5 * 'A fair jaw-cracker dwarf-language must be.' --Samwise Gamgee
7 * [p.285 of _The Lord of the Rings_, II/iii: "The Ring Goes South"]
10 /* This file contains functions for compiling a regular expression. See
11 * also regexec.c which funnily enough, contains functions for executing
12 * a regular expression.
14 * This file is also copied at build time to ext/re/re_comp.c, where
15 * it's built with -DPERL_EXT_RE_BUILD -DPERL_EXT_RE_DEBUG -DPERL_EXT.
16 * This causes the main functions to be compiled under new names and with
17 * debugging support added, which makes "use re 'debug'" work.
20 /* NOTE: this is derived from Henry Spencer's regexp code, and should not
21 * confused with the original package (see point 3 below). Thanks, Henry!
24 /* Additional note: this code is very heavily munged from Henry's version
25 * in places. In some spots I've traded clarity for efficiency, so don't
26 * blame Henry for some of the lack of readability.
29 /* The names of the functions have been changed from regcomp and
30 * regexec to pregcomp and pregexec in order to avoid conflicts
31 * with the POSIX routines of the same names.
34 #ifdef PERL_EXT_RE_BUILD
39 * pregcomp and pregexec -- regsub and regerror are not used in perl
41 * Copyright (c) 1986 by University of Toronto.
42 * Written by Henry Spencer. Not derived from licensed software.
44 * Permission is granted to anyone to use this software for any
45 * purpose on any computer system, and to redistribute it freely,
46 * subject to the following restrictions:
48 * 1. The author is not responsible for the consequences of use of
49 * this software, no matter how awful, even if they arise
52 * 2. The origin of this software must not be misrepresented, either
53 * by explicit claim or by omission.
55 * 3. Altered versions must be plainly marked as such, and must not
56 * be misrepresented as being the original software.
59 **** Alterations to Henry's code are...
61 **** Copyright (C) 1991, 1992, 1993, 1994, 1995, 1996, 1997, 1998, 1999,
62 **** 2000, 2001, 2002, 2003, 2004, 2005, 2006, 2007, 2008
63 **** by Larry Wall and others
65 **** You may distribute under the terms of either the GNU General Public
66 **** License or the Artistic License, as specified in the README file.
69 * Beware that some of this code is subtly aware of the way operator
70 * precedence is structured in regular expressions. Serious changes in
71 * regular-expression syntax might require a total rethink.
74 #define PERL_IN_REGCOMP_C
77 #ifndef PERL_IN_XSUB_RE
82 #ifdef PERL_IN_XSUB_RE
88 #include "dquote_static.c"
95 # if defined(BUGGY_MSC6)
96 /* MSC 6.00A breaks on op/regexp.t test 85 unless we turn this off */
97 # pragma optimize("a",off)
98 /* But MSC 6.00A is happy with 'w', for aliases only across function calls*/
99 # pragma optimize("w",on )
100 # endif /* BUGGY_MSC6 */
104 #define STATIC static
107 typedef struct RExC_state_t {
108 U32 flags; /* are we folding, multilining? */
109 char *precomp; /* uncompiled string. */
110 REGEXP *rx_sv; /* The SV that is the regexp. */
111 regexp *rx; /* perl core regexp structure */
112 regexp_internal *rxi; /* internal data for regexp object pprivate field */
113 char *start; /* Start of input for compile */
114 char *end; /* End of input for compile */
115 char *parse; /* Input-scan pointer. */
116 I32 whilem_seen; /* number of WHILEM in this expr */
117 regnode *emit_start; /* Start of emitted-code area */
118 regnode *emit_bound; /* First regnode outside of the allocated space */
119 regnode *emit; /* Code-emit pointer; ®dummy = don't = compiling */
120 I32 naughty; /* How bad is this pattern? */
121 I32 sawback; /* Did we see \1, ...? */
123 I32 size; /* Code size. */
124 I32 npar; /* Capture buffer count, (OPEN). */
125 I32 cpar; /* Capture buffer count, (CLOSE). */
126 I32 nestroot; /* root parens we are in - used by accept */
130 regnode **open_parens; /* pointers to open parens */
131 regnode **close_parens; /* pointers to close parens */
132 regnode *opend; /* END node in program */
133 I32 utf8; /* whether the pattern is utf8 or not */
134 I32 orig_utf8; /* whether the pattern was originally in utf8 */
135 /* XXX use this for future optimisation of case
136 * where pattern must be upgraded to utf8. */
137 I32 uni_semantics; /* If a d charset modifier should use unicode
138 rules, even if the pattern is not in
140 HV *paren_names; /* Paren names */
142 regnode **recurse; /* Recurse regops */
143 I32 recurse_count; /* Number of recurse regops */
146 I32 override_recoding;
148 char *starttry; /* -Dr: where regtry was called. */
149 #define RExC_starttry (pRExC_state->starttry)
152 const char *lastparse;
154 AV *paren_name_list; /* idx -> name */
155 #define RExC_lastparse (pRExC_state->lastparse)
156 #define RExC_lastnum (pRExC_state->lastnum)
157 #define RExC_paren_name_list (pRExC_state->paren_name_list)
161 #define RExC_flags (pRExC_state->flags)
162 #define RExC_precomp (pRExC_state->precomp)
163 #define RExC_rx_sv (pRExC_state->rx_sv)
164 #define RExC_rx (pRExC_state->rx)
165 #define RExC_rxi (pRExC_state->rxi)
166 #define RExC_start (pRExC_state->start)
167 #define RExC_end (pRExC_state->end)
168 #define RExC_parse (pRExC_state->parse)
169 #define RExC_whilem_seen (pRExC_state->whilem_seen)
170 #ifdef RE_TRACK_PATTERN_OFFSETS
171 #define RExC_offsets (pRExC_state->rxi->u.offsets) /* I am not like the others */
173 #define RExC_emit (pRExC_state->emit)
174 #define RExC_emit_start (pRExC_state->emit_start)
175 #define RExC_emit_bound (pRExC_state->emit_bound)
176 #define RExC_naughty (pRExC_state->naughty)
177 #define RExC_sawback (pRExC_state->sawback)
178 #define RExC_seen (pRExC_state->seen)
179 #define RExC_size (pRExC_state->size)
180 #define RExC_npar (pRExC_state->npar)
181 #define RExC_nestroot (pRExC_state->nestroot)
182 #define RExC_extralen (pRExC_state->extralen)
183 #define RExC_seen_zerolen (pRExC_state->seen_zerolen)
184 #define RExC_seen_evals (pRExC_state->seen_evals)
185 #define RExC_utf8 (pRExC_state->utf8)
186 #define RExC_uni_semantics (pRExC_state->uni_semantics)
187 #define RExC_orig_utf8 (pRExC_state->orig_utf8)
188 #define RExC_open_parens (pRExC_state->open_parens)
189 #define RExC_close_parens (pRExC_state->close_parens)
190 #define RExC_opend (pRExC_state->opend)
191 #define RExC_paren_names (pRExC_state->paren_names)
192 #define RExC_recurse (pRExC_state->recurse)
193 #define RExC_recurse_count (pRExC_state->recurse_count)
194 #define RExC_in_lookbehind (pRExC_state->in_lookbehind)
195 #define RExC_contains_locale (pRExC_state->contains_locale)
196 #define RExC_override_recoding (pRExC_state->override_recoding)
199 #define ISMULT1(c) ((c) == '*' || (c) == '+' || (c) == '?')
200 #define ISMULT2(s) ((*s) == '*' || (*s) == '+' || (*s) == '?' || \
201 ((*s) == '{' && regcurly(s)))
204 #undef SPSTART /* dratted cpp namespace... */
207 * Flags to be passed up and down.
209 #define WORST 0 /* Worst case. */
210 #define HASWIDTH 0x01 /* Known to match non-null strings. */
212 /* Simple enough to be STAR/PLUS operand, in an EXACT node must be a single
213 * character, and if utf8, must be invariant. Note that this is not the same thing as REGNODE_SIMPLE */
215 #define SPSTART 0x04 /* Starts with * or +. */
216 #define TRYAGAIN 0x08 /* Weeded out a declaration. */
217 #define POSTPONED 0x10 /* (?1),(?&name), (??{...}) or similar */
219 #define REG_NODE_NUM(x) ((x) ? (int)((x)-RExC_emit_start) : -1)
221 /* whether trie related optimizations are enabled */
222 #if PERL_ENABLE_EXTENDED_TRIE_OPTIMISATION
223 #define TRIE_STUDY_OPT
224 #define FULL_TRIE_STUDY
230 #define PBYTE(u8str,paren) ((U8*)(u8str))[(paren) >> 3]
231 #define PBITVAL(paren) (1 << ((paren) & 7))
232 #define PAREN_TEST(u8str,paren) ( PBYTE(u8str,paren) & PBITVAL(paren))
233 #define PAREN_SET(u8str,paren) PBYTE(u8str,paren) |= PBITVAL(paren)
234 #define PAREN_UNSET(u8str,paren) PBYTE(u8str,paren) &= (~PBITVAL(paren))
236 /* If not already in utf8, do a longjmp back to the beginning */
237 #define UTF8_LONGJMP 42 /* Choose a value not likely to ever conflict */
238 #define REQUIRE_UTF8 STMT_START { \
239 if (! UTF) JMPENV_JUMP(UTF8_LONGJMP); \
242 /* About scan_data_t.
244 During optimisation we recurse through the regexp program performing
245 various inplace (keyhole style) optimisations. In addition study_chunk
246 and scan_commit populate this data structure with information about
247 what strings MUST appear in the pattern. We look for the longest
248 string that must appear at a fixed location, and we look for the
249 longest string that may appear at a floating location. So for instance
254 Both 'FOO' and 'A' are fixed strings. Both 'B' and 'BAR' are floating
255 strings (because they follow a .* construct). study_chunk will identify
256 both FOO and BAR as being the longest fixed and floating strings respectively.
258 The strings can be composites, for instance
262 will result in a composite fixed substring 'foo'.
264 For each string some basic information is maintained:
266 - offset or min_offset
267 This is the position the string must appear at, or not before.
268 It also implicitly (when combined with minlenp) tells us how many
269 characters must match before the string we are searching for.
270 Likewise when combined with minlenp and the length of the string it
271 tells us how many characters must appear after the string we have
275 Only used for floating strings. This is the rightmost point that
276 the string can appear at. If set to I32 max it indicates that the
277 string can occur infinitely far to the right.
280 A pointer to the minimum length of the pattern that the string
281 was found inside. This is important as in the case of positive
282 lookahead or positive lookbehind we can have multiple patterns
287 The minimum length of the pattern overall is 3, the minimum length
288 of the lookahead part is 3, but the minimum length of the part that
289 will actually match is 1. So 'FOO's minimum length is 3, but the
290 minimum length for the F is 1. This is important as the minimum length
291 is used to determine offsets in front of and behind the string being
292 looked for. Since strings can be composites this is the length of the
293 pattern at the time it was committed with a scan_commit. Note that
294 the length is calculated by study_chunk, so that the minimum lengths
295 are not known until the full pattern has been compiled, thus the
296 pointer to the value.
300 In the case of lookbehind the string being searched for can be
301 offset past the start point of the final matching string.
302 If this value was just blithely removed from the min_offset it would
303 invalidate some of the calculations for how many chars must match
304 before or after (as they are derived from min_offset and minlen and
305 the length of the string being searched for).
306 When the final pattern is compiled and the data is moved from the
307 scan_data_t structure into the regexp structure the information
308 about lookbehind is factored in, with the information that would
309 have been lost precalculated in the end_shift field for the
312 The fields pos_min and pos_delta are used to store the minimum offset
313 and the delta to the maximum offset at the current point in the pattern.
317 typedef struct scan_data_t {
318 /*I32 len_min; unused */
319 /*I32 len_delta; unused */
323 I32 last_end; /* min value, <0 unless valid. */
326 SV **longest; /* Either &l_fixed, or &l_float. */
327 SV *longest_fixed; /* longest fixed string found in pattern */
328 I32 offset_fixed; /* offset where it starts */
329 I32 *minlen_fixed; /* pointer to the minlen relevant to the string */
330 I32 lookbehind_fixed; /* is the position of the string modfied by LB */
331 SV *longest_float; /* longest floating string found in pattern */
332 I32 offset_float_min; /* earliest point in string it can appear */
333 I32 offset_float_max; /* latest point in string it can appear */
334 I32 *minlen_float; /* pointer to the minlen relevant to the string */
335 I32 lookbehind_float; /* is the position of the string modified by LB */
339 struct regnode_charclass_class *start_class;
343 * Forward declarations for pregcomp()'s friends.
346 static const scan_data_t zero_scan_data =
347 { 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0 ,0};
349 #define SF_BEFORE_EOL (SF_BEFORE_SEOL|SF_BEFORE_MEOL)
350 #define SF_BEFORE_SEOL 0x0001
351 #define SF_BEFORE_MEOL 0x0002
352 #define SF_FIX_BEFORE_EOL (SF_FIX_BEFORE_SEOL|SF_FIX_BEFORE_MEOL)
353 #define SF_FL_BEFORE_EOL (SF_FL_BEFORE_SEOL|SF_FL_BEFORE_MEOL)
356 # define SF_FIX_SHIFT_EOL (0+2)
357 # define SF_FL_SHIFT_EOL (0+4)
359 # define SF_FIX_SHIFT_EOL (+2)
360 # define SF_FL_SHIFT_EOL (+4)
363 #define SF_FIX_BEFORE_SEOL (SF_BEFORE_SEOL << SF_FIX_SHIFT_EOL)
364 #define SF_FIX_BEFORE_MEOL (SF_BEFORE_MEOL << SF_FIX_SHIFT_EOL)
366 #define SF_FL_BEFORE_SEOL (SF_BEFORE_SEOL << SF_FL_SHIFT_EOL)
367 #define SF_FL_BEFORE_MEOL (SF_BEFORE_MEOL << SF_FL_SHIFT_EOL) /* 0x20 */
368 #define SF_IS_INF 0x0040
369 #define SF_HAS_PAR 0x0080
370 #define SF_IN_PAR 0x0100
371 #define SF_HAS_EVAL 0x0200
372 #define SCF_DO_SUBSTR 0x0400
373 #define SCF_DO_STCLASS_AND 0x0800
374 #define SCF_DO_STCLASS_OR 0x1000
375 #define SCF_DO_STCLASS (SCF_DO_STCLASS_AND|SCF_DO_STCLASS_OR)
376 #define SCF_WHILEM_VISITED_POS 0x2000
378 #define SCF_TRIE_RESTUDY 0x4000 /* Do restudy? */
379 #define SCF_SEEN_ACCEPT 0x8000
381 #define UTF cBOOL(RExC_utf8)
382 #define LOC (get_regex_charset(RExC_flags) == REGEX_LOCALE_CHARSET)
383 #define UNI_SEMANTICS (get_regex_charset(RExC_flags) == REGEX_UNICODE_CHARSET)
384 #define DEPENDS_SEMANTICS (get_regex_charset(RExC_flags) == REGEX_DEPENDS_CHARSET)
385 #define AT_LEAST_UNI_SEMANTICS (get_regex_charset(RExC_flags) >= REGEX_UNICODE_CHARSET)
386 #define ASCII_RESTRICTED (get_regex_charset(RExC_flags) == REGEX_ASCII_RESTRICTED_CHARSET)
387 #define MORE_ASCII_RESTRICTED (get_regex_charset(RExC_flags) == REGEX_ASCII_MORE_RESTRICTED_CHARSET)
388 #define AT_LEAST_ASCII_RESTRICTED (get_regex_charset(RExC_flags) >= REGEX_ASCII_RESTRICTED_CHARSET)
390 #define FOLD cBOOL(RExC_flags & RXf_PMf_FOLD)
392 #define OOB_UNICODE 12345678
393 #define OOB_NAMEDCLASS -1
395 #define CHR_SVLEN(sv) (UTF ? sv_len_utf8(sv) : SvCUR(sv))
396 #define CHR_DIST(a,b) (UTF ? utf8_distance(a,b) : a - b)
399 /* length of regex to show in messages that don't mark a position within */
400 #define RegexLengthToShowInErrorMessages 127
403 * If MARKER[12] are adjusted, be sure to adjust the constants at the top
404 * of t/op/regmesg.t, the tests in t/op/re_tests, and those in
405 * op/pragma/warn/regcomp.
407 #define MARKER1 "<-- HERE" /* marker as it appears in the description */
408 #define MARKER2 " <-- HERE " /* marker as it appears within the regex */
410 #define REPORT_LOCATION " in regex; marked by " MARKER1 " in m/%.*s" MARKER2 "%s/"
413 * Calls SAVEDESTRUCTOR_X if needed, then calls Perl_croak with the given
414 * arg. Show regex, up to a maximum length. If it's too long, chop and add
417 #define _FAIL(code) STMT_START { \
418 const char *ellipses = ""; \
419 IV len = RExC_end - RExC_precomp; \
422 SAVEDESTRUCTOR_X(clear_re,(void*)RExC_rx_sv); \
423 if (len > RegexLengthToShowInErrorMessages) { \
424 /* chop 10 shorter than the max, to ensure meaning of "..." */ \
425 len = RegexLengthToShowInErrorMessages - 10; \
431 #define FAIL(msg) _FAIL( \
432 Perl_croak(aTHX_ "%s in regex m/%.*s%s/", \
433 msg, (int)len, RExC_precomp, ellipses))
435 #define FAIL2(msg,arg) _FAIL( \
436 Perl_croak(aTHX_ msg " in regex m/%.*s%s/", \
437 arg, (int)len, RExC_precomp, ellipses))
440 * Simple_vFAIL -- like FAIL, but marks the current location in the scan
442 #define Simple_vFAIL(m) STMT_START { \
443 const IV offset = RExC_parse - RExC_precomp; \
444 Perl_croak(aTHX_ "%s" REPORT_LOCATION, \
445 m, (int)offset, RExC_precomp, RExC_precomp + offset); \
449 * Calls SAVEDESTRUCTOR_X if needed, then Simple_vFAIL()
451 #define vFAIL(m) STMT_START { \
453 SAVEDESTRUCTOR_X(clear_re,(void*)RExC_rx_sv); \
458 * Like Simple_vFAIL(), but accepts two arguments.
460 #define Simple_vFAIL2(m,a1) STMT_START { \
461 const IV offset = RExC_parse - RExC_precomp; \
462 S_re_croak2(aTHX_ m, REPORT_LOCATION, a1, \
463 (int)offset, RExC_precomp, RExC_precomp + offset); \
467 * Calls SAVEDESTRUCTOR_X if needed, then Simple_vFAIL2().
469 #define vFAIL2(m,a1) STMT_START { \
471 SAVEDESTRUCTOR_X(clear_re,(void*)RExC_rx_sv); \
472 Simple_vFAIL2(m, a1); \
477 * Like Simple_vFAIL(), but accepts three arguments.
479 #define Simple_vFAIL3(m, a1, a2) STMT_START { \
480 const IV offset = RExC_parse - RExC_precomp; \
481 S_re_croak2(aTHX_ m, REPORT_LOCATION, a1, a2, \
482 (int)offset, RExC_precomp, RExC_precomp + offset); \
486 * Calls SAVEDESTRUCTOR_X if needed, then Simple_vFAIL3().
488 #define vFAIL3(m,a1,a2) STMT_START { \
490 SAVEDESTRUCTOR_X(clear_re,(void*)RExC_rx_sv); \
491 Simple_vFAIL3(m, a1, a2); \
495 * Like Simple_vFAIL(), but accepts four arguments.
497 #define Simple_vFAIL4(m, a1, a2, a3) STMT_START { \
498 const IV offset = RExC_parse - RExC_precomp; \
499 S_re_croak2(aTHX_ m, REPORT_LOCATION, a1, a2, a3, \
500 (int)offset, RExC_precomp, RExC_precomp + offset); \
503 #define ckWARNreg(loc,m) STMT_START { \
504 const IV offset = loc - RExC_precomp; \
505 Perl_ck_warner(aTHX_ packWARN(WARN_REGEXP), m REPORT_LOCATION, \
506 (int)offset, RExC_precomp, RExC_precomp + offset); \
509 #define ckWARNregdep(loc,m) STMT_START { \
510 const IV offset = loc - RExC_precomp; \
511 Perl_ck_warner_d(aTHX_ packWARN2(WARN_DEPRECATED, WARN_REGEXP), \
513 (int)offset, RExC_precomp, RExC_precomp + offset); \
516 #define ckWARN2regdep(loc,m, a1) STMT_START { \
517 const IV offset = loc - RExC_precomp; \
518 Perl_ck_warner_d(aTHX_ packWARN2(WARN_DEPRECATED, WARN_REGEXP), \
520 a1, (int)offset, RExC_precomp, RExC_precomp + offset); \
523 #define ckWARN2reg(loc, m, a1) STMT_START { \
524 const IV offset = loc - RExC_precomp; \
525 Perl_ck_warner(aTHX_ packWARN(WARN_REGEXP), m REPORT_LOCATION, \
526 a1, (int)offset, RExC_precomp, RExC_precomp + offset); \
529 #define vWARN3(loc, m, a1, a2) STMT_START { \
530 const IV offset = loc - RExC_precomp; \
531 Perl_warner(aTHX_ packWARN(WARN_REGEXP), m REPORT_LOCATION, \
532 a1, a2, (int)offset, RExC_precomp, RExC_precomp + offset); \
535 #define ckWARN3reg(loc, m, a1, a2) STMT_START { \
536 const IV offset = loc - RExC_precomp; \
537 Perl_ck_warner(aTHX_ packWARN(WARN_REGEXP), m REPORT_LOCATION, \
538 a1, a2, (int)offset, RExC_precomp, RExC_precomp + offset); \
541 #define vWARN4(loc, m, a1, a2, a3) STMT_START { \
542 const IV offset = loc - RExC_precomp; \
543 Perl_warner(aTHX_ packWARN(WARN_REGEXP), m REPORT_LOCATION, \
544 a1, a2, a3, (int)offset, RExC_precomp, RExC_precomp + offset); \
547 #define ckWARN4reg(loc, m, a1, a2, a3) STMT_START { \
548 const IV offset = loc - RExC_precomp; \
549 Perl_ck_warner(aTHX_ packWARN(WARN_REGEXP), m REPORT_LOCATION, \
550 a1, a2, a3, (int)offset, RExC_precomp, RExC_precomp + offset); \
553 #define vWARN5(loc, m, a1, a2, a3, a4) STMT_START { \
554 const IV offset = loc - RExC_precomp; \
555 Perl_warner(aTHX_ packWARN(WARN_REGEXP), m REPORT_LOCATION, \
556 a1, a2, a3, a4, (int)offset, RExC_precomp, RExC_precomp + offset); \
560 /* Allow for side effects in s */
561 #define REGC(c,s) STMT_START { \
562 if (!SIZE_ONLY) *(s) = (c); else (void)(s); \
565 /* Macros for recording node offsets. 20001227 mjd@plover.com
566 * Nodes are numbered 1, 2, 3, 4. Node #n's position is recorded in
567 * element 2*n-1 of the array. Element #2n holds the byte length node #n.
568 * Element 0 holds the number n.
569 * Position is 1 indexed.
571 #ifndef RE_TRACK_PATTERN_OFFSETS
572 #define Set_Node_Offset_To_R(node,byte)
573 #define Set_Node_Offset(node,byte)
574 #define Set_Cur_Node_Offset
575 #define Set_Node_Length_To_R(node,len)
576 #define Set_Node_Length(node,len)
577 #define Set_Node_Cur_Length(node)
578 #define Node_Offset(n)
579 #define Node_Length(n)
580 #define Set_Node_Offset_Length(node,offset,len)
581 #define ProgLen(ri) ri->u.proglen
582 #define SetProgLen(ri,x) ri->u.proglen = x
584 #define ProgLen(ri) ri->u.offsets[0]
585 #define SetProgLen(ri,x) ri->u.offsets[0] = x
586 #define Set_Node_Offset_To_R(node,byte) STMT_START { \
588 MJD_OFFSET_DEBUG(("** (%d) offset of node %d is %d.\n", \
589 __LINE__, (int)(node), (int)(byte))); \
591 Perl_croak(aTHX_ "value of node is %d in Offset macro", (int)(node)); \
593 RExC_offsets[2*(node)-1] = (byte); \
598 #define Set_Node_Offset(node,byte) \
599 Set_Node_Offset_To_R((node)-RExC_emit_start, (byte)-RExC_start)
600 #define Set_Cur_Node_Offset Set_Node_Offset(RExC_emit, RExC_parse)
602 #define Set_Node_Length_To_R(node,len) STMT_START { \
604 MJD_OFFSET_DEBUG(("** (%d) size of node %d is %d.\n", \
605 __LINE__, (int)(node), (int)(len))); \
607 Perl_croak(aTHX_ "value of node is %d in Length macro", (int)(node)); \
609 RExC_offsets[2*(node)] = (len); \
614 #define Set_Node_Length(node,len) \
615 Set_Node_Length_To_R((node)-RExC_emit_start, len)
616 #define Set_Cur_Node_Length(len) Set_Node_Length(RExC_emit, len)
617 #define Set_Node_Cur_Length(node) \
618 Set_Node_Length(node, RExC_parse - parse_start)
620 /* Get offsets and lengths */
621 #define Node_Offset(n) (RExC_offsets[2*((n)-RExC_emit_start)-1])
622 #define Node_Length(n) (RExC_offsets[2*((n)-RExC_emit_start)])
624 #define Set_Node_Offset_Length(node,offset,len) STMT_START { \
625 Set_Node_Offset_To_R((node)-RExC_emit_start, (offset)); \
626 Set_Node_Length_To_R((node)-RExC_emit_start, (len)); \
630 #if PERL_ENABLE_EXPERIMENTAL_REGEX_OPTIMISATIONS
631 #define EXPERIMENTAL_INPLACESCAN
632 #endif /*PERL_ENABLE_EXPERIMENTAL_REGEX_OPTIMISATIONS*/
634 #define DEBUG_STUDYDATA(str,data,depth) \
635 DEBUG_OPTIMISE_MORE_r(if(data){ \
636 PerlIO_printf(Perl_debug_log, \
637 "%*s" str "Pos:%"IVdf"/%"IVdf \
638 " Flags: 0x%"UVXf" Whilem_c: %"IVdf" Lcp: %"IVdf" %s", \
639 (int)(depth)*2, "", \
640 (IV)((data)->pos_min), \
641 (IV)((data)->pos_delta), \
642 (UV)((data)->flags), \
643 (IV)((data)->whilem_c), \
644 (IV)((data)->last_closep ? *((data)->last_closep) : -1), \
645 is_inf ? "INF " : "" \
647 if ((data)->last_found) \
648 PerlIO_printf(Perl_debug_log, \
649 "Last:'%s' %"IVdf":%"IVdf"/%"IVdf" %sFixed:'%s' @ %"IVdf \
650 " %sFloat: '%s' @ %"IVdf"/%"IVdf"", \
651 SvPVX_const((data)->last_found), \
652 (IV)((data)->last_end), \
653 (IV)((data)->last_start_min), \
654 (IV)((data)->last_start_max), \
655 ((data)->longest && \
656 (data)->longest==&((data)->longest_fixed)) ? "*" : "", \
657 SvPVX_const((data)->longest_fixed), \
658 (IV)((data)->offset_fixed), \
659 ((data)->longest && \
660 (data)->longest==&((data)->longest_float)) ? "*" : "", \
661 SvPVX_const((data)->longest_float), \
662 (IV)((data)->offset_float_min), \
663 (IV)((data)->offset_float_max) \
665 PerlIO_printf(Perl_debug_log,"\n"); \
668 static void clear_re(pTHX_ void *r);
670 /* Mark that we cannot extend a found fixed substring at this point.
671 Update the longest found anchored substring and the longest found
672 floating substrings if needed. */
675 S_scan_commit(pTHX_ const RExC_state_t *pRExC_state, scan_data_t *data, I32 *minlenp, int is_inf)
677 const STRLEN l = CHR_SVLEN(data->last_found);
678 const STRLEN old_l = CHR_SVLEN(*data->longest);
679 GET_RE_DEBUG_FLAGS_DECL;
681 PERL_ARGS_ASSERT_SCAN_COMMIT;
683 if ((l >= old_l) && ((l > old_l) || (data->flags & SF_BEFORE_EOL))) {
684 SvSetMagicSV(*data->longest, data->last_found);
685 if (*data->longest == data->longest_fixed) {
686 data->offset_fixed = l ? data->last_start_min : data->pos_min;
687 if (data->flags & SF_BEFORE_EOL)
689 |= ((data->flags & SF_BEFORE_EOL) << SF_FIX_SHIFT_EOL);
691 data->flags &= ~SF_FIX_BEFORE_EOL;
692 data->minlen_fixed=minlenp;
693 data->lookbehind_fixed=0;
695 else { /* *data->longest == data->longest_float */
696 data->offset_float_min = l ? data->last_start_min : data->pos_min;
697 data->offset_float_max = (l
698 ? data->last_start_max
699 : data->pos_min + data->pos_delta);
700 if (is_inf || (U32)data->offset_float_max > (U32)I32_MAX)
701 data->offset_float_max = I32_MAX;
702 if (data->flags & SF_BEFORE_EOL)
704 |= ((data->flags & SF_BEFORE_EOL) << SF_FL_SHIFT_EOL);
706 data->flags &= ~SF_FL_BEFORE_EOL;
707 data->minlen_float=minlenp;
708 data->lookbehind_float=0;
711 SvCUR_set(data->last_found, 0);
713 SV * const sv = data->last_found;
714 if (SvUTF8(sv) && SvMAGICAL(sv)) {
715 MAGIC * const mg = mg_find(sv, PERL_MAGIC_utf8);
721 data->flags &= ~SF_BEFORE_EOL;
722 DEBUG_STUDYDATA("commit: ",data,0);
725 /* Can match anything (initialization) */
727 S_cl_anything(const RExC_state_t *pRExC_state, struct regnode_charclass_class *cl)
729 PERL_ARGS_ASSERT_CL_ANYTHING;
731 ANYOF_BITMAP_SETALL(cl);
732 cl->flags = ANYOF_CLASS|ANYOF_EOS|ANYOF_UNICODE_ALL
733 |ANYOF_LOC_NONBITMAP_FOLD|ANYOF_NON_UTF8_LATIN1_ALL;
735 /* If any portion of the regex is to operate under locale rules,
736 * initialization includes it. The reason this isn't done for all regexes
737 * is that the optimizer was written under the assumption that locale was
738 * all-or-nothing. Given the complexity and lack of documentation in the
739 * optimizer, and that there are inadequate test cases for locale, so many
740 * parts of it may not work properly, it is safest to avoid locale unless
742 if (RExC_contains_locale) {
743 ANYOF_CLASS_SETALL(cl); /* /l uses class */
744 cl->flags |= ANYOF_LOCALE;
747 ANYOF_CLASS_ZERO(cl); /* Only /l uses class now */
751 /* Can match anything (initialization) */
753 S_cl_is_anything(const struct regnode_charclass_class *cl)
757 PERL_ARGS_ASSERT_CL_IS_ANYTHING;
759 for (value = 0; value <= ANYOF_MAX; value += 2)
760 if (ANYOF_CLASS_TEST(cl, value) && ANYOF_CLASS_TEST(cl, value + 1))
762 if (!(cl->flags & ANYOF_UNICODE_ALL))
764 if (!ANYOF_BITMAP_TESTALLSET((const void*)cl))
769 /* Can match anything (initialization) */
771 S_cl_init(const RExC_state_t *pRExC_state, struct regnode_charclass_class *cl)
773 PERL_ARGS_ASSERT_CL_INIT;
775 Zero(cl, 1, struct regnode_charclass_class);
777 cl_anything(pRExC_state, cl);
778 ARG_SET(cl, ANYOF_NONBITMAP_EMPTY);
781 /* These two functions currently do the exact same thing */
782 #define cl_init_zero S_cl_init
784 /* 'AND' a given class with another one. Can create false positives. 'cl'
785 * should not be inverted. 'and_with->flags & ANYOF_CLASS' should be 0 if
786 * 'and_with' is a regnode_charclass instead of a regnode_charclass_class. */
788 S_cl_and(struct regnode_charclass_class *cl,
789 const struct regnode_charclass_class *and_with)
791 PERL_ARGS_ASSERT_CL_AND;
793 assert(and_with->type == ANYOF);
795 /* I (khw) am not sure all these restrictions are necessary XXX */
796 if (!(ANYOF_CLASS_TEST_ANY_SET(and_with))
797 && !(ANYOF_CLASS_TEST_ANY_SET(cl))
798 && (and_with->flags & ANYOF_LOCALE) == (cl->flags & ANYOF_LOCALE)
799 && !(and_with->flags & ANYOF_LOC_NONBITMAP_FOLD)
800 && !(cl->flags & ANYOF_LOC_NONBITMAP_FOLD)) {
803 if (and_with->flags & ANYOF_INVERT)
804 for (i = 0; i < ANYOF_BITMAP_SIZE; i++)
805 cl->bitmap[i] &= ~and_with->bitmap[i];
807 for (i = 0; i < ANYOF_BITMAP_SIZE; i++)
808 cl->bitmap[i] &= and_with->bitmap[i];
809 } /* XXXX: logic is complicated otherwise, leave it along for a moment. */
811 if (and_with->flags & ANYOF_INVERT) {
813 /* Here, the and'ed node is inverted. Get the AND of the flags that
814 * aren't affected by the inversion. Those that are affected are
815 * handled individually below */
816 U8 affected_flags = cl->flags & ~INVERSION_UNAFFECTED_FLAGS;
817 cl->flags &= (and_with->flags & INVERSION_UNAFFECTED_FLAGS);
818 cl->flags |= affected_flags;
820 /* We currently don't know how to deal with things that aren't in the
821 * bitmap, but we know that the intersection is no greater than what
822 * is already in cl, so let there be false positives that get sorted
823 * out after the synthetic start class succeeds, and the node is
824 * matched for real. */
826 /* The inversion of these two flags indicate that the resulting
827 * intersection doesn't have them */
828 if (and_with->flags & ANYOF_UNICODE_ALL) {
829 cl->flags &= ~ANYOF_UNICODE_ALL;
831 if (and_with->flags & ANYOF_NON_UTF8_LATIN1_ALL) {
832 cl->flags &= ~ANYOF_NON_UTF8_LATIN1_ALL;
835 else { /* and'd node is not inverted */
836 U8 outside_bitmap_but_not_utf8; /* Temp variable */
838 if (! ANYOF_NONBITMAP(and_with)) {
840 /* Here 'and_with' doesn't match anything outside the bitmap
841 * (except possibly ANYOF_UNICODE_ALL), which means the
842 * intersection can't either, except for ANYOF_UNICODE_ALL, in
843 * which case we don't know what the intersection is, but it's no
844 * greater than what cl already has, so can just leave it alone,
845 * with possible false positives */
846 if (! (and_with->flags & ANYOF_UNICODE_ALL)) {
847 ARG_SET(cl, ANYOF_NONBITMAP_EMPTY);
848 cl->flags &= ~ANYOF_NONBITMAP_NON_UTF8;
851 else if (! ANYOF_NONBITMAP(cl)) {
853 /* Here, 'and_with' does match something outside the bitmap, and cl
854 * doesn't have a list of things to match outside the bitmap. If
855 * cl can match all code points above 255, the intersection will
856 * be those above-255 code points that 'and_with' matches. If cl
857 * can't match all Unicode code points, it means that it can't
858 * match anything outside the bitmap (since the 'if' that got us
859 * into this block tested for that), so we leave the bitmap empty.
861 if (cl->flags & ANYOF_UNICODE_ALL) {
862 ARG_SET(cl, ARG(and_with));
864 /* and_with's ARG may match things that don't require UTF8.
865 * And now cl's will too, in spite of this being an 'and'. See
866 * the comments below about the kludge */
867 cl->flags |= and_with->flags & ANYOF_NONBITMAP_NON_UTF8;
871 /* Here, both 'and_with' and cl match something outside the
872 * bitmap. Currently we do not do the intersection, so just match
873 * whatever cl had at the beginning. */
877 /* Take the intersection of the two sets of flags. However, the
878 * ANYOF_NONBITMAP_NON_UTF8 flag is treated as an 'or'. This is a
879 * kludge around the fact that this flag is not treated like the others
880 * which are initialized in cl_anything(). The way the optimizer works
881 * is that the synthetic start class (SSC) is initialized to match
882 * anything, and then the first time a real node is encountered, its
883 * values are AND'd with the SSC's with the result being the values of
884 * the real node. However, there are paths through the optimizer where
885 * the AND never gets called, so those initialized bits are set
886 * inappropriately, which is not usually a big deal, as they just cause
887 * false positives in the SSC, which will just mean a probably
888 * imperceptible slow down in execution. However this bit has a
889 * higher false positive consequence in that it can cause utf8.pm,
890 * utf8_heavy.pl ... to be loaded when not necessary, which is a much
891 * bigger slowdown and also causes significant extra memory to be used.
