else { /* First pass */
5040 if (PL_reginterp_cnt < ++RExC_seen_evals
5042 /* No compiled RE interpolated, has runtime
5043 components ===> unsafe. */
5044 FAIL("Eval-group not allowed at runtime, use re 'eval'");
5045 if (PL_tainting && PL_tainted)
5046 FAIL("Eval-group in insecure regular expression");
5047 #if PERL_VERSION > 8
5048 if (IN_PERL_COMPILETIME)
5053 nextchar(pRExC_state);
5055 ret = reg_node(pRExC_state, LOGICAL);
5058 REGTAIL(pRExC_state, ret, reganode(pRExC_state, EVAL, n));
5059 /* deal with the length of this later - MJD */
5062 ret = reganode(pRExC_state, EVAL, n);
5063 Set_Node_Length(ret, RExC_parse - parse_start + 1);
5064 Set_Node_Offset(ret, parse_start);
5067 case '(': /* (?(?{...})...) and (?(?=...)...) */
5070 if (RExC_parse[0] == '?') { /* (?(?...)) */
5071 if (RExC_parse[1] == '=' || RExC_parse[1] == '!'
5072 || RExC_parse[1] == '<'
5073 || RExC_parse[1] == '{') { /* Lookahead or eval. */
5076 ret = reg_node(pRExC_state, LOGICAL);
5079 REGTAIL(pRExC_state, ret, reg(pRExC_state, 1, &flag,depth+1));
5083 else if ( RExC_parse[0] == '<' /* (?(<NAME>)...) */
5084 || RExC_parse[0] == '\'' ) /* (?('NAME')...) */
5086 char ch = RExC_parse[0] == '<' ? '>' : '\'';
5087 char *name_start= RExC_parse++;
5089 SV *sv_dat=reg_scan_name(pRExC_state,
5090 SIZE_ONLY ? REG_RSN_RETURN_NULL : REG_RSN_RETURN_DATA);
5091 if (RExC_parse == name_start || *RExC_parse != ch)
5092 vFAIL2("Sequence (?(%c... not terminated",
5093 (ch == '>' ? '<' : ch));
5096 num = add_data( pRExC_state, 1, "S" );
5097 RExC_rx->data->data[num]=(void*)sv_dat;
5098 SvREFCNT_inc(sv_dat);
5100 ret = reganode(pRExC_state,NGROUPP,num);
5101 goto insert_if_check_paren;
5103 else if (RExC_parse[0] == 'D' &&
5104 RExC_parse[1] == 'E' &&
5105 RExC_parse[2] == 'F' &&
5106 RExC_parse[3] == 'I' &&
5107 RExC_parse[4] == 'N' &&
5108 RExC_parse[5] == 'E')
5110 ret = reganode(pRExC_state,DEFINEP,0);
5113 goto insert_if_check_paren;
5115 else if (RExC_parse[0] == 'R') {
5118 if (RExC_parse[0] >= '1' && RExC_parse[0] <= '9' ) {
5119 parno = atoi(RExC_parse++);
5120 while (isDIGIT(*RExC_parse))
5122 } else if (RExC_parse[0] == '&') {
5125 sv_dat = reg_scan_name(pRExC_state,
5126 SIZE_ONLY ? REG_RSN_RETURN_NULL : REG_RSN_RETURN_DATA);
5127 parno = sv_dat ? *((I32 *)SvPVX(sv_dat)) : 0;
5129 ret = reganode(pRExC_state,INSUBP,parno);
5130 goto insert_if_check_paren;
5132 else if (RExC_parse[0] >= '1' && RExC_parse[0] <= '9' ) {
5135 parno = atoi(RExC_parse++);
5137 while (isDIGIT(*RExC_parse))
5139 ret = reganode(pRExC_state, GROUPP, parno);
5141 insert_if_check_paren:
5142 if ((c = *nextchar(pRExC_state)) != ')')
5143 vFAIL("Switch condition not recognized");
5145 REGTAIL(pRExC_state, ret, reganode(pRExC_state, IFTHEN, 0));
5146 br = regbranch(pRExC_state, &flags, 1,depth+1);
5148 br = reganode(pRExC_state, LONGJMP, 0);
5150 REGTAIL(pRExC_state, br, reganode(pRExC_state, LONGJMP, 0));
5151 c = *nextchar(pRExC_state);
5156 vFAIL("(?(DEFINE)....) does not allow branches");
5157 lastbr = reganode(pRExC_state, IFTHEN, 0); /* Fake one for optimizer. */
5158 regbranch(pRExC_state, &flags, 1,depth+1);
5159 REGTAIL(pRExC_state, ret, lastbr);
5162 c = *nextchar(pRExC_state);
5167 vFAIL("Switch (?(condition)... contains too many branches");
5168 ender = reg_node(pRExC_state, TAIL);
5169 REGTAIL(pRExC_state, br, ender);
5171 REGTAIL(pRExC_state, lastbr, ender);
5172 REGTAIL(pRExC_state, NEXTOPER(NEXTOPER(lastbr)), ender);
5175 REGTAIL(pRExC_state, ret, ender);
5179 vFAIL2("Unknown switch condition (?(%.2s", RExC_parse);
5183 RExC_parse--; /* for vFAIL to print correctly */
5184 vFAIL("Sequence (? incomplete");
5188 parse_flags: /* (?i) */
5189 while (*RExC_parse && strchr("iogcmsx", *RExC_parse)) {
5190 /* (?g), (?gc) and (?o) are useless here
5191 and must be globally applied -- japhy */
5193 if (*RExC_parse == 'o' || *RExC_parse == 'g') {
5194 if (SIZE_ONLY && ckWARN(WARN_REGEXP)) {
5195 const I32 wflagbit = *RExC_parse == 'o' ? WASTED_O : WASTED_G;
5196 if (! (wastedflags & wflagbit) ) {
5197 wastedflags |= wflagbit;
5200 "Useless (%s%c) - %suse /%c modifier",
5201 flagsp == &negflags ? "?-" : "?",
5203 flagsp == &negflags ? "don't " : "",
5209 else if (*RExC_parse == 'c') {
5210 if (SIZE_ONLY && ckWARN(WARN_REGEXP)) {
5211 if (! (wastedflags & WASTED_C) ) {
5212 wastedflags |= WASTED_GC;
5215 "Useless (%sc) - %suse /gc modifier",
5216 flagsp == &negflags ? "?-" : "?",
5217 flagsp == &negflags ? "don't " : ""
5222 else { pmflag(flagsp, *RExC_parse); }
5226 if (*RExC_parse == '-') {
5228 wastedflags = 0; /* reset so (?g-c) warns twice */
5232 RExC_flags |= posflags;
5233 RExC_flags &= ~negflags;
5234 if (*RExC_parse == ':') {
5240 if (*RExC_parse != ')') {
5242 vFAIL3("Sequence (%.*s...) not recognized", RExC_parse-seqstart, seqstart);
5244 nextchar(pRExC_state);
5254 ret = reganode(pRExC_state, OPEN, parno);
5257 RExC_nestroot = parno;
5258 if (RExC_seen & REG_SEEN_RECURSE) {
5259 DEBUG_OPTIMISE_MORE_r(PerlIO_printf(Perl_debug_log,
5260 "Setting open paren #%"IVdf" to %d\n",
5261 (IV)parno, REG_NODE_NUM(ret)));
5262 RExC_open_parens[parno-1]= ret;
5265 Set_Node_Length(ret, 1); /* MJD */
5266 Set_Node_Offset(ret, RExC_parse); /* MJD */
5273 /* Pick up the branches, linking them together. */
5274 parse_start = RExC_parse; /* MJD */
5275 br = regbranch(pRExC_state, &flags, 1,depth+1);
5276 /* branch_len = (paren != 0); */
5280 if (*RExC_parse == '|') {
5281 if (!SIZE_ONLY && RExC_extralen) {
5282 reginsert(pRExC_state, BRANCHJ, br, depth+1);
5285 reginsert(pRExC_state, BRANCH, br, depth+1);
5286 Set_Node_Length(br, paren != 0);
5287 Set_Node_Offset_To_R(br-RExC_emit_start, parse_start-RExC_start);
5291 RExC_extralen += 1; /* For BRANCHJ-BRANCH. */
5293 else if (paren == ':') {
5294 *flagp |= flags&SIMPLE;
5296 if (is_open) { /* Starts with OPEN. */
5297 REGTAIL(pRExC_state, ret, br); /* OPEN -> first. */
5299 else if (paren != '?') /* Not Conditional */
5301 *flagp |= flags & (SPSTART | HASWIDTH);
5303 while (*RExC_parse == '|') {
5304 if (!SIZE_ONLY && RExC_extralen) {
5305 ender = reganode(pRExC_state, LONGJMP,0);
5306 REGTAIL(pRExC_state, NEXTOPER(NEXTOPER(lastbr)), ender); /* Append to the previous. */
5309 RExC_extralen += 2; /* Account for LONGJMP. */
5310 nextchar(pRExC_state);
5311 br = regbranch(pRExC_state, &flags, 0, depth+1);
5315 REGTAIL(pRExC_state, lastbr, br); /* BRANCH -> BRANCH. */
5319 *flagp |= flags&SPSTART;
5322 if (have_branch || paren != ':') {
5323 /* Make a closing node, and hook it on the end. */
5326 ender = reg_node(pRExC_state, TAIL);
5329 ender = reganode(pRExC_state, CLOSE, parno);
5330 if (!SIZE_ONLY && RExC_seen & REG_SEEN_RECURSE) {
5331 DEBUG_OPTIMISE_MORE_r(PerlIO_printf(Perl_debug_log,
5332 "Setting close paren #%"IVdf" to %d\n",
5333 (IV)parno, REG_NODE_NUM(ender)));
5334 RExC_close_parens[parno-1]= ender;
5335 if (RExC_nestroot == parno)
5338 Set_Node_Offset(ender,RExC_parse+1); /* MJD */
5339 Set_Node_Length(ender,1); /* MJD */
5345 *flagp &= ~HASWIDTH;
5348 ender = reg_node(pRExC_state, SUCCEED);
5351 ender = reg_node(pRExC_state, END);
5353 assert(!RExC_opend); /* there can only be one! */
5358 REGTAIL(pRExC_state, lastbr, ender);
5360 if (have_branch && !SIZE_ONLY) {
5362 RExC_seen |= REG_TOP_LEVEL_BRANCHES;
5364 /* Hook the tails of the branches to the closing node. */
5365 for (br = ret; br; br = regnext(br)) {
5366 const U8 op = PL_regkind[OP(br)];
5368 REGTAIL_STUDY(pRExC_state, NEXTOPER(br), ender);
5370 else if (op == BRANCHJ) {
5371 REGTAIL_STUDY(pRExC_state, NEXTOPER(NEXTOPER(br)), ender);
5379 static const char parens[] = "=!<,>";
5381 if (paren && (p = strchr(parens, paren))) {
5382 U8 node = ((p - parens) % 2) ? UNLESSM : IFMATCH;
5383 int flag = (p - parens) > 1;
5386 node = SUSPEND, flag = 0;
5387 reginsert(pRExC_state, node,ret, depth+1);
5388 Set_Node_Cur_Length(ret);
5389 Set_Node_Offset(ret, parse_start + 1);
5391 REGTAIL_STUDY(pRExC_state, ret, reg_node(pRExC_state, TAIL));
5395 /* Check for proper termination. */
5397 RExC_flags = oregflags;
5398 if (RExC_parse >= RExC_end || *nextchar(pRExC_state) != ')') {
5399 RExC_parse = oregcomp_parse;
5400 vFAIL("Unmatched (");
5403 else if (!paren && RExC_parse < RExC_end) {
5404 if (*RExC_parse == ')') {
5406 vFAIL("Unmatched )");
5409 FAIL("Junk on end of regexp"); /* "Can't happen". */
5417 - regbranch - one alternative of an | operator
5419 * Implements the concatenation operator.
5422 S_regbranch(pTHX_ RExC_state_t *pRExC_state, I32 *flagp, I32 first, U32 depth)
5425 register regnode *ret;
5426 register regnode *chain = NULL;
5427 register regnode *latest;
5428 I32 flags = 0, c = 0;
5429 GET_RE_DEBUG_FLAGS_DECL;
5430 DEBUG_PARSE("brnc");
5434 if (!SIZE_ONLY && RExC_extralen)
5435 ret = reganode(pRExC_state, BRANCHJ,0);
5437 ret = reg_node(pRExC_state, BRANCH);
5438 Set_Node_Length(ret, 1);
5442 if (!first && SIZE_ONLY)
5443 RExC_extralen += 1; /* BRANCHJ */
5445 *flagp = WORST; /* Tentatively. */
5448 nextchar(pRExC_state);
5449 while (RExC_parse < RExC_end && *RExC_parse != '|' && *RExC_parse != ')') {
5451 latest = regpiece(pRExC_state, &flags,depth+1);
5452 if (latest == NULL) {
5453 if (flags & TRYAGAIN)
5457 else if (ret == NULL)
5459 *flagp |= flags&HASWIDTH;
5460 if (chain == NULL) /* First piece. */
5461 *flagp |= flags&SPSTART;
5464 REGTAIL(pRExC_state, chain, latest);
5469 if (chain == NULL) { /* Loop ran zero times. */
5470 chain = reg_node(pRExC_state, NOTHING);
5475 *flagp |= flags&SIMPLE;
5482 - regpiece - something followed by possible [*+?]
5484 * Note that the branching code sequences used for ? and the general cases
5485 * of * and + are somewhat optimized: they use the same NOTHING node as
5486 * both the endmarker for their branch list and the body of the last branch.
5487 * It might seem that this node could be dispensed with entirely, but the
5488 * endmarker role is not redundant.