892 * In order to prevent this, the code now takes a different tack. The
893 * bit isn't set unless some part of the regular expression needs it,
894 * but once set it won't get cleared. This means that these extra
895 * modules won't get loaded unless there was some path through the
896 * pattern that would have required them anyway, and so any false
897 * positives that occur by not ANDing them out when they could be
898 * aren't as severe as they would be if we treated this bit like all
900 outside_bitmap_but_not_utf8 = (cl->flags | and_with->flags)
901 & ANYOF_NONBITMAP_NON_UTF8;
902 cl->flags &= and_with->flags;
903 cl->flags |= outside_bitmap_but_not_utf8;
907 /* 'OR' a given class with another one. Can create false positives. 'cl'
908 * should not be inverted. 'or_with->flags & ANYOF_CLASS' should be 0 if
909 * 'or_with' is a regnode_charclass instead of a regnode_charclass_class. */
911 S_cl_or(const RExC_state_t *pRExC_state, struct regnode_charclass_class *cl, const struct regnode_charclass_class *or_with)
913 PERL_ARGS_ASSERT_CL_OR;
915 if (or_with->flags & ANYOF_INVERT) {
917 /* Here, the or'd node is to be inverted. This means we take the
918 * complement of everything not in the bitmap, but currently we don't
919 * know what that is, so give up and match anything */
920 if (ANYOF_NONBITMAP(or_with)) {
921 cl_anything(pRExC_state, cl);
924 * (B1 | CL1) | (!B2 & !CL2) = (B1 | !B2 & !CL2) | (CL1 | (!B2 & !CL2))
925 * <= (B1 | !B2) | (CL1 | !CL2)
926 * which is wasteful if CL2 is small, but we ignore CL2:
927 * (B1 | CL1) | (!B2 & !CL2) <= (B1 | CL1) | !B2 = (B1 | !B2) | CL1
928 * XXXX Can we handle case-fold? Unclear:
929 * (OK1(i) | OK1(i')) | !(OK1(i) | OK1(i')) =
930 * (OK1(i) | OK1(i')) | (!OK1(i) & !OK1(i'))
932 else if ( (or_with->flags & ANYOF_LOCALE) == (cl->flags & ANYOF_LOCALE)
933 && !(or_with->flags & ANYOF_LOC_NONBITMAP_FOLD)
934 && !(cl->flags & ANYOF_LOC_NONBITMAP_FOLD) ) {
937 for (i = 0; i < ANYOF_BITMAP_SIZE; i++)
938 cl->bitmap[i] |= ~or_with->bitmap[i];
939 } /* XXXX: logic is complicated otherwise */
941 cl_anything(pRExC_state, cl);
944 /* And, we can just take the union of the flags that aren't affected
945 * by the inversion */
946 cl->flags |= or_with->flags & INVERSION_UNAFFECTED_FLAGS;
948 /* For the remaining flags:
949 ANYOF_UNICODE_ALL and inverted means to not match anything above
950 255, which means that the union with cl should just be
951 what cl has in it, so can ignore this flag
952 ANYOF_NON_UTF8_LATIN1_ALL and inverted means if not utf8 and ord
953 is 127-255 to match them, but then invert that, so the
954 union with cl should just be what cl has in it, so can
957 } else { /* 'or_with' is not inverted */
958 /* (B1 | CL1) | (B2 | CL2) = (B1 | B2) | (CL1 | CL2)) */
959 if ( (or_with->flags & ANYOF_LOCALE) == (cl->flags & ANYOF_LOCALE)
960 && (!(or_with->flags & ANYOF_LOC_NONBITMAP_FOLD)
961 || (cl->flags & ANYOF_LOC_NONBITMAP_FOLD)) ) {
964 /* OR char bitmap and class bitmap separately */
965 for (i = 0; i < ANYOF_BITMAP_SIZE; i++)
966 cl->bitmap[i] |= or_with->bitmap[i];
967 if (ANYOF_CLASS_TEST_ANY_SET(or_with)) {
968 for (i = 0; i < ANYOF_CLASSBITMAP_SIZE; i++)
969 cl->classflags[i] |= or_with->classflags[i];
970 cl->flags |= ANYOF_CLASS;
973 else { /* XXXX: logic is complicated, leave it along for a moment. */
974 cl_anything(pRExC_state, cl);
977 if (ANYOF_NONBITMAP(or_with)) {
979 /* Use the added node's outside-the-bit-map match if there isn't a
980 * conflict. If there is a conflict (both nodes match something
981 * outside the bitmap, but what they match outside is not the same
982 * pointer, and hence not easily compared until XXX we extend
983 * inversion lists this far), give up and allow the start class to
984 * match everything outside the bitmap. If that stuff is all above
985 * 255, can just set UNICODE_ALL, otherwise caould be anything. */
986 if (! ANYOF_NONBITMAP(cl)) {
987 ARG_SET(cl, ARG(or_with));
989 else if (ARG(cl) != ARG(or_with)) {
991 if ((or_with->flags & ANYOF_NONBITMAP_NON_UTF8)) {
992 cl_anything(pRExC_state, cl);
995 cl->flags |= ANYOF_UNICODE_ALL;
1000 /* Take the union */
1001 cl->flags |= or_with->flags;
1005 #define TRIE_LIST_ITEM(state,idx) (trie->states[state].trans.list)[ idx ]
1006 #define TRIE_LIST_CUR(state) ( TRIE_LIST_ITEM( state, 0 ).forid )
1007 #define TRIE_LIST_LEN(state) ( TRIE_LIST_ITEM( state, 0 ).newstate )
1008 #define TRIE_LIST_USED(idx) ( trie->states[state].trans.list ? (TRIE_LIST_CUR( idx ) - 1) : 0 )
1013 dump_trie(trie,widecharmap,revcharmap)
1014 dump_trie_interim_list(trie,widecharmap,revcharmap,next_alloc)
1015 dump_trie_interim_table(trie,widecharmap,revcharmap,next_alloc)
1017 These routines dump out a trie in a somewhat readable format.
1018 The _interim_ variants are used for debugging the interim
1019 tables that are used to generate the final compressed
1020 representation which is what dump_trie expects.
1022 Part of the reason for their existence is to provide a form
1023 of documentation as to how the different representations function.
1028 Dumps the final compressed table form of the trie to Perl_debug_log.
1029 Used for debugging make_trie().
1033 S_dump_trie(pTHX_ const struct _reg_trie_data *trie, HV *widecharmap,
1034 AV *revcharmap, U32 depth)
1037 SV *sv=sv_newmortal();
1038 int colwidth= widecharmap ? 6 : 4;
1040 GET_RE_DEBUG_FLAGS_DECL;
1042 PERL_ARGS_ASSERT_DUMP_TRIE;
1044 PerlIO_printf( Perl_debug_log, "%*sChar : %-6s%-6s%-4s ",
1045 (int)depth * 2 + 2,"",
1046 "Match","Base","Ofs" );
1048 for( state = 0 ; state < trie->uniquecharcount ; state++ ) {
1049 SV ** const tmp = av_fetch( revcharmap, state, 0);
1051 PerlIO_printf( Perl_debug_log, "%*s",
1053 pv_pretty(sv, SvPV_nolen_const(*tmp), SvCUR(*tmp), colwidth,
1054 PL_colors[0], PL_colors[1],
1055 (SvUTF8(*tmp) ? PERL_PV_ESCAPE_UNI : 0) |
1056 PERL_PV_ESCAPE_FIRSTCHAR
1061 PerlIO_printf( Perl_debug_log, "\n%*sState|-----------------------",
1062 (int)depth * 2 + 2,"");
1064 for( state = 0 ; state < trie->uniquecharcount ; state++ )
1065 PerlIO_printf( Perl_debug_log, "%.*s", colwidth, "--------");
1066 PerlIO_printf( Perl_debug_log, "\n");
1068 for( state = 1 ; state < trie->statecount ; state++ ) {
1069 const U32 base = trie->states[ state ].trans.base;
1071 PerlIO_printf( Perl_debug_log, "%*s#%4"UVXf"|", (int)depth * 2 + 2,"", (UV)state);
1073 if ( trie->states[ state ].wordnum ) {
1074 PerlIO_printf( Perl_debug_log, " W%4X", trie->states[ state ].wordnum );
1076 PerlIO_printf( Perl_debug_log, "%6s", "" );
1079 PerlIO_printf( Perl_debug_log, " @%4"UVXf" ", (UV)base );
1084 while( ( base + ofs < trie->uniquecharcount ) ||
1085 ( base + ofs - trie->uniquecharcount < trie->lasttrans
1086 && trie->trans[ base + ofs - trie->uniquecharcount ].check != state))
1089 PerlIO_printf( Perl_debug_log, "+%2"UVXf"[ ", (UV)ofs);
1091 for ( ofs = 0 ; ofs < trie->uniquecharcount ; ofs++ ) {
1092 if ( ( base + ofs >= trie->uniquecharcount ) &&
1093 ( base + ofs - trie->uniquecharcount < trie->lasttrans ) &&
1094 trie->trans[ base + ofs - trie->uniquecharcount ].check == state )
1096 PerlIO_printf( Perl_debug_log, "%*"UVXf,
1098 (UV)trie->trans[ base + ofs - trie->uniquecharcount ].next );
1100 PerlIO_printf( Perl_debug_log, "%*s",colwidth," ." );
1104 PerlIO_printf( Perl_debug_log, "]");
1107 PerlIO_printf( Perl_debug_log, "\n" );
1109 PerlIO_printf(Perl_debug_log, "%*sword_info N:(prev,len)=", (int)depth*2, "");
1110 for (word=1; word <= trie->wordcount; word++) {
1111 PerlIO_printf(Perl_debug_log, " %d:(%d,%d)",
1112 (int)word, (int)(trie->wordinfo[word].prev),
1113 (int)(trie->wordinfo[word].len));
1115 PerlIO_printf(Perl_debug_log, "\n" );
1118 Dumps a fully constructed but uncompressed trie in list form.
1119 List tries normally only are used for construction when the number of
1120 possible chars (trie->uniquecharcount) is very high.
1121 Used for debugging make_trie().
1124 S_dump_trie_interim_list(pTHX_ const struct _reg_trie_data *trie,
1125 HV *widecharmap, AV *revcharmap, U32 next_alloc,
1129 SV *sv=sv_newmortal();
1130 int colwidth= widecharmap ? 6 : 4;
1131 GET_RE_DEBUG_FLAGS_DECL;
1133 PERL_ARGS_ASSERT_DUMP_TRIE_INTERIM_LIST;
1135 /* print out the table precompression. */
1136 PerlIO_printf( Perl_debug_log, "%*sState :Word | Transition Data\n%*s%s",
1137 (int)depth * 2 + 2,"", (int)depth * 2 + 2,"",
1138 "------:-----+-----------------\n" );
1140 for( state=1 ; state < next_alloc ; state ++ ) {
1143 PerlIO_printf( Perl_debug_log, "%*s %4"UVXf" :",
1144 (int)depth * 2 + 2,"", (UV)state );
1145 if ( ! trie->states[ state ].wordnum ) {
1146 PerlIO_printf( Perl_debug_log, "%5s| ","");
1148 PerlIO_printf( Perl_debug_log, "W%4x| ",
1149 trie->states[ state ].wordnum
1152 for( charid = 1 ; charid <= TRIE_LIST_USED( state ) ; charid++ ) {
1153 SV ** const tmp = av_fetch( revcharmap, TRIE_LIST_ITEM(state,charid).forid, 0);
1155 PerlIO_printf( Perl_debug_log, "%*s:%3X=%4"UVXf" | ",
1157 pv_pretty(sv, SvPV_nolen_const(*tmp), SvCUR(*tmp), colwidth,
1158 PL_colors[0], PL_colors[1],
1159 (SvUTF8(*tmp) ? PERL_PV_ESCAPE_UNI : 0) |
1160 PERL_PV_ESCAPE_FIRSTCHAR
1162 TRIE_LIST_ITEM(state,charid).forid,
1163 (UV)TRIE_LIST_ITEM(state,charid).newstate
1166 PerlIO_printf(Perl_debug_log, "\n%*s| ",
1167 (int)((depth * 2) + 14), "");
1170 PerlIO_printf( Perl_debug_log, "\n");
1175 Dumps a fully constructed but uncompressed trie in table form.
1176 This is the normal DFA style state transition table, with a few
1177 twists to facilitate compression later.
1178 Used for debugging make_trie().
1181 S_dump_trie_interim_table(pTHX_ const struct _reg_trie_data *trie,
1182 HV *widecharmap, AV *revcharmap, U32 next_alloc,
1187 SV *sv=sv_newmortal();
1188 int colwidth= widecharmap ? 6 : 4;
1189 GET_RE_DEBUG_FLAGS_DECL;
1191 PERL_ARGS_ASSERT_DUMP_TRIE_INTERIM_TABLE;
1194 print out the table precompression so that we can do a visual check
1195 that they are identical.
1198 PerlIO_printf( Perl_debug_log, "%*sChar : ",(int)depth * 2 + 2,"" );
1200 for( charid = 0 ; charid < trie->uniquecharcount ; charid++ ) {
1201 SV ** const tmp = av_fetch( revcharmap, charid, 0);
1203 PerlIO_printf( Perl_debug_log, "%*s",
1205 pv_pretty(sv, SvPV_nolen_const(*tmp), SvCUR(*tmp), colwidth,
1206 PL_colors[0], PL_colors[1],
1207 (SvUTF8(*tmp) ? PERL_PV_ESCAPE_UNI : 0) |
1208 PERL_PV_ESCAPE_FIRSTCHAR
1214 PerlIO_printf( Perl_debug_log, "\n%*sState+-",(int)depth * 2 + 2,"" );
1216 for( charid=0 ; charid < trie->uniquecharcount ; charid++ ) {
1217 PerlIO_printf( Perl_debug_log, "%.*s", colwidth,"--------");
1220 PerlIO_printf( Perl_debug_log, "\n" );
1222 for( state=1 ; state < next_alloc ; state += trie->uniquecharcount ) {
1224 PerlIO_printf( Perl_debug_log, "%*s%4"UVXf" : ",
1225 (int)depth * 2 + 2,"",
1226 (UV)TRIE_NODENUM( state ) );
1228 for( charid = 0 ; charid < trie->uniquecharcount ; charid++ ) {
1229 UV v=(UV)SAFE_TRIE_NODENUM( trie->trans[ state + charid ].next );
1231 PerlIO_printf( Perl_debug_log, "%*"UVXf, colwidth, v );
1233 PerlIO_printf( Perl_debug_log, "%*s", colwidth, "." );
1235 if ( ! trie->states[ TRIE_NODENUM( state ) ].wordnum ) {
1236 PerlIO_printf( Perl_debug_log, " (%4"UVXf")\n", (UV)trie->trans[ state ].check );
1238 PerlIO_printf( Perl_debug_log, " (%4"UVXf") W%4X\n", (UV)trie->trans[ state ].check,
1239 trie->states[ TRIE_NODENUM( state ) ].wordnum );
1247 /* make_trie(startbranch,first,last,tail,word_count,flags,depth)
1248 startbranch: the first branch in the whole branch sequence
1249 first : start branch of sequence of branch-exact nodes.
1250 May be the same as startbranch
1251 last : Thing following the last branch.
1252 May be the same as tail.
1253 tail : item following the branch sequence
1254 count : words in the sequence
1255 flags : currently the OP() type we will be building one of /EXACT(|F|Fl)/
1256 depth : indent depth
1258 Inplace optimizes a sequence of 2 or more Branch-Exact nodes into a TRIE node.
1260 A trie is an N'ary tree where the branches are determined by digital
1261 decomposition of the key. IE, at the root node you look up the 1st character and
1262 follow that branch repeat until you find the end of the branches. Nodes can be
1263 marked as "accepting" meaning they represent a complete word. Eg:
1267 would convert into the following structure. Numbers represent states, letters
1268 following numbers represent valid transitions on the letter from that state, if
1269 the number is in square brackets it represents an accepting state, otherwise it
1270 will be in parenthesis.
1272 +-h->+-e->[3]-+-r->(8)-+-s->[9]
1276 (1) +-i->(6)-+-s->[7]
1278 +-s->(3)-+-h->(4)-+-e->[5]
1280 Accept Word Mapping: 3=>1 (he),5=>2 (she), 7=>3 (his), 9=>4 (hers)
1282 This shows that when matching against the string 'hers' we will begin at state 1
1283 read 'h' and move to state 2, read 'e' and move to state 3 which is accepting,
1284 then read 'r' and go to state 8 followed by 's' which takes us to state 9 which
1285 is also accepting. Thus we know that we can match both 'he' and 'hers' with a
1286 single traverse. We store a mapping from accepting to state to which word was
1287 matched, and then when we have multiple possibilities we try to complete the
1288 rest of the regex in the order in which they occured in the alternation.
1290 The only prior NFA like behaviour that would be changed by the TRIE support is
1291 the silent ignoring of duplicate alternations which are of the form:
1293 / (DUPE|DUPE) X? (?{ ... }) Y /x
1295 Thus EVAL blocks following a trie may be called a different number of times with
1296 and without the optimisation. With the optimisations dupes will be silently
1297 ignored. This inconsistent behaviour of EVAL type nodes is well established as
1298 the following demonstrates:
1300 'words'=~/(word|word|word)(?{ print $1 })[xyz]/
1302 which prints out 'word' three times, but
1304 'words'=~/(word|word|word)(?{ print $1 })S/
1306 which doesnt print it out at all. This is due to other optimisations kicking in.
1308 Example of what happens on a structural level:
1310 The regexp /(ac|ad|ab)+/ will produce the following debug output:
1312 1: CURLYM[1] {1,32767}(18)
1323 This would be optimizable with startbranch=5, first=5, last=16, tail=16
1324 and should turn into:
1326 1: CURLYM[1] {1,32767}(18)
1328 [Words:3 Chars Stored:6 Unique Chars:4 States:5 NCP:1]
1336 Cases where tail != last would be like /(?foo|bar)baz/:
1346 which would be optimizable with startbranch=1, first=1, last=7, tail=8
1347 and would end up looking like:
1350 [Words:2 Chars Stored:6 Unique Chars:5 States:7 NCP:1]
1357 d = uvuni_to_utf8_flags(d, uv, 0);
1359 is the recommended Unicode-aware way of saying
1364 #define TRIE_STORE_REVCHAR \
1367 SV *zlopp = newSV(2); \
1368 unsigned char *flrbbbbb = (unsigned char *) SvPVX(zlopp); \
1369 unsigned const char *const kapow = uvuni_to_utf8(flrbbbbb, uvc & 0xFF); \
1370 SvCUR_set(zlopp, kapow - flrbbbbb); \
1373 av_push(revcharmap, zlopp); \
1375 char ooooff = (char)uvc; \
1376 av_push(revcharmap, newSVpvn(&ooooff, 1)); \
1380 #define TRIE_READ_CHAR STMT_START { \
1384 if ( foldlen > 0 ) { \
1385 uvc = utf8n_to_uvuni( scan, UTF8_MAXLEN, &len, uniflags ); \
1390 len = UTF8SKIP(uc);\
1391 uvc = to_utf8_fold( uc, foldbuf, &foldlen); \
1392 foldlen -= UNISKIP( uvc ); \
1393 scan = foldbuf + UNISKIP( uvc ); \
1396 uvc = utf8n_to_uvuni( (const U8*)uc, UTF8_MAXLEN, &len, uniflags);\
1406 #define TRIE_LIST_PUSH(state,fid,ns) STMT_START { \
1407 if ( TRIE_LIST_CUR( state ) >=TRIE_LIST_LEN( state ) ) { \
1408 U32 ging = TRIE_LIST_LEN( state ) *= 2; \
1409 Renew( trie->states[ state ].trans.list, ging, reg_trie_trans_le ); \
1411 TRIE_LIST_ITEM( state, TRIE_LIST_CUR( state ) ).forid = fid; \
1412 TRIE_LIST_ITEM( state, TRIE_LIST_CUR( state ) ).newstate = ns; \
1413 TRIE_LIST_CUR( state )++; \
1416 #define TRIE_LIST_NEW(state) STMT_START { \
1417 Newxz( trie->states[ state ].trans.list, \
1418 4, reg_trie_trans_le ); \
1419 TRIE_LIST_CUR( state ) = 1; \
1420 TRIE_LIST_LEN( state ) = 4; \
1423 #define TRIE_HANDLE_WORD(state) STMT_START { \
1424 U16 dupe= trie->states[ state ].wordnum; \
1425 regnode * const noper_next = regnext( noper ); \
1428 /* store the word for dumping */ \
1430 if (OP(noper) != NOTHING) \
1431 tmp = newSVpvn_utf8(STRING(noper), STR_LEN(noper), UTF); \
1433 tmp = newSVpvn_utf8( "", 0, UTF ); \
1434 av_push( trie_words, tmp ); \
1438 trie->wordinfo[curword].prev = 0; \
1439 trie->wordinfo[curword].len = wordlen; \
1440 trie->wordinfo[curword].accept = state; \
1442 if ( noper_next < tail ) { \
1444 trie->jump = (U16 *) PerlMemShared_calloc( word_count + 1, sizeof(U16) ); \
1445 trie->jump[curword] = (U16)(noper_next - convert); \
1447 jumper = noper_next; \
1449 nextbranch= regnext(cur); \
1453 /* It's a dupe. Pre-insert into the wordinfo[].prev */\
1454 /* chain, so that when the bits of chain are later */\
1455 /* linked together, the dups appear in the chain */\
1456 trie->wordinfo[curword].prev = trie->wordinfo[dupe].prev; \
1457 trie->wordinfo[dupe].prev = curword; \
1459 /* we haven't inserted this word yet. */ \
1460 trie->states[ state ].wordnum = curword; \
1465 #define TRIE_TRANS_STATE(state,base,ucharcount,charid,special) \
1466 ( ( base + charid >= ucharcount \
1467 && base + charid < ubound \
1468 && state == trie->trans[ base - ucharcount + charid ].check \
1469 && trie->trans[ base - ucharcount + charid ].next ) \
1470 ? trie->trans[ base - ucharcount + charid ].next \
1471 : ( state==1 ? special : 0 ) \
1475 #define MADE_JUMP_TRIE 2
1476 #define MADE_EXACT_TRIE 4
1479 S_make_trie(pTHX_ RExC_state_t *pRExC_state, regnode *startbranch, regnode *first, regnode *last, regnode *tail, U32 word_count, U32 flags, U32 depth)
1482 /* first pass, loop through and scan words */
1483 reg_trie_data *trie;
1484 HV *widecharmap = NULL;
1485 AV *revcharmap = newAV();
1487 const U32 uniflags = UTF8_ALLOW_DEFAULT;
1492 regnode *jumper = NULL;
1493 regnode *nextbranch = NULL;
1494 regnode *convert = NULL;
1495 U32 *prev_states; /* temp array mapping each state to previous one */
1496 /* we just use folder as a flag in utf8 */
1497 const U8 * folder = NULL;
1500 const U32 data_slot = add_data( pRExC_state, 4, "tuuu" );
1501 AV *trie_words = NULL;
1502 /* along with revcharmap, this only used during construction but both are
1503 * useful during debugging so we store them in the struct when debugging.
1506 const U32 data_slot = add_data( pRExC_state, 2, "tu" );
1507 STRLEN trie_charcount=0;
1509 SV *re_trie_maxbuff;
1510 GET_RE_DEBUG_FLAGS_DECL;
1512 PERL_ARGS_ASSERT_MAKE_TRIE;
1514 PERL_UNUSED_ARG(depth);
1520 case EXACTFU: folder = PL_fold_latin1; break;
1521 case EXACTF: folder = PL_fold; break;
1522 case EXACTFL: folder = PL_fold_locale; break;
1523 default: Perl_croak( aTHX_ "panic! In trie construction, unknown node type %u", (unsigned) flags );
1526 trie = (reg_trie_data *) PerlMemShared_calloc( 1, sizeof(reg_trie_data) );
1528 trie->startstate = 1;
1529 trie->wordcount = word_count;
1530 RExC_rxi->data->data[ data_slot ] = (void*)trie;
1531 trie->charmap = (U16 *) PerlMemShared_calloc( 256, sizeof(U16) );
1532 if (!(UTF && folder))
1533 trie->bitmap = (char *) PerlMemShared_calloc( ANYOF_BITMAP_SIZE, 1 );
1534 trie->wordinfo = (reg_trie_wordinfo *) PerlMemShared_calloc(
1535 trie->wordcount+1, sizeof(reg_trie_wordinfo));
1538 trie_words = newAV();
1541 re_trie_maxbuff = get_sv(RE_TRIE_MAXBUF_NAME, 1);
1542 if (!SvIOK(re_trie_maxbuff)) {
1543 sv_setiv(re_trie_maxbuff, RE_TRIE_MAXBUF_INIT);
1546 PerlIO_printf( Perl_debug_log,
1547 "%*smake_trie start==%d, first==%d, last==%d, tail==%d depth=%d\n",
1548 (int)depth * 2 + 2, "",
1549 REG_NODE_NUM(startbranch),REG_NODE_NUM(first),
1550 REG_NODE_NUM(last), REG_NODE_NUM(tail),
1554 /* Find the node we are going to overwrite */
1555 if ( first == startbranch && OP( last ) != BRANCH ) {
1556 /* whole branch chain */
1559 /* branch sub-chain */
1560 convert = NEXTOPER( first );
1563 /* -- First loop and Setup --
1565 We first traverse the branches and scan each word to determine if it
1566 contains widechars, and how many unique chars there are, this is
1567 important as we have to build a table with at least as many columns as we
1570 We use an array of integers to represent the character codes 0..255
1571 (trie->charmap) and we use a an HV* to store Unicode characters. We use the
1572 native representation of the character value as the key and IV's for the
1575 *TODO* If we keep track of how many times each character is used we can
1576 remap the columns so that the table compression later on is more
1577 efficient in terms of memory by ensuring the most common value is in the
1578 middle and the least common are on the outside. IMO this would be better
1579 than a most to least common mapping as theres a decent chance the most
1580 common letter will share a node with the least common, meaning the node
1581 will not be compressible. With a middle is most common approach the worst
1582 case is when we have the least common nodes twice.
1586 for ( cur = first ; cur < last ; cur = regnext( cur ) ) {
1587 regnode * const noper = NEXTOPER( cur );
1588 const U8 *uc = (U8*)STRING( noper );
1589 const U8 * const e = uc + STR_LEN( noper );
1591 U8 foldbuf[ UTF8_MAXBYTES_CASE + 1 ];
1592 const U8 *scan = (U8*)NULL;
1593 U32 wordlen = 0; /* required init */
1595 bool set_bit = trie->bitmap ? 1 : 0; /*store the first char in the bitmap?*/
1597 if (OP(noper) == NOTHING) {
1601 if ( set_bit ) /* bitmap only alloced when !(UTF&&Folding) */
1602 TRIE_BITMAP_SET(trie,*uc); /* store the raw first byte
1603 regardless of encoding */
1605 for ( ; uc < e ; uc += len ) {
1606 TRIE_CHARCOUNT(trie)++;
1610 if ( !trie->charmap[ uvc ] ) {
1611 trie->charmap[ uvc ]=( ++trie->uniquecharcount );
1613 trie->charmap[ folder[ uvc ] ] = trie->charmap[ uvc ];
1617 /* store the codepoint in the bitmap, and its folded
1619 TRIE_BITMAP_SET(trie,uvc);
1621 /* store the folded codepoint */
1622 if ( folder ) TRIE_BITMAP_SET(trie,folder[ uvc ]);
1625 /* store first byte of utf8 representation of
1626 variant codepoints */
1627 if (! UNI_IS_INVARIANT(uvc)) {
1628 TRIE_BITMAP_SET(trie, UTF8_TWO_BYTE_HI(uvc));
1631 set_bit = 0; /* We've done our bit :-) */
1636 widecharmap = newHV();
1638 svpp = hv_fetch( widecharmap, (char*)&uvc, sizeof( UV ), 1 );
1641 Perl_croak( aTHX_ "error creating/fetching widecharmap entry for 0x%"UVXf, uvc );
1643 if ( !SvTRUE( *svpp ) ) {
1644 sv_setiv( *svpp, ++trie->uniquecharcount );
1649 if( cur == first ) {
1652 } else if (chars < trie->minlen) {
1654 } else if (chars > trie->maxlen) {
1658 } /* end first pass */
1659 DEBUG_TRIE_COMPILE_r(
1660 PerlIO_printf( Perl_debug_log, "%*sTRIE(%s): W:%d C:%d Uq:%d Min:%d Max:%d\n",
1661 (int)depth * 2 + 2,"",
1662 ( widecharmap ? "UTF8" : "NATIVE" ), (int)word_count,
1663 (int)TRIE_CHARCOUNT(trie), trie->uniquecharcount,
1664 (int)trie->minlen, (int)trie->maxlen )
1668 We now know what we are dealing with in terms of unique chars and
1669 string sizes so we can calculate how much memory a naive
1670 representation using a flat table will take. If it's over a reasonable
1671 limit (as specified by ${^RE_TRIE_MAXBUF}) we use a more memory
1672 conservative but potentially much slower representation using an array
1675 At the end we convert both representations into the same compressed
1676 form that will be used in regexec.c for matching with. The latter
1677 is a form that cannot be used to construct with but has memory
1678 properties similar to the list form and access properties similar
1679 to the table form making it both suitable for fast searches and
1680 small enough that its feasable to store for the duration of a program.
1682 See the comment in the code where the compressed table is produced
1683 inplace from the flat tabe representation for an explanation of how
1684 the compression works.
1689 Newx(prev_states, TRIE_CHARCOUNT(trie) + 2, U32);
1692 if ( (IV)( ( TRIE_CHARCOUNT(trie) + 1 ) * trie->uniquecharcount + 1) > SvIV(re_trie_maxbuff) ) {
1694 Second Pass -- Array Of Lists Representation
1696 Each state will be represented by a list of charid:state records
1697 (reg_trie_trans_le) the first such element holds the CUR and LEN
1698 points of the allocated array. (See defines above).
1700 We build the initial structure using the lists, and then convert
1701 it into the compressed table form which allows faster lookups
1702 (but cant be modified once converted).
1705 STRLEN transcount = 1;
1707 DEBUG_TRIE_COMPILE_MORE_r( PerlIO_printf( Perl_debug_log,
1708 "%*sCompiling trie using list compiler\n",
1709 (int)depth * 2 + 2, ""));
1711 trie->states = (reg_trie_state *)
1712 PerlMemShared_calloc( TRIE_CHARCOUNT(trie) + 2,
1713 sizeof(reg_trie_state) );
1717 for ( cur = first ; cur < last ; cur = regnext( cur ) ) {
1719 regnode * const noper = NEXTOPER( cur );
1720 U8 *uc = (U8*)STRING( noper );
1721 const U8 * const e = uc + STR_LEN( noper );
1722 U32 state = 1; /* required init */
1723 U16 charid = 0; /* sanity init */
1724 U8 *scan = (U8*)NULL; /* sanity init */
1725 STRLEN foldlen = 0; /* required init */
1726 U32 wordlen = 0; /* required init */
1727 U8 foldbuf[ UTF8_MAXBYTES_CASE + 1 ];
1729 if (OP(noper) != NOTHING) {
1730 for ( ; uc < e ; uc += len ) {
1735 charid = trie->charmap[ uvc ];
1737 SV** const svpp = hv_fetch( widecharmap, (char*)&uvc, sizeof( UV ), 0);
1741 charid=(U16)SvIV( *svpp );
1744 /* charid is now 0 if we dont know the char read, or nonzero if we do */
1751 if ( !trie->states[ state ].trans.list ) {
1752 TRIE_LIST_NEW( state );
1754 for ( check = 1; check <= TRIE_LIST_USED( state ); check++ ) {
1755 if ( TRIE_LIST_ITEM( state, check ).forid == charid ) {
1756 newstate = TRIE_LIST_ITEM( state, check ).newstate;
1761 newstate = next_alloc++;
1762 prev_states[newstate] = state;
1763 TRIE_LIST_PUSH( state, charid, newstate );
1768 Perl_croak( aTHX_ "panic! In trie construction, no char mapping for %"IVdf, uvc );
1772 TRIE_HANDLE_WORD(state);
1774 } /* end second pass */
1776 /* next alloc is the NEXT state to be allocated */
1777 trie->statecount = next_alloc;
1778 trie->states = (reg_trie_state *)
1779 PerlMemShared_realloc( trie->states,
1781 * sizeof(reg_trie_state) );
1783 /* and now dump it out before we compress it */
1784 DEBUG_TRIE_COMPILE_MORE_r(dump_trie_interim_list(trie, widecharmap,
1785 revcharmap, next_alloc,
1789 trie->trans = (reg_trie_trans *)
1790 PerlMemShared_calloc( transcount, sizeof(reg_trie_trans) );
1797 for( state=1 ; state < next_alloc ; state ++ ) {
1801 DEBUG_TRIE_COMPILE_MORE_r(
1802 PerlIO_printf( Perl_debug_log, "tp: %d zp: %d ",tp,zp)
1806 if (trie->states[state].trans.list) {
1807 U16 minid=TRIE_LIST_ITEM( state, 1).forid;
1811 for( idx = 2 ; idx <= TRIE_LIST_USED( state ) ; idx++ ) {
1812 const U16 forid = TRIE_LIST_ITEM( state, idx).forid;
1813 if ( forid < minid ) {
1815 } else if ( forid > maxid ) {
1819 if ( transcount < tp + maxid - minid + 1) {
1821 trie->trans = (reg_trie_trans *)
1822 PerlMemShared_realloc( trie->trans,
1824 * sizeof(reg_trie_trans) );
1825 Zero( trie->trans + (transcount / 2), transcount / 2 , reg_trie_trans );
1827 base = trie->uniquecharcount + tp - minid;
1828 if ( maxid == minid ) {
1830 for ( ; zp < tp ; zp++ ) {
1831 if ( ! trie->trans[ zp ].next ) {
1832 base = trie->uniquecharcount + zp - minid;
1833 trie->trans[ zp ].next = TRIE_LIST_ITEM( state, 1).newstate;
1834 trie->trans[ zp ].check = state;
1840 trie->trans[ tp ].next = TRIE_LIST_ITEM( state, 1).newstate;
1841 trie->trans[ tp ].check = state;
1846 for ( idx=1; idx <= TRIE_LIST_USED( state ) ; idx++ ) {
1847 const U32 tid = base - trie->uniquecharcount + TRIE_LIST_ITEM( state, idx ).forid;
1848 trie->trans[ tid ].next = TRIE_LIST_ITEM( state, idx ).newstate;
1849 trie->trans[ tid ].check = state;
1851 tp += ( maxid - minid + 1 );
1853 Safefree(trie->states[ state ].trans.list);
1856 DEBUG_TRIE_COMPILE_MORE_r(
1857 PerlIO_printf( Perl_debug_log, " base: %d\n",base);
1860 trie->states[ state ].trans.base=base;
1862 trie->lasttrans = tp + 1;
1866 Second Pass -- Flat Table Representation.
1868 we dont use the 0 slot of either trans[] or states[] so we add 1 to each.
1869 We know that we will need Charcount+1 trans at most to store the data
1870 (one row per char at worst case) So we preallocate both structures
1871 assuming worst case.
1873 We then construct the trie using only the .next slots of the entry
1876 We use the .check field of the first entry of the node temporarily to
1877 make compression both faster and easier by keeping track of how many non
1878 zero fields are in the node.
1880 Since trans are numbered from 1 any 0 pointer in the table is a FAIL
1883 There are two terms at use here: state as a TRIE_NODEIDX() which is a
1884 number representing the first entry of the node, and state as a
1885 TRIE_NODENUM() which is the trans number. state 1 is TRIE_NODEIDX(1) and
1886 TRIE_NODENUM(1), state 2 is TRIE_NODEIDX(2) and TRIE_NODENUM(3) if there
1887 are 2 entrys per node. eg:
1895 The table is internally in the right hand, idx form. However as we also
1896 have to deal with the states array which is indexed by nodenum we have to
1897 use TRIE_NODENUM() to convert.