5491 S_regpiece(pTHX_ RExC_state_t *pRExC_state, I32 *flagp, U32 depth)
5494 register regnode *ret;
5496 register char *next;
5498 const char * const origparse = RExC_parse;
5500 I32 max = REG_INFTY;
5502 const char *maxpos = NULL;
5503 GET_RE_DEBUG_FLAGS_DECL;
5504 DEBUG_PARSE("piec");
5506 ret = regatom(pRExC_state, &flags,depth+1);
5508 if (flags & TRYAGAIN)
5515 if (op == '{' && regcurly(RExC_parse)) {
5517 parse_start = RExC_parse; /* MJD */
5518 next = RExC_parse + 1;
5519 while (isDIGIT(*next) || *next == ',') {
5528 if (*next == '}') { /* got one */
5532 min = atoi(RExC_parse);
5536 maxpos = RExC_parse;
5538 if (!max && *maxpos != '0')
5539 max = REG_INFTY; /* meaning "infinity" */
5540 else if (max >= REG_INFTY)
5541 vFAIL2("Quantifier in {,} bigger than %d", REG_INFTY - 1);
5543 nextchar(pRExC_state);
5546 if ((flags&SIMPLE)) {
5547 RExC_naughty += 2 + RExC_naughty / 2;
5548 reginsert(pRExC_state, CURLY, ret, depth+1);
5549 Set_Node_Offset(ret, parse_start+1); /* MJD */
5550 Set_Node_Cur_Length(ret);
5553 regnode * const w = reg_node(pRExC_state, WHILEM);
5556 REGTAIL(pRExC_state, ret, w);
5557 if (!SIZE_ONLY && RExC_extralen) {
5558 reginsert(pRExC_state, LONGJMP,ret, depth+1);
5559 reginsert(pRExC_state, NOTHING,ret, depth+1);
5560 NEXT_OFF(ret) = 3; /* Go over LONGJMP. */
5562 reginsert(pRExC_state, CURLYX,ret, depth+1);
5564 Set_Node_Offset(ret, parse_start+1);
5565 Set_Node_Length(ret,
5566 op == '{' ? (RExC_parse - parse_start) : 1);
5568 if (!SIZE_ONLY && RExC_extralen)
5569 NEXT_OFF(ret) = 3; /* Go over NOTHING to LONGJMP. */
5570 REGTAIL(pRExC_state, ret, reg_node(pRExC_state, NOTHING));
5572 RExC_whilem_seen++, RExC_extralen += 3;
5573 RExC_naughty += 4 + RExC_naughty; /* compound interest */
5581 if (max && max < min)
5582 vFAIL("Can't do {n,m} with n > m");
5584 ARG1_SET(ret, (U16)min);
5585 ARG2_SET(ret, (U16)max);
5597 #if 0 /* Now runtime fix should be reliable. */
5599 /* if this is reinstated, don't forget to put this back into perldiag:
5601 =item Regexp *+ operand could be empty at {#} in regex m/%s/
5603 (F) The part of the regexp subject to either the * or + quantifier
5604 could match an empty string. The {#} shows in the regular
5605 expression about where the problem was discovered.
5609 if (!(flags&HASWIDTH) && op != '?')
5610 vFAIL("Regexp *+ operand could be empty");
5613 parse_start = RExC_parse;
5614 nextchar(pRExC_state);
5616 *flagp = (op != '+') ? (WORST|SPSTART|HASWIDTH) : (WORST|HASWIDTH);
5618 if (op == '*' && (flags&SIMPLE)) {
5619 reginsert(pRExC_state, STAR, ret, depth+1);
5623 else if (op == '*') {
5627 else if (op == '+' && (flags&SIMPLE)) {
5628 reginsert(pRExC_state, PLUS, ret, depth+1);
5632 else if (op == '+') {
5636 else if (op == '?') {
5641 if (!SIZE_ONLY && !(flags&HASWIDTH) && max > REG_INFTY/3 && ckWARN(WARN_REGEXP)) {
5643 "%.*s matches null string many times",
5644 (int)(RExC_parse >= origparse ? RExC_parse - origparse : 0),
5648 if (RExC_parse < RExC_end && *RExC_parse == '?') {
5649 nextchar(pRExC_state);
5650 reginsert(pRExC_state, MINMOD, ret, depth+1);
5651 REGTAIL(pRExC_state, ret, ret + NODE_STEP_REGNODE);
5653 #ifndef REG_ALLOW_MINMOD_SUSPEND
5656 if (RExC_parse < RExC_end && *RExC_parse == '+') {
5658 nextchar(pRExC_state);
5659 ender = reg_node(pRExC_state, SUCCEED);
5660 REGTAIL(pRExC_state, ret, ender);
5661 reginsert(pRExC_state, SUSPEND, ret, depth+1);
5663 ender = reg_node(pRExC_state, TAIL);
5664 REGTAIL(pRExC_state, ret, ender);
5668 if (RExC_parse < RExC_end && ISMULT2(RExC_parse)) {
5670 vFAIL("Nested quantifiers");
5677 /* reg_namedseq(pRExC_state,UVp)
5679 This is expected to be called by a parser routine that has
5680 recognized'\N' and needs to handle the rest. RExC_parse is
5681 expected to point at the first char following the N at the time
5684 If valuep is non-null then it is assumed that we are parsing inside
5685 of a charclass definition and the first codepoint in the resolved
5686 string is returned via *valuep and the routine will return NULL.
5687 In this mode if a multichar string is returned from the charnames
5688 handler a warning will be issued, and only the first char in the
5689 sequence will be examined. If the string returned is zero length
5690 then the value of *valuep is undefined and NON-NULL will
5691 be returned to indicate failure. (This will NOT be a valid pointer
5694 If value is null then it is assumed that we are parsing normal text
5695 and inserts a new EXACT node into the program containing the resolved
5696 string and returns a pointer to the new node. If the string is
5697 zerolength a NOTHING node is emitted.
5699 On success RExC_parse is set to the char following the endbrace.
5700 Parsing failures will generate a fatal errorvia vFAIL(...)
5702 NOTE: We cache all results from the charnames handler locally in
5703 the RExC_charnames hash (created on first use) to prevent a charnames
5704 handler from playing silly-buggers and returning a short string and
5705 then a long string for a given pattern. Since the regexp program
5706 size is calculated during an initial parse this would result
5707 in a buffer overrun so we cache to prevent the charname result from
5708 changing during the course of the parse.
5712 S_reg_namedseq(pTHX_ RExC_state_t *pRExC_state, UV *valuep)
5714 char * name; /* start of the content of the name */
5715 char * endbrace; /* endbrace following the name */
5718 STRLEN len; /* this has various purposes throughout the code */
5719 bool cached = 0; /* if this is true then we shouldn't refcount dev sv_str */
5720 regnode *ret = NULL;
5722 if (*RExC_parse != '{') {
5723 vFAIL("Missing braces on \\N{}");
5725 name = RExC_parse+1;
5726 endbrace = strchr(RExC_parse, '}');
5729 vFAIL("Missing right brace on \\N{}");
5731 RExC_parse = endbrace + 1;
5734 /* RExC_parse points at the beginning brace,
5735 endbrace points at the last */
5736 if ( name[0]=='U' && name[1]=='+' ) {
5737 /* its a "unicode hex" notation {U+89AB} */
5738 I32 fl = PERL_SCAN_ALLOW_UNDERSCORES
5739 | PERL_SCAN_DISALLOW_PREFIX
5740 | (SIZE_ONLY ? PERL_SCAN_SILENT_ILLDIGIT : 0);
5742 len = (STRLEN)(endbrace - name - 2);
5743 cp = grok_hex(name + 2, &len, &fl, NULL);
5744 if ( len != (STRLEN)(endbrace - name - 2) ) {
5753 sv_str= Perl_newSVpvf_nocontext("%c",(int)cp);
5755 /* fetch the charnames handler for this scope */
5756 HV * const table = GvHV(PL_hintgv);
5758 hv_fetchs(table, "charnames", FALSE) :
5760 SV *cv= cvp ? *cvp : NULL;
5763 /* create an SV with the name as argument */
5764 sv_name = newSVpvn(name, endbrace - name);
5766 if (!table || !(PL_hints & HINT_LOCALIZE_HH)) {
5767 vFAIL2("Constant(\\N{%s}) unknown: "
5768 "(possibly a missing \"use charnames ...\")",
5771 if (!cvp || !SvOK(*cvp)) { /* when $^H{charnames} = undef; */
5772 vFAIL2("Constant(\\N{%s}): "
5773 "$^H{charnames} is not defined",SvPVX(sv_name));
5778 if (!RExC_charnames) {
5779 /* make sure our cache is allocated */
5780 RExC_charnames = newHV();
5781 sv_2mortal((SV*)RExC_charnames);
5783 /* see if we have looked this one up before */
5784 he_str = hv_fetch_ent( RExC_charnames, sv_name, 0, 0 );
5786 sv_str = HeVAL(he_str);
5799 count= call_sv(cv, G_SCALAR);
5801 if (count == 1) { /* XXXX is this right? dmq */
5803 SvREFCNT_inc_simple_void(sv_str);
5811 if ( !sv_str || !SvOK(sv_str) ) {
5812 vFAIL2("Constant(\\N{%s}): Call to &{$^H{charnames}} "
5813 "did not return a defined value",SvPVX(sv_name));
5815 if (hv_store_ent( RExC_charnames, sv_name, sv_str, 0))
5820 char *p = SvPV(sv_str, len);
5823 if ( SvUTF8(sv_str) ) {
5824 *valuep = utf8_to_uvchr((U8*)p, &numlen);
5828 We have to turn on utf8 for high bit chars otherwise
5829 we get failures with
5831 "ss" =~ /[\N{LATIN SMALL LETTER SHARP S}]/i
5832 "SS" =~ /[\N{LATIN SMALL LETTER SHARP S}]/i
5834 This is different from what \x{} would do with the same
5835 codepoint, where the condition is > 0xFF.
5842 /* warn if we havent used the whole string? */
5844 if (numlen<len && SIZE_ONLY && ckWARN(WARN_REGEXP)) {
5846 "Ignoring excess chars from \\N{%s} in character class",
5850 } else if (SIZE_ONLY && ckWARN(WARN_REGEXP)) {
5852 "Ignoring zero length \\N{%s} in character class",
5857 SvREFCNT_dec(sv_name);
5859 SvREFCNT_dec(sv_str);
5860 return len ? NULL : (regnode *)&len;
5861 } else if(SvCUR(sv_str)) {
5866 char * parse_start = name-3; /* needed for the offsets */
5867 GET_RE_DEBUG_FLAGS_DECL; /* needed for the offsets */
5869 ret = reg_node(pRExC_state,
5870 (U8)(FOLD ? (LOC ? EXACTFL : EXACTF) : EXACT));
5873 if ( RExC_utf8 && !SvUTF8(sv_str) ) {
5874 sv_utf8_upgrade(sv_str);
5875 } else if ( !RExC_utf8 && SvUTF8(sv_str) ) {
5879 p = SvPV(sv_str, len);
5881 /* len is the length written, charlen is the size the char read */
5882 for ( len = 0; p < pend; p += charlen ) {
5884 UV uvc = utf8_to_uvchr((U8*)p, &charlen);
5886 STRLEN foldlen,numlen;
5887 U8 tmpbuf[UTF8_MAXBYTES_CASE+1], *foldbuf;
5888 uvc = toFOLD_uni(uvc, tmpbuf, &foldlen);
5889 /* Emit all the Unicode characters. */
5891 for (foldbuf = tmpbuf;
5895 uvc = utf8_to_uvchr(foldbuf, &numlen);
5897 const STRLEN unilen = reguni(pRExC_state, uvc, s);
5900 /* In EBCDIC the numlen
5901 * and unilen can differ. */
5903 if (numlen >= foldlen)
5907 break; /* "Can't happen." */
5910 const STRLEN unilen = reguni(pRExC_state, uvc, s);
5922 RExC_size += STR_SZ(len);
5925 RExC_emit += STR_SZ(len);
5927 Set_Node_Cur_Length(ret); /* MJD */
5929 nextchar(pRExC_state);
5931 ret = reg_node(pRExC_state,NOTHING);
5934 SvREFCNT_dec(sv_str);
5937 SvREFCNT_dec(sv_name);
5947 * It returns the code point in utf8 for the value in *encp.
5948 * value: a code value in the source encoding
5949 * encp: a pointer to an Encode object
5951 * If the result from Encode is not a single character,
5952 * it returns U+FFFD (Replacement character) and sets *encp to NULL.
5955 S_reg_recode(pTHX_ const char value, SV **encp)
5958 SV * const sv = sv_2mortal(newSVpvn(&value, numlen));
5959 const char * const s = encp && *encp ? sv_recode_to_utf8(sv, *encp)
5961 const STRLEN newlen = SvCUR(sv);
5962 UV uv = UNICODE_REPLACEMENT;
5966 ? utf8n_to_uvchr((U8*)s, newlen, &numlen, UTF8_ALLOW_DEFAULT)
5969 if (!newlen || numlen != newlen) {
5970 uv = UNICODE_REPLACEMENT;
5979 - regatom - the lowest level
5981 * Optimization: gobbles an entire sequence of ordinary characters so that
5982 * it can turn them into a single node, which is smaller to store and
5983 * faster to run. Backslashed characters are exceptions, each becoming a
5984 * separate node; the code is simpler that way and it's not worth fixing.
5986 * [Yes, it is worth fixing, some scripts can run twice the speed.]
5987 * [It looks like its ok, as in S_study_chunk we merge adjacent EXACT nodes]
5990 S_regatom(pTHX_ RExC_state_t *pRExC_state, I32 *flagp, U32 depth)
5993 register regnode *ret = NULL;
5995 char *parse_start = RExC_parse;
5996 GET_RE_DEBUG_FLAGS_DECL;
5997 DEBUG_PARSE("atom");
5998 *flagp = WORST; /* Tentatively. */
6001 switch (*RExC_parse) {
6003 RExC_seen_zerolen++;
6004 nextchar(pRExC_state);
6005 if (RExC_flags & PMf_MULTILINE)
6006 ret = reg_node(pRExC_state, MBOL);
6007 else if (RExC_flags & PMf_SINGLELINE)
6008 ret = reg_node(pRExC_state, SBOL);
6010 ret = reg_node(pRExC_state, BOL);
6011 Set_Node_Length(ret, 1); /* MJD */
6014 nextchar(pRExC_state);
6016 RExC_seen_zerolen++;
6017 if (RExC_flags & PMf_MULTILINE)
6018 ret = reg_node(pRExC_state, MEOL);
6019 else if (RExC_flags & PMf_SINGLELINE)
6020 ret = reg_node(pRExC_state, SEOL);
6022 ret = reg_node(pRExC_state, EOL);
6023 Set_Node_Length(ret, 1); /* MJD */
6026 nextchar(pRExC_state);
6027 if (RExC_flags & PMf_SINGLELINE)
6028 ret = reg_node(pRExC_state, SANY);
6030 ret = reg_node(pRExC_state, REG_ANY);
6031 *flagp |= HASWIDTH|SIMPLE;
6033 Set_Node_Length(ret, 1); /* MJD */
6037 char * const oregcomp_parse = ++RExC_parse;
6038 ret = regclass(pRExC_state,depth+1);
6039 if (*RExC_parse != ']') {
6040 RExC_parse = oregcomp_parse;
6041 vFAIL("Unmatched [");
6043 nextchar(pRExC_state);
6044 *flagp |= HASWIDTH|SIMPLE;
6045 Set_Node_Length(ret, RExC_parse - oregcomp_parse + 1); /* MJD */
6049 nextchar(pRExC_state);
6050 ret = reg(pRExC_state, 1, &flags,depth+1);
6052 if (flags & TRYAGAIN) {
6053 if (RExC_parse == RExC_end) {
6054 /* Make parent create an empty node if needed. */
6062 *flagp |= flags&(HASWIDTH|SPSTART|SIMPLE);
6066 if (flags & TRYAGAIN) {
6070 vFAIL("Internal urp");
6071 /* Supposed to be caught earlier. */
6074 if (!regcurly(RExC_parse)) {
6083 vFAIL("Quantifier follows nothing");
6086 switch (*++RExC_parse) {
6088 RExC_seen_zerolen++;
6089 ret = reg_node(pRExC_state, SBOL);
6091 nextchar(pRExC_state);
6092 Set_Node_Length(ret, 2); /* MJD */
6095 ret = reg_node(pRExC_state, GPOS);
6096 RExC_seen |= REG_SEEN_GPOS;
6098 nextchar(pRExC_state);
6099 Set_Node_Length(ret, 2); /* MJD */
6102 ret = reg_node(pRExC_state, SEOL);
6104 RExC_seen_zerolen++; /* Do not optimize RE away */
6105 nextchar(pRExC_state);
6108 ret = reg_node(pRExC_state, EOS);
6110 RExC_seen_zerolen++; /* Do not optimize RE away */
6111 nextchar(pRExC_state);
6112 Set_Node_Length(ret, 2); /* MJD */
6115 ret = reg_node(pRExC_state, CANY);
6116 RExC_seen |= REG_SEEN_CANY;
6117 *flagp |= HASWIDTH|SIMPLE;
6118 nextchar(pRExC_state);
6119 Set_Node_Length(ret, 2); /* MJD */
6122 ret = reg_node(pRExC_state, CLUMP);
6124 nextchar(pRExC_state);
6125 Set_Node_Length(ret, 2); /* MJD */
6128 ret = reg_node(pRExC_state, (U8)(LOC ? ALNUML : ALNUM));
6129 *flagp |= HASWIDTH|SIMPLE;
6130 nextchar(pRExC_state);
6131 Set_Node_Length(ret, 2); /* MJD */
6134 ret = reg_node(pRExC_state, (U8)(LOC ? NALNUML : NALNUM));
6135 *flagp |= HASWIDTH|SIMPLE;
6136 nextchar(pRExC_state);
6137 Set_Node_Length(ret, 2); /* MJD */
6140 RExC_seen_zerolen++;
6141 RExC_seen |= REG_SEEN_LOOKBEHIND;
6142 ret = reg_node(pRExC_state, (U8)(LOC ? BOUNDL : BOUND));
6144 nextchar(pRExC_state);
6145 Set_Node_Length(ret, 2); /* MJD */
6148 RExC_seen_zerolen++;
6149 RExC_seen |= REG_SEEN_LOOKBEHIND;
6150 ret = reg_node(pRExC_state, (U8)(LOC ? NBOUNDL : NBOUND));
6152 nextchar(pRExC_state);
6153 Set_Node_Length(ret, 2); /* MJD */
6156 ret = reg_node(pRExC_state, (U8)(LOC ? SPACEL : SPACE));
6157 *flagp |= HASWIDTH|SIMPLE;
6158 nextchar(pRExC_state);
6159 Set_Node_Length(ret, 2); /* MJD */
6162 ret = reg_node(pRExC_state, (U8)(LOC ? NSPACEL : NSPACE));
6163 *flagp |= HASWIDTH|SIMPLE;
6164 nextchar(pRExC_state);
6165 Set_Node_Length(ret, 2); /* MJD */
6168 ret = reg_node(pRExC_state, DIGIT);
6169 *flagp |= HASWIDTH|SIMPLE;
6170 nextchar(pRExC_state);
6171 Set_Node_Length(ret, 2); /* MJD */
6174 ret = reg_node(pRExC_state, NDIGIT);
6175 *flagp |= HASWIDTH|SIMPLE;
6176 nextchar(pRExC_state);
6177 Set_Node_Length(ret, 2); /* MJD */
6182 char* const oldregxend = RExC_end;
6183 char* parse_start = RExC_parse - 2;
6185 if (RExC_parse[1] == '{') {
6186 /* a lovely hack--pretend we saw [\pX] instead */
6187 RExC_end = strchr(RExC_parse, '}');
6189 const U8 c = (U8)*RExC_parse;
6191 RExC_end = oldregxend;
6192 vFAIL2("Missing right brace on \\%c{}", c);
6197 RExC_end = RExC_parse + 2;
6198 if (RExC_end > oldregxend)
6199 RExC_end = oldregxend;
6203 ret = regclass(pRExC_state,depth+1);
6205 RExC_end = oldregxend;
6208 Set_Node_Offset(ret, parse_start + 2);
6209 Set_Node_Cur_Length(ret);
6210 nextchar(pRExC_state);
6211 *flagp |= HASWIDTH|SIMPLE;
6215 /* Handle \N{NAME} here and not below because it can be
6216 multicharacter. join_exact() will join them up later on.
6217 Also this makes sure that things like /\N{BLAH}+/ and
6218 \N{BLAH} being multi char Just Happen. dmq*/
6220 ret= reg_namedseq(pRExC_state, NULL);
6222 case 'k': /* Handle \k<NAME> and \k'NAME' */
6224 char ch= RExC_parse[1];
6225 if (ch != '<' && ch != '\'') {
6227 vWARN( RExC_parse + 1,
6228 "Possible broken named back reference treated as literal k");
6232 char* name_start = (RExC_parse += 2);
6234 SV *sv_dat = reg_scan_name(pRExC_state,
6235 SIZE_ONLY ? REG_RSN_RETURN_NULL : REG_RSN_RETURN_DATA);
6236 ch= (ch == '<') ? '>' : '\'';
6238 if (RExC_parse == name_start || *RExC_parse != ch)
6239 vFAIL2("Sequence \\k%c... not terminated",
6240 (ch == '>' ? '<' : ch));
6243 ret = reganode(pRExC_state,
6244 (U8)(FOLD ? (LOC ? NREFFL : NREFF) : NREF),
6250 num = add_data( pRExC_state, 1, "S" );
6252 RExC_rx->data->data[num]=(void*)sv_dat;
6253 SvREFCNT_inc(sv_dat);
6255 /* override incorrect value set in reganode MJD */
6256 Set_Node_Offset(ret, parse_start+1);
6257 Set_Node_Cur_Length(ret); /* MJD */
6258 nextchar(pRExC_state);
6273 case '1': case '2': case '3': case '4':
6274 case '5': case '6': case '7': case '8': case '9':
6276 const I32 num = atoi(RExC_parse);
6278 if (num > 9 && num >= RExC_npar)
6281 char * const parse_start = RExC_parse - 1; /* MJD */
6282 while (isDIGIT(*RExC_parse))
6285 if (!SIZE_ONLY && num > (I32)RExC_rx->nparens)
6286 vFAIL("Reference to nonexistent group");
6288 ret = reganode(pRExC_state,
6289 (U8)(FOLD ? (LOC ? REFFL : REFF) : REF),
6293 /* override incorrect value set in reganode MJD */
6294 Set_Node_Offset(ret, parse_start+1);
6295 Set_Node_Cur_Length(ret); /* MJD */
6297 nextchar(pRExC_state);
6302 if (RExC_parse >= RExC_end)
6303 FAIL("Trailing \\");
6306 /* Do not generate "unrecognized" warnings here, we fall
6307 back into the quick-grab loop below */
6314 if (RExC_flags & PMf_EXTENDED) {
6315 while (RExC_parse < RExC_end && *RExC_parse != '\n')
6317 if (RExC_parse < RExC_end)
6323 register STRLEN len;
6328 U8 tmpbuf[UTF8_MAXBYTES_CASE+1], *foldbuf;
6330 parse_start = RExC_parse - 1;
6336 ret = reg_node(pRExC_state,
6337 (U8)(FOLD ? (LOC ? EXACTFL : EXACTF) : EXACT));
6339 for (len = 0, p = RExC_parse - 1;
6340 len < 127 && p < RExC_end;
6343 char * const oldp = p;
6345 if (RExC_flags & PMf_EXTENDED)
6346 p = regwhite(p, RExC_end);
6394 ender = ASCII_TO_NATIVE('\033');
6398 ender = ASCII_TO_NATIVE('\007');
6403 char* const e = strchr(p, '}');
6407 vFAIL("Missing right brace on \\x{}");
6410 I32 flags = PERL_SCAN_ALLOW_UNDERSCORES
6411 | PERL_SCAN_DISALLOW_PREFIX;
6412 STRLEN numlen = e - p - 1;
6413 ender = grok_hex(p + 1, &numlen, &flags, NULL);
6420 I32 flags = PERL_SCAN_DISALLOW_PREFIX;
6422 ender = grok_hex(p, &numlen, &flags, NULL);
6425 if (PL_encoding && ender < 0x100)
6426 goto recode_encoding;
6430 ender = UCHARAT(p++);
6431 ender = toCTRL(ender);
6433 case '0': case '1': case '2': case '3':case '4':
6434 case '5': case '6': case '7': case '8':case '9':
6436 (isDIGIT(p[1]) && atoi(p) >= RExC_npar) ) {
6439 ender = grok_oct(p, &numlen, &flags, NULL);
6446 if (PL_encoding && ender < 0x100)
6447 goto recode_encoding;
6451 SV* enc = PL_encoding;
6452 ender = reg_recode((const char)(U8)ender, &enc);
6453 if (!enc && SIZE_ONLY && ckWARN(WARN_REGEXP))
6454 vWARN(p, "Invalid escape in the specified encoding");
6460 FAIL("Trailing \\");
6463 if (!SIZE_ONLY&& isALPHA(*p) && ckWARN(WARN_REGEXP))
6464 vWARN2(p + 1, "Unrecognized escape \\%c passed through", UCHARAT(p));
6465 goto normal_default;
6470 if (UTF8_IS_START(*p) && UTF) {
6472 ender = utf8n_to_uvchr((U8*)p, RExC_end - p,
6473 &numlen, UTF8_ALLOW_DEFAULT);
6480 if (RExC_flags & PMf_EXTENDED)
6481 p = regwhite(p, RExC_end);
6483 /* Prime the casefolded buffer. */
6484 ender = toFOLD_uni(ender, tmpbuf, &foldlen);
6486 if (ISMULT2(p)) { /* Back off on ?+*. */
6491 /* Emit all the Unicode characters. */
6493 for (foldbuf = tmpbuf;
6495 foldlen -= numlen) {
6496 ender = utf8_to_uvchr(foldbuf, &numlen);
6498 const STRLEN unilen = reguni(pRExC_state, ender, s);
6501 /* In EBCDIC the numlen
6502 * and unilen can differ. */
6504 if (numlen >= foldlen)
6508 break; /* "Can't happen." */
6512 const STRLEN unilen = reguni(pRExC_state, ender, s);
6521 REGC((char)ender, s++);
6527 /* Emit all the Unicode characters. */
6529 for (foldbuf = tmpbuf;
6531 foldlen -= numlen) {
6532 ender = utf8_to_uvchr(foldbuf, &numlen);
6534 const STRLEN unilen = reguni(pRExC_state, ender, s);
6537 /* In EBCDIC the numlen
6538 * and unilen can differ. */
6540 if (numlen >= foldlen)
6548 const STRLEN unilen = reguni(pRExC_state, ender, s);
6557 REGC((char)ender, s++);
6561 Set_Node_Cur_Length(ret); /* MJD */
6562 nextchar(pRExC_state);
6564 /* len is STRLEN which is unsigned, need to copy to signed */
6567 vFAIL("Internal disaster");
6571 if (len == 1 && UNI_IS_INVARIANT(ender))
6575 RExC_size += STR_SZ(len);
6578 RExC_emit += STR_SZ(len);
6588 S_regwhite(char *p, const char *e)
6593 else if (*p == '#') {
6596 } while (p < e && *p != '\n');
6604 /* Parse POSIX character classes: [[:foo:]], [[=foo=]], [[.foo.]].
6605 Character classes ([:foo:]) can also be negated ([:^foo:]).
6606 Returns a named class id (ANYOF_XXX) if successful, -1 otherwise.
6607 Equivalence classes ([=foo=]) and composites ([.foo.]) are parsed,
6608 but trigger failures because they are currently unimplemented. */
6610 #define POSIXCC_DONE(c) ((c) == ':')
6611 #define POSIXCC_NOTYET(c) ((c) == '=' || (c) == '.')