1900 DEBUG_TRIE_COMPILE_MORE_r( PerlIO_printf( Perl_debug_log,
1901 "%*sCompiling trie using table compiler\n",
1902 (int)depth * 2 + 2, ""));
1904 trie->trans = (reg_trie_trans *)
1905 PerlMemShared_calloc( ( TRIE_CHARCOUNT(trie) + 1 )
1906 * trie->uniquecharcount + 1,
1907 sizeof(reg_trie_trans) );
1908 trie->states = (reg_trie_state *)
1909 PerlMemShared_calloc( TRIE_CHARCOUNT(trie) + 2,
1910 sizeof(reg_trie_state) );
1911 next_alloc = trie->uniquecharcount + 1;
1914 for ( cur = first ; cur < last ; cur = regnext( cur ) ) {
1916 regnode * const noper = NEXTOPER( cur );
1917 const U8 *uc = (U8*)STRING( noper );
1918 const U8 * const e = uc + STR_LEN( noper );
1920 U32 state = 1; /* required init */
1922 U16 charid = 0; /* sanity init */
1923 U32 accept_state = 0; /* sanity init */
1924 U8 *scan = (U8*)NULL; /* sanity init */
1926 STRLEN foldlen = 0; /* required init */
1927 U32 wordlen = 0; /* required init */
1928 U8 foldbuf[ UTF8_MAXBYTES_CASE + 1 ];
1930 if ( OP(noper) != NOTHING ) {
1931 for ( ; uc < e ; uc += len ) {
1936 charid = trie->charmap[ uvc ];
1938 SV* const * const svpp = hv_fetch( widecharmap, (char*)&uvc, sizeof( UV ), 0);
1939 charid = svpp ? (U16)SvIV(*svpp) : 0;
1943 if ( !trie->trans[ state + charid ].next ) {
1944 trie->trans[ state + charid ].next = next_alloc;
1945 trie->trans[ state ].check++;
1946 prev_states[TRIE_NODENUM(next_alloc)]
1947 = TRIE_NODENUM(state);
1948 next_alloc += trie->uniquecharcount;
1950 state = trie->trans[ state + charid ].next;
1952 Perl_croak( aTHX_ "panic! In trie construction, no char mapping for %"IVdf, uvc );
1954 /* charid is now 0 if we dont know the char read, or nonzero if we do */
1957 accept_state = TRIE_NODENUM( state );
1958 TRIE_HANDLE_WORD(accept_state);
1960 } /* end second pass */
1962 /* and now dump it out before we compress it */
1963 DEBUG_TRIE_COMPILE_MORE_r(dump_trie_interim_table(trie, widecharmap,
1965 next_alloc, depth+1));
1969 * Inplace compress the table.*
1971 For sparse data sets the table constructed by the trie algorithm will
1972 be mostly 0/FAIL transitions or to put it another way mostly empty.
1973 (Note that leaf nodes will not contain any transitions.)
1975 This algorithm compresses the tables by eliminating most such
1976 transitions, at the cost of a modest bit of extra work during lookup:
1978 - Each states[] entry contains a .base field which indicates the
1979 index in the state[] array wheres its transition data is stored.
1981 - If .base is 0 there are no valid transitions from that node.
1983 - If .base is nonzero then charid is added to it to find an entry in
1986 -If trans[states[state].base+charid].check!=state then the
1987 transition is taken to be a 0/Fail transition. Thus if there are fail
1988 transitions at the front of the node then the .base offset will point
1989 somewhere inside the previous nodes data (or maybe even into a node
1990 even earlier), but the .check field determines if the transition is
1994 The following process inplace converts the table to the compressed
1995 table: We first do not compress the root node 1,and mark all its
1996 .check pointers as 1 and set its .base pointer as 1 as well. This
1997 allows us to do a DFA construction from the compressed table later,
1998 and ensures that any .base pointers we calculate later are greater
2001 - We set 'pos' to indicate the first entry of the second node.
2003 - We then iterate over the columns of the node, finding the first and
2004 last used entry at l and m. We then copy l..m into pos..(pos+m-l),
2005 and set the .check pointers accordingly, and advance pos
2006 appropriately and repreat for the next node. Note that when we copy
2007 the next pointers we have to convert them from the original
2008 NODEIDX form to NODENUM form as the former is not valid post
2011 - If a node has no transitions used we mark its base as 0 and do not
2012 advance the pos pointer.
2014 - If a node only has one transition we use a second pointer into the
2015 structure to fill in allocated fail transitions from other states.
2016 This pointer is independent of the main pointer and scans forward
2017 looking for null transitions that are allocated to a state. When it
2018 finds one it writes the single transition into the "hole". If the
2019 pointer doesnt find one the single transition is appended as normal.
2021 - Once compressed we can Renew/realloc the structures to release the
2024 See "Table-Compression Methods" in sec 3.9 of the Red Dragon,
2025 specifically Fig 3.47 and the associated pseudocode.
2029 const U32 laststate = TRIE_NODENUM( next_alloc );
2032 trie->statecount = laststate;
2034 for ( state = 1 ; state < laststate ; state++ ) {
2036 const U32 stateidx = TRIE_NODEIDX( state );
2037 const U32 o_used = trie->trans[ stateidx ].check;
2038 U32 used = trie->trans[ stateidx ].check;
2039 trie->trans[ stateidx ].check = 0;
2041 for ( charid = 0 ; used && charid < trie->uniquecharcount ; charid++ ) {
2042 if ( flag || trie->trans[ stateidx + charid ].next ) {
2043 if ( trie->trans[ stateidx + charid ].next ) {
2045 for ( ; zp < pos ; zp++ ) {
2046 if ( ! trie->trans[ zp ].next ) {
2050 trie->states[ state ].trans.base = zp + trie->uniquecharcount - charid ;
2051 trie->trans[ zp ].next = SAFE_TRIE_NODENUM( trie->trans[ stateidx + charid ].next );
2052 trie->trans[ zp ].check = state;
2053 if ( ++zp > pos ) pos = zp;
2060 trie->states[ state ].trans.base = pos + trie->uniquecharcount - charid ;
2062 trie->trans[ pos ].next = SAFE_TRIE_NODENUM( trie->trans[ stateidx + charid ].next );
2063 trie->trans[ pos ].check = state;
2068 trie->lasttrans = pos + 1;
2069 trie->states = (reg_trie_state *)
2070 PerlMemShared_realloc( trie->states, laststate
2071 * sizeof(reg_trie_state) );
2072 DEBUG_TRIE_COMPILE_MORE_r(
2073 PerlIO_printf( Perl_debug_log,
2074 "%*sAlloc: %d Orig: %"IVdf" elements, Final:%"IVdf". Savings of %%%5.2f\n",
2075 (int)depth * 2 + 2,"",
2076 (int)( ( TRIE_CHARCOUNT(trie) + 1 ) * trie->uniquecharcount + 1 ),
2079 ( ( next_alloc - pos ) * 100 ) / (double)next_alloc );
2082 } /* end table compress */
2084 DEBUG_TRIE_COMPILE_MORE_r(
2085 PerlIO_printf(Perl_debug_log, "%*sStatecount:%"UVxf" Lasttrans:%"UVxf"\n",
2086 (int)depth * 2 + 2, "",
2087 (UV)trie->statecount,
2088 (UV)trie->lasttrans)
2090 /* resize the trans array to remove unused space */
2091 trie->trans = (reg_trie_trans *)
2092 PerlMemShared_realloc( trie->trans, trie->lasttrans
2093 * sizeof(reg_trie_trans) );
2095 { /* Modify the program and insert the new TRIE node */
2096 U8 nodetype =(U8)(flags & 0xFF);
2100 regnode *optimize = NULL;
2101 #ifdef RE_TRACK_PATTERN_OFFSETS
2104 U32 mjd_nodelen = 0;
2105 #endif /* RE_TRACK_PATTERN_OFFSETS */
2106 #endif /* DEBUGGING */
2108 This means we convert either the first branch or the first Exact,
2109 depending on whether the thing following (in 'last') is a branch
2110 or not and whther first is the startbranch (ie is it a sub part of
2111 the alternation or is it the whole thing.)
2112 Assuming its a sub part we convert the EXACT otherwise we convert
2113 the whole branch sequence, including the first.
2115 /* Find the node we are going to overwrite */
2116 if ( first != startbranch || OP( last ) == BRANCH ) {
2117 /* branch sub-chain */
2118 NEXT_OFF( first ) = (U16)(last - first);
2119 #ifdef RE_TRACK_PATTERN_OFFSETS
2121 mjd_offset= Node_Offset((convert));
2122 mjd_nodelen= Node_Length((convert));
2125 /* whole branch chain */
2127 #ifdef RE_TRACK_PATTERN_OFFSETS
2130 const regnode *nop = NEXTOPER( convert );
2131 mjd_offset= Node_Offset((nop));
2132 mjd_nodelen= Node_Length((nop));
2136 PerlIO_printf(Perl_debug_log, "%*sMJD offset:%"UVuf" MJD length:%"UVuf"\n",
2137 (int)depth * 2 + 2, "",
2138 (UV)mjd_offset, (UV)mjd_nodelen)
2141 /* But first we check to see if there is a common prefix we can
2142 split out as an EXACT and put in front of the TRIE node. */
2143 trie->startstate= 1;
2144 if ( trie->bitmap && !widecharmap && !trie->jump ) {
2146 for ( state = 1 ; state < trie->statecount-1 ; state++ ) {
2150 const U32 base = trie->states[ state ].trans.base;
2152 if ( trie->states[state].wordnum )
2155 for ( ofs = 0 ; ofs < trie->uniquecharcount ; ofs++ ) {
2156 if ( ( base + ofs >= trie->uniquecharcount ) &&
2157 ( base + ofs - trie->uniquecharcount < trie->lasttrans ) &&
2158 trie->trans[ base + ofs - trie->uniquecharcount ].check == state )
2160 if ( ++count > 1 ) {
2161 SV **tmp = av_fetch( revcharmap, ofs, 0);
2162 const U8 *ch = (U8*)SvPV_nolen_const( *tmp );
2163 if ( state == 1 ) break;
2165 Zero(trie->bitmap, ANYOF_BITMAP_SIZE, char);
2167 PerlIO_printf(Perl_debug_log,
2168 "%*sNew Start State=%"UVuf" Class: [",
2169 (int)depth * 2 + 2, "",
2172 SV ** const tmp = av_fetch( revcharmap, idx, 0);
2173 const U8 * const ch = (U8*)SvPV_nolen_const( *tmp );
2175 TRIE_BITMAP_SET(trie,*ch);
2177 TRIE_BITMAP_SET(trie, folder[ *ch ]);
2179 PerlIO_printf(Perl_debug_log, "%s", (char*)ch)
2183 TRIE_BITMAP_SET(trie,*ch);
2185 TRIE_BITMAP_SET(trie,folder[ *ch ]);
2186 DEBUG_OPTIMISE_r(PerlIO_printf( Perl_debug_log,"%s", ch));
2192 SV **tmp = av_fetch( revcharmap, idx, 0);
2194 char *ch = SvPV( *tmp, len );
2196 SV *sv=sv_newmortal();
2197 PerlIO_printf( Perl_debug_log,
2198 "%*sPrefix State: %"UVuf" Idx:%"UVuf" Char='%s'\n",
2199 (int)depth * 2 + 2, "",
2201 pv_pretty(sv, SvPV_nolen_const(*tmp), SvCUR(*tmp), 6,
2202 PL_colors[0], PL_colors[1],
2203 (SvUTF8(*tmp) ? PERL_PV_ESCAPE_UNI : 0) |
2204 PERL_PV_ESCAPE_FIRSTCHAR
2209 OP( convert ) = nodetype;
2210 str=STRING(convert);
2213 STR_LEN(convert) += len;
2219 DEBUG_OPTIMISE_r(PerlIO_printf( Perl_debug_log,"]\n"));
2224 trie->prefixlen = (state-1);
2226 regnode *n = convert+NODE_SZ_STR(convert);
2227 NEXT_OFF(convert) = NODE_SZ_STR(convert);
2228 trie->startstate = state;
2229 trie->minlen -= (state - 1);
2230 trie->maxlen -= (state - 1);
2232 /* At least the UNICOS C compiler choked on this
2233 * being argument to DEBUG_r(), so let's just have
2236 #ifdef PERL_EXT_RE_BUILD
2242 regnode *fix = convert;
2243 U32 word = trie->wordcount;
2245 Set_Node_Offset_Length(convert, mjd_offset, state - 1);
2246 while( ++fix < n ) {
2247 Set_Node_Offset_Length(fix, 0, 0);
2250 SV ** const tmp = av_fetch( trie_words, word, 0 );
2252 if ( STR_LEN(convert) <= SvCUR(*tmp) )
2253 sv_chop(*tmp, SvPV_nolen(*tmp) + STR_LEN(convert));
2255 sv_chop(*tmp, SvPV_nolen(*tmp) + SvCUR(*tmp));
2263 NEXT_OFF(convert) = (U16)(tail - convert);
2264 DEBUG_r(optimize= n);
2270 if ( trie->maxlen ) {
2271 NEXT_OFF( convert ) = (U16)(tail - convert);
2272 ARG_SET( convert, data_slot );
2273 /* Store the offset to the first unabsorbed branch in
2274 jump[0], which is otherwise unused by the jump logic.
2275 We use this when dumping a trie and during optimisation. */
2277 trie->jump[0] = (U16)(nextbranch - convert);
2279 /* If the start state is not accepting (meaning there is no empty string/NOTHING)
2280 * and there is a bitmap
2281 * and the first "jump target" node we found leaves enough room
2282 * then convert the TRIE node into a TRIEC node, with the bitmap
2283 * embedded inline in the opcode - this is hypothetically faster.
2285 if ( !trie->states[trie->startstate].wordnum
2287 && ( (char *)jumper - (char *)convert) >= (int)sizeof(struct regnode_charclass) )
2289 OP( convert ) = TRIEC;
2290 Copy(trie->bitmap, ((struct regnode_charclass *)convert)->bitmap, ANYOF_BITMAP_SIZE, char);
2291 PerlMemShared_free(trie->bitmap);
2294 OP( convert ) = TRIE;
2296 /* store the type in the flags */
2297 convert->flags = nodetype;
2301 + regarglen[ OP( convert ) ];
2303 /* XXX We really should free up the resource in trie now,
2304 as we won't use them - (which resources?) dmq */
2306 /* needed for dumping*/
2307 DEBUG_r(if (optimize) {
2308 regnode *opt = convert;
2310 while ( ++opt < optimize) {
2311 Set_Node_Offset_Length(opt,0,0);
2314 Try to clean up some of the debris left after the
2317 while( optimize < jumper ) {
2318 mjd_nodelen += Node_Length((optimize));
2319 OP( optimize ) = OPTIMIZED;
2320 Set_Node_Offset_Length(optimize,0,0);
2323 Set_Node_Offset_Length(convert,mjd_offset,mjd_nodelen);
2325 } /* end node insert */
2326 REH_CALL_COMP_NODE_HOOK(pRExC_state->rx, convert);
2328 /* Finish populating the prev field of the wordinfo array. Walk back
2329 * from each accept state until we find another accept state, and if
2330 * so, point the first word's .prev field at the second word. If the
2331 * second already has a .prev field set, stop now. This will be the
2332 * case either if we've already processed that word's accept state,
2333 * or that state had multiple words, and the overspill words were
2334 * already linked up earlier.
2341 for (word=1; word <= trie->wordcount; word++) {
2343 if (trie->wordinfo[word].prev)
2345 state = trie->wordinfo[word].accept;
2347 state = prev_states[state];
2350 prev = trie->states[state].wordnum;
2354 trie->wordinfo[word].prev = prev;
2356 Safefree(prev_states);
2360 /* and now dump out the compressed format */
2361 DEBUG_TRIE_COMPILE_r(dump_trie(trie, widecharmap, revcharmap, depth+1));
2363 RExC_rxi->data->data[ data_slot + 1 ] = (void*)widecharmap;
2365 RExC_rxi->data->data[ data_slot + TRIE_WORDS_OFFSET ] = (void*)trie_words;
2366 RExC_rxi->data->data[ data_slot + 3 ] = (void*)revcharmap;
2368 SvREFCNT_dec(revcharmap);
2372 : trie->startstate>1
2378 S_make_trie_failtable(pTHX_ RExC_state_t *pRExC_state, regnode *source, regnode *stclass, U32 depth)
2380 /* The Trie is constructed and compressed now so we can build a fail array if it's needed
2382 This is basically the Aho-Corasick algorithm. Its from exercise 3.31 and 3.32 in the
2383 "Red Dragon" -- Compilers, principles, techniques, and tools. Aho, Sethi, Ullman 1985/88
2386 We find the fail state for each state in the trie, this state is the longest proper
2387 suffix of the current state's 'word' that is also a proper prefix of another word in our
2388 trie. State 1 represents the word '' and is thus the default fail state. This allows
2389 the DFA not to have to restart after its tried and failed a word at a given point, it
2390 simply continues as though it had been matching the other word in the first place.
2392 'abcdgu'=~/abcdefg|cdgu/
2393 When we get to 'd' we are still matching the first word, we would encounter 'g' which would
2394 fail, which would bring us to the state representing 'd' in the second word where we would
2395 try 'g' and succeed, proceeding to match 'cdgu'.
2397 /* add a fail transition */
2398 const U32 trie_offset = ARG(source);
2399 reg_trie_data *trie=(reg_trie_data *)RExC_rxi->data->data[trie_offset];
2401 const U32 ucharcount = trie->uniquecharcount;
2402 const U32 numstates = trie->statecount;
2403 const U32 ubound = trie->lasttrans + ucharcount;
2407 U32 base = trie->states[ 1 ].trans.base;
2410 const U32 data_slot = add_data( pRExC_state, 1, "T" );
2411 GET_RE_DEBUG_FLAGS_DECL;
2413 PERL_ARGS_ASSERT_MAKE_TRIE_FAILTABLE;
2415 PERL_UNUSED_ARG(depth);
2419 ARG_SET( stclass, data_slot );
2420 aho = (reg_ac_data *) PerlMemShared_calloc( 1, sizeof(reg_ac_data) );
2421 RExC_rxi->data->data[ data_slot ] = (void*)aho;
2422 aho->trie=trie_offset;
2423 aho->states=(reg_trie_state *)PerlMemShared_malloc( numstates * sizeof(reg_trie_state) );
2424 Copy( trie->states, aho->states, numstates, reg_trie_state );
2425 Newxz( q, numstates, U32);
2426 aho->fail = (U32 *) PerlMemShared_calloc( numstates, sizeof(U32) );
2429 /* initialize fail[0..1] to be 1 so that we always have
2430 a valid final fail state */
2431 fail[ 0 ] = fail[ 1 ] = 1;
2433 for ( charid = 0; charid < ucharcount ; charid++ ) {
2434 const U32 newstate = TRIE_TRANS_STATE( 1, base, ucharcount, charid, 0 );
2436 q[ q_write ] = newstate;
2437 /* set to point at the root */
2438 fail[ q[ q_write++ ] ]=1;
2441 while ( q_read < q_write) {
2442 const U32 cur = q[ q_read++ % numstates ];
2443 base = trie->states[ cur ].trans.base;
2445 for ( charid = 0 ; charid < ucharcount ; charid++ ) {
2446 const U32 ch_state = TRIE_TRANS_STATE( cur, base, ucharcount, charid, 1 );
2448 U32 fail_state = cur;
2451 fail_state = fail[ fail_state ];
2452 fail_base = aho->states[ fail_state ].trans.base;
2453 } while ( !TRIE_TRANS_STATE( fail_state, fail_base, ucharcount, charid, 1 ) );
2455 fail_state = TRIE_TRANS_STATE( fail_state, fail_base, ucharcount, charid, 1 );
2456 fail[ ch_state ] = fail_state;
2457 if ( !aho->states[ ch_state ].wordnum && aho->states[ fail_state ].wordnum )
2459 aho->states[ ch_state ].wordnum = aho->states[ fail_state ].wordnum;
2461 q[ q_write++ % numstates] = ch_state;
2465 /* restore fail[0..1] to 0 so that we "fall out" of the AC loop
2466 when we fail in state 1, this allows us to use the
2467 charclass scan to find a valid start char. This is based on the principle
2468 that theres a good chance the string being searched contains lots of stuff
2469 that cant be a start char.
2471 fail[ 0 ] = fail[ 1 ] = 0;
2472 DEBUG_TRIE_COMPILE_r({
2473 PerlIO_printf(Perl_debug_log,
2474 "%*sStclass Failtable (%"UVuf" states): 0",
2475 (int)(depth * 2), "", (UV)numstates
2477 for( q_read=1; q_read<numstates; q_read++ ) {
2478 PerlIO_printf(Perl_debug_log, ", %"UVuf, (UV)fail[q_read]);
2480 PerlIO_printf(Perl_debug_log, "\n");
2483 /*RExC_seen |= REG_SEEN_TRIEDFA;*/
2488 * There are strange code-generation bugs caused on sparc64 by gcc-2.95.2.
2489 * These need to be revisited when a newer toolchain becomes available.
2491 #if defined(__sparc64__) && defined(__GNUC__)
2492 # if __GNUC__ < 2 || (__GNUC__ == 2 && __GNUC_MINOR__ < 96)
2493 # undef SPARC64_GCC_WORKAROUND
2494 # define SPARC64_GCC_WORKAROUND 1
2498 #define DEBUG_PEEP(str,scan,depth) \
2499 DEBUG_OPTIMISE_r({if (scan){ \
2500 SV * const mysv=sv_newmortal(); \
2501 regnode *Next = regnext(scan); \
2502 regprop(RExC_rx, mysv, scan); \
2503 PerlIO_printf(Perl_debug_log, "%*s" str ">%3d: %s (%d)\n", \
2504 (int)depth*2, "", REG_NODE_NUM(scan), SvPV_nolen_const(mysv),\
2505 Next ? (REG_NODE_NUM(Next)) : 0 ); \
2509 /* The below joins as many adjacent EXACTish nodes as possible into a single
2510 * one, and looks for problematic sequences of characters whose folds vs.
2511 * non-folds have sufficiently different lengths, that the optimizer would be
2512 * fooled into rejecting legitimate matches of them, and the trie construction
2513 * code can't cope with them. The joining is only done if:
2514 * 1) there is room in the current conglomerated node to entirely contain the
2516 * 2) they are the exact same node type
2518 * The adjacent nodes actually may be separated by NOTHING kind nodes, and
2519 * these get optimized out
2521 * If there are problematic code sequences, *min_subtract is set to the delta
2522 * that the minimum size of the node can be less than its actual size. And,
2523 * the node type of the result is changed to reflect that it contains these
2526 * And *has_exactf_sharp_s is set to indicate whether or not the node is EXACTF
2527 * and contains LATIN SMALL LETTER SHARP S
2529 * This is as good a place as any to discuss the design of handling these
2530 * problematic sequences. It's been wrong in Perl for a very long time. There
2531 * are three code points in Unicode whose folded lengths differ so much from
2532 * the un-folded lengths that it causes problems for the optimizer and trie
2533 * construction. Why only these are problematic, and not others where lengths
2534 * also differ is something I (khw) do not understand. New versions of Unicode
2535 * might add more such code points. Hopefully the logic in fold_grind.t that
2536 * figures out what to test (in part by verifying that each size-combination
2537 * gets tested) will catch any that do come along, so they can be added to the
2538 * special handling below. The chances of new ones are actually rather small,
2539 * as most, if not all, of the world's scripts that have casefolding have
2540 * already been encoded by Unicode. Also, a number of Unicode's decisions were
2541 * made to allow compatibility with pre-existing standards, and almost all of
2542 * those have already been dealt with. These would otherwise be the most
2543 * likely candidates for generating further tricky sequences. In other words,
2544 * Unicode by itself is unlikely to add new ones unless it is for compatibility
2545 * with pre-existing standards, and there aren't many of those left.
2547 * The previous designs for dealing with these involved assigning a special
2548 * node for them. This approach doesn't work, as evidenced by this example:
2549 * "\xDFs" =~ /s\xDF/ui # Used to fail before these patches
2550 * Both these fold to "sss", but if the pattern is parsed to create a node of
2551 * that would match just the \xDF, it won't be able to handle the case where a
2552 * successful match would have to cross the node's boundary. The new approach
2553 * that hopefully generally solves the problem generates an EXACTFU_SS node
2556 * There are a number of components to the approach (a lot of work for just
2557 * three code points!):
2558 * 1) This routine examines each EXACTFish node that could contain the
2559 * problematic sequences. It returns in *min_subtract how much to
2560 * subtract from the the actual length of the string to get a real minimum
2561 * for one that could match it. This number is usually 0 except for the
2562 * problematic sequences. This delta is used by the caller to adjust the
2563 * min length of the match, and the delta between min and max, so that the
2564 * optimizer doesn't reject these possibilities based on size constraints.
2565 * 2) These sequences are not currently correctly handled by the trie code
2566 * either, so it changes the joined node type to ops that are not handled
2567 * by trie's, those new ops being EXACTFU_SS and EXACTFU_NO_TRIE.
2568 * 3) This is sufficient for the two Greek sequences (described below), but
2569 * the one involving the Sharp s (\xDF) needs more. The node type
2570 * EXACTFU_SS is used for an EXACTFU node that contains at least one "ss"
2571 * sequence in it. For non-UTF-8 patterns and strings, this is the only
2572 * case where there is a possible fold length change. That means that a
2573 * regular EXACTFU node without UTF-8 involvement doesn't have to concern
2574 * itself with length changes, and so can be processed faster. regexec.c
2575 * takes advantage of this. Generally, an EXACTFish node that is in UTF-8
2576 * is pre-folded by regcomp.c. This saves effort in regex matching.
2577 * However, probably mostly for historical reasons, the pre-folding isn't
2578 * done for non-UTF8 patterns (and it can't be for EXACTF and EXACTFL
2579 * nodes, as what they fold to isn't known until runtime.) The fold
2580 * possibilities for the non-UTF8 patterns are quite simple, except for
2581 * the sharp s. All the ones that don't involve a UTF-8 target string
2582 * are members of a fold-pair, and arrays are set up for all of them
2583 * that quickly find the other member of the pair. It might actually
2584 * be faster to pre-fold these, but it isn't currently done, except for
2585 * the sharp s. Code elsewhere in this file makes sure that it gets
2586 * folded to 'ss', even if the pattern isn't UTF-8. This avoids the
2587 * issues described in the next item.
2588 * 4) A problem remains for the sharp s in EXACTF nodes. Whether it matches
2589 * 'ss' or not is not knowable at compile time. It will match iff the
2590 * target string is in UTF-8, unlike the EXACTFU nodes, where it always
2591 * matches; and the EXACTFL and EXACTFA nodes where it never does. Thus
2592 * it can't be folded to "ss" at compile time, unlike EXACTFU does as
2593 * described in item 3). An assumption that the optimizer part of
2594 * regexec.c (probably unwittingly) makes is that a character in the
2595 * pattern corresponds to at most a single character in the target string.
2596 * (And I do mean character, and not byte here, unlike other parts of the
2597 * documentation that have never been updated to account for multibyte
2598 * Unicode.) This assumption is wrong only in this case, as all other
2599 * cases are either 1-1 folds when no UTF-8 is involved; or is true by
2600 * virtue of having this file pre-fold UTF-8 patterns. I'm
2601 * reluctant to try to change this assumption, so instead the code punts.
2602 * This routine examines EXACTF nodes for the sharp s, and returns a
2603 * boolean indicating whether or not the node is an EXACTF node that
2604 * contains a sharp s. When it is true, the caller sets a flag that later
2605 * causes the optimizer in this file to not set values for the floating
2606 * and fixed string lengths, and thus avoids the optimizer code in
2607 * regexec.c that makes the invalid assumption. Thus, there is no
2608 * optimization based on string lengths for EXACTF nodes that contain the
2609 * sharp s. This only happens for /id rules (which means the pattern
2613 #define JOIN_EXACT(scan,min_subtract,has_exactf_sharp_s, flags) \
2614 if (PL_regkind[OP(scan)] == EXACT) \
2615 join_exact(pRExC_state,(scan),(min_subtract),has_exactf_sharp_s, (flags),NULL,depth+1)
2618 S_join_exact(pTHX_ RExC_state_t *pRExC_state, regnode *scan, UV *min_subtract, bool *has_exactf_sharp_s, U32 flags,regnode *val, U32 depth) {
2619 /* Merge several consecutive EXACTish nodes into one. */
2620 regnode *n = regnext(scan);
2622 regnode *next = scan + NODE_SZ_STR(scan);
2626 regnode *stop = scan;
2627 GET_RE_DEBUG_FLAGS_DECL;
2629 PERL_UNUSED_ARG(depth);
2632 PERL_ARGS_ASSERT_JOIN_EXACT;
2633 #ifndef EXPERIMENTAL_INPLACESCAN
2634 PERL_UNUSED_ARG(flags);
2635 PERL_UNUSED_ARG(val);
2637 DEBUG_PEEP("join",scan,depth);
2639 /* Look through the subsequent nodes in the chain. Skip NOTHING, merge
2640 * EXACT ones that are mergeable to the current one. */
2642 && (PL_regkind[OP(n)] == NOTHING
2643 || (stringok && OP(n) == OP(scan)))
2645 && NEXT_OFF(scan) + NEXT_OFF(n) < I16_MAX)
2648 if (OP(n) == TAIL || n > next)
2650 if (PL_regkind[OP(n)] == NOTHING) {
2651 DEBUG_PEEP("skip:",n,depth);
2652 NEXT_OFF(scan) += NEXT_OFF(n);
2653 next = n + NODE_STEP_REGNODE;
2660 else if (stringok) {
2661 const unsigned int oldl = STR_LEN(scan);
2662 regnode * const nnext = regnext(n);
2664 if (oldl + STR_LEN(n) > U8_MAX)
2667 DEBUG_PEEP("merg",n,depth);
2670 NEXT_OFF(scan) += NEXT_OFF(n);
2671 STR_LEN(scan) += STR_LEN(n);
2672 next = n + NODE_SZ_STR(n);
2673 /* Now we can overwrite *n : */
2674 Move(STRING(n), STRING(scan) + oldl, STR_LEN(n), char);
2682 #ifdef EXPERIMENTAL_INPLACESCAN
2683 if (flags && !NEXT_OFF(n)) {
2684 DEBUG_PEEP("atch", val, depth);
2685 if (reg_off_by_arg[OP(n)]) {
2686 ARG_SET(n, val - n);
2689 NEXT_OFF(n) = val - n;
2697 *has_exactf_sharp_s = FALSE;
2699 /* Here, all the adjacent mergeable EXACTish nodes have been merged. We
2700 * can now analyze for sequences of problematic code points. (Prior to
2701 * this final joining, sequences could have been split over boundaries, and
2702 * hence missed). The sequences only happen in folding, hence for any
2703 * non-EXACT EXACTish node */
2704 if (OP(scan) != EXACT) {
2706 U8 * s0 = (U8*) STRING(scan);
2707 U8 * const s_end = s0 + STR_LEN(scan);
2709 /* The below is perhaps overboard, but this allows us to save a test
2710 * each time through the loop at the expense of a mask. This is
2711 * because on both EBCDIC and ASCII machines, 'S' and 's' differ by a
2712 * single bit. On ASCII they are 32 apart; on EBCDIC, they are 64.
2713 * This uses an exclusive 'or' to find that bit and then inverts it to
2714 * form a mask, with just a single 0, in the bit position where 'S' and
2716 const U8 S_or_s_mask = ~ ('S' ^ 's');
2717 const U8 s_masked = 's' & S_or_s_mask;
2719 /* One pass is made over the node's string looking for all the
2720 * possibilities. to avoid some tests in the loop, there are two main
2721 * cases, for UTF-8 patterns (which can't have EXACTF nodes) and
2725 /* There are two problematic Greek code points in Unicode
2728 * U+0390 - GREEK SMALL LETTER IOTA WITH DIALYTIKA AND TONOS
2729 * U+03B0 - GREEK SMALL LETTER UPSILON WITH DIALYTIKA AND TONOS
2735 * U+03B9 U+0308 U+0301 0xCE 0xB9 0xCC 0x88 0xCC 0x81
2736 * U+03C5 U+0308 U+0301 0xCF 0x85 0xCC 0x88 0xCC 0x81
2738 * This means that in case-insensitive matching (or "loose
2739 * matching", as Unicode calls it), an EXACTF of length six (the
2740 * UTF-8 encoded byte length of the above casefolded versions) can
2741 * match a target string of length two (the byte length of UTF-8
2742 * encoded U+0390 or U+03B0). This would rather mess up the
2743 * minimum length computation. (there are other code points that
2744 * also fold to these two sequences, but the delta is smaller)
2746 * If these sequences are found, the minimum length is decreased by
2747 * four (six minus two).
2749 * Similarly, 'ss' may match the single char and byte LATIN SMALL
2750 * LETTER SHARP S. We decrease the min length by 1 for each
2751 * occurrence of 'ss' found */
2753 #ifdef EBCDIC /* RD tunifold greek 0390 and 03B0 */
2754 # define U390_first_byte 0xb4
2755 const U8 U390_tail[] = "\x68\xaf\x49\xaf\x42";
2756 # define U3B0_first_byte 0xb5
2757 const U8 U3B0_tail[] = "\x46\xaf\x49\xaf\x42";
2759 # define U390_first_byte 0xce
2760 const U8 U390_tail[] = "\xb9\xcc\x88\xcc\x81";
2761 # define U3B0_first_byte 0xcf
2762 const U8 U3B0_tail[] = "\x85\xcc\x88\xcc\x81";
2764 const U8 len = sizeof(U390_tail); /* (-1 for NUL; +1 for 1st byte;
2765 yields a net of 0 */
2766 /* Examine the string for one of the problematic sequences */
2768 s < s_end - 1; /* Can stop 1 before the end, as minimum length
2769 * sequence we are looking for is 2 */
2773 /* Look for the first byte in each problematic sequence */
2775 /* We don't have to worry about other things that fold to
2776 * 's' (such as the long s, U+017F), as all above-latin1
2777 * code points have been pre-folded */
2781 /* Current character is an 's' or 'S'. If next one is
2782 * as well, we have the dreaded sequence */
2783 if (((*(s+1) & S_or_s_mask) == s_masked)
2784 /* These two node types don't have special handling
2786 && OP(scan) != EXACTFL && OP(scan) != EXACTFA)
2789 OP(scan) = EXACTFU_SS;
2790 s++; /* No need to look at this character again */
2794 case U390_first_byte:
2795 if (s_end - s >= len
2797 /* The 1's are because are skipping comparing the
2799 && memEQ(s + 1, U390_tail, len - 1))
2801 goto greek_sequence;
2805 case U3B0_first_byte:
2806 if (! (s_end - s >= len
2807 && memEQ(s + 1, U3B0_tail, len - 1)))
2814 /* This can't currently be handled by trie's, so change
2815 * the node type to indicate this. If EXACTFA and
2816 * EXACTFL were ever to be handled by trie's, this
2817 * would have to be changed. If this node has already
2818 * been changed to EXACTFU_SS in this loop, leave it as
2819 * is. (I (khw) think it doesn't matter in regexec.c
2820 * for UTF patterns, but no need to change it */
2821 if (OP(scan) == EXACTFU) {
2822 OP(scan) = EXACTFU_NO_TRIE;
2824 s += 6; /* We already know what this sequence is. Skip
2830 else if (OP(scan) != EXACTFL && OP(scan) != EXACTFA) {
2832 /* Here, the pattern is not UTF-8. We need to look only for the
2833 * 'ss' sequence, and in the EXACTF case, the sharp s, which can be
2834 * in the final position. Otherwise we can stop looking 1 byte
2835 * earlier because have to find both the first and second 's' */
2836 const U8* upper = (OP(scan) == EXACTF) ? s_end : s_end -1;
2838 for (s = s0; s < upper; s++) {
2843 && ((*(s+1) & S_or_s_mask) == s_masked))
2847 /* EXACTF nodes need to know that the minimum
2848 * length changed so that a sharp s in the string
2849 * can match this ss in the pattern, but they
2850 * remain EXACTF nodes, as they are not trie'able,
2851 * so don't have to invent a new node type to
2852 * exclude them from the trie code */
2853 if (OP(scan) != EXACTF) {
2854 OP(scan) = EXACTFU_SS;
2859 case LATIN_SMALL_LETTER_SHARP_S:
2860 if (OP(scan) == EXACTF) {
2861 *has_exactf_sharp_s = TRUE;
2870 /* Allow dumping but overwriting the collection of skipped
2871 * ops and/or strings with fake optimized ops */
2872 n = scan + NODE_SZ_STR(scan);
2880 DEBUG_OPTIMISE_r(if (merged){DEBUG_PEEP("finl",scan,depth)});
2884 /* REx optimizer. Converts nodes into quicker variants "in place".
2885 Finds fixed substrings. */
2887 /* Stops at toplevel WHILEM as well as at "last". At end *scanp is set
2888 to the position after last scanned or to NULL. */
2890 #define INIT_AND_WITHP \
2891 assert(!and_withp); \
2892 Newx(and_withp,1,struct regnode_charclass_class); \
2893 SAVEFREEPV(and_withp)
2895 /* this is a chain of data about sub patterns we are processing that
2896 need to be handled separately/specially in study_chunk. Its so
2897 we can simulate recursion without losing state. */
2899 typedef struct scan_frame {
2900 regnode *last; /* last node to process in this frame */
2901 regnode *next; /* next node to process when last is reached */
2902 struct scan_frame *prev; /*previous frame*/
2903 I32 stop; /* what stopparen do we use */
2907 #define SCAN_COMMIT(s, data, m) scan_commit(s, data, m, is_inf)
2909 #define CASE_SYNST_FNC(nAmE) \
2911 if (flags & SCF_DO_STCLASS_AND) { \
2912 for (value = 0; value < 256; value++) \
2913 if (!is_ ## nAmE ## _cp(value)) \
2914 ANYOF_BITMAP_CLEAR(data->start_class, value); \
2917 for (value = 0; value < 256; value++) \
2918 if (is_ ## nAmE ## _cp(value)) \
2919 ANYOF_BITMAP_SET(data->start_class, value); \
2923 if (flags & SCF_DO_STCLASS_AND) { \
2924 for (value = 0; value < 256; value++) \
2925 if (is_ ## nAmE ## _cp(value)) \
2926 ANYOF_BITMAP_CLEAR(data->start_class, value); \
2929 for (value = 0; value < 256; value++) \
2930 if (!is_ ## nAmE ## _cp(value)) \
2931 ANYOF_BITMAP_SET(data->start_class, value); \
2938 S_study_chunk(pTHX_ RExC_state_t *pRExC_state, regnode **scanp,
2939 I32 *minlenp, I32 *deltap,
2944 struct regnode_charclass_class *and_withp,
2945 U32 flags, U32 depth)
2946 /* scanp: Start here (read-write). */
2947 /* deltap: Write maxlen-minlen here. */
2948 /* last: Stop before this one. */
2949 /* data: string data about the pattern */
2950 /* stopparen: treat close N as END */
2951 /* recursed: which subroutines have we recursed into */
2952 /* and_withp: Valid if flags & SCF_DO_STCLASS_OR */
2955 I32 min = 0, pars = 0, code;
2956 regnode *scan = *scanp, *next;
2958 int is_inf = (flags & SCF_DO_SUBSTR) && (data->flags & SF_IS_INF);
2959 int is_inf_internal = 0; /* The studied chunk is infinite */
2960 I32 is_par = OP(scan) == OPEN ? ARG(scan) : 0;
2961 scan_data_t data_fake;
2962 SV *re_trie_maxbuff = NULL;
2963 regnode *first_non_open = scan;
2964 I32 stopmin = I32_MAX;
2965 scan_frame *frame = NULL;
2966 GET_RE_DEBUG_FLAGS_DECL;
2968 PERL_ARGS_ASSERT_STUDY_CHUNK;
2971 StructCopy(&zero_scan_data, &data_fake, scan_data_t);
2975 while (first_non_open && OP(first_non_open) == OPEN)
2976 first_non_open=regnext(first_non_open);
2981 while ( scan && OP(scan) != END && scan < last ){
2982 UV min_subtract = 0; /* How much to subtract from the minimum node
2983 length to get a real minimum (because the
2984 folded version may be shorter) */
2985 bool has_exactf_sharp_s = FALSE;
2986 /* Peephole optimizer: */
2987 DEBUG_STUDYDATA("Peep:", data,depth);
2988 DEBUG_PEEP("Peep",scan,depth);
2990 /* Its not clear to khw or hv why this is done here, and not in the
2991 * clauses that deal with EXACT nodes. khw's guess is that it's
2992 * because of a previous design */
2993 JOIN_EXACT(scan,&min_subtract, &has_exactf_sharp_s, 0);
2995 /* Follow the next-chain of the current node and optimize
2996 away all the NOTHINGs from it. */
2997 if (OP(scan) != CURLYX) {
2998 const int max = (reg_off_by_arg[OP(scan)]
3000 /* I32 may be smaller than U16 on CRAYs! */
3001 : (I32_MAX < U16_MAX ? I32_MAX : U16_MAX));
3002 int off = (reg_off_by_arg[OP(scan)] ? ARG(scan) : NEXT_OFF(scan));
3006 /* Skip NOTHING and LONGJMP. */
3007 while ((n = regnext(n))
3008 && ((PL_regkind[OP(n)] == NOTHING && (noff = NEXT_OFF(n)))
3009 || ((OP(n) == LONGJMP) && (noff = ARG(n))))
3010 && off + noff < max)
3012 if (reg_off_by_arg[OP(scan)])
3015 NEXT_OFF(scan) = off;
3020 /* The principal pseudo-switch. Cannot be a switch, since we
3021 look into several different things. */
3022 if (OP(scan) == BRANCH || OP(scan) == BRANCHJ
3023 || OP(scan) == IFTHEN) {
3024 next = regnext(scan);
3026 /* demq: the op(next)==code check is to see if we have "branch-branch" AFAICT */
3028 if (OP(next) == code || code == IFTHEN) {
3029 /* NOTE - There is similar code to this block below for handling
3030 TRIE nodes on a re-study. If you change stuff here check there
3032 I32 max1 = 0, min1 = I32_MAX, num = 0;
3033 struct regnode_charclass_class accum;
3034 regnode * const startbranch=scan;
3036 if (flags & SCF_DO_SUBSTR)
3037 SCAN_COMMIT(pRExC_state, data, minlenp); /* Cannot merge strings after this. */
3038 if (flags & SCF_DO_STCLASS)
3039 cl_init_zero(pRExC_state, &accum);
3041 while (OP(scan) == code) {
3042 I32 deltanext, minnext, f = 0, fake;
3043 struct regnode_charclass_class this_class;
3046 data_fake.flags = 0;
3048 data_fake.whilem_c = data->whilem_c;
3049 data_fake.last_closep = data->last_closep;
3052 data_fake.last_closep = &fake;
3054 data_fake.pos_delta = delta;
3055 next = regnext(scan);
3056 scan = NEXTOPER(scan);
3058 scan = NEXTOPER(scan);
3059 if (flags & SCF_DO_STCLASS) {
3060 cl_init(pRExC_state, &this_class);
3061 data_fake.start_class = &this_class;
3062 f = SCF_DO_STCLASS_AND;
3064 if (flags & SCF_WHILEM_VISITED_POS)
3065 f |= SCF_WHILEM_VISITED_POS;
3067 /* we suppose the run is continuous, last=next...*/
3068 minnext = study_chunk(pRExC_state, &scan, minlenp, &deltanext,
3070 stopparen, recursed, NULL, f,depth+1);
3073 if (max1 < minnext + deltanext)
3074 max1 = minnext + deltanext;
3075 if (deltanext == I32_MAX)
3076 is_inf = is_inf_internal = 1;
3078 if (data_fake.flags & (SF_HAS_PAR|SF_IN_PAR))
3080 if (data_fake.flags & SCF_SEEN_ACCEPT) {
3081 if ( stopmin > minnext)
3082 stopmin = min + min1;
3083 flags &= ~SCF_DO_SUBSTR;
3085 data->flags |= SCF_SEEN_ACCEPT;
3088 if (data_fake.flags & SF_HAS_EVAL)
3089 data->flags |= SF_HAS_EVAL;
3090 data->whilem_c = data_fake.whilem_c;
3092 if (flags & SCF_DO_STCLASS)
3093 cl_or(pRExC_state, &accum, &this_class);
3095 if (code == IFTHEN && num < 2) /* Empty ELSE branch */
3097 if (flags & SCF_DO_SUBSTR) {
3098 data->pos_min += min1;
3099 data->pos_delta += max1 - min1;
3100 if (max1 != min1 || is_inf)
3101 data->longest = &(data->longest_float);
3104 delta += max1 - min1;
3105 if (flags & SCF_DO_STCLASS_OR) {
3106 cl_or(pRExC_state, data->start_class, &accum);
3108 cl_and(data->start_class, and_withp);
3109 flags &= ~SCF_DO_STCLASS;
3112 else if (flags & SCF_DO_STCLASS_AND) {
3114 cl_and(data->start_class, &accum);
3115 flags &= ~SCF_DO_STCLASS;
3118 /* Switch to OR mode: cache the old value of
3119 * data->start_class */
3121 StructCopy(data->start_class, and_withp,
3122 struct regnode_charclass_class);
3123 flags &= ~SCF_DO_STCLASS_AND;
3124 StructCopy(&accum, data->start_class,
3125 struct regnode_charclass_class);
3126 flags |= SCF_DO_STCLASS_OR;
3127 data->start_class->flags |= ANYOF_EOS;
3131 if (PERL_ENABLE_TRIE_OPTIMISATION && OP( startbranch ) == BRANCH ) {
3134 Assuming this was/is a branch we are dealing with: 'scan' now
3135 points at the item that follows the branch sequence, whatever
3136 it is. We now start at the beginning of the sequence and look
3143 which would be constructed from a pattern like /A|LIST|OF|WORDS/
3145 If we can find such a subsequence we need to turn the first
3146 element into a trie and then add the subsequent branch exact
3147 strings to the trie.
3151 1. patterns where the whole set of branches can be converted.
3153 2. patterns where only a subset can be converted.
3155 In case 1 we can replace the whole set with a single regop
3156 for the trie. In case 2 we need to keep the start and end
3159 'BRANCH EXACT; BRANCH EXACT; BRANCH X'
3160 becomes BRANCH TRIE; BRANCH X;
3162 There is an additional case, that being where there is a
3163 common prefix, which gets split out into an EXACT like node
3164 preceding the TRIE node.
3166 If x(1..n)==tail then we can do a simple trie, if not we make
3167 a "jump" trie, such that when we match the appropriate word
3168 we "jump" to the appropriate tail node. Essentially we turn
3169 a nested if into a case structure of sorts.
3174 if (!re_trie_maxbuff) {
3175 re_trie_maxbuff = get_sv(RE_TRIE_MAXBUF_NAME, 1);
3176 if (!SvIOK(re_trie_maxbuff))
3177 sv_setiv(re_trie_maxbuff, RE_TRIE_MAXBUF_INIT);
3179 if ( SvIV(re_trie_maxbuff)>=0 ) {
3181 regnode *first = (regnode *)NULL;
3182 regnode *last = (regnode *)NULL;
3183 regnode *tail = scan;
3188 SV * const mysv = sv_newmortal(); /* for dumping */
3190 /* var tail is used because there may be a TAIL
3191 regop in the way. Ie, the exacts will point to the
3192 thing following the TAIL, but the last branch will
3193 point at the TAIL. So we advance tail. If we
3194 have nested (?:) we may have to move through several
3198 while ( OP( tail ) == TAIL ) {
3199 /* this is the TAIL generated by (?:) */
3200 tail = regnext( tail );
3205 regprop(RExC_rx, mysv, tail );
3206 PerlIO_printf( Perl_debug_log, "%*s%s%s\n",
3207 (int)depth * 2 + 2, "",
3208 "Looking for TRIE'able sequences. Tail node is: ",
3209 SvPV_nolen_const( mysv )
3215 step through the branches, cur represents each
3216 branch, noper is the first thing to be matched
3217 as part of that branch and noper_next is the
3218 regnext() of that node. if noper is an EXACT
3219 and noper_next is the same as scan (our current
3220 position in the regex) then the EXACT branch is
3221 a possible optimization target. Once we have
3222 two or more consecutive such branches we can
3223 create a trie of the EXACT's contents and stich
3224 it in place. If the sequence represents all of
3225 the branches we eliminate the whole thing and
3226 replace it with a single TRIE. If it is a
3227 subsequence then we need to stitch it in. This
3228 means the first branch has to remain, and needs
3229 to be repointed at the item on the branch chain
3230 following the last branch optimized. This could
3231 be either a BRANCH, in which case the
3232 subsequence is internal, or it could be the
3233 item following the branch sequence in which
3234 case the subsequence is at the end.
3238 /* dont use tail as the end marker for this traverse */
3239 for ( cur = startbranch ; cur != scan ; cur = regnext( cur ) ) {
3240 regnode * const noper = NEXTOPER( cur );
3241 #if defined(DEBUGGING) || defined(NOJUMPTRIE)
3242 regnode * const noper_next = regnext( noper );
3246 regprop(RExC_rx, mysv, cur);
3247 PerlIO_printf( Perl_debug_log, "%*s- %s (%d)",
3248 (int)depth * 2 + 2,"", SvPV_nolen_const( mysv ), REG_NODE_NUM(cur) );
3250 regprop(RExC_rx, mysv, noper);
3251 PerlIO_printf( Perl_debug_log, " -> %s",
3252 SvPV_nolen_const(mysv));
3255 regprop(RExC_rx, mysv, noper_next );
3256 PerlIO_printf( Perl_debug_log,"\t=> %s\t",
3257 SvPV_nolen_const(mysv));
3259 PerlIO_printf( Perl_debug_log, "(First==%d,Last==%d,Cur==%d)\n",
3260 REG_NODE_NUM(first), REG_NODE_NUM(last), REG_NODE_NUM(cur) );
3262 if ( (((first && optype!=NOTHING) ? OP( noper ) == optype
3263 : PL_regkind[ OP( noper ) ] == EXACT )
3264 || OP(noper) == NOTHING )
3266 && noper_next == tail
3271 if ( !first || optype == NOTHING ) {
3272 if (!first) first = cur;
3273 optype = OP( noper );
3279 Currently the trie logic handles case insensitive matching properly only
3280 when the pattern is UTF-8 and the node is EXACTFU (thus forcing unicode
3283 If/when this is fixed the following define can be swapped
3284 in below to fully enable trie logic.
3286 #define TRIE_TYPE_IS_SAFE 1
3288 Note that join_exact() assumes that the other types of EXACTFish nodes are not
3289 used in tries, so that would have to be updated if this changed
3292 #define TRIE_TYPE_IS_SAFE ((UTF && optype == EXACTFU) || optype==EXACT)
3294 if ( last && TRIE_TYPE_IS_SAFE ) {
3295 make_trie( pRExC_state,
3296 startbranch, first, cur, tail, count,
3299 if ( PL_regkind[ OP( noper ) ] == EXACT
3301 && noper_next == tail
3306 optype = OP( noper );
3316 regprop(RExC_rx, mysv, cur);
3317 PerlIO_printf( Perl_debug_log,
3318 "%*s- %s (%d) <SCAN FINISHED>\n", (int)depth * 2 + 2,
3319 "", SvPV_nolen_const( mysv ),REG_NODE_NUM(cur));
3323 if ( last && TRIE_TYPE_IS_SAFE ) {
3324 made= make_trie( pRExC_state, startbranch, first, scan, tail, count, optype, depth+1 );
3325 #ifdef TRIE_STUDY_OPT
3326 if ( ((made == MADE_EXACT_TRIE &&
3327 startbranch == first)
3328 || ( first_non_open == first )) &&
3330 flags |= SCF_TRIE_RESTUDY;
3331 if ( startbranch == first
3334 RExC_seen &=~REG_TOP_LEVEL_BRANCHES;
3344 else if ( code == BRANCHJ ) { /* single branch is optimized. */
3345 scan = NEXTOPER(NEXTOPER(scan));
3346 } else /* single branch is optimized. */
3347 scan = NEXTOPER(scan);
3349 } else if (OP(scan) == SUSPEND || OP(scan) == GOSUB || OP(scan) == GOSTART) {
3350 scan_frame *newframe = NULL;
3355 if (OP(scan) != SUSPEND) {
3356 /* set the pointer */
3357 if (OP(scan) == GOSUB) {
3359 RExC_recurse[ARG2L(scan)] = scan;
3360 start = RExC_open_parens[paren-1];
3361 end = RExC_close_parens[paren-1];
3364 start = RExC_rxi->program + 1;
3368 Newxz(recursed, (((RExC_npar)>>3) +1), U8);
3369 SAVEFREEPV(recursed);
3371 if (!PAREN_TEST(recursed,paren+1)) {
3372 PAREN_SET(recursed,paren+1);
3373 Newx(newframe,1,scan_frame);
3375 if (flags & SCF_DO_SUBSTR) {
3376 SCAN_COMMIT(pRExC_state,data,minlenp);
3377 data->longest = &(data->longest_float);
3379 is_inf = is_inf_internal = 1;
3380 if (flags & SCF_DO_STCLASS_OR) /* Allow everything */
3381 cl_anything(pRExC_state, data->start_class);
3382 flags &= ~SCF_DO_STCLASS;
3385 Newx(newframe,1,scan_frame);
3388 end = regnext(scan);
3393 SAVEFREEPV(newframe);
3394 newframe->next = regnext(scan);
3395 newframe->last = last;
3396 newframe->stop = stopparen;
3397 newframe->prev = frame;
3407 else if (OP(scan) == EXACT) {
3408 I32 l = STR_LEN(scan);
3411 const U8 * const s = (U8*)STRING(scan);
3412 l = utf8_length(s, s + l);
3413 uc = utf8_to_uvchr(s, NULL);
3415 uc = *((U8*)STRING(scan));
3418 if (flags & SCF_DO_SUBSTR) { /* Update longest substr. */
3419 /* The code below prefers earlier match for fixed
3420 offset, later match for variable offset. */
3421 if (data->last_end == -1) { /* Update the start info. */
3422 data->last_start_min = data->pos_min;
3423 data->last_start_max = is_inf
3424 ? I32_MAX : data->pos_min + data->pos_delta;
3426 sv_catpvn(data->last_found, STRING(scan), STR_LEN(scan));
3428 SvUTF8_on(data->last_found);
3430 SV * const sv = data->last_found;
3431 MAGIC * const mg = SvUTF8(sv) && SvMAGICAL(sv) ?
3432 mg_find(sv, PERL_MAGIC_utf8) : NULL;
3433 if (mg && mg->mg_len >= 0)
3434 mg->mg_len += utf8_length((U8*)STRING(scan),
3435 (U8*)STRING(scan)+STR_LEN(scan));
3437 data->last_end = data->pos_min + l;
3438 data->pos_min += l; /* As in the first entry. */
3439 data->flags &= ~SF_BEFORE_EOL;
3441 if (flags & SCF_DO_STCLASS_AND) {
3442 /* Check whether it is compatible with what we know already! */
3446 /* If compatible, we or it in below. It is compatible if is
3447 * in the bitmp and either 1) its bit or its fold is set, or 2)
3448 * it's for a locale. Even if there isn't unicode semantics
3449 * here, at runtime there may be because of matching against a
3450 * utf8 string, so accept a possible false positive for
3451 * latin1-range folds */
3453 (!(data->start_class->flags & (ANYOF_CLASS | ANYOF_LOCALE))
3454 && !ANYOF_BITMAP_TEST(data->start_class, uc)
3455 && (!(data->start_class->flags & ANYOF_LOC_NONBITMAP_FOLD)
3456 || !ANYOF_BITMAP_TEST(data->start_class, PL_fold_latin1[uc])))
3461 ANYOF_CLASS_ZERO(data->start_class);
3462 ANYOF_BITMAP_ZERO(data->start_class);
3464 ANYOF_BITMAP_SET(data->start_class, uc);
3465 else if (uc >= 0x100) {
3468 /* Some Unicode code points fold to the Latin1 range; as
3469 * XXX temporary code, instead of figuring out if this is
3470 * one, just assume it is and set all the start class bits
3471 * that could be some such above 255 code point's fold
3472 * which will generate fals positives. As the code
3473 * elsewhere that does compute the fold settles down, it
3474 * can be extracted out and re-used here */
3475 for (i = 0; i < 256; i++){
3476 if (_HAS_NONLATIN1_FOLD_CLOSURE_ONLY_FOR_USE_BY_REGCOMP_DOT_C_AND_REGEXEC_DOT_C(i)) {
3477 ANYOF_BITMAP_SET(data->start_class, i);
3481 data->start_class->flags &= ~ANYOF_EOS;
3483 data->start_class->flags &= ~ANYOF_UNICODE_ALL;
3485 else if (flags & SCF_DO_STCLASS_OR) {
3486 /* false positive possible if the class is case-folded */
3488 ANYOF_BITMAP_SET(data->start_class, uc);
3490 data->start_class->flags |= ANYOF_UNICODE_ALL;
3491 data->start_class->flags &= ~ANYOF_EOS;
3492 cl_and(data->start_class, and_withp);
3494 flags &= ~SCF_DO_STCLASS;
3496 else if (PL_regkind[OP(scan)] == EXACT) { /* But OP != EXACT! */
3497 I32 l = STR_LEN(scan);
3498 UV uc = *((U8*)STRING(scan));
3500 /* Search for fixed substrings supports EXACT only. */
3501 if (flags & SCF_DO_SUBSTR) {
3503 SCAN_COMMIT(pRExC_state, data, minlenp);
3506 const U8 * const s = (U8 *)STRING(scan);
3507 l = utf8_length(s, s + l);
3508 uc = utf8_to_uvchr(s, NULL);
3510 else if (has_exactf_sharp_s) {
3511 RExC_seen |= REG_SEEN_EXACTF_SHARP_S;
3513 min += l - min_subtract;
3517 delta += min_subtract;
3518 if (flags & SCF_DO_SUBSTR) {
3519 data->pos_min += l - min_subtract;
3520 if (data->pos_min < 0) {
3523 data->pos_delta += min_subtract;
3525 data->longest = &(data->longest_float);
3528 if (flags & SCF_DO_STCLASS_AND) {
3529 /* Check whether it is compatible with what we know already! */
3532 (!(data->start_class->flags & (ANYOF_CLASS | ANYOF_LOCALE))
3533 && !ANYOF_BITMAP_TEST(data->start_class, uc)
3534 && !ANYOF_BITMAP_TEST(data->start_class, PL_fold_latin1[uc])))
3538 ANYOF_CLASS_ZERO(data->start_class);
3539 ANYOF_BITMAP_ZERO(data->start_class);
3541 ANYOF_BITMAP_SET(data->start_class, uc);
3542 data->start_class->flags &= ~ANYOF_EOS;
3543 data->start_class->flags |= ANYOF_LOC_NONBITMAP_FOLD;
3544 if (OP(scan) == EXACTFL) {
3545 /* XXX This set is probably no longer necessary, and
3546 * probably wrong as LOCALE now is on in the initial
3548 data->start_class->flags |= ANYOF_LOCALE;
3552 /* Also set the other member of the fold pair. In case
3553 * that unicode semantics is called for at runtime, use
3554 * the full latin1 fold. (Can't do this for locale,
3555 * because not known until runtime) */
3556 ANYOF_BITMAP_SET(data->start_class, PL_fold_latin1[uc]);
3558 /* All other (EXACTFL handled above) folds except under
3559 * /iaa that include s, S, and sharp_s also may include
3561 if (OP(scan) != EXACTFA) {
3562 if (uc == 's' || uc == 'S') {
3563 ANYOF_BITMAP_SET(data->start_class,
3564 LATIN_SMALL_LETTER_SHARP_S);
3566 else if (uc == LATIN_SMALL_LETTER_SHARP_S) {
3567 ANYOF_BITMAP_SET(data->start_class, 's');
3568 ANYOF_BITMAP_SET(data->start_class, 'S');
3573 else if (uc >= 0x100) {
3575 for (i = 0; i < 256; i++){
3576 if (_HAS_NONLATIN1_FOLD_CLOSURE_ONLY_FOR_USE_BY_REGCOMP_DOT_C_AND_REGEXEC_DOT_C(i)) {
3577 ANYOF_BITMAP_SET(data->start_class, i);
3582 else if (flags & SCF_DO_STCLASS_OR) {
3583 if (data->start_class->flags & ANYOF_LOC_NONBITMAP_FOLD) {
3584 /* false positive possible if the class is case-folded.
3585 Assume that the locale settings are the same... */
3587 ANYOF_BITMAP_SET(data->start_class, uc);
3588 if (OP(scan) != EXACTFL) {
3590 /* And set the other member of the fold pair, but
3591 * can't do that in locale because not known until
3593 ANYOF_BITMAP_SET(data->start_class,
3594 PL_fold_latin1[uc]);
3596 /* All folds except under /iaa that include s, S,
3597 * and sharp_s also may include the others */
3598 if (OP(scan) != EXACTFA) {
3599 if (uc == 's' || uc == 'S') {
3600 ANYOF_BITMAP_SET(data->start_class,
3601 LATIN_SMALL_LETTER_SHARP_S);
3603 else if (uc == LATIN_SMALL_LETTER_SHARP_S) {
3604 ANYOF_BITMAP_SET(data->start_class, 's');
3605 ANYOF_BITMAP_SET(data->start_class, 'S');
3610 data->start_class->flags &= ~ANYOF_EOS;
3612 cl_and(data->start_class, and_withp);
3614 flags &= ~SCF_DO_STCLASS;
3616 else if (REGNODE_VARIES(OP(scan))) {
3617 I32 mincount, maxcount, minnext, deltanext, fl = 0;
3618 I32 f = flags, pos_before = 0;
3619 regnode * const oscan = scan;
3620 struct regnode_charclass_class this_class;
3621 struct regnode_charclass_class *oclass = NULL;
3622 I32 next_is_eval = 0;
3624 switch (PL_regkind[OP(scan)]) {
3625 case WHILEM: /* End of (?:...)* . */
3626 scan = NEXTOPER(scan);
3629 if (flags & (SCF_DO_SUBSTR | SCF_DO_STCLASS)) {
3630 next = NEXTOPER(scan);
3631 if (OP(next) == EXACT || (flags & SCF_DO_STCLASS)) {
3633 maxcount = REG_INFTY;
3634 next = regnext(scan);
3635 scan = NEXTOPER(scan);
3639 if (flags & SCF_DO_SUBSTR)
3644 if (flags & SCF_DO_STCLASS) {
3646 maxcount = REG_INFTY;
3647 next = regnext(scan);
3648 scan = NEXTOPER(scan);
3651 is_inf = is_inf_internal = 1;
3652 scan = regnext(scan);
3653 if (flags & SCF_DO_SUBSTR) {
3654 SCAN_COMMIT(pRExC_state, data, minlenp); /* Cannot extend fixed substrings */
3655 data->longest = &(data->longest_float);
3657 goto optimize_curly_tail;
3659 if (stopparen>0 && (OP(scan)==CURLYN || OP(scan)==CURLYM)
3660 && (scan->flags == stopparen))
3665 mincount = ARG1(scan);
3666 maxcount = ARG2(scan);
3668 next = regnext(scan);
3669 if (OP(scan) == CURLYX) {
3670 I32 lp = (data ? *(data->last_closep) : 0);
3671 scan->flags = ((lp <= (I32)U8_MAX) ? (U8)lp : U8_MAX);
3673 scan = NEXTOPER(scan) + EXTRA_STEP_2ARGS;
3674 next_is_eval = (OP(scan) == EVAL);
3676 if (flags & SCF_DO_SUBSTR) {
3677 if (mincount == 0) SCAN_COMMIT(pRExC_state,data,minlenp); /* Cannot extend fixed substrings */
3678 pos_before = data->pos_min;
3682 data->flags &= ~(SF_HAS_PAR|SF_IN_PAR|SF_HAS_EVAL);
3684 data->flags |= SF_IS_INF;
3686 if (flags & SCF_DO_STCLASS) {
3687 cl_init(pRExC_state, &this_class);
3688 oclass = data->start_class;
3689 data->start_class = &this_class;
3690 f |= SCF_DO_STCLASS_AND;
3691 f &= ~SCF_DO_STCLASS_OR;
3693 /* Exclude from super-linear cache processing any {n,m}
3694 regops for which the combination of input pos and regex
3695 pos is not enough information to determine if a match
3698 For example, in the regex /foo(bar\s*){4,8}baz/ with the
3699 regex pos at the \s*, the prospects for a match depend not
3700 only on the input position but also on how many (bar\s*)
3701 repeats into the {4,8} we are. */
3702 if ((mincount > 1) || (maxcount > 1 && maxcount != REG_INFTY))
3703 f &= ~SCF_WHILEM_VISITED_POS;
3705 /* This will finish on WHILEM, setting scan, or on NULL: */
3706 minnext = study_chunk(pRExC_state, &scan, minlenp, &deltanext,
3707 last, data, stopparen, recursed, NULL,
3709 ? (f & ~SCF_DO_SUBSTR) : f),depth+1);
3711 if (flags & SCF_DO_STCLASS)
3712 data->start_class = oclass;
3713 if (mincount == 0 || minnext == 0) {
3714 if (flags & SCF_DO_STCLASS_OR) {
3715 cl_or(pRExC_state, data->start_class, &this_class);
3717 else if (flags & SCF_DO_STCLASS_AND) {
3718 /* Switch to OR mode: cache the old value of
3719 * data->start_class */
3721 StructCopy(data->start_class, and_withp,
3722 struct regnode_charclass_class);
3723 flags &= ~SCF_DO_STCLASS_AND;
3724 StructCopy(&this_class, data->start_class,
3725 struct regnode_charclass_class);
3726 flags |= SCF_DO_STCLASS_OR;
3727 data->start_class->flags |= ANYOF_EOS;
3729 } else { /* Non-zero len */
3730 if (flags & SCF_DO_STCLASS_OR) {
3731 cl_or(pRExC_state, data->start_class, &this_class);
3732 cl_and(data->start_class, and_withp);
3734 else if (flags & SCF_DO_STCLASS_AND)
3735 cl_and(data->start_class, &this_class);
3736 flags &= ~SCF_DO_STCLASS;
3738 if (!scan) /* It was not CURLYX, but CURLY. */
3740 if ( /* ? quantifier ok, except for (?{ ... }) */
3741 (next_is_eval || !(mincount == 0 && maxcount == 1))
3742 && (minnext == 0) && (deltanext == 0)
3743 && data && !(data->flags & (SF_HAS_PAR|SF_IN_PAR))
3744 && maxcount <= REG_INFTY/3) /* Complement check for big count */
3746 ckWARNreg(RExC_parse,
3747 "Quantifier unexpected on zero-length expression");
3750 min += minnext * mincount;
3751 is_inf_internal |= ((maxcount == REG_INFTY
3752 && (minnext + deltanext) > 0)
3753 || deltanext == I32_MAX);
3754 is_inf |= is_inf_internal;
3755 delta += (minnext + deltanext) * maxcount - minnext * mincount;
3757 /* Try powerful optimization CURLYX => CURLYN. */
3758 if ( OP(oscan) == CURLYX && data
3759 && data->flags & SF_IN_PAR
3760 && !(data->flags & SF_HAS_EVAL)
3761 && !deltanext && minnext == 1 ) {
3762 /* Try to optimize to CURLYN. */
3763 regnode *nxt = NEXTOPER(oscan) + EXTRA_STEP_2ARGS;
3764 regnode * const nxt1 = nxt;
3771 if (!REGNODE_SIMPLE(OP(nxt))
3772 && !(PL_regkind[OP(nxt)] == EXACT
3773 && STR_LEN(nxt) == 1))
3779 if (OP(nxt) != CLOSE)
3781 if (RExC_open_parens) {
3782 RExC_open_parens[ARG(nxt1)-1]=oscan; /*open->CURLYM*/
3783 RExC_close_parens[ARG(nxt1)-1]=nxt+2; /*close->while*/
3785 /* Now we know that nxt2 is the only contents: */
3786 oscan->flags = (U8)ARG(nxt);
3788 OP(nxt1) = NOTHING; /* was OPEN. */
3791 OP(nxt1 + 1) = OPTIMIZED; /* was count. */
3792 NEXT_OFF(nxt1+ 1) = 0; /* just for consistency. */
3793 NEXT_OFF(nxt2) = 0; /* just for consistency with CURLY. */
3794 OP(nxt) = OPTIMIZED; /* was CLOSE. */
3795 OP(nxt + 1) = OPTIMIZED; /* was count. */
3796 NEXT_OFF(nxt+ 1) = 0; /* just for consistency. */
3801 /* Try optimization CURLYX => CURLYM. */
3802 if ( OP(oscan) == CURLYX && data
3803 && !(data->flags & SF_HAS_PAR)
3804 && !(data->flags & SF_HAS_EVAL)
3805 && !deltanext /* atom is fixed width */
3806 && minnext != 0 /* CURLYM can't handle zero width */
3808 /* XXXX How to optimize if data == 0? */
3809 /* Optimize to a simpler form. */
3810 regnode *nxt = NEXTOPER(oscan) + EXTRA_STEP_2ARGS; /* OPEN */
3814 while ( (nxt2 = regnext(nxt)) /* skip over embedded stuff*/
3815 && (OP(nxt2) != WHILEM))
3817 OP(nxt2) = SUCCEED; /* Whas WHILEM */
3818 /* Need to optimize away parenths. */
3819 if ((data->flags & SF_IN_PAR) && OP(nxt) == CLOSE) {
3820 /* Set the parenth number. */
3821 regnode *nxt1 = NEXTOPER(oscan) + EXTRA_STEP_2ARGS; /* OPEN*/
3823 oscan->flags = (U8)ARG(nxt);
3824 if (RExC_open_parens) {
3825 RExC_open_parens[ARG(nxt1)-1]=oscan; /*open->CURLYM*/
3826 RExC_close_parens[ARG(nxt1)-1]=nxt2+1; /*close->NOTHING*/
3828 OP(nxt1) = OPTIMIZED; /* was OPEN. */
3829 OP(nxt) = OPTIMIZED; /* was CLOSE. */
3832 OP(nxt1 + 1) = OPTIMIZED; /* was count. */
3833 OP(nxt + 1) = OPTIMIZED; /* was count. */
3834 NEXT_OFF(nxt1 + 1) = 0; /* just for consistency. */
3835 NEXT_OFF(nxt + 1) = 0; /* just for consistency. */
3838 while ( nxt1 && (OP(nxt1) != WHILEM)) {
3839 regnode *nnxt = regnext(nxt1);
3841 if (reg_off_by_arg[OP(nxt1)])
3842 ARG_SET(nxt1, nxt2 - nxt1);
3843 else if (nxt2 - nxt1 < U16_MAX)
3844 NEXT_OFF(nxt1) = nxt2 - nxt1;
3846 OP(nxt) = NOTHING; /* Cannot beautify */
3851 /* Optimize again: */
3852 study_chunk(pRExC_state, &nxt1, minlenp, &deltanext, nxt,
3853 NULL, stopparen, recursed, NULL, 0,depth+1);
3858 else if ((OP(oscan) == CURLYX)
3859 && (flags & SCF_WHILEM_VISITED_POS)
3860 /* See the comment on a similar expression above.