6612 #define POSIXCC(c) (POSIXCC_DONE(c) || POSIXCC_NOTYET(c))
6615 S_regpposixcc(pTHX_ RExC_state_t *pRExC_state, I32 value)
6618 I32 namedclass = OOB_NAMEDCLASS;
6620 if (value == '[' && RExC_parse + 1 < RExC_end &&
6621 /* I smell either [: or [= or [. -- POSIX has been here, right? */
6622 POSIXCC(UCHARAT(RExC_parse))) {
6623 const char c = UCHARAT(RExC_parse);
6624 char* const s = RExC_parse++;
6626 while (RExC_parse < RExC_end && UCHARAT(RExC_parse) != c)
6628 if (RExC_parse == RExC_end)
6629 /* Grandfather lone [:, [=, [. */
6632 const char* const t = RExC_parse++; /* skip over the c */
6635 if (UCHARAT(RExC_parse) == ']') {
6636 const char *posixcc = s + 1;
6637 RExC_parse++; /* skip over the ending ] */
6640 const I32 complement = *posixcc == '^' ? *posixcc++ : 0;
6641 const I32 skip = t - posixcc;
6643 /* Initially switch on the length of the name. */
6646 if (memEQ(posixcc, "word", 4)) /* this is not POSIX, this is the Perl \w */
6647 namedclass = complement ? ANYOF_NALNUM : ANYOF_ALNUM;
6650 /* Names all of length 5. */
6651 /* alnum alpha ascii blank cntrl digit graph lower
6652 print punct space upper */
6653 /* Offset 4 gives the best switch position. */
6654 switch (posixcc[4]) {
6656 if (memEQ(posixcc, "alph", 4)) /* alpha */
6657 namedclass = complement ? ANYOF_NALPHA : ANYOF_ALPHA;
6660 if (memEQ(posixcc, "spac", 4)) /* space */
6661 namedclass = complement ? ANYOF_NPSXSPC : ANYOF_PSXSPC;
6664 if (memEQ(posixcc, "grap", 4)) /* graph */
6665 namedclass = complement ? ANYOF_NGRAPH : ANYOF_GRAPH;
6668 if (memEQ(posixcc, "asci", 4)) /* ascii */
6669 namedclass = complement ? ANYOF_NASCII : ANYOF_ASCII;
6672 if (memEQ(posixcc, "blan", 4)) /* blank */
6673 namedclass = complement ? ANYOF_NBLANK : ANYOF_BLANK;
6676 if (memEQ(posixcc, "cntr", 4)) /* cntrl */
6677 namedclass = complement ? ANYOF_NCNTRL : ANYOF_CNTRL;
6680 if (memEQ(posixcc, "alnu", 4)) /* alnum */
6681 namedclass = complement ? ANYOF_NALNUMC : ANYOF_ALNUMC;
6684 if (memEQ(posixcc, "lowe", 4)) /* lower */
6685 namedclass = complement ? ANYOF_NLOWER : ANYOF_LOWER;
6686 else if (memEQ(posixcc, "uppe", 4)) /* upper */
6687 namedclass = complement ? ANYOF_NUPPER : ANYOF_UPPER;
6690 if (memEQ(posixcc, "digi", 4)) /* digit */
6691 namedclass = complement ? ANYOF_NDIGIT : ANYOF_DIGIT;
6692 else if (memEQ(posixcc, "prin", 4)) /* print */
6693 namedclass = complement ? ANYOF_NPRINT : ANYOF_PRINT;
6694 else if (memEQ(posixcc, "punc", 4)) /* punct */
6695 namedclass = complement ? ANYOF_NPUNCT : ANYOF_PUNCT;
6700 if (memEQ(posixcc, "xdigit", 6))
6701 namedclass = complement ? ANYOF_NXDIGIT : ANYOF_XDIGIT;
6705 if (namedclass == OOB_NAMEDCLASS)
6706 Simple_vFAIL3("POSIX class [:%.*s:] unknown",
6708 assert (posixcc[skip] == ':');
6709 assert (posixcc[skip+1] == ']');
6710 } else if (!SIZE_ONLY) {
6711 /* [[=foo=]] and [[.foo.]] are still future. */
6713 /* adjust RExC_parse so the warning shows after
6715 while (UCHARAT(RExC_parse) && UCHARAT(RExC_parse) != ']')
6717 Simple_vFAIL3("POSIX syntax [%c %c] is reserved for future extensions", c, c);
6720 /* Maternal grandfather:
6721 * "[:" ending in ":" but not in ":]" */
6731 S_checkposixcc(pTHX_ RExC_state_t *pRExC_state)
6734 if (POSIXCC(UCHARAT(RExC_parse))) {
6735 const char *s = RExC_parse;
6736 const char c = *s++;
6740 if (*s && c == *s && s[1] == ']') {
6741 if (ckWARN(WARN_REGEXP))
6743 "POSIX syntax [%c %c] belongs inside character classes",
6746 /* [[=foo=]] and [[.foo.]] are still future. */
6747 if (POSIXCC_NOTYET(c)) {
6748 /* adjust RExC_parse so the error shows after
6750 while (UCHARAT(RExC_parse) && UCHARAT(RExC_parse++) != ']')
6752 Simple_vFAIL3("POSIX syntax [%c %c] is reserved for future extensions", c, c);
6760 parse a class specification and produce either an ANYOF node that
6761 matches the pattern. If the pattern matches a single char only and
6762 that char is < 256 then we produce an EXACT node instead.
6765 S_regclass(pTHX_ RExC_state_t *pRExC_state, U32 depth)
6768 register UV value = 0;
6769 register UV nextvalue;
6770 register IV prevvalue = OOB_UNICODE;
6771 register IV range = 0;
6772 register regnode *ret;
6775 char *rangebegin = NULL;
6776 bool need_class = 0;
6779 bool optimize_invert = TRUE;
6780 AV* unicode_alternate = NULL;
6782 UV literal_endpoint = 0;
6784 UV stored = 0; /* number of chars stored in the class */
6786 regnode * const orig_emit = RExC_emit; /* Save the original RExC_emit in
6787 case we need to change the emitted regop to an EXACT. */
6788 const char * orig_parse = RExC_parse;
6789 GET_RE_DEBUG_FLAGS_DECL;
6791 PERL_UNUSED_ARG(depth);
6794 DEBUG_PARSE("clas");
6796 /* Assume we are going to generate an ANYOF node. */
6797 ret = reganode(pRExC_state, ANYOF, 0);
6800 ANYOF_FLAGS(ret) = 0;
6802 if (UCHARAT(RExC_parse) == '^') { /* Complement of range. */
6806 ANYOF_FLAGS(ret) |= ANYOF_INVERT;
6810 RExC_size += ANYOF_SKIP;
6811 listsv = &PL_sv_undef; /* For code scanners: listsv always non-NULL. */
6814 RExC_emit += ANYOF_SKIP;
6816 ANYOF_FLAGS(ret) |= ANYOF_FOLD;
6818 ANYOF_FLAGS(ret) |= ANYOF_LOCALE;
6819 ANYOF_BITMAP_ZERO(ret);
6820 listsv = newSVpvs("# comment\n");
6823 nextvalue = RExC_parse < RExC_end ? UCHARAT(RExC_parse) : 0;
6825 if (!SIZE_ONLY && POSIXCC(nextvalue))
6826 checkposixcc(pRExC_state);
6828 /* allow 1st char to be ] (allowing it to be - is dealt with later) */
6829 if (UCHARAT(RExC_parse) == ']')
6833 while (RExC_parse < RExC_end && UCHARAT(RExC_parse) != ']') {
6837 namedclass = OOB_NAMEDCLASS; /* initialize as illegal */
6840 rangebegin = RExC_parse;
6842 value = utf8n_to_uvchr((U8*)RExC_parse,
6843 RExC_end - RExC_parse,
6844 &numlen, UTF8_ALLOW_DEFAULT);
6845 RExC_parse += numlen;
6848 value = UCHARAT(RExC_parse++);
6850 nextvalue = RExC_parse < RExC_end ? UCHARAT(RExC_parse) : 0;
6851 if (value == '[' && POSIXCC(nextvalue))
6852 namedclass = regpposixcc(pRExC_state, value);
6853 else if (value == '\\') {
6855 value = utf8n_to_uvchr((U8*)RExC_parse,
6856 RExC_end - RExC_parse,
6857 &numlen, UTF8_ALLOW_DEFAULT);
6858 RExC_parse += numlen;
6861 value = UCHARAT(RExC_parse++);
6862 /* Some compilers cannot handle switching on 64-bit integer
6863 * values, therefore value cannot be an UV. Yes, this will
6864 * be a problem later if we want switch on Unicode.
6865 * A similar issue a little bit later when switching on
6866 * namedclass. --jhi */
6867 switch ((I32)value) {
6868 case 'w': namedclass = ANYOF_ALNUM; break;
6869 case 'W': namedclass = ANYOF_NALNUM; break;
6870 case 's': namedclass = ANYOF_SPACE; break;
6871 case 'S': namedclass = ANYOF_NSPACE; break;
6872 case 'd': namedclass = ANYOF_DIGIT; break;
6873 case 'D': namedclass = ANYOF_NDIGIT; break;
6874 case 'N': /* Handle \N{NAME} in class */
6876 /* We only pay attention to the first char of
6877 multichar strings being returned. I kinda wonder
6878 if this makes sense as it does change the behaviour
6879 from earlier versions, OTOH that behaviour was broken
6881 UV v; /* value is register so we cant & it /grrr */
6882 if (reg_namedseq(pRExC_state, &v)) {
6892 if (RExC_parse >= RExC_end)
6893 vFAIL2("Empty \\%c{}", (U8)value);
6894 if (*RExC_parse == '{') {
6895 const U8 c = (U8)value;
6896 e = strchr(RExC_parse++, '}');
6898 vFAIL2("Missing right brace on \\%c{}", c);
6899 while (isSPACE(UCHARAT(RExC_parse)))
6901 if (e == RExC_parse)
6902 vFAIL2("Empty \\%c{}", c);
6904 while (isSPACE(UCHARAT(RExC_parse + n - 1)))
6912 if (UCHARAT(RExC_parse) == '^') {
6915 value = value == 'p' ? 'P' : 'p'; /* toggle */
6916 while (isSPACE(UCHARAT(RExC_parse))) {
6921 Perl_sv_catpvf(aTHX_ listsv, "%cutf8::%.*s\n",
6922 (value=='p' ? '+' : '!'), (int)n, RExC_parse);
6925 ANYOF_FLAGS(ret) |= ANYOF_UNICODE;
6926 namedclass = ANYOF_MAX; /* no official name, but it's named */
6929 case 'n': value = '\n'; break;
6930 case 'r': value = '\r'; break;
6931 case 't': value = '\t'; break;
6932 case 'f': value = '\f'; break;
6933 case 'b': value = '\b'; break;
6934 case 'e': value = ASCII_TO_NATIVE('\033');break;
6935 case 'a': value = ASCII_TO_NATIVE('\007');break;
6937 if (*RExC_parse == '{') {
6938 I32 flags = PERL_SCAN_ALLOW_UNDERSCORES
6939 | PERL_SCAN_DISALLOW_PREFIX;
6940 char * const e = strchr(RExC_parse++, '}');
6942 vFAIL("Missing right brace on \\x{}");
6944 numlen = e - RExC_parse;
6945 value = grok_hex(RExC_parse, &numlen, &flags, NULL);
6949 I32 flags = PERL_SCAN_DISALLOW_PREFIX;
6951 value = grok_hex(RExC_parse, &numlen, &flags, NULL);
6952 RExC_parse += numlen;
6954 if (PL_encoding && value < 0x100)
6955 goto recode_encoding;
6958 value = UCHARAT(RExC_parse++);
6959 value = toCTRL(value);
6961 case '0': case '1': case '2': case '3': case '4':
6962 case '5': case '6': case '7': case '8': case '9':
6966 value = grok_oct(--RExC_parse, &numlen, &flags, NULL);
6967 RExC_parse += numlen;
6968 if (PL_encoding && value < 0x100)
6969 goto recode_encoding;
6974 SV* enc = PL_encoding;
6975 value = reg_recode((const char)(U8)value, &enc);
6976 if (!enc && SIZE_ONLY && ckWARN(WARN_REGEXP))
6978 "Invalid escape in the specified encoding");
6982 if (!SIZE_ONLY && isALPHA(value) && ckWARN(WARN_REGEXP))
6984 "Unrecognized escape \\%c in character class passed through",
6988 } /* end of \blah */
6994 if (namedclass > OOB_NAMEDCLASS) { /* this is a named class \blah */
6996 if (!SIZE_ONLY && !need_class)
6997 ANYOF_CLASS_ZERO(ret);
7001 /* a bad range like a-\d, a-[:digit:] ? */
7004 if (ckWARN(WARN_REGEXP)) {
7006 RExC_parse >= rangebegin ?
7007 RExC_parse - rangebegin : 0;
7009 "False [] range \"%*.*s\"",
7012 if (prevvalue < 256) {
7013 ANYOF_BITMAP_SET(ret, prevvalue);
7014 ANYOF_BITMAP_SET(ret, '-');
7017 ANYOF_FLAGS(ret) |= ANYOF_UNICODE;
7018 Perl_sv_catpvf(aTHX_ listsv,
7019 "%04"UVxf"\n%04"UVxf"\n", (UV)prevvalue, (UV) '-');
7023 range = 0; /* this was not a true range */
7027 const char *what = NULL;
7030 if (namedclass > OOB_NAMEDCLASS)
7031 optimize_invert = FALSE;
7032 /* Possible truncation here but in some 64-bit environments
7033 * the compiler gets heartburn about switch on 64-bit values.
7034 * A similar issue a little earlier when switching on value.
7036 switch ((I32)namedclass) {
7039 ANYOF_CLASS_SET(ret, ANYOF_ALNUM);
7041 for (value = 0; value < 256; value++)
7043 ANYOF_BITMAP_SET(ret, value);
7050 ANYOF_CLASS_SET(ret, ANYOF_NALNUM);
7052 for (value = 0; value < 256; value++)
7053 if (!isALNUM(value))
7054 ANYOF_BITMAP_SET(ret, value);
7061 ANYOF_CLASS_SET(ret, ANYOF_ALNUMC);
7063 for (value = 0; value < 256; value++)
7064 if (isALNUMC(value))
7065 ANYOF_BITMAP_SET(ret, value);
7072 ANYOF_CLASS_SET(ret, ANYOF_NALNUMC);
7074 for (value = 0; value < 256; value++)
7075 if (!isALNUMC(value))
7076 ANYOF_BITMAP_SET(ret, value);
7083 ANYOF_CLASS_SET(ret, ANYOF_ALPHA);
7085 for (value = 0; value < 256; value++)
7087 ANYOF_BITMAP_SET(ret, value);
7094 ANYOF_CLASS_SET(ret, ANYOF_NALPHA);
7096 for (value = 0; value < 256; value++)
7097 if (!isALPHA(value))
7098 ANYOF_BITMAP_SET(ret, value);
7105 ANYOF_CLASS_SET(ret, ANYOF_ASCII);
7108 for (value = 0; value < 128; value++)
7109 ANYOF_BITMAP_SET(ret, value);
7111 for (value = 0; value < 256; value++) {
7113 ANYOF_BITMAP_SET(ret, value);
7122 ANYOF_CLASS_SET(ret, ANYOF_NASCII);
7125 for (value = 128; value < 256; value++)
7126 ANYOF_BITMAP_SET(ret, value);
7128 for (value = 0; value < 256; value++) {
7129 if (!isASCII(value))
7130 ANYOF_BITMAP_SET(ret, value);
7139 ANYOF_CLASS_SET(ret, ANYOF_BLANK);
7141 for (value = 0; value < 256; value++)
7143 ANYOF_BITMAP_SET(ret, value);
7150 ANYOF_CLASS_SET(ret, ANYOF_NBLANK);
7152 for (value = 0; value < 256; value++)
7153 if (!isBLANK(value))
7154 ANYOF_BITMAP_SET(ret, value);
7161 ANYOF_CLASS_SET(ret, ANYOF_CNTRL);
7163 for (value = 0; value < 256; value++)
7165 ANYOF_BITMAP_SET(ret, value);
7172 ANYOF_CLASS_SET(ret, ANYOF_NCNTRL);
7174 for (value = 0; value < 256; value++)
7175 if (!isCNTRL(value))
7176 ANYOF_BITMAP_SET(ret, value);
7183 ANYOF_CLASS_SET(ret, ANYOF_DIGIT);
7185 /* consecutive digits assumed */
7186 for (value = '0'; value <= '9'; value++)
7187 ANYOF_BITMAP_SET(ret, value);
7194 ANYOF_CLASS_SET(ret, ANYOF_NDIGIT);
7196 /* consecutive digits assumed */
7197 for (value = 0; value < '0'; value++)
7198 ANYOF_BITMAP_SET(ret, value);
7199 for (value = '9' + 1; value < 256; value++)
7200 ANYOF_BITMAP_SET(ret, value);
7207 ANYOF_CLASS_SET(ret, ANYOF_GRAPH);
7209 for (value = 0; value < 256; value++)
7211 ANYOF_BITMAP_SET(ret, value);
7218 ANYOF_CLASS_SET(ret, ANYOF_NGRAPH);
7220 for (value = 0; value < 256; value++)
7221 if (!isGRAPH(value))
7222 ANYOF_BITMAP_SET(ret, value);
7229 ANYOF_CLASS_SET(ret, ANYOF_LOWER);
7231 for (value = 0; value < 256; value++)
7233 ANYOF_BITMAP_SET(ret, value);
7240 ANYOF_CLASS_SET(ret, ANYOF_NLOWER);
7242 for (value = 0; value < 256; value++)
7243 if (!isLOWER(value))
7244 ANYOF_BITMAP_SET(ret, value);
7251 ANYOF_CLASS_SET(ret, ANYOF_PRINT);
7253 for (value = 0; value < 256; value++)
7255 ANYOF_BITMAP_SET(ret, value);
7262 ANYOF_CLASS_SET(ret, ANYOF_NPRINT);
7264 for (value = 0; value < 256; value++)
7265 if (!isPRINT(value))
7266 ANYOF_BITMAP_SET(ret, value);
7273 ANYOF_CLASS_SET(ret, ANYOF_PSXSPC);
7275 for (value = 0; value < 256; value++)
7276 if (isPSXSPC(value))
7277 ANYOF_BITMAP_SET(ret, value);
7284 ANYOF_CLASS_SET(ret, ANYOF_NPSXSPC);
7286 for (value = 0; value < 256; value++)
7287 if (!isPSXSPC(value))
7288 ANYOF_BITMAP_SET(ret, value);
7295 ANYOF_CLASS_SET(ret, ANYOF_PUNCT);
7297 for (value = 0; value < 256; value++)
7299 ANYOF_BITMAP_SET(ret, value);
7306 ANYOF_CLASS_SET(ret, ANYOF_NPUNCT);
7308 for (value = 0; value < 256; value++)
7309 if (!isPUNCT(value))
7310 ANYOF_BITMAP_SET(ret, value);
7317 ANYOF_CLASS_SET(ret, ANYOF_SPACE);
7319 for (value = 0; value < 256; value++)
7321 ANYOF_BITMAP_SET(ret, value);
7328 ANYOF_CLASS_SET(ret, ANYOF_NSPACE);
7330 for (value = 0; value < 256; value++)
7331 if (!isSPACE(value))
7332 ANYOF_BITMAP_SET(ret, value);
7339 ANYOF_CLASS_SET(ret, ANYOF_UPPER);
7341 for (value = 0; value < 256; value++)
7343 ANYOF_BITMAP_SET(ret, value);
7350 ANYOF_CLASS_SET(ret, ANYOF_NUPPER);
7352 for (value = 0; value < 256; value++)
7353 if (!isUPPER(value))
7354 ANYOF_BITMAP_SET(ret, value);
7361 ANYOF_CLASS_SET(ret, ANYOF_XDIGIT);
7363 for (value = 0; value < 256; value++)
7364 if (isXDIGIT(value))
7365 ANYOF_BITMAP_SET(ret, value);
7372 ANYOF_CLASS_SET(ret, ANYOF_NXDIGIT);
7374 for (value = 0; value < 256; value++)
7375 if (!isXDIGIT(value))
7376 ANYOF_BITMAP_SET(ret, value);
7382 /* this is to handle \p and \P */
7385 vFAIL("Invalid [::] class");
7389 /* Strings such as "+utf8::isWord\n" */
7390 Perl_sv_catpvf(aTHX_ listsv, "%cutf8::Is%s\n", yesno, what);
7393 ANYOF_FLAGS(ret) |= ANYOF_CLASS;
7396 } /* end of namedclass \blah */
7399 if (prevvalue > (IV)value) /* b-a */ {
7400 const int w = RExC_parse - rangebegin;
7401 Simple_vFAIL4("Invalid [] range \"%*.*s\"", w, w, rangebegin);
7402 range = 0; /* not a valid range */
7406 prevvalue = value; /* save the beginning of the range */
7407 if (*RExC_parse == '-' && RExC_parse+1 < RExC_end &&
7408 RExC_parse[1] != ']') {
7411 /* a bad range like \w-, [:word:]- ? */
7412 if (namedclass > OOB_NAMEDCLASS) {
7413 if (ckWARN(WARN_REGEXP)) {
7415 RExC_parse >= rangebegin ?
7416 RExC_parse - rangebegin : 0;
7418 "False [] range \"%*.*s\"",
7422 ANYOF_BITMAP_SET(ret, '-');
7424 range = 1; /* yeah, it's a range! */
7425 continue; /* but do it the next time */
7429 /* now is the next time */
7430 /*stored += (value - prevvalue + 1);*/
7432 if (prevvalue < 256) {
7433 const IV ceilvalue = value < 256 ? value : 255;
7436 /* In EBCDIC [\x89-\x91] should include
7437 * the \x8e but [i-j] should not. */
7438 if (literal_endpoint == 2 &&
7439 ((isLOWER(prevvalue) && isLOWER(ceilvalue)) ||
7440 (isUPPER(prevvalue) && isUPPER(ceilvalue))))
7442 if (isLOWER(prevvalue)) {
7443 for (i = prevvalue; i <= ceilvalue; i++)
7445 ANYOF_BITMAP_SET(ret, i);
7447 for (i = prevvalue; i <= ceilvalue; i++)
7449 ANYOF_BITMAP_SET(ret, i);
7454 for (i = prevvalue; i <= ceilvalue; i++) {
7455 if (!ANYOF_BITMAP_TEST(ret,i)) {
7457 ANYOF_BITMAP_SET(ret, i);
7461 if (value > 255 || UTF) {
7462 const UV prevnatvalue = NATIVE_TO_UNI(prevvalue);
7463 const UV natvalue = NATIVE_TO_UNI(value);
7464 stored+=2; /* can't optimize this class */
7465 ANYOF_FLAGS(ret) |= ANYOF_UNICODE;
7466 if (prevnatvalue < natvalue) { /* what about > ? */
7467 Perl_sv_catpvf(aTHX_ listsv, "%04"UVxf"\t%04"UVxf"\n",
7468 prevnatvalue, natvalue);
7470 else if (prevnatvalue == natvalue) {
7471 Perl_sv_catpvf(aTHX_ listsv, "%04"UVxf"\n", natvalue);
7473 U8 foldbuf[UTF8_MAXBYTES_CASE+1];
7475 const UV f = to_uni_fold(natvalue, foldbuf, &foldlen);
7477 #ifdef EBCDIC /* RD t/uni/fold ff and 6b */
7478 if (RExC_precomp[0] == ':' &&
7479 RExC_precomp[1] == '[' &&
7480 (f == 0xDF || f == 0x92)) {
7481 f = NATIVE_TO_UNI(f);
7484 /* If folding and foldable and a single
7485 * character, insert also the folded version
7486 * to the charclass. */
7488 #ifdef EBCDIC /* RD tunifold ligatures s,t fb05, fb06 */
7489 if ((RExC_precomp[0] == ':' &&
7490 RExC_precomp[1] == '[' &&
7492 (value == 0xFB05 || value == 0xFB06))) ?
7493 foldlen == ((STRLEN)UNISKIP(f) - 1) :
7494 foldlen == (STRLEN)UNISKIP(f) )
7496 if (foldlen == (STRLEN)UNISKIP(f))
7498 Perl_sv_catpvf(aTHX_ listsv,
7501 /* Any multicharacter foldings
7502 * require the following transform:
7503 * [ABCDEF] -> (?:[ABCabcDEFd]|pq|rst)
7504 * where E folds into "pq" and F folds
7505 * into "rst", all other characters
7506 * fold to single characters. We save
7507 * away these multicharacter foldings,
7508 * to be later saved as part of the
7509 * additional "s" data. */
7512 if (!unicode_alternate)
7513 unicode_alternate = newAV();
7514 sv = newSVpvn((char*)foldbuf, foldlen);
7516 av_push(unicode_alternate, sv);
7520 /* If folding and the value is one of the Greek
7521 * sigmas insert a few more sigmas to make the
7522 * folding rules of the sigmas to work right.
7523 * Note that not all the possible combinations
7524 * are handled here: some of them are handled
7525 * by the standard folding rules, and some of
7526 * them (literal or EXACTF cases) are handled
7527 * during runtime in regexec.c:S_find_byclass(). */
7528 if (value == UNICODE_GREEK_SMALL_LETTER_FINAL_SIGMA) {
7529 Perl_sv_catpvf(aTHX_ listsv, "%04"UVxf"\n",
7530 (UV)UNICODE_GREEK_CAPITAL_LETTER_SIGMA);
7531 Perl_sv_catpvf(aTHX_ listsv, "%04"UVxf"\n",
7532 (UV)UNICODE_GREEK_SMALL_LETTER_SIGMA);
7534 else if (value == UNICODE_GREEK_CAPITAL_LETTER_SIGMA)
7535 Perl_sv_catpvf(aTHX_ listsv, "%04"UVxf"\n",
7536 (UV)UNICODE_GREEK_SMALL_LETTER_SIGMA);
7541 literal_endpoint = 0;
7545 range = 0; /* this range (if it was one) is done now */
7549 ANYOF_FLAGS(ret) |= ANYOF_LARGE;
7551 RExC_size += ANYOF_CLASS_ADD_SKIP;
7553 RExC_emit += ANYOF_CLASS_ADD_SKIP;
7559 /****** !SIZE_ONLY AFTER HERE *********/
7561 if( stored == 1 && value < 256
7562 && !( ANYOF_FLAGS(ret) & ( ANYOF_FLAGS_ALL ^ ANYOF_FOLD ) )
7564 /* optimize single char class to an EXACT node
7565 but *only* when its not a UTF/high char */
7566 const char * cur_parse= RExC_parse;
7567 RExC_emit = (regnode *)orig_emit;
7568 RExC_parse = (char *)orig_parse;
7569 ret = reg_node(pRExC_state,
7570 (U8)((ANYOF_FLAGS(ret) & ANYOF_FOLD) ? EXACTF : EXACT));
7571 RExC_parse = (char *)cur_parse;
7572 *STRING(ret)= (char)value;
7574 RExC_emit += STR_SZ(1);
7577 /* optimize case-insensitive simple patterns (e.g. /[a-z]/i) */
7578 if ( /* If the only flag is folding (plus possibly inversion). */
7579 ((ANYOF_FLAGS(ret) & (ANYOF_FLAGS_ALL ^ ANYOF_INVERT)) == ANYOF_FOLD)
7581 for (value = 0; value < 256; ++value) {
7582 if (ANYOF_BITMAP_TEST(ret, value)) {
7583 UV fold = PL_fold[value];
7586 ANYOF_BITMAP_SET(ret, fold);
7589 ANYOF_FLAGS(ret) &= ~ANYOF_FOLD;
7592 /* optimize inverted simple patterns (e.g. [^a-z]) */
7593 if (optimize_invert &&
7594 /* If the only flag is inversion. */
7595 (ANYOF_FLAGS(ret) & ANYOF_FLAGS_ALL) == ANYOF_INVERT) {
7596 for (value = 0; value < ANYOF_BITMAP_SIZE; ++value)
7597 ANYOF_BITMAP(ret)[value] ^= ANYOF_FLAGS_ALL;
7598 ANYOF_FLAGS(ret) = ANYOF_UNICODE_ALL;
7601 AV * const av = newAV();
7603 /* The 0th element stores the character class description
7604 * in its textual form: used later (regexec.c:Perl_regclass_swash())
7605 * to initialize the appropriate swash (which gets stored in
7606 * the 1st element), and also useful for dumping the regnode.