3861 However, this time it's not a subexpression
3862 we care about, but the expression itself. */
3863 && (maxcount == REG_INFTY)
3864 && data && ++data->whilem_c < 16) {
3865 /* This stays as CURLYX, we can put the count/of pair. */
3866 /* Find WHILEM (as in regexec.c) */
3867 regnode *nxt = oscan + NEXT_OFF(oscan);
3869 if (OP(PREVOPER(nxt)) == NOTHING) /* LONGJMP */
3871 PREVOPER(nxt)->flags = (U8)(data->whilem_c
3872 | (RExC_whilem_seen << 4)); /* On WHILEM */
3874 if (data && fl & (SF_HAS_PAR|SF_IN_PAR))
3876 if (flags & SCF_DO_SUBSTR) {
3877 SV *last_str = NULL;
3878 int counted = mincount != 0;
3880 if (data->last_end > 0 && mincount != 0) { /* Ends with a string. */
3881 #if defined(SPARC64_GCC_WORKAROUND)
3884 const char *s = NULL;
3887 if (pos_before >= data->last_start_min)
3890 b = data->last_start_min;
3893 s = SvPV_const(data->last_found, l);
3894 old = b - data->last_start_min;
3897 I32 b = pos_before >= data->last_start_min
3898 ? pos_before : data->last_start_min;
3900 const char * const s = SvPV_const(data->last_found, l);
3901 I32 old = b - data->last_start_min;
3905 old = utf8_hop((U8*)s, old) - (U8*)s;
3907 /* Get the added string: */
3908 last_str = newSVpvn_utf8(s + old, l, UTF);
3909 if (deltanext == 0 && pos_before == b) {
3910 /* What was added is a constant string */
3912 SvGROW(last_str, (mincount * l) + 1);
3913 repeatcpy(SvPVX(last_str) + l,
3914 SvPVX_const(last_str), l, mincount - 1);
3915 SvCUR_set(last_str, SvCUR(last_str) * mincount);
3916 /* Add additional parts. */
3917 SvCUR_set(data->last_found,
3918 SvCUR(data->last_found) - l);
3919 sv_catsv(data->last_found, last_str);
3921 SV * sv = data->last_found;
3923 SvUTF8(sv) && SvMAGICAL(sv) ?
3924 mg_find(sv, PERL_MAGIC_utf8) : NULL;
3925 if (mg && mg->mg_len >= 0)
3926 mg->mg_len += CHR_SVLEN(last_str) - l;
3928 data->last_end += l * (mincount - 1);
3931 /* start offset must point into the last copy */
3932 data->last_start_min += minnext * (mincount - 1);
3933 data->last_start_max += is_inf ? I32_MAX
3934 : (maxcount - 1) * (minnext + data->pos_delta);
3937 /* It is counted once already... */
3938 data->pos_min += minnext * (mincount - counted);
3939 data->pos_delta += - counted * deltanext +
3940 (minnext + deltanext) * maxcount - minnext * mincount;
3941 if (mincount != maxcount) {
3942 /* Cannot extend fixed substrings found inside
3944 SCAN_COMMIT(pRExC_state,data,minlenp);
3945 if (mincount && last_str) {
3946 SV * const sv = data->last_found;
3947 MAGIC * const mg = SvUTF8(sv) && SvMAGICAL(sv) ?
3948 mg_find(sv, PERL_MAGIC_utf8) : NULL;
3952 sv_setsv(sv, last_str);
3953 data->last_end = data->pos_min;
3954 data->last_start_min =
3955 data->pos_min - CHR_SVLEN(last_str);
3956 data->last_start_max = is_inf
3958 : data->pos_min + data->pos_delta
3959 - CHR_SVLEN(last_str);
3961 data->longest = &(data->longest_float);
3963 SvREFCNT_dec(last_str);
3965 if (data && (fl & SF_HAS_EVAL))
3966 data->flags |= SF_HAS_EVAL;
3967 optimize_curly_tail:
3968 if (OP(oscan) != CURLYX) {
3969 while (PL_regkind[OP(next = regnext(oscan))] == NOTHING
3971 NEXT_OFF(oscan) += NEXT_OFF(next);
3974 default: /* REF, ANYOFV, and CLUMP only? */
3975 if (flags & SCF_DO_SUBSTR) {
3976 SCAN_COMMIT(pRExC_state,data,minlenp); /* Cannot expect anything... */
3977 data->longest = &(data->longest_float);
3979 is_inf = is_inf_internal = 1;
3980 if (flags & SCF_DO_STCLASS_OR)
3981 cl_anything(pRExC_state, data->start_class);
3982 flags &= ~SCF_DO_STCLASS;
3986 else if (OP(scan) == LNBREAK) {
3987 if (flags & SCF_DO_STCLASS) {
3989 data->start_class->flags &= ~ANYOF_EOS; /* No match on empty */
3990 if (flags & SCF_DO_STCLASS_AND) {
3991 for (value = 0; value < 256; value++)
3992 if (!is_VERTWS_cp(value))
3993 ANYOF_BITMAP_CLEAR(data->start_class, value);
3996 for (value = 0; value < 256; value++)
3997 if (is_VERTWS_cp(value))
3998 ANYOF_BITMAP_SET(data->start_class, value);
4000 if (flags & SCF_DO_STCLASS_OR)
4001 cl_and(data->start_class, and_withp);
4002 flags &= ~SCF_DO_STCLASS;
4006 if (flags & SCF_DO_SUBSTR) {
4007 SCAN_COMMIT(pRExC_state,data,minlenp); /* Cannot expect anything... */
4009 data->pos_delta += 1;
4010 data->longest = &(data->longest_float);
4013 else if (REGNODE_SIMPLE(OP(scan))) {
4016 if (flags & SCF_DO_SUBSTR) {
4017 SCAN_COMMIT(pRExC_state,data,minlenp);
4021 if (flags & SCF_DO_STCLASS) {
4022 data->start_class->flags &= ~ANYOF_EOS; /* No match on empty */
4024 /* Some of the logic below assumes that switching
4025 locale on will only add false positives. */
4026 switch (PL_regkind[OP(scan)]) {
4030 /* Perl_croak(aTHX_ "panic: unexpected simple REx opcode %d", OP(scan)); */
4031 if (flags & SCF_DO_STCLASS_OR) /* Allow everything */
4032 cl_anything(pRExC_state, data->start_class);
4035 if (OP(scan) == SANY)
4037 if (flags & SCF_DO_STCLASS_OR) { /* Everything but \n */
4038 value = (ANYOF_BITMAP_TEST(data->start_class,'\n')
4039 || ANYOF_CLASS_TEST_ANY_SET(data->start_class));
4040 cl_anything(pRExC_state, data->start_class);
4042 if (flags & SCF_DO_STCLASS_AND || !value)
4043 ANYOF_BITMAP_CLEAR(data->start_class,'\n');
4046 if (flags & SCF_DO_STCLASS_AND)
4047 cl_and(data->start_class,
4048 (struct regnode_charclass_class*)scan);
4050 cl_or(pRExC_state, data->start_class,
4051 (struct regnode_charclass_class*)scan);
4054 if (flags & SCF_DO_STCLASS_AND) {
4055 if (!(data->start_class->flags & ANYOF_LOCALE)) {
4056 ANYOF_CLASS_CLEAR(data->start_class,ANYOF_NALNUM);
4057 if (OP(scan) == ALNUMU) {
4058 for (value = 0; value < 256; value++) {
4059 if (!isWORDCHAR_L1(value)) {
4060 ANYOF_BITMAP_CLEAR(data->start_class, value);
4064 for (value = 0; value < 256; value++) {
4065 if (!isALNUM(value)) {
4066 ANYOF_BITMAP_CLEAR(data->start_class, value);
4073 if (data->start_class->flags & ANYOF_LOCALE)
4074 ANYOF_CLASS_SET(data->start_class,ANYOF_ALNUM);
4076 /* Even if under locale, set the bits for non-locale
4077 * in case it isn't a true locale-node. This will
4078 * create false positives if it truly is locale */
4079 if (OP(scan) == ALNUMU) {
4080 for (value = 0; value < 256; value++) {
4081 if (isWORDCHAR_L1(value)) {
4082 ANYOF_BITMAP_SET(data->start_class, value);
4086 for (value = 0; value < 256; value++) {
4087 if (isALNUM(value)) {
4088 ANYOF_BITMAP_SET(data->start_class, value);
4095 if (flags & SCF_DO_STCLASS_AND) {
4096 if (!(data->start_class->flags & ANYOF_LOCALE)) {
4097 ANYOF_CLASS_CLEAR(data->start_class,ANYOF_ALNUM);
4098 if (OP(scan) == NALNUMU) {
4099 for (value = 0; value < 256; value++) {
4100 if (isWORDCHAR_L1(value)) {
4101 ANYOF_BITMAP_CLEAR(data->start_class, value);
4105 for (value = 0; value < 256; value++) {
4106 if (isALNUM(value)) {
4107 ANYOF_BITMAP_CLEAR(data->start_class, value);
4114 if (data->start_class->flags & ANYOF_LOCALE)
4115 ANYOF_CLASS_SET(data->start_class,ANYOF_NALNUM);
4117 /* Even if under locale, set the bits for non-locale in
4118 * case it isn't a true locale-node. This will create
4119 * false positives if it truly is locale */
4120 if (OP(scan) == NALNUMU) {
4121 for (value = 0; value < 256; value++) {
4122 if (! isWORDCHAR_L1(value)) {
4123 ANYOF_BITMAP_SET(data->start_class, value);
4127 for (value = 0; value < 256; value++) {
4128 if (! isALNUM(value)) {
4129 ANYOF_BITMAP_SET(data->start_class, value);
4136 if (flags & SCF_DO_STCLASS_AND) {
4137 if (!(data->start_class->flags & ANYOF_LOCALE)) {
4138 ANYOF_CLASS_CLEAR(data->start_class,ANYOF_NSPACE);
4139 if (OP(scan) == SPACEU) {
4140 for (value = 0; value < 256; value++) {
4141 if (!isSPACE_L1(value)) {
4142 ANYOF_BITMAP_CLEAR(data->start_class, value);
4146 for (value = 0; value < 256; value++) {
4147 if (!isSPACE(value)) {
4148 ANYOF_BITMAP_CLEAR(data->start_class, value);
4155 if (data->start_class->flags & ANYOF_LOCALE) {
4156 ANYOF_CLASS_SET(data->start_class,ANYOF_SPACE);
4158 if (OP(scan) == SPACEU) {
4159 for (value = 0; value < 256; value++) {
4160 if (isSPACE_L1(value)) {
4161 ANYOF_BITMAP_SET(data->start_class, value);
4165 for (value = 0; value < 256; value++) {
4166 if (isSPACE(value)) {
4167 ANYOF_BITMAP_SET(data->start_class, value);
4174 if (flags & SCF_DO_STCLASS_AND) {
4175 if (!(data->start_class->flags & ANYOF_LOCALE)) {
4176 ANYOF_CLASS_CLEAR(data->start_class,ANYOF_SPACE);
4177 if (OP(scan) == NSPACEU) {
4178 for (value = 0; value < 256; value++) {
4179 if (isSPACE_L1(value)) {
4180 ANYOF_BITMAP_CLEAR(data->start_class, value);
4184 for (value = 0; value < 256; value++) {
4185 if (isSPACE(value)) {
4186 ANYOF_BITMAP_CLEAR(data->start_class, value);
4193 if (data->start_class->flags & ANYOF_LOCALE)
4194 ANYOF_CLASS_SET(data->start_class,ANYOF_NSPACE);
4195 if (OP(scan) == NSPACEU) {
4196 for (value = 0; value < 256; value++) {
4197 if (!isSPACE_L1(value)) {
4198 ANYOF_BITMAP_SET(data->start_class, value);
4203 for (value = 0; value < 256; value++) {
4204 if (!isSPACE(value)) {
4205 ANYOF_BITMAP_SET(data->start_class, value);
4212 if (flags & SCF_DO_STCLASS_AND) {
4213 if (!(data->start_class->flags & ANYOF_LOCALE)) {
4214 ANYOF_CLASS_CLEAR(data->start_class,ANYOF_NDIGIT);
4215 for (value = 0; value < 256; value++)
4216 if (!isDIGIT(value))
4217 ANYOF_BITMAP_CLEAR(data->start_class, value);
4221 if (data->start_class->flags & ANYOF_LOCALE)
4222 ANYOF_CLASS_SET(data->start_class,ANYOF_DIGIT);
4223 for (value = 0; value < 256; value++)
4225 ANYOF_BITMAP_SET(data->start_class, value);
4229 if (flags & SCF_DO_STCLASS_AND) {
4230 if (!(data->start_class->flags & ANYOF_LOCALE))
4231 ANYOF_CLASS_CLEAR(data->start_class,ANYOF_DIGIT);
4232 for (value = 0; value < 256; value++)
4234 ANYOF_BITMAP_CLEAR(data->start_class, value);
4237 if (data->start_class->flags & ANYOF_LOCALE)
4238 ANYOF_CLASS_SET(data->start_class,ANYOF_NDIGIT);
4239 for (value = 0; value < 256; value++)
4240 if (!isDIGIT(value))
4241 ANYOF_BITMAP_SET(data->start_class, value);
4244 CASE_SYNST_FNC(VERTWS);
4245 CASE_SYNST_FNC(HORIZWS);
4248 if (flags & SCF_DO_STCLASS_OR)
4249 cl_and(data->start_class, and_withp);
4250 flags &= ~SCF_DO_STCLASS;
4253 else if (PL_regkind[OP(scan)] == EOL && flags & SCF_DO_SUBSTR) {
4254 data->flags |= (OP(scan) == MEOL
4258 else if ( PL_regkind[OP(scan)] == BRANCHJ
4259 /* Lookbehind, or need to calculate parens/evals/stclass: */
4260 && (scan->flags || data || (flags & SCF_DO_STCLASS))
4261 && (OP(scan) == IFMATCH || OP(scan) == UNLESSM)) {
4262 if ( !PERL_ENABLE_POSITIVE_ASSERTION_STUDY
4263 || OP(scan) == UNLESSM )
4265 /* Negative Lookahead/lookbehind
4266 In this case we can't do fixed string optimisation.
4269 I32 deltanext, minnext, fake = 0;
4271 struct regnode_charclass_class intrnl;
4274 data_fake.flags = 0;
4276 data_fake.whilem_c = data->whilem_c;
4277 data_fake.last_closep = data->last_closep;
4280 data_fake.last_closep = &fake;
4281 data_fake.pos_delta = delta;
4282 if ( flags & SCF_DO_STCLASS && !scan->flags
4283 && OP(scan) == IFMATCH ) { /* Lookahead */
4284 cl_init(pRExC_state, &intrnl);
4285 data_fake.start_class = &intrnl;
4286 f |= SCF_DO_STCLASS_AND;
4288 if (flags & SCF_WHILEM_VISITED_POS)
4289 f |= SCF_WHILEM_VISITED_POS;
4290 next = regnext(scan);
4291 nscan = NEXTOPER(NEXTOPER(scan));
4292 minnext = study_chunk(pRExC_state, &nscan, minlenp, &deltanext,
4293 last, &data_fake, stopparen, recursed, NULL, f, depth+1);
4296 FAIL("Variable length lookbehind not implemented");
4298 else if (minnext > (I32)U8_MAX) {
4299 FAIL2("Lookbehind longer than %"UVuf" not implemented", (UV)U8_MAX);
4301 scan->flags = (U8)minnext;
4304 if (data_fake.flags & (SF_HAS_PAR|SF_IN_PAR))
4306 if (data_fake.flags & SF_HAS_EVAL)
4307 data->flags |= SF_HAS_EVAL;
4308 data->whilem_c = data_fake.whilem_c;
4310 if (f & SCF_DO_STCLASS_AND) {
4311 if (flags & SCF_DO_STCLASS_OR) {
4312 /* OR before, AND after: ideally we would recurse with
4313 * data_fake to get the AND applied by study of the
4314 * remainder of the pattern, and then derecurse;
4315 * *** HACK *** for now just treat as "no information".
4316 * See [perl #56690].
4318 cl_init(pRExC_state, data->start_class);
4320 /* AND before and after: combine and continue */
4321 const int was = (data->start_class->flags & ANYOF_EOS);
4323 cl_and(data->start_class, &intrnl);
4325 data->start_class->flags |= ANYOF_EOS;
4329 #if PERL_ENABLE_POSITIVE_ASSERTION_STUDY
4331 /* Positive Lookahead/lookbehind
4332 In this case we can do fixed string optimisation,
4333 but we must be careful about it. Note in the case of
4334 lookbehind the positions will be offset by the minimum
4335 length of the pattern, something we won't know about
4336 until after the recurse.
4338 I32 deltanext, fake = 0;
4340 struct regnode_charclass_class intrnl;
4342 /* We use SAVEFREEPV so that when the full compile
4343 is finished perl will clean up the allocated
4344 minlens when it's all done. This way we don't
4345 have to worry about freeing them when we know
4346 they wont be used, which would be a pain.
4349 Newx( minnextp, 1, I32 );
4350 SAVEFREEPV(minnextp);
4353 StructCopy(data, &data_fake, scan_data_t);
4354 if ((flags & SCF_DO_SUBSTR) && data->last_found) {
4357 SCAN_COMMIT(pRExC_state, &data_fake,minlenp);
4358 data_fake.last_found=newSVsv(data->last_found);
4362 data_fake.last_closep = &fake;
4363 data_fake.flags = 0;
4364 data_fake.pos_delta = delta;
4366 data_fake.flags |= SF_IS_INF;
4367 if ( flags & SCF_DO_STCLASS && !scan->flags
4368 && OP(scan) == IFMATCH ) { /* Lookahead */
4369 cl_init(pRExC_state, &intrnl);
4370 data_fake.start_class = &intrnl;
4371 f |= SCF_DO_STCLASS_AND;
4373 if (flags & SCF_WHILEM_VISITED_POS)
4374 f |= SCF_WHILEM_VISITED_POS;
4375 next = regnext(scan);
4376 nscan = NEXTOPER(NEXTOPER(scan));
4378 *minnextp = study_chunk(pRExC_state, &nscan, minnextp, &deltanext,
4379 last, &data_fake, stopparen, recursed, NULL, f,depth+1);
4382 FAIL("Variable length lookbehind not implemented");
4384 else if (*minnextp > (I32)U8_MAX) {
4385 FAIL2("Lookbehind longer than %"UVuf" not implemented", (UV)U8_MAX);
4387 scan->flags = (U8)*minnextp;
4392 if (f & SCF_DO_STCLASS_AND) {
4393 const int was = (data->start_class->flags & ANYOF_EOS);
4395 cl_and(data->start_class, &intrnl);
4397 data->start_class->flags |= ANYOF_EOS;
4400 if (data_fake.flags & (SF_HAS_PAR|SF_IN_PAR))
4402 if (data_fake.flags & SF_HAS_EVAL)
4403 data->flags |= SF_HAS_EVAL;
4404 data->whilem_c = data_fake.whilem_c;
4405 if ((flags & SCF_DO_SUBSTR) && data_fake.last_found) {
4406 if (RExC_rx->minlen<*minnextp)
4407 RExC_rx->minlen=*minnextp;
4408 SCAN_COMMIT(pRExC_state, &data_fake, minnextp);
4409 SvREFCNT_dec(data_fake.last_found);
4411 if ( data_fake.minlen_fixed != minlenp )
4413 data->offset_fixed= data_fake.offset_fixed;
4414 data->minlen_fixed= data_fake.minlen_fixed;
4415 data->lookbehind_fixed+= scan->flags;
4417 if ( data_fake.minlen_float != minlenp )
4419 data->minlen_float= data_fake.minlen_float;
4420 data->offset_float_min=data_fake.offset_float_min;
4421 data->offset_float_max=data_fake.offset_float_max;
4422 data->lookbehind_float+= scan->flags;
4431 else if (OP(scan) == OPEN) {
4432 if (stopparen != (I32)ARG(scan))
4435 else if (OP(scan) == CLOSE) {
4436 if (stopparen == (I32)ARG(scan)) {
4439 if ((I32)ARG(scan) == is_par) {
4440 next = regnext(scan);
4442 if ( next && (OP(next) != WHILEM) && next < last)
4443 is_par = 0; /* Disable optimization */
4446 *(data->last_closep) = ARG(scan);
4448 else if (OP(scan) == EVAL) {
4450 data->flags |= SF_HAS_EVAL;
4452 else if ( PL_regkind[OP(scan)] == ENDLIKE ) {
4453 if (flags & SCF_DO_SUBSTR) {
4454 SCAN_COMMIT(pRExC_state,data,minlenp);
4455 flags &= ~SCF_DO_SUBSTR;
4457 if (data && OP(scan)==ACCEPT) {
4458 data->flags |= SCF_SEEN_ACCEPT;
4463 else if (OP(scan) == LOGICAL && scan->flags == 2) /* Embedded follows */
4465 if (flags & SCF_DO_SUBSTR) {
4466 SCAN_COMMIT(pRExC_state,data,minlenp);
4467 data->longest = &(data->longest_float);
4469 is_inf = is_inf_internal = 1;
4470 if (flags & SCF_DO_STCLASS_OR) /* Allow everything */
4471 cl_anything(pRExC_state, data->start_class);
4472 flags &= ~SCF_DO_STCLASS;
4474 else if (OP(scan) == GPOS) {
4475 if (!(RExC_rx->extflags & RXf_GPOS_FLOAT) &&
4476 !(delta || is_inf || (data && data->pos_delta)))
4478 if (!(RExC_rx->extflags & RXf_ANCH) && (flags & SCF_DO_SUBSTR))
4479 RExC_rx->extflags |= RXf_ANCH_GPOS;
4480 if (RExC_rx->gofs < (U32)min)
4481 RExC_rx->gofs = min;
4483 RExC_rx->extflags |= RXf_GPOS_FLOAT;
4487 #ifdef TRIE_STUDY_OPT
4488 #ifdef FULL_TRIE_STUDY
4489 else if (PL_regkind[OP(scan)] == TRIE) {
4490 /* NOTE - There is similar code to this block above for handling
4491 BRANCH nodes on the initial study. If you change stuff here
4493 regnode *trie_node= scan;
4494 regnode *tail= regnext(scan);
4495 reg_trie_data *trie = (reg_trie_data*)RExC_rxi->data->data[ ARG(scan) ];
4496 I32 max1 = 0, min1 = I32_MAX;
4497 struct regnode_charclass_class accum;
4499 if (flags & SCF_DO_SUBSTR) /* XXXX Add !SUSPEND? */
4500 SCAN_COMMIT(pRExC_state, data,minlenp); /* Cannot merge strings after this. */
4501 if (flags & SCF_DO_STCLASS)
4502 cl_init_zero(pRExC_state, &accum);
4508 const regnode *nextbranch= NULL;
4511 for ( word=1 ; word <= trie->wordcount ; word++)
4513 I32 deltanext=0, minnext=0, f = 0, fake;
4514 struct regnode_charclass_class this_class;
4516 data_fake.flags = 0;
4518 data_fake.whilem_c = data->whilem_c;
4519 data_fake.last_closep = data->last_closep;
4522 data_fake.last_closep = &fake;
4523 data_fake.pos_delta = delta;
4524 if (flags & SCF_DO_STCLASS) {
4525 cl_init(pRExC_state, &this_class);
4526 data_fake.start_class = &this_class;
4527 f = SCF_DO_STCLASS_AND;
4529 if (flags & SCF_WHILEM_VISITED_POS)
4530 f |= SCF_WHILEM_VISITED_POS;
4532 if (trie->jump[word]) {
4534 nextbranch = trie_node + trie->jump[0];
4535 scan= trie_node + trie->jump[word];
4536 /* We go from the jump point to the branch that follows
4537 it. Note this means we need the vestigal unused branches
4538 even though they arent otherwise used.
4540 minnext = study_chunk(pRExC_state, &scan, minlenp,
4541 &deltanext, (regnode *)nextbranch, &data_fake,
4542 stopparen, recursed, NULL, f,depth+1);
4544 if (nextbranch && PL_regkind[OP(nextbranch)]==BRANCH)
4545 nextbranch= regnext((regnode*)nextbranch);
4547 if (min1 > (I32)(minnext + trie->minlen))
4548 min1 = minnext + trie->minlen;
4549 if (max1 < (I32)(minnext + deltanext + trie->maxlen))
4550 max1 = minnext + deltanext + trie->maxlen;
4551 if (deltanext == I32_MAX)
4552 is_inf = is_inf_internal = 1;
4554 if (data_fake.flags & (SF_HAS_PAR|SF_IN_PAR))
4556 if (data_fake.flags & SCF_SEEN_ACCEPT) {
4557 if ( stopmin > min + min1)
4558 stopmin = min + min1;
4559 flags &= ~SCF_DO_SUBSTR;
4561 data->flags |= SCF_SEEN_ACCEPT;
4564 if (data_fake.flags & SF_HAS_EVAL)
4565 data->flags |= SF_HAS_EVAL;
4566 data->whilem_c = data_fake.whilem_c;
4568 if (flags & SCF_DO_STCLASS)
4569 cl_or(pRExC_state, &accum, &this_class);
4572 if (flags & SCF_DO_SUBSTR) {
4573 data->pos_min += min1;
4574 data->pos_delta += max1 - min1;
4575 if (max1 != min1 || is_inf)
4576 data->longest = &(data->longest_float);
4579 delta += max1 - min1;
4580 if (flags & SCF_DO_STCLASS_OR) {
4581 cl_or(pRExC_state, data->start_class, &accum);
4583 cl_and(data->start_class, and_withp);
4584 flags &= ~SCF_DO_STCLASS;
4587 else if (flags & SCF_DO_STCLASS_AND) {
4589 cl_and(data->start_class, &accum);
4590 flags &= ~SCF_DO_STCLASS;
4593 /* Switch to OR mode: cache the old value of
4594 * data->start_class */
4596 StructCopy(data->start_class, and_withp,
4597 struct regnode_charclass_class);
4598 flags &= ~SCF_DO_STCLASS_AND;
4599 StructCopy(&accum, data->start_class,
4600 struct regnode_charclass_class);
4601 flags |= SCF_DO_STCLASS_OR;
4602 data->start_class->flags |= ANYOF_EOS;
4609 else if (PL_regkind[OP(scan)] == TRIE) {
4610 reg_trie_data *trie = (reg_trie_data*)RExC_rxi->data->data[ ARG(scan) ];
4613 min += trie->minlen;
4614 delta += (trie->maxlen - trie->minlen);
4615 flags &= ~SCF_DO_STCLASS; /* xxx */
4616 if (flags & SCF_DO_SUBSTR) {
4617 SCAN_COMMIT(pRExC_state,data,minlenp); /* Cannot expect anything... */
4618 data->pos_min += trie->minlen;
4619 data->pos_delta += (trie->maxlen - trie->minlen);
4620 if (trie->maxlen != trie->minlen)
4621 data->longest = &(data->longest_float);
4623 if (trie->jump) /* no more substrings -- for now /grr*/
4624 flags &= ~SCF_DO_SUBSTR;
4626 #endif /* old or new */
4627 #endif /* TRIE_STUDY_OPT */
4629 /* Else: zero-length, ignore. */
4630 scan = regnext(scan);
4635 stopparen = frame->stop;
4636 frame = frame->prev;
4637 goto fake_study_recurse;
4642 DEBUG_STUDYDATA("pre-fin:",data,depth);
4645 *deltap = is_inf_internal ? I32_MAX : delta;
4646 if (flags & SCF_DO_SUBSTR && is_inf)
4647 data->pos_delta = I32_MAX - data->pos_min;
4648 if (is_par > (I32)U8_MAX)
4650 if (is_par && pars==1 && data) {
4651 data->flags |= SF_IN_PAR;
4652 data->flags &= ~SF_HAS_PAR;
4654 else if (pars && data) {
4655 data->flags |= SF_HAS_PAR;
4656 data->flags &= ~SF_IN_PAR;
4658 if (flags & SCF_DO_STCLASS_OR)
4659 cl_and(data->start_class, and_withp);
4660 if (flags & SCF_TRIE_RESTUDY)
4661 data->flags |= SCF_TRIE_RESTUDY;
4663 DEBUG_STUDYDATA("post-fin:",data,depth);
4665 return min < stopmin ? min : stopmin;
4669 S_add_data(RExC_state_t *pRExC_state, U32 n, const char *s)
4671 U32 count = RExC_rxi->data ? RExC_rxi->data->count : 0;
4673 PERL_ARGS_ASSERT_ADD_DATA;
4675 Renewc(RExC_rxi->data,
4676 sizeof(*RExC_rxi->data) + sizeof(void*) * (count + n - 1),
4677 char, struct reg_data);
4679 Renew(RExC_rxi->data->what, count + n, U8);
4681 Newx(RExC_rxi->data->what, n, U8);
4682 RExC_rxi->data->count = count + n;
4683 Copy(s, RExC_rxi->data->what + count, n, U8);
4687 /*XXX: todo make this not included in a non debugging perl */
4688 #ifndef PERL_IN_XSUB_RE
4690 Perl_reginitcolors(pTHX)
4693 const char * const s = PerlEnv_getenv("PERL_RE_COLORS");
4695 char *t = savepv(s);
4699 t = strchr(t, '\t');
4705 PL_colors[i] = t = (char *)"";
4710 PL_colors[i++] = (char *)"";
4717 #ifdef TRIE_STUDY_OPT
4718 #define CHECK_RESTUDY_GOTO \
4720 (data.flags & SCF_TRIE_RESTUDY) \
4724 #define CHECK_RESTUDY_GOTO
4728 - pregcomp - compile a regular expression into internal code
4730 * We can't allocate space until we know how big the compiled form will be,
4731 * but we can't compile it (and thus know how big it is) until we've got a
4732 * place to put the code. So we cheat: we compile it twice, once with code
4733 * generation turned off and size counting turned on, and once "for real".
4734 * This also means that we don't allocate space until we are sure that the
4735 * thing really will compile successfully, and we never have to move the
4736 * code and thus invalidate pointers into it. (Note that it has to be in
4737 * one piece because free() must be able to free it all.) [NB: not true in perl]
4739 * Beware that the optimization-preparation code in here knows about some
4740 * of the structure of the compiled regexp. [I'll say.]
4745 #ifndef PERL_IN_XSUB_RE
4746 #define RE_ENGINE_PTR &reh_regexp_engine
4748 extern const struct regexp_engine my_reg_engine;
4749 #define RE_ENGINE_PTR &my_reg_engine
4752 #ifndef PERL_IN_XSUB_RE
4754 Perl_pregcomp(pTHX_ SV * const pattern, const U32 flags)
4757 HV * const table = GvHV(PL_hintgv);
4759 PERL_ARGS_ASSERT_PREGCOMP;
4761 /* Dispatch a request to compile a regexp to correct
4764 SV **ptr= hv_fetchs(table, "regcomp", FALSE);
4765 GET_RE_DEBUG_FLAGS_DECL;
4766 if (ptr && SvIOK(*ptr) && SvIV(*ptr)) {
4767 const regexp_engine *eng=INT2PTR(regexp_engine*,SvIV(*ptr));
4769 PerlIO_printf(Perl_debug_log, "Using engine %"UVxf"\n",
4772 return CALLREGCOMP_ENG(eng, pattern, flags);
4775 return Perl_re_compile(aTHX_ pattern, flags);
4780 Perl_re_compile(pTHX_ SV * const pattern, U32 orig_pm_flags)
4785 register regexp_internal *ri;
4794 /* these are all flags - maybe they should be turned
4795 * into a single int with different bit masks */
4796 I32 sawlookahead = 0;
4799 bool used_setjump = FALSE;
4800 regex_charset initial_charset = get_regex_charset(orig_pm_flags);
4805 RExC_state_t RExC_state;
4806 RExC_state_t * const pRExC_state = &RExC_state;
4807 #ifdef TRIE_STUDY_OPT
4809 RExC_state_t copyRExC_state;
4811 GET_RE_DEBUG_FLAGS_DECL;
4813 PERL_ARGS_ASSERT_RE_COMPILE;
4815 DEBUG_r(if (!PL_colorset) reginitcolors());
4817 exp = SvPV(pattern, plen);
4819 if (plen == 0) { /* ignore the utf8ness if the pattern is 0 length */
4820 RExC_utf8 = RExC_orig_utf8 = 0;
4823 RExC_utf8 = RExC_orig_utf8 = SvUTF8(pattern);
4825 RExC_uni_semantics = 0;
4826 RExC_contains_locale = 0;
4828 /****************** LONG JUMP TARGET HERE***********************/
4829 /* Longjmp back to here if have to switch in midstream to utf8 */
4830 if (! RExC_orig_utf8) {
4831 JMPENV_PUSH(jump_ret);
4832 used_setjump = TRUE;
4835 if (jump_ret == 0) { /* First time through */
4839 SV *dsv= sv_newmortal();
4840 RE_PV_QUOTED_DECL(s, RExC_utf8,
4841 dsv, exp, plen, 60);
4842 PerlIO_printf(Perl_debug_log, "%sCompiling REx%s %s\n",
4843 PL_colors[4],PL_colors[5],s);
4846 else { /* longjumped back */
4849 /* If the cause for the longjmp was other than changing to utf8, pop
4850 * our own setjmp, and longjmp to the correct handler */
4851 if (jump_ret != UTF8_LONGJMP) {
4853 JMPENV_JUMP(jump_ret);
4858 /* It's possible to write a regexp in ascii that represents Unicode
4859 codepoints outside of the byte range, such as via \x{100}. If we
4860 detect such a sequence we have to convert the entire pattern to utf8
4861 and then recompile, as our sizing calculation will have been based
4862 on 1 byte == 1 character, but we will need to use utf8 to encode
4863 at least some part of the pattern, and therefore must convert the whole
4866 DEBUG_PARSE_r(PerlIO_printf(Perl_debug_log,
4867 "UTF8 mismatch! Converting to utf8 for resizing and compile\n"));
4868 exp = (char*)Perl_bytes_to_utf8(aTHX_
4869 (U8*)SvPV_nomg(pattern, plen),
4872 RExC_orig_utf8 = RExC_utf8 = 1;
4876 #ifdef TRIE_STUDY_OPT
4880 pm_flags = orig_pm_flags;
4882 if (initial_charset == REGEX_LOCALE_CHARSET) {
4883 RExC_contains_locale = 1;
4885 else if (RExC_utf8 && initial_charset == REGEX_DEPENDS_CHARSET) {
4887 /* Set to use unicode semantics if the pattern is in utf8 and has the
4888 * 'depends' charset specified, as it means unicode when utf8 */
4889 set_regex_charset(&pm_flags, REGEX_UNICODE_CHARSET);
4893 RExC_flags = pm_flags;
4897 RExC_in_lookbehind = 0;
4898 RExC_seen_zerolen = *exp == '^' ? -1 : 0;
4899 RExC_seen_evals = 0;
4901 RExC_override_recoding = 0;
4903 /* First pass: determine size, legality. */
4911 RExC_emit = &PL_regdummy;
4912 RExC_whilem_seen = 0;
4913 RExC_open_parens = NULL;
4914 RExC_close_parens = NULL;
4916 RExC_paren_names = NULL;
4918 RExC_paren_name_list = NULL;
4920 RExC_recurse = NULL;
4921 RExC_recurse_count = 0;
4923 #if 0 /* REGC() is (currently) a NOP at the first pass.
4924 * Clever compilers notice this and complain. --jhi */
4925 REGC((U8)REG_MAGIC, (char*)RExC_emit);
4928 PerlIO_printf(Perl_debug_log, "Starting first pass (sizing)\n");
4930 RExC_lastparse=NULL;
4932 if (reg(pRExC_state, 0, &flags,1) == NULL) {
4933 RExC_precomp = NULL;
4937 /* Here, finished first pass. Get rid of any added setjmp */
4943 PerlIO_printf(Perl_debug_log,
4944 "Required size %"IVdf" nodes\n"
4945 "Starting second pass (creation)\n",
4948 RExC_lastparse=NULL;
4951 /* The first pass could have found things that force Unicode semantics */
4952 if ((RExC_utf8 || RExC_uni_semantics)
4953 && get_regex_charset(pm_flags) == REGEX_DEPENDS_CHARSET)
4955 set_regex_charset(&pm_flags, REGEX_UNICODE_CHARSET);
4958 /* Small enough for pointer-storage convention?
4959 If extralen==0, this means that we will not need long jumps. */
4960 if (RExC_size >= 0x10000L && RExC_extralen)
4961 RExC_size += RExC_extralen;
4964 if (RExC_whilem_seen > 15)
4965 RExC_whilem_seen = 15;
4967 /* Allocate space and zero-initialize. Note, the two step process
4968 of zeroing when in debug mode, thus anything assigned has to
4969 happen after that */
4970 rx = (REGEXP*) newSV_type(SVt_REGEXP);
4971 r = (struct regexp*)SvANY(rx);
4972 Newxc(ri, sizeof(regexp_internal) + (unsigned)RExC_size * sizeof(regnode),
4973 char, regexp_internal);
4974 if ( r == NULL || ri == NULL )
4975 FAIL("Regexp out of space");
4977 /* avoid reading uninitialized memory in DEBUGGING code in study_chunk() */
4978 Zero(ri, sizeof(regexp_internal) + (unsigned)RExC_size * sizeof(regnode), char);
4980 /* bulk initialize base fields with 0. */
4981 Zero(ri, sizeof(regexp_internal), char);
4984 /* non-zero initialization begins here */
4986 r->engine= RE_ENGINE_PTR;
4987 r->extflags = pm_flags;
4989 bool has_p = ((r->extflags & RXf_PMf_KEEPCOPY) == RXf_PMf_KEEPCOPY);
4990 bool has_charset = (get_regex_charset(r->extflags) != REGEX_DEPENDS_CHARSET);
4992 /* The caret is output if there are any defaults: if not all the STD
4993 * flags are set, or if no character set specifier is needed */
4995 (((r->extflags & RXf_PMf_STD_PMMOD) != RXf_PMf_STD_PMMOD)
4997 bool has_runon = ((RExC_seen & REG_SEEN_RUN_ON_COMMENT)==REG_SEEN_RUN_ON_COMMENT);
4998 U16 reganch = (U16)((r->extflags & RXf_PMf_STD_PMMOD)
4999 >> RXf_PMf_STD_PMMOD_SHIFT);
5000 const char *fptr = STD_PAT_MODS; /*"msix"*/
5002 /* Allocate for the worst case, which is all the std flags are turned
5003 * on. If more precision is desired, we could do a population count of
5004 * the flags set. This could be done with a small lookup table, or by
5005 * shifting, masking and adding, or even, when available, assembly
5006 * language for a machine-language population count.