7607 * The 2nd element stores the multicharacter foldings,
7608 * used later (regexec.c:S_reginclass()). */
7609 av_store(av, 0, listsv);
7610 av_store(av, 1, NULL);
7611 av_store(av, 2, (SV*)unicode_alternate);
7612 rv = newRV_noinc((SV*)av);
7613 n = add_data(pRExC_state, 1, "s");
7614 RExC_rx->data->data[n] = (void*)rv;
7621 S_nextchar(pTHX_ RExC_state_t *pRExC_state)
7623 char* const retval = RExC_parse++;
7626 if (*RExC_parse == '(' && RExC_parse[1] == '?' &&
7627 RExC_parse[2] == '#') {
7628 while (*RExC_parse != ')') {
7629 if (RExC_parse == RExC_end)
7630 FAIL("Sequence (?#... not terminated");
7636 if (RExC_flags & PMf_EXTENDED) {
7637 if (isSPACE(*RExC_parse)) {
7641 else if (*RExC_parse == '#') {
7642 while (RExC_parse < RExC_end)
7643 if (*RExC_parse++ == '\n') break;
7652 - reg_node - emit a node
7654 STATIC regnode * /* Location. */
7655 S_reg_node(pTHX_ RExC_state_t *pRExC_state, U8 op)
7658 register regnode *ptr;
7659 regnode * const ret = RExC_emit;
7660 GET_RE_DEBUG_FLAGS_DECL;
7663 SIZE_ALIGN(RExC_size);
7668 if (OP(RExC_emit) == 255)
7669 Perl_croak(aTHX_ "panic: reg_node overrun trying to emit %s: %d ",
7670 reg_name[op], OP(RExC_emit));
7672 NODE_ALIGN_FILL(ret);
7674 FILL_ADVANCE_NODE(ptr, op);
7675 if (RExC_offsets) { /* MJD */
7676 MJD_OFFSET_DEBUG(("%s:%d: (op %s) %s %"UVuf" (len %"UVuf") (max %"UVuf").\n",
7677 "reg_node", __LINE__,
7679 (UV)(RExC_emit - RExC_emit_start) > RExC_offsets[0]
7680 ? "Overwriting end of array!\n" : "OK",
7681 (UV)(RExC_emit - RExC_emit_start),
7682 (UV)(RExC_parse - RExC_start),
7683 (UV)RExC_offsets[0]));
7684 Set_Node_Offset(RExC_emit, RExC_parse + (op == END));
7692 - reganode - emit a node with an argument
7694 STATIC regnode * /* Location. */
7695 S_reganode(pTHX_ RExC_state_t *pRExC_state, U8 op, U32 arg)
7698 register regnode *ptr;
7699 regnode * const ret = RExC_emit;
7700 GET_RE_DEBUG_FLAGS_DECL;
7703 SIZE_ALIGN(RExC_size);
7708 assert(2==regarglen[op]+1);
7710 Anything larger than this has to allocate the extra amount.
7711 If we changed this to be:
7713 RExC_size += (1 + regarglen[op]);
7715 then it wouldn't matter. Its not clear what side effect
7716 might come from that so its not done so far.
7722 if (OP(RExC_emit) == 255)
7723 Perl_croak(aTHX_ "panic: reganode overwriting end of allocated program space");
7725 NODE_ALIGN_FILL(ret);
7727 FILL_ADVANCE_NODE_ARG(ptr, op, arg);
7728 if (RExC_offsets) { /* MJD */
7729 MJD_OFFSET_DEBUG(("%s(%d): (op %s) %s %"UVuf" <- %"UVuf" (max %"UVuf").\n",
7733 (UV)(RExC_emit - RExC_emit_start) > RExC_offsets[0] ?
7734 "Overwriting end of array!\n" : "OK",
7735 (UV)(RExC_emit - RExC_emit_start),
7736 (UV)(RExC_parse - RExC_start),
7737 (UV)RExC_offsets[0]));
7738 Set_Cur_Node_Offset;
7746 - reguni - emit (if appropriate) a Unicode character
7749 S_reguni(pTHX_ const RExC_state_t *pRExC_state, UV uv, char* s)
7752 return SIZE_ONLY ? UNISKIP(uv) : (uvchr_to_utf8((U8*)s, uv) - (U8*)s);
7756 - reginsert - insert an operator in front of already-emitted operand
7758 * Means relocating the operand.
7761 S_reginsert(pTHX_ RExC_state_t *pRExC_state, U8 op, regnode *opnd, U32 depth)
7764 register regnode *src;
7765 register regnode *dst;
7766 register regnode *place;
7767 const int offset = regarglen[(U8)op];
7768 const int size = NODE_STEP_REGNODE + offset;
7769 GET_RE_DEBUG_FLAGS_DECL;
7770 /* (PL_regkind[(U8)op] == CURLY ? EXTRA_STEP_2ARGS : 0); */
7771 DEBUG_PARSE_FMT("inst"," - %s",reg_name[op]);
7780 if (RExC_open_parens) {
7782 DEBUG_PARSE_FMT("inst"," - %"IVdf, (IV)RExC_npar);
7783 for ( paren=0 ; paren < RExC_npar ; paren++ ) {
7784 if ( RExC_open_parens[paren] >= opnd ) {
7785 DEBUG_PARSE_FMT("open"," - %d",size);
7786 RExC_open_parens[paren] += size;
7788 DEBUG_PARSE_FMT("open"," - %s","ok");
7790 if ( RExC_close_parens[paren] >= opnd ) {
7791 DEBUG_PARSE_FMT("close"," - %d",size);
7792 RExC_close_parens[paren] += size;
7794 DEBUG_PARSE_FMT("close"," - %s","ok");
7799 while (src > opnd) {
7800 StructCopy(--src, --dst, regnode);
7801 if (RExC_offsets) { /* MJD 20010112 */
7802 MJD_OFFSET_DEBUG(("%s(%d): (op %s) %s copy %"UVuf" -> %"UVuf" (max %"UVuf").\n",
7806 (UV)(dst - RExC_emit_start) > RExC_offsets[0]
7807 ? "Overwriting end of array!\n" : "OK",
7808 (UV)(src - RExC_emit_start),
7809 (UV)(dst - RExC_emit_start),
7810 (UV)RExC_offsets[0]));
7811 Set_Node_Offset_To_R(dst-RExC_emit_start, Node_Offset(src));
7812 Set_Node_Length_To_R(dst-RExC_emit_start, Node_Length(src));
7817 place = opnd; /* Op node, where operand used to be. */
7818 if (RExC_offsets) { /* MJD */
7819 MJD_OFFSET_DEBUG(("%s(%d): (op %s) %s %"UVuf" <- %"UVuf" (max %"UVuf").\n",
7823 (UV)(place - RExC_emit_start) > RExC_offsets[0]
7824 ? "Overwriting end of array!\n" : "OK",
7825 (UV)(place - RExC_emit_start),
7826 (UV)(RExC_parse - RExC_start),
7827 (UV)RExC_offsets[0]));
7828 Set_Node_Offset(place, RExC_parse);
7829 Set_Node_Length(place, 1);
7831 src = NEXTOPER(place);
7832 FILL_ADVANCE_NODE(place, op);
7833 Zero(src, offset, regnode);
7837 - regtail - set the next-pointer at the end of a node chain of p to val.
7838 - SEE ALSO: regtail_study
7840 /* TODO: All three parms should be const */
7842 S_regtail(pTHX_ RExC_state_t *pRExC_state, regnode *p, const regnode *val,U32 depth)
7845 register regnode *scan;
7846 GET_RE_DEBUG_FLAGS_DECL;
7848 PERL_UNUSED_ARG(depth);
7854 /* Find last node. */
7857 regnode * const temp = regnext(scan);
7859 SV * const mysv=sv_newmortal();
7860 DEBUG_PARSE_MSG((scan==p ? "tail" : ""));
7861 regprop(RExC_rx, mysv, scan);
7862 PerlIO_printf(Perl_debug_log, "~ %s (%d) %s %s\n",
7863 SvPV_nolen_const(mysv), REG_NODE_NUM(scan),
7864 (temp == NULL ? "->" : ""),
7865 (temp == NULL ? reg_name[OP(val)] : "")
7873 if (reg_off_by_arg[OP(scan)]) {
7874 ARG_SET(scan, val - scan);
7877 NEXT_OFF(scan) = val - scan;
7883 - regtail_study - set the next-pointer at the end of a node chain of p to val.
7884 - Look for optimizable sequences at the same time.
7885 - currently only looks for EXACT chains.
7887 This is expermental code. The idea is to use this routine to perform
7888 in place optimizations on branches and groups as they are constructed,
7889 with the long term intention of removing optimization from study_chunk so
7890 that it is purely analytical.
7892 Currently only used when in DEBUG mode. The macro REGTAIL_STUDY() is used
7893 to control which is which.
7896 /* TODO: All four parms should be const */
7899 S_regtail_study(pTHX_ RExC_state_t *pRExC_state, regnode *p, const regnode *val,U32 depth)
7902 register regnode *scan;
7904 #ifdef EXPERIMENTAL_INPLACESCAN
7908 GET_RE_DEBUG_FLAGS_DECL;
7914 /* Find last node. */
7918 regnode * const temp = regnext(scan);
7919 #ifdef EXPERIMENTAL_INPLACESCAN
7920 if (PL_regkind[OP(scan)] == EXACT)
7921 if (join_exact(pRExC_state,scan,&min,1,val,depth+1))
7929 if( exact == PSEUDO )
7931 else if ( exact != OP(scan) )
7940 SV * const mysv=sv_newmortal();
7941 DEBUG_PARSE_MSG((scan==p ? "tsdy" : ""));
7942 regprop(RExC_rx, mysv, scan);
7943 PerlIO_printf(Perl_debug_log, "~ %s (%d) -> %s\n",
7944 SvPV_nolen_const(mysv),
7953 SV * const mysv_val=sv_newmortal();
7954 DEBUG_PARSE_MSG("");
7955 regprop(RExC_rx, mysv_val, val);
7956 PerlIO_printf(Perl_debug_log, "~ attach to %s (%d) offset to %d\n",
7957 SvPV_nolen_const(mysv_val),
7962 if (reg_off_by_arg[OP(scan)]) {
7963 ARG_SET(scan, val - scan);
7966 NEXT_OFF(scan) = val - scan;
7974 - regcurly - a little FSA that accepts {\d+,?\d*}
7977 S_regcurly(register const char *s)
7996 - regdump - dump a regexp onto Perl_debug_log in vaguely comprehensible form
7999 Perl_regdump(pTHX_ const regexp *r)
8003 SV * const sv = sv_newmortal();
8004 SV *dsv= sv_newmortal();
8006 (void)dumpuntil(r, r->program, r->program + 1, NULL, NULL, sv, 0, 0);
8008 /* Header fields of interest. */
8009 if (r->anchored_substr) {
8010 RE_PV_QUOTED_DECL(s, 0, dsv, SvPVX_const(r->anchored_substr),
8011 RE_SV_DUMPLEN(r->anchored_substr), 30);
8012 PerlIO_printf(Perl_debug_log,
8013 "anchored %s%s at %"IVdf" ",
8014 s, RE_SV_TAIL(r->anchored_substr),
8015 (IV)r->anchored_offset);
8016 } else if (r->anchored_utf8) {
8017 RE_PV_QUOTED_DECL(s, 1, dsv, SvPVX_const(r->anchored_utf8),
8018 RE_SV_DUMPLEN(r->anchored_utf8), 30);
8019 PerlIO_printf(Perl_debug_log,
8020 "anchored utf8 %s%s at %"IVdf" ",
8021 s, RE_SV_TAIL(r->anchored_utf8),
8022 (IV)r->anchored_offset);
8024 if (r->float_substr) {
8025 RE_PV_QUOTED_DECL(s, 0, dsv, SvPVX_const(r->float_substr),
8026 RE_SV_DUMPLEN(r->float_substr), 30);
8027 PerlIO_printf(Perl_debug_log,
8028 "floating %s%s at %"IVdf"..%"UVuf" ",
8029 s, RE_SV_TAIL(r->float_substr),
8030 (IV)r->float_min_offset, (UV)r->float_max_offset);
8031 } else if (r->float_utf8) {
8032 RE_PV_QUOTED_DECL(s, 1, dsv, SvPVX_const(r->float_utf8),
8033 RE_SV_DUMPLEN(r->float_utf8), 30);
8034 PerlIO_printf(Perl_debug_log,
8035 "floating utf8 %s%s at %"IVdf"..%"UVuf" ",
8036 s, RE_SV_TAIL(r->float_utf8),
8037 (IV)r->float_min_offset, (UV)r->float_max_offset);
8039 if (r->check_substr || r->check_utf8)
8040 PerlIO_printf(Perl_debug_log,
8042 (r->check_substr == r->float_substr
8043 && r->check_utf8 == r->float_utf8
8044 ? "(checking floating" : "(checking anchored"));
8045 if (r->reganch & ROPT_NOSCAN)
8046 PerlIO_printf(Perl_debug_log, " noscan");
8047 if (r->reganch & ROPT_CHECK_ALL)
8048 PerlIO_printf(Perl_debug_log, " isall");
8049 if (r->check_substr || r->check_utf8)
8050 PerlIO_printf(Perl_debug_log, ") ");
8052 if (r->regstclass) {
8053 regprop(r, sv, r->regstclass);
8054 PerlIO_printf(Perl_debug_log, "stclass %s ", SvPVX_const(sv));
8056 if (r->reganch & ROPT_ANCH) {
8057 PerlIO_printf(Perl_debug_log, "anchored");
8058 if (r->reganch & ROPT_ANCH_BOL)
8059 PerlIO_printf(Perl_debug_log, "(BOL)");
8060 if (r->reganch & ROPT_ANCH_MBOL)
8061 PerlIO_printf(Perl_debug_log, "(MBOL)");
8062 if (r->reganch & ROPT_ANCH_SBOL)
8063 PerlIO_printf(Perl_debug_log, "(SBOL)");
8064 if (r->reganch & ROPT_ANCH_GPOS)
8065 PerlIO_printf(Perl_debug_log, "(GPOS)");
8066 PerlIO_putc(Perl_debug_log, ' ');
8068 if (r->reganch & ROPT_GPOS_SEEN)
8069 PerlIO_printf(Perl_debug_log, "GPOS ");
8070 if (r->reganch & ROPT_SKIP)
8071 PerlIO_printf(Perl_debug_log, "plus ");
8072 if (r->reganch & ROPT_IMPLICIT)
8073 PerlIO_printf(Perl_debug_log, "implicit ");
8074 PerlIO_printf(Perl_debug_log, "minlen %ld ", (long) r->minlen);
8075 if (r->reganch & ROPT_EVAL_SEEN)
8076 PerlIO_printf(Perl_debug_log, "with eval ");
8077 PerlIO_printf(Perl_debug_log, "\n");
8079 PERL_UNUSED_CONTEXT;
8081 #endif /* DEBUGGING */
8085 - regprop - printable representation of opcode
8088 Perl_regprop(pTHX_ const regexp *prog, SV *sv, const regnode *o)
8093 GET_RE_DEBUG_FLAGS_DECL;
8095 sv_setpvn(sv, "", 0);
8096 if (OP(o) > REGNODE_MAX) /* regnode.type is unsigned */
8097 /* It would be nice to FAIL() here, but this may be called from
8098 regexec.c, and it would be hard to supply pRExC_state. */
8099 Perl_croak(aTHX_ "Corrupted regexp opcode %d > %d", (int)OP(o), (int)REGNODE_MAX);
8100 sv_catpv(sv, reg_name[OP(o)]); /* Take off const! */
8102 k = PL_regkind[OP(o)];
8105 SV * const dsv = sv_2mortal(newSVpvs(""));
8106 /* Using is_utf8_string() (via PERL_PV_UNI_DETECT)
8107 * is a crude hack but it may be the best for now since
8108 * we have no flag "this EXACTish node was UTF-8"
8110 const char * const s =
8111 pv_pretty(dsv, STRING(o), STR_LEN(o), 60,
8112 PL_colors[0], PL_colors[1],
8113 PERL_PV_ESCAPE_UNI_DETECT |
8114 PERL_PV_PRETTY_ELIPSES |
8117 Perl_sv_catpvf(aTHX_ sv, " %s", s );
8118 } else if (k == TRIE) {
8119 /* print the details of the trie in dumpuntil instead, as
8120 * prog->data isn't available here */
8121 const char op = OP(o);
8122 const I32 n = ARG(o);
8123 const reg_ac_data * const ac = IS_TRIE_AC(op) ?