5007 * We never output a minus, as all those are defaults, so are
5008 * covered by the caret */
5009 const STRLEN wraplen = plen + has_p + has_runon
5010 + has_default /* If needs a caret */
5012 /* If needs a character set specifier */
5013 + ((has_charset) ? MAX_CHARSET_NAME_LENGTH : 0)
5014 + (sizeof(STD_PAT_MODS) - 1)
5015 + (sizeof("(?:)") - 1);
5017 p = sv_grow(MUTABLE_SV(rx), wraplen + 1); /* +1 for the ending NUL */
5019 SvFLAGS(rx) |= SvUTF8(pattern);
5022 /* If a default, cover it using the caret */
5024 *p++= DEFAULT_PAT_MOD;
5028 const char* const name = get_regex_charset_name(r->extflags, &len);
5029 Copy(name, p, len, char);
5033 *p++ = KEEPCOPY_PAT_MOD; /*'p'*/
5036 while((ch = *fptr++)) {
5044 Copy(RExC_precomp, p, plen, char);
5045 assert ((RX_WRAPPED(rx) - p) < 16);
5046 r->pre_prefix = p - RX_WRAPPED(rx);
5052 SvCUR_set(rx, p - SvPVX_const(rx));
5056 r->nparens = RExC_npar - 1; /* set early to validate backrefs */
5058 if (RExC_seen & REG_SEEN_RECURSE) {
5059 Newxz(RExC_open_parens, RExC_npar,regnode *);
5060 SAVEFREEPV(RExC_open_parens);
5061 Newxz(RExC_close_parens,RExC_npar,regnode *);
5062 SAVEFREEPV(RExC_close_parens);
5065 /* Useful during FAIL. */
5066 #ifdef RE_TRACK_PATTERN_OFFSETS
5067 Newxz(ri->u.offsets, 2*RExC_size+1, U32); /* MJD 20001228 */
5068 DEBUG_OFFSETS_r(PerlIO_printf(Perl_debug_log,
5069 "%s %"UVuf" bytes for offset annotations.\n",
5070 ri->u.offsets ? "Got" : "Couldn't get",
5071 (UV)((2*RExC_size+1) * sizeof(U32))));
5073 SetProgLen(ri,RExC_size);
5077 REH_CALL_COMP_BEGIN_HOOK(pRExC_state->rx);
5079 /* Second pass: emit code. */
5080 RExC_flags = pm_flags; /* don't let top level (?i) bleed */
5085 RExC_emit_start = ri->program;
5086 RExC_emit = ri->program;
5087 RExC_emit_bound = ri->program + RExC_size + 1;
5089 /* Store the count of eval-groups for security checks: */
5090 RExC_rx->seen_evals = RExC_seen_evals;
5091 REGC((U8)REG_MAGIC, (char*) RExC_emit++);
5092 if (reg(pRExC_state, 0, &flags,1) == NULL) {
5096 /* XXXX To minimize changes to RE engine we always allocate
5097 3-units-long substrs field. */
5098 Newx(r->substrs, 1, struct reg_substr_data);
5099 if (RExC_recurse_count) {
5100 Newxz(RExC_recurse,RExC_recurse_count,regnode *);
5101 SAVEFREEPV(RExC_recurse);
5105 r->minlen = minlen = sawlookahead = sawplus = sawopen = 0;
5106 Zero(r->substrs, 1, struct reg_substr_data);
5108 #ifdef TRIE_STUDY_OPT
5110 StructCopy(&zero_scan_data, &data, scan_data_t);
5111 copyRExC_state = RExC_state;
5114 DEBUG_OPTIMISE_r(PerlIO_printf(Perl_debug_log,"Restudying\n"));
5116 RExC_state = copyRExC_state;
5117 if (seen & REG_TOP_LEVEL_BRANCHES)
5118 RExC_seen |= REG_TOP_LEVEL_BRANCHES;
5120 RExC_seen &= ~REG_TOP_LEVEL_BRANCHES;
5121 if (data.last_found) {
5122 SvREFCNT_dec(data.longest_fixed);
5123 SvREFCNT_dec(data.longest_float);
5124 SvREFCNT_dec(data.last_found);
5126 StructCopy(&zero_scan_data, &data, scan_data_t);
5129 StructCopy(&zero_scan_data, &data, scan_data_t);
5132 /* Dig out information for optimizations. */
5133 r->extflags = RExC_flags; /* was pm_op */
5134 /*dmq: removed as part of de-PMOP: pm->op_pmflags = RExC_flags; */
5137 SvUTF8_on(rx); /* Unicode in it? */
5138 ri->regstclass = NULL;
5139 if (RExC_naughty >= 10) /* Probably an expensive pattern. */
5140 r->intflags |= PREGf_NAUGHTY;
5141 scan = ri->program + 1; /* First BRANCH. */
5143 /* testing for BRANCH here tells us whether there is "must appear"
5144 data in the pattern. If there is then we can use it for optimisations */
5145 if (!(RExC_seen & REG_TOP_LEVEL_BRANCHES)) { /* Only one top-level choice. */
5147 STRLEN longest_float_length, longest_fixed_length;
5148 struct regnode_charclass_class ch_class; /* pointed to by data */
5150 I32 last_close = 0; /* pointed to by data */
5151 regnode *first= scan;
5152 regnode *first_next= regnext(first);
5154 * Skip introductions and multiplicators >= 1
5155 * so that we can extract the 'meat' of the pattern that must
5156 * match in the large if() sequence following.
5157 * NOTE that EXACT is NOT covered here, as it is normally
5158 * picked up by the optimiser separately.
5160 * This is unfortunate as the optimiser isnt handling lookahead
5161 * properly currently.
5164 while ((OP(first) == OPEN && (sawopen = 1)) ||
5165 /* An OR of *one* alternative - should not happen now. */
5166 (OP(first) == BRANCH && OP(first_next) != BRANCH) ||
5167 /* for now we can't handle lookbehind IFMATCH*/
5168 (OP(first) == IFMATCH && !first->flags && (sawlookahead = 1)) ||
5169 (OP(first) == PLUS) ||
5170 (OP(first) == MINMOD) ||
5171 /* An {n,m} with n>0 */
5172 (PL_regkind[OP(first)] == CURLY && ARG1(first) > 0) ||
5173 (OP(first) == NOTHING && PL_regkind[OP(first_next)] != END ))
5176 * the only op that could be a regnode is PLUS, all the rest
5177 * will be regnode_1 or regnode_2.
5180 if (OP(first) == PLUS)
5183 first += regarglen[OP(first)];
5185 first = NEXTOPER(first);
5186 first_next= regnext(first);
5189 /* Starting-point info. */
5191 DEBUG_PEEP("first:",first,0);
5192 /* Ignore EXACT as we deal with it later. */
5193 if (PL_regkind[OP(first)] == EXACT) {
5194 if (OP(first) == EXACT)
5195 NOOP; /* Empty, get anchored substr later. */
5197 ri->regstclass = first;
5200 else if (PL_regkind[OP(first)] == TRIE &&
5201 ((reg_trie_data *)ri->data->data[ ARG(first) ])->minlen>0)
5204 /* this can happen only on restudy */
5205 if ( OP(first) == TRIE ) {
5206 struct regnode_1 *trieop = (struct regnode_1 *)
5207 PerlMemShared_calloc(1, sizeof(struct regnode_1));
5208 StructCopy(first,trieop,struct regnode_1);
5209 trie_op=(regnode *)trieop;
5211 struct regnode_charclass *trieop = (struct regnode_charclass *)
5212 PerlMemShared_calloc(1, sizeof(struct regnode_charclass));
5213 StructCopy(first,trieop,struct regnode_charclass);
5214 trie_op=(regnode *)trieop;
5217 make_trie_failtable(pRExC_state, (regnode *)first, trie_op, 0);
5218 ri->regstclass = trie_op;
5221 else if (REGNODE_SIMPLE(OP(first)))
5222 ri->regstclass = first;
5223 else if (PL_regkind[OP(first)] == BOUND ||
5224 PL_regkind[OP(first)] == NBOUND)
5225 ri->regstclass = first;
5226 else if (PL_regkind[OP(first)] == BOL) {
5227 r->extflags |= (OP(first) == MBOL
5229 : (OP(first) == SBOL
5232 first = NEXTOPER(first);
5235 else if (OP(first) == GPOS) {
5236 r->extflags |= RXf_ANCH_GPOS;
5237 first = NEXTOPER(first);
5240 else if ((!sawopen || !RExC_sawback) &&
5241 (OP(first) == STAR &&
5242 PL_regkind[OP(NEXTOPER(first))] == REG_ANY) &&
5243 !(r->extflags & RXf_ANCH) && !(RExC_seen & REG_SEEN_EVAL))
5245 /* turn .* into ^.* with an implied $*=1 */
5247 (OP(NEXTOPER(first)) == REG_ANY)
5250 r->extflags |= type;
5251 r->intflags |= PREGf_IMPLICIT;
5252 first = NEXTOPER(first);
5255 if (sawplus && !sawlookahead && (!sawopen || !RExC_sawback)
5256 && !(RExC_seen & REG_SEEN_EVAL)) /* May examine pos and $& */
5257 /* x+ must match at the 1st pos of run of x's */
5258 r->intflags |= PREGf_SKIP;
5260 /* Scan is after the zeroth branch, first is atomic matcher. */
5261 #ifdef TRIE_STUDY_OPT
5264 PerlIO_printf(Perl_debug_log, "first at %"IVdf"\n",
5265 (IV)(first - scan + 1))
5269 PerlIO_printf(Perl_debug_log, "first at %"IVdf"\n",
5270 (IV)(first - scan + 1))
5276 * If there's something expensive in the r.e., find the
5277 * longest literal string that must appear and make it the
5278 * regmust. Resolve ties in favor of later strings, since
5279 * the regstart check works with the beginning of the r.e.
5280 * and avoiding duplication strengthens checking. Not a
5281 * strong reason, but sufficient in the absence of others.
5282 * [Now we resolve ties in favor of the earlier string if
5283 * it happens that c_offset_min has been invalidated, since the
5284 * earlier string may buy us something the later one won't.]
5287 data.longest_fixed = newSVpvs("");
5288 data.longest_float = newSVpvs("");
5289 data.last_found = newSVpvs("");
5290 data.longest = &(data.longest_fixed);
5292 if (!ri->regstclass) {
5293 cl_init(pRExC_state, &ch_class);
5294 data.start_class = &ch_class;
5295 stclass_flag = SCF_DO_STCLASS_AND;
5296 } else /* XXXX Check for BOUND? */
5298 data.last_closep = &last_close;
5300 minlen = study_chunk(pRExC_state, &first, &minlen, &fake, scan + RExC_size, /* Up to end */
5301 &data, -1, NULL, NULL,
5302 SCF_DO_SUBSTR | SCF_WHILEM_VISITED_POS | stclass_flag,0);
5308 if ( RExC_npar == 1 && data.longest == &(data.longest_fixed)
5309 && data.last_start_min == 0 && data.last_end > 0
5310 && !RExC_seen_zerolen
5311 && !(RExC_seen & REG_SEEN_VERBARG)
5312 && (!(RExC_seen & REG_SEEN_GPOS) || (r->extflags & RXf_ANCH_GPOS)))
5313 r->extflags |= RXf_CHECK_ALL;
5314 scan_commit(pRExC_state, &data,&minlen,0);
5315 SvREFCNT_dec(data.last_found);
5317 /* Note that code very similar to this but for anchored string
5318 follows immediately below, changes may need to be made to both.
5321 longest_float_length = CHR_SVLEN(data.longest_float);
5322 if (longest_float_length
5323 || (data.flags & SF_FL_BEFORE_EOL
5324 && (!(data.flags & SF_FL_BEFORE_MEOL)
5325 || (RExC_flags & RXf_PMf_MULTILINE))))
5329 /* See comments for join_exact for why REG_SEEN_EXACTF_SHARP_S */
5330 if ((RExC_seen & REG_SEEN_EXACTF_SHARP_S)
5331 || (SvCUR(data.longest_fixed) /* ok to leave SvCUR */
5332 && data.offset_fixed == data.offset_float_min
5333 && SvCUR(data.longest_fixed) == SvCUR(data.longest_float)))
5334 goto remove_float; /* As in (a)+. */
5336 /* copy the information about the longest float from the reg_scan_data
5337 over to the program. */
5338 if (SvUTF8(data.longest_float)) {
5339 r->float_utf8 = data.longest_float;
5340 r->float_substr = NULL;
5342 r->float_substr = data.longest_float;
5343 r->float_utf8 = NULL;
5345 /* float_end_shift is how many chars that must be matched that
5346 follow this item. We calculate it ahead of time as once the
5347 lookbehind offset is added in we lose the ability to correctly
5349 ml = data.minlen_float ? *(data.minlen_float)
5350 : (I32)longest_float_length;
5351 r->float_end_shift = ml - data.offset_float_min
5352 - longest_float_length + (SvTAIL(data.longest_float) != 0)
5353 + data.lookbehind_float;
5354 r->float_min_offset = data.offset_float_min - data.lookbehind_float;
5355 r->float_max_offset = data.offset_float_max;
5356 if (data.offset_float_max < I32_MAX) /* Don't offset infinity */
5357 r->float_max_offset -= data.lookbehind_float;
5359 t = (data.flags & SF_FL_BEFORE_EOL /* Can't have SEOL and MULTI */
5360 && (!(data.flags & SF_FL_BEFORE_MEOL)
5361 || (RExC_flags & RXf_PMf_MULTILINE)));
5362 fbm_compile(data.longest_float, t ? FBMcf_TAIL : 0);
5366 r->float_substr = r->float_utf8 = NULL;
5367 SvREFCNT_dec(data.longest_float);
5368 longest_float_length = 0;
5371 /* Note that code very similar to this but for floating string
5372 is immediately above, changes may need to be made to both.
5375 longest_fixed_length = CHR_SVLEN(data.longest_fixed);
5377 /* See comments for join_exact for why REG_SEEN_EXACTF_SHARP_S */
5378 if (! (RExC_seen & REG_SEEN_EXACTF_SHARP_S)
5379 && (longest_fixed_length
5380 || (data.flags & SF_FIX_BEFORE_EOL /* Cannot have SEOL and MULTI */
5381 && (!(data.flags & SF_FIX_BEFORE_MEOL)
5382 || (RExC_flags & RXf_PMf_MULTILINE)))) )
5386 /* copy the information about the longest fixed
5387 from the reg_scan_data over to the program. */
5388 if (SvUTF8(data.longest_fixed)) {
5389 r->anchored_utf8 = data.longest_fixed;
5390 r->anchored_substr = NULL;
5392 r->anchored_substr = data.longest_fixed;
5393 r->anchored_utf8 = NULL;
5395 /* fixed_end_shift is how many chars that must be matched that
5396 follow this item. We calculate it ahead of time as once the
5397 lookbehind offset is added in we lose the ability to correctly
5399 ml = data.minlen_fixed ? *(data.minlen_fixed)
5400 : (I32)longest_fixed_length;
5401 r->anchored_end_shift = ml - data.offset_fixed
5402 - longest_fixed_length + (SvTAIL(data.longest_fixed) != 0)
5403 + data.lookbehind_fixed;
5404 r->anchored_offset = data.offset_fixed - data.lookbehind_fixed;
5406 t = (data.flags & SF_FIX_BEFORE_EOL /* Can't have SEOL and MULTI */
5407 && (!(data.flags & SF_FIX_BEFORE_MEOL)
5408 || (RExC_flags & RXf_PMf_MULTILINE)));
5409 fbm_compile(data.longest_fixed, t ? FBMcf_TAIL : 0);
5412 r->anchored_substr = r->anchored_utf8 = NULL;
5413 SvREFCNT_dec(data.longest_fixed);
5414 longest_fixed_length = 0;
5417 && (OP(ri->regstclass) == REG_ANY || OP(ri->regstclass) == SANY))
5418 ri->regstclass = NULL;
5420 if ((!(r->anchored_substr || r->anchored_utf8) || r->anchored_offset)
5422 && !(data.start_class->flags & ANYOF_EOS)
5423 && !cl_is_anything(data.start_class))
5425 const U32 n = add_data(pRExC_state, 1, "f");
5426 data.start_class->flags |= ANYOF_IS_SYNTHETIC;
5428 Newx(RExC_rxi->data->data[n], 1,
5429 struct regnode_charclass_class);
5430 StructCopy(data.start_class,
5431 (struct regnode_charclass_class*)RExC_rxi->data->data[n],
5432 struct regnode_charclass_class);
5433 ri->regstclass = (regnode*)RExC_rxi->data->data[n];
5434 r->intflags &= ~PREGf_SKIP; /* Used in find_byclass(). */
5435 DEBUG_COMPILE_r({ SV *sv = sv_newmortal();
5436 regprop(r, sv, (regnode*)data.start_class);
5437 PerlIO_printf(Perl_debug_log,
5438 "synthetic stclass \"%s\".\n",
5439 SvPVX_const(sv));});
5442 /* A temporary algorithm prefers floated substr to fixed one to dig more info. */
5443 if (longest_fixed_length > longest_float_length) {
5444 r->check_end_shift = r->anchored_end_shift;
5445 r->check_substr = r->anchored_substr;
5446 r->check_utf8 = r->anchored_utf8;
5447 r->check_offset_min = r->check_offset_max = r->anchored_offset;
5448 if (r->extflags & RXf_ANCH_SINGLE)
5449 r->extflags |= RXf_NOSCAN;
5452 r->check_end_shift = r->float_end_shift;
5453 r->check_substr = r->float_substr;
5454 r->check_utf8 = r->float_utf8;
5455 r->check_offset_min = r->float_min_offset;
5456 r->check_offset_max = r->float_max_offset;
5458 /* XXXX Currently intuiting is not compatible with ANCH_GPOS.
5459 This should be changed ASAP! */
5460 if ((r->check_substr || r->check_utf8) && !(r->extflags & RXf_ANCH_GPOS)) {
5461 r->extflags |= RXf_USE_INTUIT;
5462 if (SvTAIL(r->check_substr ? r->check_substr : r->check_utf8))
5463 r->extflags |= RXf_INTUIT_TAIL;
5465 /* XXX Unneeded? dmq (shouldn't as this is handled elsewhere)
5466 if ( (STRLEN)minlen < longest_float_length )
5467 minlen= longest_float_length;
5468 if ( (STRLEN)minlen < longest_fixed_length )
5469 minlen= longest_fixed_length;
5473 /* Several toplevels. Best we can is to set minlen. */
5475 struct regnode_charclass_class ch_class;
5478 DEBUG_PARSE_r(PerlIO_printf(Perl_debug_log, "\nMulti Top Level\n"));
5480 scan = ri->program + 1;
5481 cl_init(pRExC_state, &ch_class);
5482 data.start_class = &ch_class;
5483 data.last_closep = &last_close;
5486 minlen = study_chunk(pRExC_state, &scan, &minlen, &fake, scan + RExC_size,
5487 &data, -1, NULL, NULL, SCF_DO_STCLASS_AND|SCF_WHILEM_VISITED_POS,0);
5491 r->check_substr = r->check_utf8 = r->anchored_substr = r->anchored_utf8
5492 = r->float_substr = r->float_utf8 = NULL;
5494 if (!(data.start_class->flags & ANYOF_EOS)
5495 && !cl_is_anything(data.start_class))
5497 const U32 n = add_data(pRExC_state, 1, "f");
5498 data.start_class->flags |= ANYOF_IS_SYNTHETIC;
5500 Newx(RExC_rxi->data->data[n], 1,
5501 struct regnode_charclass_class);
5502 StructCopy(data.start_class,
5503 (struct regnode_charclass_class*)RExC_rxi->data->data[n],
5504 struct regnode_charclass_class);
5505 ri->regstclass = (regnode*)RExC_rxi->data->data[n];
5506 r->intflags &= ~PREGf_SKIP; /* Used in find_byclass(). */
5507 DEBUG_COMPILE_r({ SV* sv = sv_newmortal();
5508 regprop(r, sv, (regnode*)data.start_class);
5509 PerlIO_printf(Perl_debug_log,
5510 "synthetic stclass \"%s\".\n",
5511 SvPVX_const(sv));});
5515 /* Guard against an embedded (?=) or (?<=) with a longer minlen than
5516 the "real" pattern. */
5518 PerlIO_printf(Perl_debug_log,"minlen: %"IVdf" r->minlen:%"IVdf"\n",
5519 (IV)minlen, (IV)r->minlen);
5521 r->minlenret = minlen;
5522 if (r->minlen < minlen)
5525 if (RExC_seen & REG_SEEN_GPOS)
5526 r->extflags |= RXf_GPOS_SEEN;
5527 if (RExC_seen & REG_SEEN_LOOKBEHIND)
5528 r->extflags |= RXf_LOOKBEHIND_SEEN;
5529 if (RExC_seen & REG_SEEN_EVAL)
5530 r->extflags |= RXf_EVAL_SEEN;
5531 if (RExC_seen & REG_SEEN_CANY)
5532 r->extflags |= RXf_CANY_SEEN;
5533 if (RExC_seen & REG_SEEN_VERBARG)
5534 r->intflags |= PREGf_VERBARG_SEEN;
5535 if (RExC_seen & REG_SEEN_CUTGROUP)
5536 r->intflags |= PREGf_CUTGROUP_SEEN;
5537 if (RExC_paren_names)
5538 RXp_PAREN_NAMES(r) = MUTABLE_HV(SvREFCNT_inc(RExC_paren_names));
5540 RXp_PAREN_NAMES(r) = NULL;
5542 #ifdef STUPID_PATTERN_CHECKS
5543 if (RX_PRELEN(rx) == 0)
5544 r->extflags |= RXf_NULL;
5545 if (r->extflags & RXf_SPLIT && RX_PRELEN(rx) == 1 && RX_PRECOMP(rx)[0] == ' ')
5546 /* XXX: this should happen BEFORE we compile */
5547 r->extflags |= (RXf_SKIPWHITE|RXf_WHITE);
5548 else if (RX_PRELEN(rx) == 3 && memEQ("\\s+", RX_PRECOMP(rx), 3))
5549 r->extflags |= RXf_WHITE;
5550 else if (RX_PRELEN(rx) == 1 && RXp_PRECOMP(rx)[0] == '^')
5551 r->extflags |= RXf_START_ONLY;
5553 if (r->extflags & RXf_SPLIT && RX_PRELEN(rx) == 1 && RX_PRECOMP(rx)[0] == ' ')
5554 /* XXX: this should happen BEFORE we compile */
5555 r->extflags |= (RXf_SKIPWHITE|RXf_WHITE);
5557 regnode *first = ri->program + 1;
5560 if (PL_regkind[fop] == NOTHING && OP(NEXTOPER(first)) == END)
5561 r->extflags |= RXf_NULL;
5562 else if (PL_regkind[fop] == BOL && OP(NEXTOPER(first)) == END)
5563 r->extflags |= RXf_START_ONLY;
5564 else if (fop == PLUS && OP(NEXTOPER(first)) == SPACE
5565 && OP(regnext(first)) == END)
5566 r->extflags |= RXf_WHITE;
5570 if (RExC_paren_names) {
5571 ri->name_list_idx = add_data( pRExC_state, 1, "a" );
5572 ri->data->data[ri->name_list_idx] = (void*)SvREFCNT_inc(RExC_paren_name_list);
5575 ri->name_list_idx = 0;
5577 if (RExC_recurse_count) {
5578 for ( ; RExC_recurse_count ; RExC_recurse_count-- ) {
5579 const regnode *scan = RExC_recurse[RExC_recurse_count-1];
5580 ARG2L_SET( scan, RExC_open_parens[ARG(scan)-1] - scan );
5583 Newxz(r->offs, RExC_npar, regexp_paren_pair);
5584 /* assume we don't need to swap parens around before we match */
5587 PerlIO_printf(Perl_debug_log,"Final program:\n");
5590 #ifdef RE_TRACK_PATTERN_OFFSETS
5591 DEBUG_OFFSETS_r(if (ri->u.offsets) {
5592 const U32 len = ri->u.offsets[0];
5594 GET_RE_DEBUG_FLAGS_DECL;
5595 PerlIO_printf(Perl_debug_log, "Offsets: [%"UVuf"]\n\t", (UV)ri->u.offsets[0]);
5596 for (i = 1; i <= len; i++) {
5597 if (ri->u.offsets[i*2-1] || ri->u.offsets[i*2])
5598 PerlIO_printf(Perl_debug_log, "%"UVuf":%"UVuf"[%"UVuf"] ",
5599 (UV)i, (UV)ri->u.offsets[i*2-1], (UV)ri->u.offsets[i*2]);
5601 PerlIO_printf(Perl_debug_log, "\n");
5607 #undef RE_ENGINE_PTR
5611 Perl_reg_named_buff(pTHX_ REGEXP * const rx, SV * const key, SV * const value,
5614 PERL_ARGS_ASSERT_REG_NAMED_BUFF;
5616 PERL_UNUSED_ARG(value);
5618 if (flags & RXapif_FETCH) {
5619 return reg_named_buff_fetch(rx, key, flags);
5620 } else if (flags & (RXapif_STORE | RXapif_DELETE | RXapif_CLEAR)) {
5621 Perl_croak_no_modify(aTHX);
5623 } else if (flags & RXapif_EXISTS) {
5624 return reg_named_buff_exists(rx, key, flags)
5627 } else if (flags & RXapif_REGNAMES) {
5628 return reg_named_buff_all(rx, flags);
5629 } else if (flags & (RXapif_SCALAR | RXapif_REGNAMES_COUNT)) {
5630 return reg_named_buff_scalar(rx, flags);
5632 Perl_croak(aTHX_ "panic: Unknown flags %d in named_buff", (int)flags);
5638 Perl_reg_named_buff_iter(pTHX_ REGEXP * const rx, const SV * const lastkey,
5641 PERL_ARGS_ASSERT_REG_NAMED_BUFF_ITER;
5642 PERL_UNUSED_ARG(lastkey);
5644 if (flags & RXapif_FIRSTKEY)
5645 return reg_named_buff_firstkey(rx, flags);
5646 else if (flags & RXapif_NEXTKEY)
5647 return reg_named_buff_nextkey(rx, flags);
5649 Perl_croak(aTHX_ "panic: Unknown flags %d in named_buff_iter", (int)flags);
5655 Perl_reg_named_buff_fetch(pTHX_ REGEXP * const r, SV * const namesv,
5658 AV *retarray = NULL;
5660 struct regexp *const rx = (struct regexp *)SvANY(r);
5662 PERL_ARGS_ASSERT_REG_NAMED_BUFF_FETCH;
5664 if (flags & RXapif_ALL)
5667 if (rx && RXp_PAREN_NAMES(rx)) {
5668 HE *he_str = hv_fetch_ent( RXp_PAREN_NAMES(rx), namesv, 0, 0 );
5671 SV* sv_dat=HeVAL(he_str);
5672 I32 *nums=(I32*)SvPVX(sv_dat);
5673 for ( i=0; i<SvIVX(sv_dat); i++ ) {
5674 if ((I32)(rx->nparens) >= nums[i]
5675 && rx->offs[nums[i]].start != -1
5676 && rx->offs[nums[i]].end != -1)
5679 CALLREG_NUMBUF_FETCH(r,nums[i],ret);
5684 ret = newSVsv(&PL_sv_undef);
5687 av_push(retarray, ret);
5690 return newRV_noinc(MUTABLE_SV(retarray));
5697 Perl_reg_named_buff_exists(pTHX_ REGEXP * const r, SV * const key,
5700 struct regexp *const rx = (struct regexp *)SvANY(r);
5702 PERL_ARGS_ASSERT_REG_NAMED_BUFF_EXISTS;
5704 if (rx && RXp_PAREN_NAMES(rx)) {
5705 if (flags & RXapif_ALL) {
5706 return hv_exists_ent(RXp_PAREN_NAMES(rx), key, 0);
5708 SV *sv = CALLREG_NAMED_BUFF_FETCH(r, key, flags);
5722 Perl_reg_named_buff_firstkey(pTHX_ REGEXP * const r, const U32 flags)
5724 struct regexp *const rx = (struct regexp *)SvANY(r);
5726 PERL_ARGS_ASSERT_REG_NAMED_BUFF_FIRSTKEY;
5728 if ( rx && RXp_PAREN_NAMES(rx) ) {
5729 (void)hv_iterinit(RXp_PAREN_NAMES(rx));
5731 return CALLREG_NAMED_BUFF_NEXTKEY(r, NULL, flags & ~RXapif_FIRSTKEY);
5738 Perl_reg_named_buff_nextkey(pTHX_ REGEXP * const r, const U32 flags)
5740 struct regexp *const rx = (struct regexp *)SvANY(r);
5741 GET_RE_DEBUG_FLAGS_DECL;
5743 PERL_ARGS_ASSERT_REG_NAMED_BUFF_NEXTKEY;
5745 if (rx && RXp_PAREN_NAMES(rx)) {
5746 HV *hv = RXp_PAREN_NAMES(rx);
5748 while ( (temphe = hv_iternext_flags(hv,0)) ) {
5751 SV* sv_dat = HeVAL(temphe);
5752 I32 *nums = (I32*)SvPVX(sv_dat);
5753 for ( i = 0; i < SvIVX(sv_dat); i++ ) {
5754 if ((I32)(rx->lastparen) >= nums[i] &&
5755 rx->offs[nums[i]].start != -1 &&
5756 rx->offs[nums[i]].end != -1)
5762 if (parno || flags & RXapif_ALL) {
5763 return newSVhek(HeKEY_hek(temphe));
5771 Perl_reg_named_buff_scalar(pTHX_ REGEXP * const r, const U32 flags)
5776 struct regexp *const rx = (struct regexp *)SvANY(r);
5778 PERL_ARGS_ASSERT_REG_NAMED_BUFF_SCALAR;
5780 if (rx && RXp_PAREN_NAMES(rx)) {
5781 if (flags & (RXapif_ALL | RXapif_REGNAMES_COUNT)) {
5782 return newSViv(HvTOTALKEYS(RXp_PAREN_NAMES(rx)));
5783 } else if (flags & RXapif_ONE) {
5784 ret = CALLREG_NAMED_BUFF_ALL(r, (flags | RXapif_REGNAMES));
5785 av = MUTABLE_AV(SvRV(ret));
5786 length = av_len(av);
5788 return newSViv(length + 1);
5790 Perl_croak(aTHX_ "panic: Unknown flags %d in named_buff_scalar", (int)flags);
5794 return &PL_sv_undef;
5798 Perl_reg_named_buff_all(pTHX_ REGEXP * const r, const U32 flags)
5800 struct regexp *const rx = (struct regexp *)SvANY(r);
5803 PERL_ARGS_ASSERT_REG_NAMED_BUFF_ALL;
5805 if (rx && RXp_PAREN_NAMES(rx)) {
5806 HV *hv= RXp_PAREN_NAMES(rx);
5808 (void)hv_iterinit(hv);
5809 while ( (temphe = hv_iternext_flags(hv,0)) ) {
5812 SV* sv_dat = HeVAL(temphe);
5813 I32 *nums = (I32*)SvPVX(sv_dat);
5814 for ( i = 0; i < SvIVX(sv_dat); i++ ) {
5815 if ((I32)(rx->lastparen) >= nums[i] &&
5816 rx->offs[nums[i]].start != -1 &&
5817 rx->offs[nums[i]].end != -1)
5823 if (parno || flags & RXapif_ALL) {
5824 av_push(av, newSVhek(HeKEY_hek(temphe)));
5829 return newRV_noinc(MUTABLE_SV(av));
5833 Perl_reg_numbered_buff_fetch(pTHX_ REGEXP * const r, const I32 paren,
5836 struct regexp *const rx = (struct regexp *)SvANY(r);
5841 PERL_ARGS_ASSERT_REG_NUMBERED_BUFF_FETCH;
5844 sv_setsv(sv,&PL_sv_undef);
5848 if (paren == RX_BUFF_IDX_PREMATCH && rx->offs[0].start != -1) {
5850 i = rx->offs[0].start;
5854 if (paren == RX_BUFF_IDX_POSTMATCH && rx->offs[0].end != -1) {
5856 s = rx->subbeg + rx->offs[0].end;
5857 i = rx->sublen - rx->offs[0].end;
5860 if ( 0 <= paren && paren <= (I32)rx->nparens &&
5861 (s1 = rx->offs[paren].start) != -1 &&
5862 (t1 = rx->offs[paren].end) != -1)
5866 s = rx->subbeg + s1;
5868 sv_setsv(sv,&PL_sv_undef);
5871 assert(rx->sublen >= (s - rx->subbeg) + i );
5873 const int oldtainted = PL_tainted;
5875 sv_setpvn(sv, s, i);
5876 PL_tainted = oldtainted;
5877 if ( (rx->extflags & RXf_CANY_SEEN)
5878 ? (RXp_MATCH_UTF8(rx)
5879 && (!i || is_utf8_string((U8*)s, i)))
5880 : (RXp_MATCH_UTF8(rx)) )
5887 if (RXp_MATCH_TAINTED(rx)) {
5888 if (SvTYPE(sv) >= SVt_PVMG) {
5889 MAGIC* const mg = SvMAGIC(sv);
5892 SvMAGIC_set(sv, mg->mg_moremagic);
5894 if ((mgt = SvMAGIC(sv))) {
5895 mg->mg_moremagic = mgt;
5896 SvMAGIC_set(sv, mg);
5906 sv_setsv(sv,&PL_sv_undef);
5912 Perl_reg_numbered_buff_store(pTHX_ REGEXP * const rx, const I32 paren,
5913 SV const * const value)
5915 PERL_ARGS_ASSERT_REG_NUMBERED_BUFF_STORE;
5917 PERL_UNUSED_ARG(rx);
5918 PERL_UNUSED_ARG(paren);
5919 PERL_UNUSED_ARG(value);
5922 Perl_croak_no_modify(aTHX);
5926 Perl_reg_numbered_buff_length(pTHX_ REGEXP * const r, const SV * const sv,
5929 struct regexp *const rx = (struct regexp *)SvANY(r);
5933 PERL_ARGS_ASSERT_REG_NUMBERED_BUFF_LENGTH;
5935 /* Some of this code was originally in C<Perl_magic_len> in F<mg.c> */
5937 /* $` / ${^PREMATCH} */
5938 case RX_BUFF_IDX_PREMATCH:
5939 if (rx->offs[0].start != -1) {
5940 i = rx->offs[0].start;
5948 /* $' / ${^POSTMATCH} */
5949 case RX_BUFF_IDX_POSTMATCH:
5950 if (rx->offs[0].end != -1) {
5951 i = rx->sublen - rx->offs[0].end;
5953 s1 = rx->offs[0].end;
5959 /* $& / ${^MATCH}, $1, $2, ... */
5961 if (paren <= (I32)rx->nparens &&
5962 (s1 = rx->offs[paren].start) != -1 &&
5963 (t1 = rx->offs[paren].end) != -1)
5968 if (ckWARN(WARN_UNINITIALIZED))
5969 report_uninit((const SV *)sv);
5974 if (i > 0 && RXp_MATCH_UTF8(rx)) {
5975 const char * const s = rx->subbeg + s1;
5980 if (is_utf8_string_loclen((U8*)s, i, &ep, &el))
5987 Perl_reg_qr_package(pTHX_ REGEXP * const rx)
5989 PERL_ARGS_ASSERT_REG_QR_PACKAGE;
5990 PERL_UNUSED_ARG(rx);
5994 return newSVpvs("Regexp");
5997 /* Scans the name of a named buffer from the pattern.
5998 * If flags is REG_RSN_RETURN_NULL returns null.
5999 * If flags is REG_RSN_RETURN_NAME returns an SV* containing the name
6000 * If flags is REG_RSN_RETURN_DATA returns the data SV* corresponding
6001 * to the parsed name as looked up in the RExC_paren_names hash.
6002 * If there is an error throws a vFAIL().. type exception.
6005 #define REG_RSN_RETURN_NULL 0
6006 #define REG_RSN_RETURN_NAME 1
6007 #define REG_RSN_RETURN_DATA 2
6010 S_reg_scan_name(pTHX_ RExC_state_t *pRExC_state, U32 flags)
6012 char *name_start = RExC_parse;
6014 PERL_ARGS_ASSERT_REG_SCAN_NAME;
6016 if (isIDFIRST_lazy_if(RExC_parse, UTF)) {
6017 /* skip IDFIRST by using do...while */
6020 RExC_parse += UTF8SKIP(RExC_parse);
6021 } while (isALNUM_utf8((U8*)RExC_parse));
6025 } while (isALNUM(*RExC_parse));
6030 = newSVpvn_flags(name_start, (int)(RExC_parse - name_start),
6031 SVs_TEMP | (UTF ? SVf_UTF8 : 0));
6032 if ( flags == REG_RSN_RETURN_NAME)
6034 else if (flags==REG_RSN_RETURN_DATA) {
6037 if ( ! sv_name ) /* should not happen*/
6038 Perl_croak(aTHX_ "panic: no svname in reg_scan_name");
6039 if (RExC_paren_names)
6040 he_str = hv_fetch_ent( RExC_paren_names, sv_name, 0, 0 );
6042 sv_dat = HeVAL(he_str);
6044 vFAIL("Reference to nonexistent named group");
6048 Perl_croak(aTHX_ "panic: bad flag %lx in reg_scan_name",
6049 (unsigned long) flags);
6056 #define DEBUG_PARSE_MSG(funcname) DEBUG_PARSE_r({ \
6057 int rem=(int)(RExC_end - RExC_parse); \
6066 if (RExC_lastparse!=RExC_parse) \
6067 PerlIO_printf(Perl_debug_log," >%.*s%-*s", \
6070 iscut ? "..." : "<" \
6073 PerlIO_printf(Perl_debug_log,"%16s",""); \
6076 num = RExC_size + 1; \
6078 num=REG_NODE_NUM(RExC_emit); \
6079 if (RExC_lastnum!=num) \
6080 PerlIO_printf(Perl_debug_log,"|%4d",num); \
6082 PerlIO_printf(Perl_debug_log,"|%4s",""); \
6083 PerlIO_printf(Perl_debug_log,"|%*s%-4s", \
6084 (int)((depth*2)), "", \
6088 RExC_lastparse=RExC_parse; \
6093 #define DEBUG_PARSE(funcname) DEBUG_PARSE_r({ \
6094 DEBUG_PARSE_MSG((funcname)); \
6095 PerlIO_printf(Perl_debug_log,"%4s","\n"); \
6097 #define DEBUG_PARSE_FMT(funcname,fmt,args) DEBUG_PARSE_r({ \
6098 DEBUG_PARSE_MSG((funcname)); \
6099 PerlIO_printf(Perl_debug_log,fmt "\n",args); \
6102 /* This section of code defines the inversion list object and its methods. The
6103 * interfaces are highly subject to change, so as much as possible is static to
6104 * this file. An inversion list is here implemented as a malloc'd C UV array
6105 * with some added info that is placed as UVs at the beginning in a header
6106 * portion. An inversion list for Unicode is an array of code points, sorted
6107 * by ordinal number. The zeroth element is the first code point in the list.
6108 * The 1th element is the first element beyond that not in the list. In other
6109 * words, the first range is
6110 * invlist[0]..(invlist[1]-1)
6111 * The other ranges follow. Thus every element whose index is divisible by two
6112 * marks the beginning of a range that is in the list, and every element not
6113 * divisible by two marks the beginning of a range not in the list. A single
6114 * element inversion list that contains the single code point N generally
6115 * consists of two elements
6118 * (The exception is when N is the highest representable value on the
6119 * machine, in which case the list containing just it would be a single
6120 * element, itself. By extension, if the last range in the list extends to
6121 * infinity, then the first element of that range will be in the inversion list
6122 * at a position that is divisible by two, and is the final element in the
6124 * Taking the complement (inverting) an inversion list is quite simple, if the
6125 * first element is 0, remove it; otherwise add a 0 element at the beginning.
6126 * This implementation reserves an element at the beginning of each inversion list
6127 * to contain 0 when the list contains 0, and contains 1 otherwise. The actual
6128 * beginning of the list is either that element if 0, or the next one if 1.
6130 * More about inversion lists can be found in "Unicode Demystified"
6131 * Chapter 13 by Richard Gillam, published by Addison-Wesley.
6132 * More will be coming when functionality is added later.
6134 * The inversion list data structure is currently implemented as an SV pointing
6135 * to an array of UVs that the SV thinks are bytes. This allows us to have an
6136 * array of UV whose memory management is automatically handled by the existing
6137 * facilities for SV's.
6139 * Some of the methods should always be private to the implementation, and some
6140 * should eventually be made public */
6142 #define INVLIST_LEN_OFFSET 0 /* Number of elements in the inversion list */
6143 #define INVLIST_ITER_OFFSET 1 /* Current iteration position */
6145 #define INVLIST_ZERO_OFFSET 2 /* 0 or 1; must be last element in header */
6146 /* The UV at position ZERO contains either 0 or 1. If 0, the inversion list
6147 * contains the code point U+00000, and begins here. If 1, the inversion list
6148 * doesn't contain U+0000, and it begins at the next UV in the array.
6149 * Inverting an inversion list consists of adding or removing the 0 at the
6150 * beginning of it. By reserving a space for that 0, inversion can be made
6153 #define HEADER_LENGTH (INVLIST_ZERO_OFFSET + 1)
6155 /* Internally things are UVs */
6156 #define TO_INTERNAL_SIZE(x) ((x + HEADER_LENGTH) * sizeof(UV))
6157 #define FROM_INTERNAL_SIZE(x) ((x / sizeof(UV)) - HEADER_LENGTH)
6159 #define INVLIST_INITIAL_LEN 10
6161 PERL_STATIC_INLINE UV*
6162 S__invlist_array_init(pTHX_ SV* const invlist, const bool will_have_0)
6164 /* Returns a pointer to the first element in the inversion list's array.
6165 * This is called upon initialization of an inversion list. Where the
6166 * array begins depends on whether the list has the code point U+0000
6167 * in it or not. The other parameter tells it whether the code that
6168 * follows this call is about to put a 0 in the inversion list or not.
6169 * The first element is either the element with 0, if 0, or the next one,
6172 UV* zero = get_invlist_zero_addr(invlist);
6174 PERL_ARGS_ASSERT__INVLIST_ARRAY_INIT;
6177 assert(! *get_invlist_len_addr(invlist));
6179 /* 1^1 = 0; 1^0 = 1 */
6180 *zero = 1 ^ will_have_0;
6181 return zero + *zero;
6184 PERL_STATIC_INLINE UV*
6185 S_invlist_array(pTHX_ SV* const invlist)
6187 /* Returns the pointer to the inversion list's array. Every time the
6188 * length changes, this needs to be called in case malloc or realloc moved
6191 PERL_ARGS_ASSERT_INVLIST_ARRAY;
6193 /* Must not be empty. If these fail, you probably didn't check for <len>
6194 * being non-zero before trying to get the array */
6195 assert(*get_invlist_len_addr(invlist));
6196 assert(*get_invlist_zero_addr(invlist) == 0
6197 || *get_invlist_zero_addr(invlist) == 1);
6199 /* The array begins either at the element reserved for zero if the
6200 * list contains 0 (that element will be set to 0), or otherwise the next
6201 * element (in which case the reserved element will be set to 1). */
6202 return (UV *) (get_invlist_zero_addr(invlist)
6203 + *get_invlist_zero_addr(invlist));
6206 PERL_STATIC_INLINE UV*
6207 S_get_invlist_len_addr(pTHX_ SV* invlist)
6209 /* Return the address of the UV that contains the current number
6210 * of used elements in the inversion list */
6212 PERL_ARGS_ASSERT_GET_INVLIST_LEN_ADDR;
6214 return (UV *) (SvPVX(invlist) + (INVLIST_LEN_OFFSET * sizeof (UV)));
6217 PERL_STATIC_INLINE UV
6218 S_invlist_len(pTHX_ SV* const invlist)
6220 /* Returns the current number of elements stored in the inversion list's
6223 PERL_ARGS_ASSERT_INVLIST_LEN;
6225 return *get_invlist_len_addr(invlist);
6228 PERL_STATIC_INLINE void
6229 S_invlist_set_len(pTHX_ SV* const invlist, const UV len)
6231 /* Sets the current number of elements stored in the inversion list */
6233 PERL_ARGS_ASSERT_INVLIST_SET_LEN;
6235 *get_invlist_len_addr(invlist) = len;
6237 assert(len <= SvLEN(invlist));
6239 SvCUR_set(invlist, TO_INTERNAL_SIZE(len));
6240 /* If the list contains U+0000, that element is part of the header,
6241 * and should not be counted as part of the array. It will contain
6242 * 0 in that case, and 1 otherwise. So we could flop 0=>1, 1=>0 and
6244 * SvCUR_set(invlist,
6245 * TO_INTERNAL_SIZE(len
6246 * - (*get_invlist_zero_addr(inv_list) ^ 1)));
6247 * But, this is only valid if len is not 0. The consequences of not doing
6248 * this is that the memory allocation code may think that 1 more UV is
6249 * being used than actually is, and so might do an unnecessary grow. That
6250 * seems worth not bothering to make this the precise amount.
6252 * Note that when inverting, SvCUR shouldn't change */
6255 PERL_STATIC_INLINE UV
6256 S_invlist_max(pTHX_ SV* const invlist)
6258 /* Returns the maximum number of elements storable in the inversion list's
6259 * array, without having to realloc() */
6261 PERL_ARGS_ASSERT_INVLIST_MAX;
6263 return FROM_INTERNAL_SIZE(SvLEN(invlist));
6266 PERL_STATIC_INLINE UV*
6267 S_get_invlist_zero_addr(pTHX_ SV* invlist)
6269 /* Return the address of the UV that is reserved to hold 0 if the inversion
6270 * list contains 0. This has to be the last element of the heading, as the
6271 * list proper starts with either it if 0, or the next element if not.
6272 * (But we force it to contain either 0 or 1) */
6274 PERL_ARGS_ASSERT_GET_INVLIST_ZERO_ADDR;
6276 return (UV *) (SvPVX(invlist) + (INVLIST_ZERO_OFFSET * sizeof (UV)));
6279 #ifndef PERL_IN_XSUB_RE
6281 Perl__new_invlist(pTHX_ IV initial_size)
6284 /* Return a pointer to a newly constructed inversion list, with enough
6285 * space to store 'initial_size' elements. If that number is negative, a
6286 * system default is used instead */
6290 if (initial_size < 0) {
6291 initial_size = INVLIST_INITIAL_LEN;
6294 /* Allocate the initial space */
6295 new_list = newSV(TO_INTERNAL_SIZE(initial_size));
6296 invlist_set_len(new_list, 0);
6298 /* Force iterinit() to be used to get iteration to work */
6299 *get_invlist_iter_addr(new_list) = UV_MAX;
6301 /* This should force a segfault if a method doesn't initialize this
6303 *get_invlist_zero_addr(new_list) = UV_MAX;
6310 S_invlist_extend(pTHX_ SV* const invlist, const UV new_max)
6312 /* Grow the maximum size of an inversion list */
6314 PERL_ARGS_ASSERT_INVLIST_EXTEND;
6316 SvGROW((SV *)invlist, TO_INTERNAL_SIZE(new_max));
6319 PERL_STATIC_INLINE void
6320 S_invlist_trim(pTHX_ SV* const invlist)
6322 PERL_ARGS_ASSERT_INVLIST_TRIM;
6324 /* Change the length of the inversion list to how many entries it currently
6327 SvPV_shrink_to_cur((SV *) invlist);
6330 /* An element is in an inversion list iff its index is even numbered: 0, 2, 4,
6332 #define ELEMENT_RANGE_MATCHES_INVLIST(i) (! ((i) & 1))
6333 #define PREV_RANGE_MATCHES_INVLIST(i) (! ELEMENT_RANGE_MATCHES_INVLIST(i))
6335 #ifndef PERL_IN_XSUB_RE
6337 Perl__append_range_to_invlist(pTHX_ SV* const invlist, const UV start, const UV end)
6339 /* Subject to change or removal. Append the range from 'start' to 'end' at
6340 * the end of the inversion list. The range must be above any existing
6344 UV max = invlist_max(invlist);
6345 UV len = invlist_len(invlist);
6347 PERL_ARGS_ASSERT__APPEND_RANGE_TO_INVLIST;
6349 if (len == 0) { /* Empty lists must be initialized */
6350 array = _invlist_array_init(invlist, start == 0);
6353 /* Here, the existing list is non-empty. The current max entry in the
6354 * list is generally the first value not in the set, except when the
6355 * set extends to the end of permissible values, in which case it is
6356 * the first entry in that final set, and so this call is an attempt to
6357 * append out-of-order */
6359 UV final_element = len - 1;
6360 array = invlist_array(invlist);
6361 if (array[final_element] > start
6362 || ELEMENT_RANGE_MATCHES_INVLIST(final_element))
6364 Perl_croak(aTHX_ "panic: attempting to append to an inversion list, but wasn't at the end of the list, final=%"UVuf", start=%"UVuf", match=%c",
6365 array[final_element], start,
6366 ELEMENT_RANGE_MATCHES_INVLIST(final_element) ? 't' : 'f');
6369 /* Here, it is a legal append. If the new range begins with the first
6370 * value not in the set, it is extending the set, so the new first
6371 * value not in the set is one greater than the newly extended range.
6373 if (array[final_element] == start) {
6374 if (end != UV_MAX) {
6375 array[final_element] = end + 1;
6378 /* But if the end is the maximum representable on the machine,
6379 * just let the range that this would extend to have no end */
6380 invlist_set_len(invlist, len - 1);
6386 /* Here the new range doesn't extend any existing set. Add it */
6388 len += 2; /* Includes an element each for the start and end of range */
6390 /* If overflows the existing space, extend, which may cause the array to be
6393 invlist_extend(invlist, len);
6394 invlist_set_len(invlist, len); /* Have to set len here to avoid assert
6395 failure in invlist_array() */
6396 array = invlist_array(invlist);
6399 invlist_set_len(invlist, len);
6402 /* The next item on the list starts the range, the one after that is
6403 * one past the new range. */
6404 array[len - 2] = start;
6405 if (end != UV_MAX) {
6406 array[len - 1] = end + 1;
6409 /* But if the end is the maximum representable on the machine, just let
6410 * the range have no end */
6411 invlist_set_len(invlist, len - 1);
6416 S_invlist_search(pTHX_ SV* const invlist, const UV cp)
6418 /* Searches the inversion list for the entry that contains the input code
6419 * point <cp>. If <cp> is not in the list, -1 is returned. Otherwise, the
6420 * return value is the index into the list's array of the range that
6424 IV high = invlist_len(invlist);
6425 const UV * const array = invlist_array(invlist);
6427 PERL_ARGS_ASSERT_INVLIST_SEARCH;
6429 /* If list is empty or the code point is before the first element, return
6431 if (high == 0 || cp < array[0]) {
6435 /* Binary search. What we are looking for is <i> such that
6436 * array[i] <= cp < array[i+1]
6437 * The loop below converges on the i+1. */
6438 while (low < high) {
6439 IV mid = (low + high) / 2;
6440 if (array[mid] <= cp) {
6443 /* We could do this extra test to exit the loop early.
6444 if (cp < array[low]) {
6449 else { /* cp < array[mid] */
6458 Perl__invlist_populate_swatch(pTHX_ SV* const invlist, const UV start, const UV end, U8* swatch)
6460 /* populates a swatch of a swash the same way swatch_get() does in utf8.c,
6461 * but is used when the swash has an inversion list. This makes this much
6462 * faster, as it uses a binary search instead of a linear one. This is
6463 * intimately tied to that function, and perhaps should be in utf8.c,
6464 * except it is intimately tied to inversion lists as well. It assumes
6465 * that <swatch> is all 0's on input */
6468 const IV len = invlist_len(invlist);
6472 PERL_ARGS_ASSERT__INVLIST_POPULATE_SWATCH;
6474 if (len == 0) { /* Empty inversion list */
6478 array = invlist_array(invlist);
6480 /* Find which element it is */
6481 i = invlist_search(invlist, start);
6483 /* We populate from <start> to <end> */
6484 while (current < end) {
6487 /* The inversion list gives the results for every possible code point
6488 * after the first one in the list. Only those ranges whose index is
6489 * even are ones that the inversion list matches. For the odd ones,
6490 * and if the initial code point is not in the list, we have to skip
6491 * forward to the next element */
6492 if (i == -1 || ! ELEMENT_RANGE_MATCHES_INVLIST(i)) {
6494 if (i >= len) { /* Finished if beyond the end of the array */
6498 if (current >= end) { /* Finished if beyond the end of what we
6503 assert(current >= start);
6505 /* The current range ends one below the next one, except don't go past
6508 upper = (i < len && array[i] < end) ? array[i] : end;
6510 /* Here we are in a range that matches. Populate a bit in the 3-bit U8
6511 * for each code point in it */
6512 for (; current < upper; current++) {
6513 const STRLEN offset = (STRLEN)(current - start);
6514 swatch[offset >> 3] |= 1 << (offset & 7);
6517 /* Quit if at the end of the list */
6520 /* But first, have to deal with the highest possible code point on
6521 * the platform. The previous code assumes that <end> is one
6522 * beyond where we want to populate, but that is impossible at the
6523 * platform's infinity, so have to handle it specially */
6524 if (UNLIKELY(end == UV_MAX && ELEMENT_RANGE_MATCHES_INVLIST(len-1)))
6526 const STRLEN offset = (STRLEN)(end - start);
6527 swatch[offset >> 3] |= 1 << (offset & 7);
6532 /* Advance to the next range, which will be for code points not in the
6541 Perl__invlist_union(pTHX_ SV* const a, SV* const b, SV** output)
6543 /* Take the union of two inversion lists and point <output> to it. *output
6544 * should be defined upon input, and if it points to one of the two lists,
6545 * the reference count to that list will be decremented.
6546 * The basis for this comes from "Unicode Demystified" Chapter 13 by
6547 * Richard Gillam, published by Addison-Wesley, and explained at some
6548 * length there. The preface says to incorporate its examples into your
6549 * code at your own risk.
6551 * The algorithm is like a merge sort.
6553 * XXX A potential performance improvement is to keep track as we go along
6554 * if only one of the inputs contributes to the result, meaning the other
6555 * is a subset of that one. In that case, we can skip the final copy and
6556 * return the larger of the input lists, but then outside code might need
6557 * to keep track of whether to free the input list or not */
6559 UV* array_a; /* a's array */
6561 UV len_a; /* length of a's array */
6564 SV* u; /* the resulting union */
6568 UV i_a = 0; /* current index into a's array */
6572 /* running count, as explained in the algorithm source book; items are
6573 * stopped accumulating and are output when the count changes to/from 0.
6574 * The count is incremented when we start a range that's in the set, and
6575 * decremented when we start a range that's not in the set. So its range
6576 * is 0 to 2. Only when the count is zero is something not in the set.
6580 PERL_ARGS_ASSERT__INVLIST_UNION;
6583 /* If either one is empty, the union is the other one */
6584 len_a = invlist_len(a);
6590 *output = invlist_clone(b);
6591 } /* else *output already = b; */
6594 else if ((len_b = invlist_len(b)) == 0) {
6599 *output = invlist_clone(a);
6601 /* else *output already = a; */
6605 /* Here both lists exist and are non-empty */
6606 array_a = invlist_array(a);
6607 array_b = invlist_array(b);
6609 /* Size the union for the worst case: that the sets are completely
6611 u = _new_invlist(len_a + len_b);
6613 /* Will contain U+0000 if either component does */
6614 array_u = _invlist_array_init(u, (len_a > 0 && array_a[0] == 0)
6615 || (len_b > 0 && array_b[0] == 0));
6617 /* Go through each list item by item, stopping when exhausted one of
6619 while (i_a < len_a && i_b < len_b) {
6620 UV cp; /* The element to potentially add to the union's array */
6621 bool cp_in_set; /* is it in the the input list's set or not */
6623 /* We need to take one or the other of the two inputs for the union.
6624 * Since we are merging two sorted lists, we take the smaller of the
6625 * next items. In case of a tie, we take the one that is in its set
6626 * first. If we took one not in the set first, it would decrement the
6627 * count, possibly to 0 which would cause it to be output as ending the
6628 * range, and the next time through we would take the same number, and
6629 * output it again as beginning the next range. By doing it the
6630 * opposite way, there is no possibility that the count will be
6631 * momentarily decremented to 0, and thus the two adjoining ranges will
6632 * be seamlessly merged. (In a tie and both are in the set or both not
6633 * in the set, it doesn't matter which we take first.) */
6634 if (array_a[i_a] < array_b[i_b]
6635 || (array_a[i_a] == array_b[i_b]
6636 && ELEMENT_RANGE_MATCHES_INVLIST(i_a)))
6638 cp_in_set = ELEMENT_RANGE_MATCHES_INVLIST(i_a);
6642 cp_in_set = ELEMENT_RANGE_MATCHES_INVLIST(i_b);
6646 /* Here, have chosen which of the two inputs to look at. Only output
6647 * if the running count changes to/from 0, which marks the
6648 * beginning/end of a range in that's in the set */
6651 array_u[i_u++] = cp;
6658 array_u[i_u++] = cp;
6663 /* Here, we are finished going through at least one of the lists, which
6664 * means there is something remaining in at most one. We check if the list
6665 * that hasn't been exhausted is positioned such that we are in the middle
6666 * of a range in its set or not. (i_a and i_b point to the element beyond
6667 * the one we care about.) If in the set, we decrement 'count'; if 0, there
6668 * is potentially more to output.
6669 * There are four cases:
6670 * 1) Both weren't in their sets, count is 0, and remains 0. What's left
6671 * in the union is entirely from the non-exhausted set.
6672 * 2) Both were in their sets, count is 2. Nothing further should
6673 * be output, as everything that remains will be in the exhausted
6674 * list's set, hence in the union; decrementing to 1 but not 0 insures
6676 * 3) the exhausted was in its set, non-exhausted isn't, count is 1.
6677 * Nothing further should be output because the union includes
6678 * everything from the exhausted set. Not decrementing ensures that.
6679 * 4) the exhausted wasn't in its set, non-exhausted is, count is 1;
6680 * decrementing to 0 insures that we look at the remainder of the
6681 * non-exhausted set */
6682 if ((i_a != len_a && PREV_RANGE_MATCHES_INVLIST(i_a))
6683 || (i_b != len_b && PREV_RANGE_MATCHES_INVLIST(i_b)))
6688 /* The final length is what we've output so far, plus what else is about to
6689 * be output. (If 'count' is non-zero, then the input list we exhausted
6690 * has everything remaining up to the machine's limit in its set, and hence
6691 * in the union, so there will be no further output. */
6694 /* At most one of the subexpressions will be non-zero */
6695 len_u += (len_a - i_a) + (len_b - i_b);
6698 /* Set result to final length, which can change the pointer to array_u, so
6700 if (len_u != invlist_len(u)) {
6701 invlist_set_len(u, len_u);
6703 array_u = invlist_array(u);
6706 /* When 'count' is 0, the list that was exhausted (if one was shorter than
6707 * the other) ended with everything above it not in its set. That means
6708 * that the remaining part of the union is precisely the same as the
6709 * non-exhausted list, so can just copy it unchanged. (If both list were
6710 * exhausted at the same time, then the operations below will be both 0.)
6713 IV copy_count; /* At most one will have a non-zero copy count */
6714 if ((copy_count = len_a - i_a) > 0) {
6715 Copy(array_a + i_a, array_u + i_u, copy_count, UV);
6717 else if ((copy_count = len_b - i_b) > 0) {
6718 Copy(array_b + i_b, array_u + i_u, copy_count, UV);
6722 /* We may be removing a reference to one of the inputs */
6723 if (a == *output || b == *output) {
6724 SvREFCNT_dec(*output);
6732 Perl__invlist_intersection(pTHX_ SV* const a, SV* const b, SV** i)
6734 /* Take the intersection of two inversion lists and point <i> to it. *i
6735 * should be defined upon input, and if it points to one of the two lists,
6736 * the reference count to that list will be decremented.
6737 * The basis for this comes from "Unicode Demystified" Chapter 13 by
6738 * Richard Gillam, published by Addison-Wesley, and explained at some
6739 * length there. The preface says to incorporate its examples into your
6740 * code at your own risk. In fact, it had bugs
6742 * The algorithm is like a merge sort, and is essentially the same as the
6746 UV* array_a; /* a's array */
6748 UV len_a; /* length of a's array */
6751 SV* r; /* the resulting intersection */
6755 UV i_a = 0; /* current index into a's array */
6759 /* running count, as explained in the algorithm source book; items are
6760 * stopped accumulating and are output when the count changes to/from 2.
6761 * The count is incremented when we start a range that's in the set, and
6762 * decremented when we start a range that's not in the set. So its range
6763 * is 0 to 2. Only when the count is 2 is something in the intersection.
6767 PERL_ARGS_ASSERT__INVLIST_INTERSECTION;
6770 /* If either one is empty, the intersection is null */
6771 len_a = invlist_len(a);
6772 if ((len_a == 0) || ((len_b = invlist_len(b)) == 0)) {
6774 /* If the result is the same as one of the inputs, the input is being
6783 *i = _new_invlist(0);
6787 /* Here both lists exist and are non-empty */
6788 array_a = invlist_array(a);
6789 array_b = invlist_array(b);
6791 /* Size the intersection for the worst case: that the intersection ends up
6792 * fragmenting everything to be completely disjoint */
6793 r= _new_invlist(len_a + len_b);
6795 /* Will contain U+0000 iff both components do */
6796 array_r = _invlist_array_init(r, len_a > 0 && array_a[0] == 0
6797 && len_b > 0 && array_b[0] == 0);
6799 /* Go through each list item by item, stopping when exhausted one of
6801 while (i_a < len_a && i_b < len_b) {
6802 UV cp; /* The element to potentially add to the intersection's
6804 bool cp_in_set; /* Is it in the input list's set or not */
6806 /* We need to take one or the other of the two inputs for the
6807 * intersection. Since we are merging two sorted lists, we take the
6808 * smaller of the next items. In case of a tie, we take the one that
6809 * is not in its set first (a difference from the union algorithm). If
6810 * we took one in the set first, it would increment the count, possibly
6811 * to 2 which would cause it to be output as starting a range in the
6812 * intersection, and the next time through we would take that same
6813 * number, and output it again as ending the set. By doing it the
6814 * opposite of this, there is no possibility that the count will be
6815 * momentarily incremented to 2. (In a tie and both are in the set or
6816 * both not in the set, it doesn't matter which we take first.) */
6817 if (array_a[i_a] < array_b[i_b]
6818 || (array_a[i_a] == array_b[i_b]
6819 && ! ELEMENT_RANGE_MATCHES_INVLIST(i_a)))
6821 cp_in_set = ELEMENT_RANGE_MATCHES_INVLIST(i_a);
6825 cp_in_set = ELEMENT_RANGE_MATCHES_INVLIST(i_b);
6829 /* Here, have chosen which of the two inputs to look at. Only output
6830 * if the running count changes to/from 2, which marks the
6831 * beginning/end of a range that's in the intersection */
6835 array_r[i_r++] = cp;
6840 array_r[i_r++] = cp;
6846 /* Here, we are finished going through at least one of the lists, which
6847 * means there is something remaining in at most one. We check if the list
6848 * that has been exhausted is positioned such that we are in the middle
6849 * of a range in its set or not. (i_a and i_b point to elements 1 beyond
6850 * the ones we care about.) There are four cases:
6851 * 1) Both weren't in their sets, count is 0, and remains 0. There's
6852 * nothing left in the intersection.
6853 * 2) Both were in their sets, count is 2 and perhaps is incremented to
6854 * above 2. What should be output is exactly that which is in the
6855 * non-exhausted set, as everything it has is also in the intersection
6856 * set, and everything it doesn't have can't be in the intersection
6857 * 3) The exhausted was in its set, non-exhausted isn't, count is 1, and
6858 * gets incremented to 2. Like the previous case, the intersection is
6859 * everything that remains in the non-exhausted set.
6860 * 4) the exhausted wasn't in its set, non-exhausted is, count is 1, and
6861 * remains 1. And the intersection has nothing more. */
6862 if ((i_a == len_a && PREV_RANGE_MATCHES_INVLIST(i_a))
6863 || (i_b == len_b && PREV_RANGE_MATCHES_INVLIST(i_b)))
6868 /* The final length is what we've output so far plus what else is in the
6869 * intersection. At most one of the subexpressions below will be non-zero */
6872 len_r += (len_a - i_a) + (len_b - i_b);
6875 /* Set result to final length, which can change the pointer to array_r, so
6877 if (len_r != invlist_len(r)) {
6878 invlist_set_len(r, len_r);
6880 array_r = invlist_array(r);
6883 /* Finish outputting any remaining */
6884 if (count >= 2) { /* At most one will have a non-zero copy count */
6886 if ((copy_count = len_a - i_a) > 0) {
6887 Copy(array_a + i_a, array_r + i_r, copy_count, UV);
6889 else if ((copy_count = len_b - i_b) > 0) {
6890 Copy(array_b + i_b, array_r + i_r, copy_count, UV);
6894 /* We may be removing a reference to one of the inputs */
6895 if (a == *i || b == *i) {
6906 S_add_range_to_invlist(pTHX_ SV* invlist, const UV start, const UV end)
6908 /* Add the range from 'start' to 'end' inclusive to the inversion list's
6909 * set. A pointer to the inversion list is returned. This may actually be
6910 * a new list, in which case the passed in one has been destroyed. The
6911 * passed in inversion list can be NULL, in which case a new one is created
6912 * with just the one range in it */
6917 if (invlist == NULL) {
6918 invlist = _new_invlist(2);
6922 len = invlist_len(invlist);
6925 /* If comes after the final entry, can just append it to the end */
6927 || start >= invlist_array(invlist)
6928 [invlist_len(invlist) - 1])
6930 _append_range_to_invlist(invlist, start, end);
6934 /* Here, can't just append things, create and return a new inversion list
6935 * which is the union of this range and the existing inversion list */
6936 range_invlist = _new_invlist(2);
6937 _append_range_to_invlist(range_invlist, start, end);
6939 _invlist_union(invlist, range_invlist, &invlist);
6941 /* The temporary can be freed */
6942 SvREFCNT_dec(range_invlist);
6947 PERL_STATIC_INLINE SV*
6948 S_add_cp_to_invlist(pTHX_ SV* invlist, const UV cp) {
6949 return add_range_to_invlist(invlist, cp, cp);
6952 #ifndef PERL_IN_XSUB_RE
6954 Perl__invlist_invert(pTHX_ SV* const invlist)
6956 /* Complement the input inversion list. This adds a 0 if the list didn't
6957 * have a zero; removes it otherwise. As described above, the data
6958 * structure is set up so that this is very efficient */
6960 UV* len_pos = get_invlist_len_addr(invlist);
6962 PERL_ARGS_ASSERT__INVLIST_INVERT;
6964 /* The inverse of matching nothing is matching everything */
6965 if (*len_pos == 0) {
6966 _append_range_to_invlist(invlist, 0, UV_MAX);
6970 /* The exclusive or complents 0 to 1; and 1 to 0. If the result is 1, the
6971 * zero element was a 0, so it is being removed, so the length decrements
6972 * by 1; and vice-versa. SvCUR is unaffected */
6973 if (*get_invlist_zero_addr(invlist) ^= 1) {
6982 Perl__invlist_invert_prop(pTHX_ SV* const invlist)
6984 /* Complement the input inversion list (which must be a Unicode property,
6985 * all of which don't match above the Unicode maximum code point.) And
6986 * Perl has chosen to not have the inversion match above that either. This
6987 * adds a 0x110000 if the list didn't end with it, and removes it if it did
6993 PERL_ARGS_ASSERT__INVLIST_INVERT_PROP;
6995 _invlist_invert(invlist);
6997 len = invlist_len(invlist);
6999 if (len != 0) { /* If empty do nothing */
7000 array = invlist_array(invlist);
7001 if (array[len - 1] != PERL_UNICODE_MAX + 1) {
7002 /* Add 0x110000. First, grow if necessary */
7004 if (invlist_max(invlist) < len) {
7005 invlist_extend(invlist, len);
7006 array = invlist_array(invlist);
7008 invlist_set_len(invlist, len);
7009 array[len - 1] = PERL_UNICODE_MAX + 1;
7011 else { /* Remove the 0x110000 */
7012 invlist_set_len(invlist, len - 1);
7020 PERL_STATIC_INLINE SV*
7021 S_invlist_clone(pTHX_ SV* const invlist)
7024 /* Return a new inversion list that is a copy of the input one, which is
7027 /* Need to allocate extra space to accommodate Perl's addition of a
7028 * trailing NUL to SvPV's, since it thinks they are always strings */
7029 SV* new_invlist = _new_invlist(invlist_len(invlist) + 1);
7030 STRLEN length = SvCUR(invlist);
7032 PERL_ARGS_ASSERT_INVLIST_CLONE;
7034 SvCUR_set(new_invlist, length); /* This isn't done automatically */
7035 Copy(SvPVX(invlist), SvPVX(new_invlist), length, char);
7040 #ifndef PERL_IN_XSUB_RE
7042 Perl__invlist_subtract(pTHX_ SV* const a, SV* const b, SV** result)
7044 /* Point <result> to an inversion list which consists of all elements in
7045 * <a> that aren't also in <b>. *result should be defined upon input, and
7046 * if it points to C<b> its reference count will be decremented. */
7048 PERL_ARGS_ASSERT__INVLIST_SUBTRACT;
7051 /* Subtracting nothing retains the original */
7052 if (invlist_len(b) == 0) {
7058 /* If the result is not to be the same variable as the original, create
7061 *result = invlist_clone(a);
7064 SV *b_copy = invlist_clone(b);
7065 _invlist_invert(b_copy); /* Everything not in 'b' */
7071 _invlist_intersection(a, b_copy, result); /* Everything in 'a' not in
7073 SvREFCNT_dec(b_copy);
7080 PERL_STATIC_INLINE UV*
7081 S_get_invlist_iter_addr(pTHX_ SV* invlist)
7083 /* Return the address of the UV that contains the current iteration
7086 PERL_ARGS_ASSERT_GET_INVLIST_ITER_ADDR;
7088 return (UV *) (SvPVX(invlist) + (INVLIST_ITER_OFFSET * sizeof (UV)));
7091 PERL_STATIC_INLINE void
7092 S_invlist_iterinit(pTHX_ SV* invlist) /* Initialize iterator for invlist */
7094 PERL_ARGS_ASSERT_INVLIST_ITERINIT;
7096 *get_invlist_iter_addr(invlist) = 0;
7100 S_invlist_iternext(pTHX_ SV* invlist, UV* start, UV* end)
7102 /* An C<invlist_iterinit> call on <invlist> must be used to set this up.
7103 * This call sets in <*start> and <*end>, the next range in <invlist>.
7104 * Returns <TRUE> if successful and the next call will return the next
7105 * range; <FALSE> if was already at the end of the list. If the latter,
7106 * <*start> and <*end> are unchanged, and the next call to this function
7107 * will start over at the beginning of the list */
7109 UV* pos = get_invlist_iter_addr(invlist);
7110 UV len = invlist_len(invlist);
7113 PERL_ARGS_ASSERT_INVLIST_ITERNEXT;
7116 *pos = UV_MAX; /* Force iternit() to be required next time */
7120 array = invlist_array(invlist);
7122 *start = array[(*pos)++];
7128 *end = array[(*pos)++] - 1;
7134 #ifndef PERL_IN_XSUB_RE
7136 Perl__invlist_contents(pTHX_ SV* const invlist)
7138 /* Get the contents of an inversion list into a string SV so that they can
7139 * be printed out. It uses the format traditionally done for debug tracing
7143 SV* output = newSVpvs("\n");
7145 PERL_ARGS_ASSERT__INVLIST_CONTENTS;
7147 invlist_iterinit(invlist);
7148 while (invlist_iternext(invlist, &start, &end)) {
7149 if (end == UV_MAX) {
7150 Perl_sv_catpvf(aTHX_ output, "%04"UVXf"\tINFINITY\n", start);
7152 else if (end != start) {
7153 Perl_sv_catpvf(aTHX_ output, "%04"UVXf"\t%04"UVXf"\n",
7157 Perl_sv_catpvf(aTHX_ output, "%04"UVXf"\n", start);
7167 S_invlist_dump(pTHX_ SV* const invlist, const char * const header)
7169 /* Dumps out the ranges in an inversion list. The string 'header'
7170 * if present is output on a line before the first range */
7174 if (header && strlen(header)) {
7175 PerlIO_printf(Perl_debug_log, "%s\n", header);
7177 invlist_iterinit(invlist);
7178 while (invlist_iternext(invlist, &start, &end)) {
7179 if (end == UV_MAX) {
7180 PerlIO_printf(Perl_debug_log, "0x%04"UVXf" .. INFINITY\n", start);
7183 PerlIO_printf(Perl_debug_log, "0x%04"UVXf" .. 0x%04"UVXf"\n", start, end);
7189 #undef HEADER_LENGTH
7190 #undef INVLIST_INITIAL_LENGTH
7191 #undef TO_INTERNAL_SIZE
7192 #undef FROM_INTERNAL_SIZE
7193 #undef INVLIST_LEN_OFFSET
7194 #undef INVLIST_ZERO_OFFSET
7195 #undef INVLIST_ITER_OFFSET
7197 /* End of inversion list object */
7200 - reg - regular expression, i.e. main body or parenthesized thing
7202 * Caller must absorb opening parenthesis.