8124 (reg_ac_data *)prog->data->data[n] :
8126 const reg_trie_data * const trie = !IS_TRIE_AC(op) ?
8127 (reg_trie_data*)prog->data->data[n] :
8130 Perl_sv_catpvf(aTHX_ sv, "-%s",reg_name[o->flags]);
8131 DEBUG_TRIE_COMPILE_r(
8132 Perl_sv_catpvf(aTHX_ sv,
8133 "<S:%"UVuf"/%"IVdf" W:%"UVuf" L:%"UVuf"/%"UVuf" C:%"UVuf"/%"UVuf">",
8134 (UV)trie->startstate,
8135 (IV)trie->statecount-1, /* -1 because of the unused 0 element */
8136 (UV)trie->wordcount,
8139 (UV)TRIE_CHARCOUNT(trie),
8140 (UV)trie->uniquecharcount
8143 if ( IS_ANYOF_TRIE(op) || trie->bitmap ) {
8145 int rangestart = -1;
8146 U8* bitmap = IS_ANYOF_TRIE(op) ? (U8*)ANYOF_BITMAP(o) : (U8*)TRIE_BITMAP(trie);
8147 Perl_sv_catpvf(aTHX_ sv, "[");
8148 for (i = 0; i <= 256; i++) {
8149 if (i < 256 && BITMAP_TEST(bitmap,i)) {
8150 if (rangestart == -1)
8152 } else if (rangestart != -1) {
8153 if (i <= rangestart + 3)
8154 for (; rangestart < i; rangestart++)
8155 put_byte(sv, rangestart);
8157 put_byte(sv, rangestart);
8159 put_byte(sv, i - 1);
8164 Perl_sv_catpvf(aTHX_ sv, "]");
8167 } else if (k == CURLY) {
8168 if (OP(o) == CURLYM || OP(o) == CURLYN || OP(o) == CURLYX)
8169 Perl_sv_catpvf(aTHX_ sv, "[%d]", o->flags); /* Parenth number */
8170 Perl_sv_catpvf(aTHX_ sv, " {%d,%d}", ARG1(o), ARG2(o));
8172 else if (k == WHILEM && o->flags) /* Ordinal/of */
8173 Perl_sv_catpvf(aTHX_ sv, "[%d/%d]", o->flags & 0xf, o->flags>>4);
8174 else if (k == REF || k == OPEN || k == CLOSE || k == GROUPP || OP(o)==ACCEPT)
8175 Perl_sv_catpvf(aTHX_ sv, "%d", (int)ARG(o)); /* Parenth number */
8176 else if (k == GOSUB)
8177 Perl_sv_catpvf(aTHX_ sv, "%d[%+d]", (int)ARG(o),(int)ARG2L(o)); /* Paren and offset */
8178 else if (k == VERB) {
8180 Perl_sv_catpvf(aTHX_ sv, ":%"SVf,
8181 (SV*)prog->data->data[ ARG( o ) ]);
8182 } else if (k == LOGICAL)
8183 Perl_sv_catpvf(aTHX_ sv, "[%d]", o->flags); /* 2: embedded, otherwise 1 */
8184 else if (k == ANYOF) {
8185 int i, rangestart = -1;
8186 const U8 flags = ANYOF_FLAGS(o);
8188 /* Should be synchronized with * ANYOF_ #xdefines in regcomp.h */
8189 static const char * const anyofs[] = {
8222 if (flags & ANYOF_LOCALE)
8223 sv_catpvs(sv, "{loc}");
8224 if (flags & ANYOF_FOLD)
8225 sv_catpvs(sv, "{i}");
8226 Perl_sv_catpvf(aTHX_ sv, "[%s", PL_colors[0]);
8227 if (flags & ANYOF_INVERT)
8229 for (i = 0; i <= 256; i++) {
8230 if (i < 256 && ANYOF_BITMAP_TEST(o,i)) {
8231 if (rangestart == -1)
8233 } else if (rangestart != -1) {
8234 if (i <= rangestart + 3)
8235 for (; rangestart < i; rangestart++)
8236 put_byte(sv, rangestart);
8238 put_byte(sv, rangestart);
8240 put_byte(sv, i - 1);
8246 if (o->flags & ANYOF_CLASS)
8247 for (i = 0; i < (int)(sizeof(anyofs)/sizeof(char*)); i++)
8248 if (ANYOF_CLASS_TEST(o,i))
8249 sv_catpv(sv, anyofs[i]);
8251 if (flags & ANYOF_UNICODE)
8252 sv_catpvs(sv, "{unicode}");
8253 else if (flags & ANYOF_UNICODE_ALL)
8254 sv_catpvs(sv, "{unicode_all}");
8258 SV * const sw = regclass_swash(prog, o, FALSE, &lv, 0);
8262 U8 s[UTF8_MAXBYTES_CASE+1];
8264 for (i = 0; i <= 256; i++) { /* just the first 256 */
8265 uvchr_to_utf8(s, i);
8267 if (i < 256 && swash_fetch(sw, s, TRUE)) {
8268 if (rangestart == -1)
8270 } else if (rangestart != -1) {
8271 if (i <= rangestart + 3)
8272 for (; rangestart < i; rangestart++) {
8273 const U8 * const e = uvchr_to_utf8(s,rangestart);
8275 for(p = s; p < e; p++)
8279 const U8 *e = uvchr_to_utf8(s,rangestart);
8281 for (p = s; p < e; p++)
8284 e = uvchr_to_utf8(s, i-1);
8285 for (p = s; p < e; p++)
8292 sv_catpvs(sv, "..."); /* et cetera */
8296 char *s = savesvpv(lv);
8297 char * const origs = s;
8299 while (*s && *s != '\n')
8303 const char * const t = ++s;
8321 Perl_sv_catpvf(aTHX_ sv, "%s]", PL_colors[1]);
8323 else if (k == BRANCHJ && (OP(o) == UNLESSM || OP(o) == IFMATCH))
8324 Perl_sv_catpvf(aTHX_ sv, "[%d]", -(o->flags));
8326 PERL_UNUSED_CONTEXT;
8327 PERL_UNUSED_ARG(sv);
8329 PERL_UNUSED_ARG(prog);
8330 #endif /* DEBUGGING */
8334 Perl_re_intuit_string(pTHX_ regexp *prog)
8335 { /* Assume that RE_INTUIT is set */
8337 GET_RE_DEBUG_FLAGS_DECL;
8338 PERL_UNUSED_CONTEXT;
8342 const char * const s = SvPV_nolen_const(prog->check_substr
8343 ? prog->check_substr : prog->check_utf8);
8345 if (!PL_colorset) reginitcolors();
8346 PerlIO_printf(Perl_debug_log,
8347 "%sUsing REx %ssubstr:%s \"%s%.60s%s%s\"\n",
8349 prog->check_substr ? "" : "utf8 ",
8350 PL_colors[5],PL_colors[0],
8353 (strlen(s) > 60 ? "..." : ""));
8356 return prog->check_substr ? prog->check_substr : prog->check_utf8;
8360 pregfree - free a regexp
8362 See regdupe below if you change anything here.
8366 Perl_pregfree(pTHX_ struct regexp *r)
8370 GET_RE_DEBUG_FLAGS_DECL;
8372 if (!r || (--r->refcnt > 0))
8378 SV *dsv= sv_newmortal();
8379 RE_PV_QUOTED_DECL(s, (r->reganch & ROPT_UTF8),
8380 dsv, r->precomp, r->prelen, 60);
8381 PerlIO_printf(Perl_debug_log,"%sFreeing REx:%s %s\n",
8382 PL_colors[4],PL_colors[5],s);
8386 /* gcov results gave these as non-null 100% of the time, so there's no
8387 optimisation in checking them before calling Safefree */
8388 Safefree(r->precomp);
8389 Safefree(r->offsets); /* 20010421 MJD */
8390 RX_MATCH_COPY_FREE(r);
8391 #ifdef PERL_OLD_COPY_ON_WRITE
8393 SvREFCNT_dec(r->saved_copy);
8396 if (r->anchored_substr)
8397 SvREFCNT_dec(r->anchored_substr);
8398 if (r->anchored_utf8)
8399 SvREFCNT_dec(r->anchored_utf8);
8400 if (r->float_substr)
8401 SvREFCNT_dec(r->float_substr);
8403 SvREFCNT_dec(r->float_utf8);
8404 Safefree(r->substrs);
8407 SvREFCNT_dec(r->paren_names);
8409 int n = r->data->count;
8410 PAD* new_comppad = NULL;
8415 /* If you add a ->what type here, update the comment in regcomp.h */
8416 switch (r->data->what[n]) {
8419 SvREFCNT_dec((SV*)r->data->data[n]);
8422 Safefree(r->data->data[n]);
8425 new_comppad = (AV*)r->data->data[n];
8428 if (new_comppad == NULL)
8429 Perl_croak(aTHX_ "panic: pregfree comppad");
8430 PAD_SAVE_LOCAL(old_comppad,
8431 /* Watch out for global destruction's random ordering. */
8432 (SvTYPE(new_comppad) == SVt_PVAV) ? new_comppad : NULL
8435 refcnt = OpREFCNT_dec((OP_4tree*)r->data->data[n]);
8438 op_free((OP_4tree*)r->data->data[n]);
8440 PAD_RESTORE_LOCAL(old_comppad);
8441 SvREFCNT_dec((SV*)new_comppad);
8447 { /* Aho Corasick add-on structure for a trie node.
8448 Used in stclass optimization only */
8450 reg_ac_data *aho=(reg_ac_data*)r->data->data[n];
8452 refcount = --aho->refcount;
8455 Safefree(aho->states);
8456 Safefree(aho->fail);
8457 aho->trie=NULL; /* not necessary to free this as it is
8458 handled by the 't' case */
8459 Safefree(r->data->data[n]); /* do this last!!!! */
8460 Safefree(r->regstclass);
8466 /* trie structure. */
8468 reg_trie_data *trie=(reg_trie_data*)r->data->data[n];
8470 refcount = --trie->refcount;
8473 Safefree(trie->charmap);
8474 if (trie->widecharmap)
8475 SvREFCNT_dec((SV*)trie->widecharmap);
8476 Safefree(trie->states);
8477 Safefree(trie->trans);
8479 Safefree(trie->bitmap);
8481 Safefree(trie->wordlen);
8483 Safefree(trie->jump);
8485 Safefree(trie->nextword);
8488 SvREFCNT_dec((SV*)trie->words);
8489 if (trie->revcharmap)
8490 SvREFCNT_dec((SV*)trie->revcharmap);
8492 Safefree(r->data->data[n]); /* do this last!!!! */
8497 Perl_croak(aTHX_ "panic: regfree data code '%c'", r->data->what[n]);
8500 Safefree(r->data->what);
8503 Safefree(r->startp);
8508 #define sv_dup_inc(s,t) SvREFCNT_inc(sv_dup(s,t))
8509 #define av_dup_inc(s,t) (AV*)SvREFCNT_inc(sv_dup((SV*)s,t))
8510 #define hv_dup_inc(s,t) (HV*)SvREFCNT_inc(sv_dup((SV*)s,t))
8511 #define SAVEPVN(p,n) ((p) ? savepvn(p,n) : NULL)
8514 regdupe - duplicate a regexp.
8516 This routine is called by sv.c's re_dup and is expected to clone a
8517 given regexp structure. It is a no-op when not under USE_ITHREADS.
8518 (Originally this *was* re_dup() for change history see sv.c)
8520 See pregfree() above if you change anything here.
8522 #if defined(USE_ITHREADS)
8524 Perl_regdupe(pTHX_ const regexp *r, CLONE_PARAMS *param)
8529 struct reg_substr_datum *s;
8532 return (REGEXP *)NULL;
8534 if ((ret = (REGEXP *)ptr_table_fetch(PL_ptr_table, r)))
8537 len = r->offsets[0];
8538 npar = r->nparens+1;
8540 Newxc(ret, sizeof(regexp) + (len+1)*sizeof(regnode), char, regexp);
8541 Copy(r->program, ret->program, len+1, regnode);
8543 Newx(ret->startp, npar, I32);
8544 Copy(r->startp, ret->startp, npar, I32);
8545 Newx(ret->endp, npar, I32);
8546 Copy(r->startp, ret->startp, npar, I32);
8548 Newx(ret->substrs, 1, struct reg_substr_data);
8549 for (s = ret->substrs->data, i = 0; i < 3; i++, s++) {
8550 s->min_offset = r->substrs->data[i].min_offset;
8551 s->max_offset = r->substrs->data[i].max_offset;
8552 s->end_shift = r->substrs->data[i].end_shift;
8553 s->substr = sv_dup_inc(r->substrs->data[i].substr, param);
8554 s->utf8_substr = sv_dup_inc(r->substrs->data[i].utf8_substr, param);
8557 ret->regstclass = NULL;
8560 const int count = r->data->count;
8563 Newxc(d, sizeof(struct reg_data) + count*sizeof(void *),
8564 char, struct reg_data);
8565 Newx(d->what, count, U8);
8568 for (i = 0; i < count; i++) {
8569 d->what[i] = r->data->what[i];
8570 switch (d->what[i]) {
8571 /* legal options are one of: sSfpont
8572 see also regcomp.h and pregfree() */
8575 d->data[i] = sv_dup_inc((SV *)r->data->data[i], param);
8578 d->data[i] = av_dup_inc((AV *)r->data->data[i], param);
8581 /* This is cheating. */
8582 Newx(d->data[i], 1, struct regnode_charclass_class);
8583 StructCopy(r->data->data[i], d->data[i],
8584 struct regnode_charclass_class);
8585 ret->regstclass = (regnode*)d->data[i];
8588 /* Compiled op trees are readonly, and can thus be
8589 shared without duplication. */
8591 d->data[i] = (void*)OpREFCNT_inc((OP*)r->data->data[i]);
8595 d->data[i] = r->data->data[i];
8598 d->data[i] = r->data->data[i];
8600 ((reg_trie_data*)d->data[i])->refcount++;
8604 d->data[i] = r->data->data[i];
8606 ((reg_ac_data*)d->data[i])->refcount++;
8608 /* Trie stclasses are readonly and can thus be shared
8609 * without duplication. We free the stclass in pregfree
8610 * when the corresponding reg_ac_data struct is freed.
8612 ret->regstclass= r->regstclass;
8615 Perl_croak(aTHX_ "panic: re_dup unknown data code '%c'", r->data->what[i]);
8624 Newx(ret->offsets, 2*len+1, U32);
8625 Copy(r->offsets, ret->offsets, 2*len+1, U32);
8627 ret->precomp = SAVEPVN(r->precomp, r->prelen);
8628 ret->refcnt = r->refcnt;
8629 ret->minlen = r->minlen;
8630 ret->minlenret = r->minlenret;
8631 ret->prelen = r->prelen;
8632 ret->nparens = r->nparens;
8633 ret->lastparen = r->lastparen;
8634 ret->lastcloseparen = r->lastcloseparen;
8635 ret->reganch = r->reganch;
8637 ret->sublen = r->sublen;
8639 ret->engine = r->engine;
8641 ret->paren_names = hv_dup_inc(r->paren_names, param);
8643 if (RX_MATCH_COPIED(ret))
8644 ret->subbeg = SAVEPVN(r->subbeg, r->sublen);
8647 #ifdef PERL_OLD_COPY_ON_WRITE
8648 ret->saved_copy = NULL;
8651 ptr_table_store(PL_ptr_table, r, ret);
8659 converts a regexp embedded in a MAGIC struct to its stringified form,
8660 caching the converted form in the struct and returns the cached
8663 If lp is nonnull then it is used to return the length of the
8666 If flags is nonnull and the returned string contains UTF8 then
8667 (flags & 1) will be true.
8669 If haseval is nonnull then it is used to return whether the pattern
8672 Normally called via macro:
8674 CALLREG_STRINGIFY(mg,0,0);
8678 CALLREG_AS_STR(mg,lp,flags,haseval)
8680 See sv_2pv_flags() in sv.c for an example of internal usage.
8685 Perl_reg_stringify(pTHX_ MAGIC *mg, STRLEN *lp, U32 *flags, I32 *haseval ) {
8687 const regexp * const re = (regexp *)mg->mg_obj;
8690 const char *fptr = "msix";
8695 bool need_newline = 0;
8696 U16 reganch = (U16)((re->reganch & PMf_COMPILETIME) >> 12);
8698 while((ch = *fptr++)) {
8700 reflags[left++] = ch;
8703 reflags[right--] = ch;
8708 reflags[left] = '-';
8712 mg->mg_len = re->prelen + 4 + left;
8714 * If /x was used, we have to worry about a regex ending with a
8715 * comment later being embedded within another regex. If so, we don't
8716 * want this regex's "commentization" to leak out to the right part of
8717 * the enclosing regex, we must cap it with a newline.
8719 * So, if /x was used, we scan backwards from the end of the regex. If
8720 * we find a '#' before we find a newline, we need to add a newline
8721 * ourself. If we find a '\n' first (or if we don't find '#' or '\n'),
8722 * we don't need to add anything. -jfriedl
8724 if (PMf_EXTENDED & re->reganch) {
8725 const char *endptr = re->precomp + re->prelen;
8726 while (endptr >= re->precomp) {
8727 const char c = *(endptr--);
8729 break; /* don't need another */
8731 /* we end while in a comment, so we need a newline */
8732 mg->mg_len++; /* save space for it */
8733 need_newline = 1; /* note to add it */
8739 Newx(mg->mg_ptr, mg->mg_len + 1 + left, char);
8740 mg->mg_ptr[0] = '(';
8741 mg->mg_ptr[1] = '?';
8742 Copy(reflags, mg->mg_ptr+2, left, char);
8743 *(mg->mg_ptr+left+2) = ':';
8744 Copy(re->precomp, mg->mg_ptr+3+left, re->prelen, char);
8746 mg->mg_ptr[mg->mg_len - 2] = '\n';
8747 mg->mg_ptr[mg->mg_len - 1] = ')';
8748 mg->mg_ptr[mg->mg_len] = 0;
8751 *haseval = re->program[0].next_off;
8753 *flags = ((re->reganch & ROPT_UTF8) ? 1 : 0);
8761 #ifndef PERL_IN_XSUB_RE
8763 - regnext - dig the "next" pointer out of a node
8766 Perl_regnext(pTHX_ register regnode *p)
8769 register I32 offset;
8771 if (p == &PL_regdummy)
8774 offset = (reg_off_by_arg[OP(p)] ? ARG(p) : NEXT_OFF(p));
8783 S_re_croak2(pTHX_ const char* pat1,const char* pat2,...)
8786 STRLEN l1 = strlen(pat1);
8787 STRLEN l2 = strlen(pat2);
8790 const char *message;
8796 Copy(pat1, buf, l1 , char);
8797 Copy(pat2, buf + l1, l2 , char);
8798 buf[l1 + l2] = '\n';
8799 buf[l1 + l2 + 1] = '\0';
8801 /* ANSI variant takes additional second argument */
8802 va_start(args, pat2);
8806 msv = vmess(buf, &args);
8808 message = SvPV_const(msv,l1);
8811 Copy(message, buf, l1 , char);
8812 buf[l1-1] = '\0'; /* Overwrite \n */
8813 Perl_croak(aTHX_ "%s", buf);
8816 /* XXX Here's a total kludge. But we need to re-enter for swash routines. */
8818 #ifndef PERL_IN_XSUB_RE
8820 Perl_save_re_context(pTHX)
8824 struct re_save_state *state;
8826 SAVEVPTR(PL_curcop);
8827 SSGROW(SAVESTACK_ALLOC_FOR_RE_SAVE_STATE + 1);
8829 state = (struct re_save_state *)(PL_savestack + PL_savestack_ix);
8830 PL_savestack_ix += SAVESTACK_ALLOC_FOR_RE_SAVE_STATE;
8831 SSPUSHINT(SAVEt_RE_STATE);
8833 Copy(&PL_reg_state, state, 1, struct re_save_state);
8835 PL_reg_start_tmp = 0;
8836 PL_reg_start_tmpl = 0;
8837 PL_reg_oldsaved = NULL;
8838 PL_reg_oldsavedlen = 0;
8840 PL_reg_leftiter = 0;
8841 PL_reg_poscache = NULL;
8842 PL_reg_poscache_size = 0;
8843 #ifdef PERL_OLD_COPY_ON_WRITE
8847 /* Save $1..$n (#18107: UTF-8 s/(\w+)/uc($1)/e); AMS 20021106. */
8849 const REGEXP * const rx = PM_GETRE(PL_curpm);
8852 for (i = 1; i <= rx->nparens; i++) {
8853 char digits[TYPE_CHARS(long)];
8854 const STRLEN len = my_snprintf(digits, sizeof(digits), "%lu", (long)i);
8855 GV *const *const gvp
8856 = (GV**)hv_fetch(PL_defstash, digits, len, 0);
8859 GV * const gv = *gvp;
8860 if (SvTYPE(gv) == SVt_PVGV && GvSV(gv))
8870 clear_re(pTHX_ void *r)
8873 ReREFCNT_dec((regexp *)r);
8879 S_put_byte(pTHX_ SV *sv, int c)
8881 if (isCNTRL(c) || c == 255 || !isPRINT(c))
8882 Perl_sv_catpvf(aTHX_ sv, "\\%o", c);
8883 else if (c == '-' || c == ']' || c == '\\' || c == '^')
8884 Perl_sv_catpvf(aTHX_ sv, "\\%c", c);
8886 Perl_sv_catpvf(aTHX_ sv, "%c", c);
8890 #define CLEAR_OPTSTART \
8891 if (optstart) STMT_START { \
8892 DEBUG_OPTIMISE_r(PerlIO_printf(Perl_debug_log, " (%d nodes)\n", node - optstart)); \
8896 #define DUMPUNTIL(b,e) CLEAR_OPTSTART; node=dumpuntil(r,start,(b),(e),last,sv,indent+1,depth+1);
8898 STATIC const regnode *
8899 S_dumpuntil(pTHX_ const regexp *r, const regnode *start, const regnode *node,
8900 const regnode *last, const regnode *plast,
8901 SV* sv, I32 indent, U32 depth)
8904 register U8 op = PSEUDO; /* Arbitrary non-END op. */
8905 register const regnode *next;
8906 const regnode *optstart= NULL;
8907 GET_RE_DEBUG_FLAGS_DECL;
8909 #ifdef DEBUG_DUMPUNTIL
8910 PerlIO_printf(Perl_debug_log, "--- %d : %d - %d - %d\n",indent,node-start,
8911 last ? last-start : 0,plast ? plast-start : 0);
8914 if (plast && plast < last)
8917 while (PL_regkind[op] != END && (!last || node < last)) {
8918 /* While that wasn't END last time... */
8924 next = regnext((regnode *)node);
8927 if (OP(node) == OPTIMIZED) {
8928 if (!optstart && RE_DEBUG_FLAG(RE_DEBUG_COMPILE_OPTIMISE))
8935 regprop(r, sv, node);
8936 PerlIO_printf(Perl_debug_log, "%4"IVdf":%*s%s", (IV)(node - start),
8937 (int)(2*indent + 1), "", SvPVX_const(sv));
8939 if (OP(node) != OPTIMIZED) {
8940 if (next == NULL) /* Next ptr. */
8941 PerlIO_printf(Perl_debug_log, "(0)");
8942 else if (PL_regkind[(U8)op] == BRANCH && PL_regkind[OP(next)] != BRANCH )
8943 PerlIO_printf(Perl_debug_log, "(FAIL)");
8945 PerlIO_printf(Perl_debug_log, "(%"IVdf")", (IV)(next - start));
8947 /*if (PL_regkind[(U8)op] != TRIE)*/
8948 (void)PerlIO_putc(Perl_debug_log, '\n');
8952 if (PL_regkind[(U8)op] == BRANCHJ) {
8955 register const regnode *nnode = (OP(next) == LONGJMP
8956 ? regnext((regnode *)next)
8958 if (last && nnode > last)
8960 DUMPUNTIL(NEXTOPER(NEXTOPER(node)), nnode);
8963 else if (PL_regkind[(U8)op] == BRANCH) {
8965 DUMPUNTIL(NEXTOPER(node), next);
8967 else if ( PL_regkind[(U8)op] == TRIE ) {
8968 const regnode *this_trie = node;
8969 const char op = OP(node);
8970 const I32 n = ARG(node);
8971 const reg_ac_data * const ac = op>=AHOCORASICK ?
8972 (reg_ac_data *)r->data->data[n] :
8974 const reg_trie_data * const trie = op<AHOCORASICK ?
8975 (reg_trie_data*)r->data->data[n] :
8977 const regnode *nextbranch= NULL;
8979 sv_setpvn(sv, "", 0);
8980 for (word_idx= 0; word_idx < (I32)trie->wordcount; word_idx++) {
8981 SV ** const elem_ptr = av_fetch(trie->words,word_idx,0);
8983 PerlIO_printf(Perl_debug_log, "%*s%s ",
8984 (int)(2*(indent+3)), "",
8985 elem_ptr ? pv_pretty(sv, SvPV_nolen_const(*elem_ptr), SvCUR(*elem_ptr), 60,
8986 PL_colors[0], PL_colors[1],
8987 (SvUTF8(*elem_ptr) ? PERL_PV_ESCAPE_UNI : 0) |
8988 PERL_PV_PRETTY_ELIPSES |
8994 U16 dist= trie->jump[word_idx+1];
8995 PerlIO_printf(Perl_debug_log, "(%u)\n",
8996 (dist ? this_trie + dist : next) - start);
8999 nextbranch= this_trie + trie->jump[0];
9000 DUMPUNTIL(this_trie + dist, nextbranch);
9002 if (nextbranch && PL_regkind[OP(nextbranch)]==BRANCH)
9003 nextbranch= regnext((regnode *)nextbranch);
9005 PerlIO_printf(Perl_debug_log, "\n");
9008 if (last && next > last)
9013 else if ( op == CURLY ) { /* "next" might be very big: optimizer */
9014 DUMPUNTIL(NEXTOPER(node) + EXTRA_STEP_2ARGS,
9015 NEXTOPER(node) + EXTRA_STEP_2ARGS + 1);
9017 else if (PL_regkind[(U8)op] == CURLY && op != CURLYX) {
9019 DUMPUNTIL(NEXTOPER(node) + EXTRA_STEP_2ARGS, next);
9021 else if ( op == PLUS || op == STAR) {
9022 DUMPUNTIL(NEXTOPER(node), NEXTOPER(node) + 1);
9024 else if (op == ANYOF) {
9025 /* arglen 1 + class block */
9026 node += 1 + ((ANYOF_FLAGS(node) & ANYOF_LARGE)
9027 ? ANYOF_CLASS_SKIP : ANYOF_SKIP);
9028 node = NEXTOPER(node);
9030 else if (PL_regkind[(U8)op] == EXACT) {
9031 /* Literal string, where present. */
9032 node += NODE_SZ_STR(node) - 1;
9033 node = NEXTOPER(node);
9036 node = NEXTOPER(node);
9037 node += regarglen[(U8)op];
9039 if (op == CURLYX || op == OPEN)
9041 else if (op == WHILEM)
9045 #ifdef DEBUG_DUMPUNTIL
9046 PerlIO_printf(Perl_debug_log, "--- %d\n",indent);
9051 #endif /* DEBUGGING */
9055 * c-indentation-style: bsd
9057 * indent-tabs-mode: t
9060 * ex: set ts=8 sts=4 sw=4 noet:
projects / p5sagit/p5-mst-13.2.git / blob