7204 * Combining parenthesis handling with the base level of regular expression
7205 * is a trifle forced, but the need to tie the tails of the branches to what
7206 * follows makes it hard to avoid.
7208 #define REGTAIL(x,y,z) regtail((x),(y),(z),depth+1)
7210 #define REGTAIL_STUDY(x,y,z) regtail_study((x),(y),(z),depth+1)
7212 #define REGTAIL_STUDY(x,y,z) regtail((x),(y),(z),depth+1)
7216 S_reg(pTHX_ RExC_state_t *pRExC_state, I32 paren, I32 *flagp,U32 depth)
7217 /* paren: Parenthesized? 0=top, 1=(, inside: changed to letter. */
7220 register regnode *ret; /* Will be the head of the group. */
7221 register regnode *br;
7222 register regnode *lastbr;
7223 register regnode *ender = NULL;
7224 register I32 parno = 0;
7226 U32 oregflags = RExC_flags;
7227 bool have_branch = 0;
7229 I32 freeze_paren = 0;
7230 I32 after_freeze = 0;
7232 /* for (?g), (?gc), and (?o) warnings; warning
7233 about (?c) will warn about (?g) -- japhy */
7235 #define WASTED_O 0x01
7236 #define WASTED_G 0x02
7237 #define WASTED_C 0x04
7238 #define WASTED_GC (0x02|0x04)
7239 I32 wastedflags = 0x00;
7241 char * parse_start = RExC_parse; /* MJD */
7242 char * const oregcomp_parse = RExC_parse;
7244 GET_RE_DEBUG_FLAGS_DECL;
7246 PERL_ARGS_ASSERT_REG;
7247 DEBUG_PARSE("reg ");
7249 *flagp = 0; /* Tentatively. */
7252 /* Make an OPEN node, if parenthesized. */
7254 if ( *RExC_parse == '*') { /* (*VERB:ARG) */
7255 char *start_verb = RExC_parse;
7256 STRLEN verb_len = 0;
7257 char *start_arg = NULL;
7258 unsigned char op = 0;
7260 int internal_argval = 0; /* internal_argval is only useful if !argok */
7261 while ( *RExC_parse && *RExC_parse != ')' ) {
7262 if ( *RExC_parse == ':' ) {
7263 start_arg = RExC_parse + 1;
7269 verb_len = RExC_parse - start_verb;
7272 while ( *RExC_parse && *RExC_parse != ')' )
7274 if ( *RExC_parse != ')' )
7275 vFAIL("Unterminated verb pattern argument");
7276 if ( RExC_parse == start_arg )
7279 if ( *RExC_parse != ')' )
7280 vFAIL("Unterminated verb pattern");
7283 switch ( *start_verb ) {
7284 case 'A': /* (*ACCEPT) */
7285 if ( memEQs(start_verb,verb_len,"ACCEPT") ) {
7287 internal_argval = RExC_nestroot;
7290 case 'C': /* (*COMMIT) */
7291 if ( memEQs(start_verb,verb_len,"COMMIT") )
7294 case 'F': /* (*FAIL) */
7295 if ( verb_len==1 || memEQs(start_verb,verb_len,"FAIL") ) {
7300 case ':': /* (*:NAME) */
7301 case 'M': /* (*MARK:NAME) */
7302 if ( verb_len==0 || memEQs(start_verb,verb_len,"MARK") ) {
7307 case 'P': /* (*PRUNE) */
7308 if ( memEQs(start_verb,verb_len,"PRUNE") )
7311 case 'S': /* (*SKIP) */
7312 if ( memEQs(start_verb,verb_len,"SKIP") )
7315 case 'T': /* (*THEN) */
7316 /* [19:06] <TimToady> :: is then */
7317 if ( memEQs(start_verb,verb_len,"THEN") ) {
7319 RExC_seen |= REG_SEEN_CUTGROUP;
7325 vFAIL3("Unknown verb pattern '%.*s'",
7326 verb_len, start_verb);
7329 if ( start_arg && internal_argval ) {
7330 vFAIL3("Verb pattern '%.*s' may not have an argument",
7331 verb_len, start_verb);
7332 } else if ( argok < 0 && !start_arg ) {
7333 vFAIL3("Verb pattern '%.*s' has a mandatory argument",
7334 verb_len, start_verb);
7336 ret = reganode(pRExC_state, op, internal_argval);
7337 if ( ! internal_argval && ! SIZE_ONLY ) {
7339 SV *sv = newSVpvn( start_arg, RExC_parse - start_arg);
7340 ARG(ret) = add_data( pRExC_state, 1, "S" );
7341 RExC_rxi->data->data[ARG(ret)]=(void*)sv;
7348 if (!internal_argval)
7349 RExC_seen |= REG_SEEN_VERBARG;
7350 } else if ( start_arg ) {
7351 vFAIL3("Verb pattern '%.*s' may not have an argument",
7352 verb_len, start_verb);
7354 ret = reg_node(pRExC_state, op);
7356 nextchar(pRExC_state);
7359 if (*RExC_parse == '?') { /* (?...) */
7360 bool is_logical = 0;
7361 const char * const seqstart = RExC_parse;
7362 bool has_use_defaults = FALSE;
7365 paren = *RExC_parse++;
7366 ret = NULL; /* For look-ahead/behind. */
7369 case 'P': /* (?P...) variants for those used to PCRE/Python */
7370 paren = *RExC_parse++;
7371 if ( paren == '<') /* (?P<...>) named capture */
7373 else if (paren == '>') { /* (?P>name) named recursion */
7374 goto named_recursion;
7376 else if (paren == '=') { /* (?P=...) named backref */
7377 /* this pretty much dupes the code for \k<NAME> in regatom(), if
7378 you change this make sure you change that */
7379 char* name_start = RExC_parse;
7381 SV *sv_dat = reg_scan_name(pRExC_state,
7382 SIZE_ONLY ? REG_RSN_RETURN_NULL : REG_RSN_RETURN_DATA);
7383 if (RExC_parse == name_start || *RExC_parse != ')')
7384 vFAIL2("Sequence %.3s... not terminated",parse_start);
7387 num = add_data( pRExC_state, 1, "S" );
7388 RExC_rxi->data->data[num]=(void*)sv_dat;
7389 SvREFCNT_inc_simple_void(sv_dat);
7392 ret = reganode(pRExC_state,
7395 : (MORE_ASCII_RESTRICTED)
7397 : (AT_LEAST_UNI_SEMANTICS)
7405 Set_Node_Offset(ret, parse_start+1);
7406 Set_Node_Cur_Length(ret); /* MJD */
7408 nextchar(pRExC_state);
7412 vFAIL3("Sequence (%.*s...) not recognized", RExC_parse-seqstart, seqstart);
7414 case '<': /* (?<...) */
7415 if (*RExC_parse == '!')
7417 else if (*RExC_parse != '=')
7423 case '\'': /* (?'...') */
7424 name_start= RExC_parse;
7425 svname = reg_scan_name(pRExC_state,
7426 SIZE_ONLY ? /* reverse test from the others */
7427 REG_RSN_RETURN_NAME :
7428 REG_RSN_RETURN_NULL);
7429 if (RExC_parse == name_start) {
7431 vFAIL3("Sequence (%.*s...) not recognized", RExC_parse-seqstart, seqstart);
7434 if (*RExC_parse != paren)
7435 vFAIL2("Sequence (?%c... not terminated",
7436 paren=='>' ? '<' : paren);
7440 if (!svname) /* shouldn't happen */
7442 "panic: reg_scan_name returned NULL");
7443 if (!RExC_paren_names) {
7444 RExC_paren_names= newHV();
7445 sv_2mortal(MUTABLE_SV(RExC_paren_names));
7447 RExC_paren_name_list= newAV();
7448 sv_2mortal(MUTABLE_SV(RExC_paren_name_list));
7451 he_str = hv_fetch_ent( RExC_paren_names, svname, 1, 0 );
7453 sv_dat = HeVAL(he_str);
7455 /* croak baby croak */
7457 "panic: paren_name hash element allocation failed");
7458 } else if ( SvPOK(sv_dat) ) {
7459 /* (?|...) can mean we have dupes so scan to check
7460 its already been stored. Maybe a flag indicating
7461 we are inside such a construct would be useful,
7462 but the arrays are likely to be quite small, so
7463 for now we punt -- dmq */
7464 IV count = SvIV(sv_dat);
7465 I32 *pv = (I32*)SvPVX(sv_dat);
7467 for ( i = 0 ; i < count ; i++ ) {
7468 if ( pv[i] == RExC_npar ) {
7474 pv = (I32*)SvGROW(sv_dat, SvCUR(sv_dat) + sizeof(I32)+1);
7475 SvCUR_set(sv_dat, SvCUR(sv_dat) + sizeof(I32));
7476 pv[count] = RExC_npar;
7477 SvIV_set(sv_dat, SvIVX(sv_dat) + 1);
7480 (void)SvUPGRADE(sv_dat,SVt_PVNV);
7481 sv_setpvn(sv_dat, (char *)&(RExC_npar), sizeof(I32));
7483 SvIV_set(sv_dat, 1);
7486 /* Yes this does cause a memory leak in debugging Perls */
7487 if (!av_store(RExC_paren_name_list, RExC_npar, SvREFCNT_inc(svname)))
7488 SvREFCNT_dec(svname);
7491 /*sv_dump(sv_dat);*/
7493 nextchar(pRExC_state);
7495 goto capturing_parens;
7497 RExC_seen |= REG_SEEN_LOOKBEHIND;
7498 RExC_in_lookbehind++;
7500 case '=': /* (?=...) */
7501 RExC_seen_zerolen++;
7503 case '!': /* (?!...) */
7504 RExC_seen_zerolen++;
7505 if (*RExC_parse == ')') {
7506 ret=reg_node(pRExC_state, OPFAIL);
7507 nextchar(pRExC_state);
7511 case '|': /* (?|...) */
7512 /* branch reset, behave like a (?:...) except that
7513 buffers in alternations share the same numbers */
7515 after_freeze = freeze_paren = RExC_npar;
7517 case ':': /* (?:...) */
7518 case '>': /* (?>...) */
7520 case '$': /* (?$...) */
7521 case '@': /* (?@...) */
7522 vFAIL2("Sequence (?%c...) not implemented", (int)paren);
7524 case '#': /* (?#...) */
7525 while (*RExC_parse && *RExC_parse != ')')
7527 if (*RExC_parse != ')')
7528 FAIL("Sequence (?#... not terminated");
7529 nextchar(pRExC_state);
7532 case '0' : /* (?0) */
7533 case 'R' : /* (?R) */
7534 if (*RExC_parse != ')')
7535 FAIL("Sequence (?R) not terminated");
7536 ret = reg_node(pRExC_state, GOSTART);
7537 *flagp |= POSTPONED;
7538 nextchar(pRExC_state);
7541 { /* named and numeric backreferences */
7543 case '&': /* (?&NAME) */
7544 parse_start = RExC_parse - 1;
7547 SV *sv_dat = reg_scan_name(pRExC_state,
7548 SIZE_ONLY ? REG_RSN_RETURN_NULL : REG_RSN_RETURN_DATA);
7549 num = sv_dat ? *((I32 *)SvPVX(sv_dat)) : 0;
7551 goto gen_recurse_regop;
7554 if (!(RExC_parse[0] >= '1' && RExC_parse[0] <= '9')) {
7556 vFAIL("Illegal pattern");
7558 goto parse_recursion;
7560 case '-': /* (?-1) */
7561 if (!(RExC_parse[0] >= '1' && RExC_parse[0] <= '9')) {
7562 RExC_parse--; /* rewind to let it be handled later */
7566 case '1': case '2': case '3': case '4': /* (?1) */
7567 case '5': case '6': case '7': case '8': case '9':
7570 num = atoi(RExC_parse);
7571 parse_start = RExC_parse - 1; /* MJD */
7572 if (*RExC_parse == '-')
7574 while (isDIGIT(*RExC_parse))
7576 if (*RExC_parse!=')')
7577 vFAIL("Expecting close bracket");
7580 if ( paren == '-' ) {
7582 Diagram of capture buffer numbering.
7583 Top line is the normal capture buffer numbers
7584 Bottom line is the negative indexing as from
7588 /(a(x)y)(a(b(c(?-2)d)e)f)(g(h))/
7592 num = RExC_npar + num;
7595 vFAIL("Reference to nonexistent group");
7597 } else if ( paren == '+' ) {
7598 num = RExC_npar + num - 1;
7601 ret = reganode(pRExC_state, GOSUB, num);
7603 if (num > (I32)RExC_rx->nparens) {
7605 vFAIL("Reference to nonexistent group");
7607 ARG2L_SET( ret, RExC_recurse_count++);
7609 DEBUG_OPTIMISE_MORE_r(PerlIO_printf(Perl_debug_log,
7610 "Recurse #%"UVuf" to %"IVdf"\n", (UV)ARG(ret), (IV)ARG2L(ret)));
7614 RExC_seen |= REG_SEEN_RECURSE;
7615 Set_Node_Length(ret, 1 + regarglen[OP(ret)]); /* MJD */
7616 Set_Node_Offset(ret, parse_start); /* MJD */
7618 *flagp |= POSTPONED;
7619 nextchar(pRExC_state);
7621 } /* named and numeric backreferences */
7624 case '?': /* (??...) */
7626 if (*RExC_parse != '{') {
7628 vFAIL3("Sequence (%.*s...) not recognized", RExC_parse-seqstart, seqstart);
7631 *flagp |= POSTPONED;
7632 paren = *RExC_parse++;
7634 case '{': /* (?{...}) */
7639 char *s = RExC_parse;
7641 RExC_seen_zerolen++;
7642 RExC_seen |= REG_SEEN_EVAL;
7643 while (count && (c = *RExC_parse)) {
7654 if (*RExC_parse != ')') {
7656 vFAIL("Sequence (?{...}) not terminated or not {}-balanced");
7660 OP_4tree *sop, *rop;
7661 SV * const sv = newSVpvn(s, RExC_parse - 1 - s);
7664 Perl_save_re_context(aTHX);
7665 rop = Perl_sv_compile_2op_is_broken(aTHX_ sv, &sop, "re", &pad);
7666 sop->op_private |= OPpREFCOUNTED;
7667 /* re_dup will OpREFCNT_inc */
7668 OpREFCNT_set(sop, 1);
7671 n = add_data(pRExC_state, 3, "nop");
7672 RExC_rxi->data->data[n] = (void*)rop;
7673 RExC_rxi->data->data[n+1] = (void*)sop;
7674 RExC_rxi->data->data[n+2] = (void*)pad;
7677 else { /* First pass */
7678 if (PL_reginterp_cnt < ++RExC_seen_evals
7680 /* No compiled RE interpolated, has runtime
7681 components ===> unsafe. */
7682 FAIL("Eval-group not allowed at runtime, use re 'eval'");
7683 if (PL_tainting && PL_tainted)
7684 FAIL("Eval-group in insecure regular expression");
7685 #if PERL_VERSION > 8
7686 if (IN_PERL_COMPILETIME)
7691 nextchar(pRExC_state);
7693 ret = reg_node(pRExC_state, LOGICAL);
7696 REGTAIL(pRExC_state, ret, reganode(pRExC_state, EVAL, n));
7697 /* deal with the length of this later - MJD */
7700 ret = reganode(pRExC_state, EVAL, n);
7701 Set_Node_Length(ret, RExC_parse - parse_start + 1);
7702 Set_Node_Offset(ret, parse_start);
7705 case '(': /* (?(?{...})...) and (?(?=...)...) */
7708 if (RExC_parse[0] == '?') { /* (?(?...)) */
7709 if (RExC_parse[1] == '=' || RExC_parse[1] == '!'
7710 || RExC_parse[1] == '<'
7711 || RExC_parse[1] == '{') { /* Lookahead or eval. */
7714 ret = reg_node(pRExC_state, LOGICAL);
7717 REGTAIL(pRExC_state, ret, reg(pRExC_state, 1, &flag,depth+1));
7721 else if ( RExC_parse[0] == '<' /* (?(<NAME>)...) */
7722 || RExC_parse[0] == '\'' ) /* (?('NAME')...) */
7724 char ch = RExC_parse[0] == '<' ? '>' : '\'';
7725 char *name_start= RExC_parse++;
7727 SV *sv_dat=reg_scan_name(pRExC_state,
7728 SIZE_ONLY ? REG_RSN_RETURN_NULL : REG_RSN_RETURN_DATA);
7729 if (RExC_parse == name_start || *RExC_parse != ch)
7730 vFAIL2("Sequence (?(%c... not terminated",
7731 (ch == '>' ? '<' : ch));
7734 num = add_data( pRExC_state, 1, "S" );
7735 RExC_rxi->data->data[num]=(void*)sv_dat;
7736 SvREFCNT_inc_simple_void(sv_dat);
7738 ret = reganode(pRExC_state,NGROUPP,num);
7739 goto insert_if_check_paren;
7741 else if (RExC_parse[0] == 'D' &&
7742 RExC_parse[1] == 'E' &&
7743 RExC_parse[2] == 'F' &&
7744 RExC_parse[3] == 'I' &&
7745 RExC_parse[4] == 'N' &&
7746 RExC_parse[5] == 'E')
7748 ret = reganode(pRExC_state,DEFINEP,0);
7751 goto insert_if_check_paren;
7753 else if (RExC_parse[0] == 'R') {
7756 if (RExC_parse[0] >= '1' && RExC_parse[0] <= '9' ) {
7757 parno = atoi(RExC_parse++);
7758 while (isDIGIT(*RExC_parse))
7760 } else if (RExC_parse[0] == '&') {
7763 sv_dat = reg_scan_name(pRExC_state,
7764 SIZE_ONLY ? REG_RSN_RETURN_NULL : REG_RSN_RETURN_DATA);
7765 parno = sv_dat ? *((I32 *)SvPVX(sv_dat)) : 0;
7767 ret = reganode(pRExC_state,INSUBP,parno);
7768 goto insert_if_check_paren;
7770 else if (RExC_parse[0] >= '1' && RExC_parse[0] <= '9' ) {
7773 parno = atoi(RExC_parse++);
7775 while (isDIGIT(*RExC_parse))
7777 ret = reganode(pRExC_state, GROUPP, parno);
7779 insert_if_check_paren:
7780 if ((c = *nextchar(pRExC_state)) != ')')
7781 vFAIL("Switch condition not recognized");
7783 REGTAIL(pRExC_state, ret, reganode(pRExC_state, IFTHEN, 0));
7784 br = regbranch(pRExC_state, &flags, 1,depth+1);
7786 br = reganode(pRExC_state, LONGJMP, 0);
7788 REGTAIL(pRExC_state, br, reganode(pRExC_state, LONGJMP, 0));
7789 c = *nextchar(pRExC_state);
7794 vFAIL("(?(DEFINE)....) does not allow branches");
7795 lastbr = reganode(pRExC_state, IFTHEN, 0); /* Fake one for optimizer. */
7796 regbranch(pRExC_state, &flags, 1,depth+1);
7797 REGTAIL(pRExC_state, ret, lastbr);
7800 c = *nextchar(pRExC_state);
7805 vFAIL("Switch (?(condition)... contains too many branches");
7806 ender = reg_node(pRExC_state, TAIL);
7807 REGTAIL(pRExC_state, br, ender);
7809 REGTAIL(pRExC_state, lastbr, ender);
7810 REGTAIL(pRExC_state, NEXTOPER(NEXTOPER(lastbr)), ender);
7813 REGTAIL(pRExC_state, ret, ender);
7814 RExC_size++; /* XXX WHY do we need this?!!
7815 For large programs it seems to be required
7816 but I can't figure out why. -- dmq*/
7820 vFAIL2("Unknown switch condition (?(%.2s", RExC_parse);
7824 RExC_parse--; /* for vFAIL to print correctly */
7825 vFAIL("Sequence (? incomplete");
7827 case DEFAULT_PAT_MOD: /* Use default flags with the exceptions
7829 has_use_defaults = TRUE;
7830 STD_PMMOD_FLAGS_CLEAR(&RExC_flags);
7831 set_regex_charset(&RExC_flags, (RExC_utf8 || RExC_uni_semantics)
7832 ? REGEX_UNICODE_CHARSET
7833 : REGEX_DEPENDS_CHARSET);
7837 parse_flags: /* (?i) */
7839 U32 posflags = 0, negflags = 0;
7840 U32 *flagsp = &posflags;
7841 char has_charset_modifier = '\0';
7842 regex_charset cs = (RExC_utf8 || RExC_uni_semantics)
7843 ? REGEX_UNICODE_CHARSET
7844 : REGEX_DEPENDS_CHARSET;
7846 while (*RExC_parse) {
7847 /* && strchr("iogcmsx", *RExC_parse) */
7848 /* (?g), (?gc) and (?o) are useless here
7849 and must be globally applied -- japhy */
7850 switch (*RExC_parse) {
7851 CASE_STD_PMMOD_FLAGS_PARSE_SET(flagsp);
7852 case LOCALE_PAT_MOD:
7853 if (has_charset_modifier) {
7854 goto excess_modifier;
7856 else if (flagsp == &negflags) {
7859 cs = REGEX_LOCALE_CHARSET;
7860 has_charset_modifier = LOCALE_PAT_MOD;
7861 RExC_contains_locale = 1;
7863 case UNICODE_PAT_MOD:
7864 if (has_charset_modifier) {
7865 goto excess_modifier;
7867 else if (flagsp == &negflags) {
7870 cs = REGEX_UNICODE_CHARSET;
7871 has_charset_modifier = UNICODE_PAT_MOD;
7873 case ASCII_RESTRICT_PAT_MOD:
7874 if (flagsp == &negflags) {
7877 if (has_charset_modifier) {
7878 if (cs != REGEX_ASCII_RESTRICTED_CHARSET) {
7879 goto excess_modifier;
7881 /* Doubled modifier implies more restricted */
7882 cs = REGEX_ASCII_MORE_RESTRICTED_CHARSET;
7885 cs = REGEX_ASCII_RESTRICTED_CHARSET;
7887 has_charset_modifier = ASCII_RESTRICT_PAT_MOD;
7889 case DEPENDS_PAT_MOD:
7890 if (has_use_defaults) {
7891 goto fail_modifiers;
7893 else if (flagsp == &negflags) {
7896 else if (has_charset_modifier) {
7897 goto excess_modifier;
7900 /* The dual charset means unicode semantics if the
7901 * pattern (or target, not known until runtime) are
7902 * utf8, or something in the pattern indicates unicode
7904 cs = (RExC_utf8 || RExC_uni_semantics)
7905 ? REGEX_UNICODE_CHARSET
7906 : REGEX_DEPENDS_CHARSET;
7907 has_charset_modifier = DEPENDS_PAT_MOD;
7911 if (has_charset_modifier == ASCII_RESTRICT_PAT_MOD) {
7912 vFAIL2("Regexp modifier \"%c\" may appear a maximum of twice", ASCII_RESTRICT_PAT_MOD);
7914 else if (has_charset_modifier == *(RExC_parse - 1)) {
7915 vFAIL2("Regexp modifier \"%c\" may not appear twice", *(RExC_parse - 1));
7918 vFAIL3("Regexp modifiers \"%c\" and \"%c\" are mutually exclusive", has_charset_modifier, *(RExC_parse - 1));
7923 vFAIL2("Regexp modifier \"%c\" may not appear after the \"-\"", *(RExC_parse - 1));
7925 case ONCE_PAT_MOD: /* 'o' */
7926 case GLOBAL_PAT_MOD: /* 'g' */
7927 if (SIZE_ONLY && ckWARN(WARN_REGEXP)) {
7928 const I32 wflagbit = *RExC_parse == 'o' ? WASTED_O : WASTED_G;
7929 if (! (wastedflags & wflagbit) ) {
7930 wastedflags |= wflagbit;
7933 "Useless (%s%c) - %suse /%c modifier",
7934 flagsp == &negflags ? "?-" : "?",
7936 flagsp == &negflags ? "don't " : "",
7943 case CONTINUE_PAT_MOD: /* 'c' */
7944 if (SIZE_ONLY && ckWARN(WARN_REGEXP)) {
7945 if (! (wastedflags & WASTED_C) ) {
7946 wastedflags |= WASTED_GC;
7949 "Useless (%sc) - %suse /gc modifier",
7950 flagsp == &negflags ? "?-" : "?",
7951 flagsp == &negflags ? "don't " : ""
7956 case KEEPCOPY_PAT_MOD: /* 'p' */
7957 if (flagsp == &negflags) {
7959 ckWARNreg(RExC_parse + 1,"Useless use of (?-p)");
7961 *flagsp |= RXf_PMf_KEEPCOPY;
7965 /* A flag is a default iff it is following a minus, so
7966 * if there is a minus, it means will be trying to
7967 * re-specify a default which is an error */
7968 if (has_use_defaults || flagsp == &negflags) {
7971 vFAIL3("Sequence (%.*s...) not recognized", RExC_parse-seqstart, seqstart);
7975 wastedflags = 0; /* reset so (?g-c) warns twice */
7981 RExC_flags |= posflags;
7982 RExC_flags &= ~negflags;
7983 set_regex_charset(&RExC_flags, cs);
7985 oregflags |= posflags;
7986 oregflags &= ~negflags;
7987 set_regex_charset(&oregflags, cs);
7989 nextchar(pRExC_state);
8000 vFAIL3("Sequence (%.*s...) not recognized", RExC_parse-seqstart, seqstart);
8005 }} /* one for the default block, one for the switch */
8012 ret = reganode(pRExC_state, OPEN, parno);
8015 RExC_nestroot = parno;
8016 if (RExC_seen & REG_SEEN_RECURSE
8017 && !RExC_open_parens[parno-1])
8019 DEBUG_OPTIMISE_MORE_r(PerlIO_printf(Perl_debug_log,
8020 "Setting open paren #%"IVdf" to %d\n",
8021 (IV)parno, REG_NODE_NUM(ret)));
8022 RExC_open_parens[parno-1]= ret;
8025 Set_Node_Length(ret, 1); /* MJD */
8026 Set_Node_Offset(ret, RExC_parse); /* MJD */
8034 /* Pick up the branches, linking them together. */
8035 parse_start = RExC_parse; /* MJD */
8036 br = regbranch(pRExC_state, &flags, 1,depth+1);
8038 /* branch_len = (paren != 0); */
8042 if (*RExC_parse == '|') {
8043 if (!SIZE_ONLY && RExC_extralen) {
8044 reginsert(pRExC_state, BRANCHJ, br, depth+1);
8047 reginsert(pRExC_state, BRANCH, br, depth+1);
8048 Set_Node_Length(br, paren != 0);
8049 Set_Node_Offset_To_R(br-RExC_emit_start, parse_start-RExC_start);
8053 RExC_extralen += 1; /* For BRANCHJ-BRANCH. */
8055 else if (paren == ':') {
8056 *flagp |= flags&SIMPLE;
8058 if (is_open) { /* Starts with OPEN. */
8059 REGTAIL(pRExC_state, ret, br); /* OPEN -> first. */
8061 else if (paren != '?') /* Not Conditional */
8063 *flagp |= flags & (SPSTART | HASWIDTH | POSTPONED);
8065 while (*RExC_parse == '|') {
8066 if (!SIZE_ONLY && RExC_extralen) {
8067 ender = reganode(pRExC_state, LONGJMP,0);
8068 REGTAIL(pRExC_state, NEXTOPER(NEXTOPER(lastbr)), ender); /* Append to the previous. */
8071 RExC_extralen += 2; /* Account for LONGJMP. */
8072 nextchar(pRExC_state);
8074 if (RExC_npar > after_freeze)
8075 after_freeze = RExC_npar;
8076 RExC_npar = freeze_paren;
8078 br = regbranch(pRExC_state, &flags, 0, depth+1);
8082 REGTAIL(pRExC_state, lastbr, br); /* BRANCH -> BRANCH. */
8084 *flagp |= flags & (SPSTART | HASWIDTH | POSTPONED);
8087 if (have_branch || paren != ':') {
8088 /* Make a closing node, and hook it on the end. */
8091 ender = reg_node(pRExC_state, TAIL);
8094 ender = reganode(pRExC_state, CLOSE, parno);
8095 if (!SIZE_ONLY && RExC_seen & REG_SEEN_RECURSE) {
8096 DEBUG_OPTIMISE_MORE_r(PerlIO_printf(Perl_debug_log,
8097 "Setting close paren #%"IVdf" to %d\n",
8098 (IV)parno, REG_NODE_NUM(ender)));
8099 RExC_close_parens[parno-1]= ender;
8100 if (RExC_nestroot == parno)
8103 Set_Node_Offset(ender,RExC_parse+1); /* MJD */
8104 Set_Node_Length(ender,1); /* MJD */
8110 *flagp &= ~HASWIDTH;
8113 ender = reg_node(pRExC_state, SUCCEED);
8116 ender = reg_node(pRExC_state, END);
8118 assert(!RExC_opend); /* there can only be one! */
8123 REGTAIL(pRExC_state, lastbr, ender);
8125 if (have_branch && !SIZE_ONLY) {
8127 RExC_seen |= REG_TOP_LEVEL_BRANCHES;
8129 /* Hook the tails of the branches to the closing node. */
8130 for (br = ret; br; br = regnext(br)) {
8131 const U8 op = PL_regkind[OP(br)];
8133 REGTAIL_STUDY(pRExC_state, NEXTOPER(br), ender);
8135 else if (op == BRANCHJ) {
8136 REGTAIL_STUDY(pRExC_state, NEXTOPER(NEXTOPER(br)), ender);
8144 static const char parens[] = "=!<,>";
8146 if (paren && (p = strchr(parens, paren))) {
8147 U8 node = ((p - parens) % 2) ? UNLESSM : IFMATCH;
8148 int flag = (p - parens) > 1;
8151 node = SUSPEND, flag = 0;
8152 reginsert(pRExC_state, node,ret, depth+1);
8153 Set_Node_Cur_Length(ret);
8154 Set_Node_Offset(ret, parse_start + 1);
8156 REGTAIL_STUDY(pRExC_state, ret, reg_node(pRExC_state, TAIL));
8160 /* Check for proper termination. */
8162 RExC_flags = oregflags;
8163 if (RExC_parse >= RExC_end || *nextchar(pRExC_state) != ')') {
8164 RExC_parse = oregcomp_parse;
8165 vFAIL("Unmatched (");
8168 else if (!paren && RExC_parse < RExC_end) {
8169 if (*RExC_parse == ')') {
8171 vFAIL("Unmatched )");
8174 FAIL("Junk on end of regexp"); /* "Can't happen". */
8178 if (RExC_in_lookbehind) {
8179 RExC_in_lookbehind--;
8181 if (after_freeze > RExC_npar)
8182 RExC_npar = after_freeze;
8187 - regbranch - one alternative of an | operator
8189 * Implements the concatenation operator.
8192 S_regbranch(pTHX_ RExC_state_t *pRExC_state, I32 *flagp, I32 first, U32 depth)
8195 register regnode *ret;
8196 register regnode *chain = NULL;
8197 register regnode *latest;
8198 I32 flags = 0, c = 0;
8199 GET_RE_DEBUG_FLAGS_DECL;
8201 PERL_ARGS_ASSERT_REGBRANCH;
8203 DEBUG_PARSE("brnc");
8208 if (!SIZE_ONLY && RExC_extralen)
8209 ret = reganode(pRExC_state, BRANCHJ,0);
8211 ret = reg_node(pRExC_state, BRANCH);
8212 Set_Node_Length(ret, 1);
8216 if (!first && SIZE_ONLY)
8217 RExC_extralen += 1; /* BRANCHJ */
8219 *flagp = WORST; /* Tentatively. */
8222 nextchar(pRExC_state);
8223 while (RExC_parse < RExC_end && *RExC_parse != '|' && *RExC_parse != ')') {
8225 latest = regpiece(pRExC_state, &flags,depth+1);
8226 if (latest == NULL) {
8227 if (flags & TRYAGAIN)
8231 else if (ret == NULL)
8233 *flagp |= flags&(HASWIDTH|POSTPONED);
8234 if (chain == NULL) /* First piece. */
8235 *flagp |= flags&SPSTART;
8238 REGTAIL(pRExC_state, chain, latest);
8243 if (chain == NULL) { /* Loop ran zero times. */
8244 chain = reg_node(pRExC_state, NOTHING);
8249 *flagp |= flags&SIMPLE;
8256 - regpiece - something followed by possible [*+?]
8258 * Note that the branching code sequences used for ? and the general cases
8259 * of * and + are somewhat optimized: they use the same NOTHING node as
8260 * both the endmarker for their branch list and the body of the last branch.
8261 * It might seem that this node could be dispensed with entirely, but the
8262 * endmarker role is not redundant.
8265 S_regpiece(pTHX_ RExC_state_t *pRExC_state, I32 *flagp, U32 depth)
8268 register regnode *ret;
8270 register char *next;
8272 const char * const origparse = RExC_parse;
8274 I32 max = REG_INFTY;
8275 #ifdef RE_TRACK_PATTERN_OFFSETS
8278 const char *maxpos = NULL;
8279 GET_RE_DEBUG_FLAGS_DECL;
8281 PERL_ARGS_ASSERT_REGPIECE;
8283 DEBUG_PARSE("piec");
8285 ret = regatom(pRExC_state, &flags,depth+1);
8287 if (flags & TRYAGAIN)
8294 if (op == '{' && regcurly(RExC_parse)) {
8296 #ifdef RE_TRACK_PATTERN_OFFSETS
8297 parse_start = RExC_parse; /* MJD */
8299 next = RExC_parse + 1;
8300 while (isDIGIT(*next) || *next == ',') {
8309 if (*next == '}') { /* got one */
8313 min = atoi(RExC_parse);
8317 maxpos = RExC_parse;
8319 if (!max && *maxpos != '0')
8320 max = REG_INFTY; /* meaning "infinity" */
8321 else if (max >= REG_INFTY)
8322 vFAIL2("Quantifier in {,} bigger than %d", REG_INFTY - 1);
8324 nextchar(pRExC_state);
8327 if ((flags&SIMPLE)) {
8328 RExC_naughty += 2 + RExC_naughty / 2;
8329 reginsert(pRExC_state, CURLY, ret, depth+1);
8330 Set_Node_Offset(ret, parse_start+1); /* MJD */
8331 Set_Node_Cur_Length(ret);
8334 regnode * const w = reg_node(pRExC_state, WHILEM);
8337 REGTAIL(pRExC_state, ret, w);
8338 if (!SIZE_ONLY && RExC_extralen) {
8339 reginsert(pRExC_state, LONGJMP,ret, depth+1);
8340 reginsert(pRExC_state, NOTHING,ret, depth+1);
8341 NEXT_OFF(ret) = 3; /* Go over LONGJMP. */
8343 reginsert(pRExC_state, CURLYX,ret, depth+1);
8345 Set_Node_Offset(ret, parse_start+1);
8346 Set_Node_Length(ret,
8347 op == '{' ? (RExC_parse - parse_start) : 1);
8349 if (!SIZE_ONLY && RExC_extralen)
8350 NEXT_OFF(ret) = 3; /* Go over NOTHING to LONGJMP. */
8351 REGTAIL(pRExC_state, ret, reg_node(pRExC_state, NOTHING));
8353 RExC_whilem_seen++, RExC_extralen += 3;
8354 RExC_naughty += 4 + RExC_naughty; /* compound interest */
8363 vFAIL("Can't do {n,m} with n > m");
8365 ARG1_SET(ret, (U16)min);
8366 ARG2_SET(ret, (U16)max);
8378 #if 0 /* Now runtime fix should be reliable. */
8380 /* if this is reinstated, don't forget to put this back into perldiag:
8382 =item Regexp *+ operand could be empty at {#} in regex m/%s/
8384 (F) The part of the regexp subject to either the * or + quantifier
8385 could match an empty string. The {#} shows in the regular
8386 expression about where the problem was discovered.
8390 if (!(flags&HASWIDTH) && op != '?')
8391 vFAIL("Regexp *+ operand could be empty");
8394 #ifdef RE_TRACK_PATTERN_OFFSETS
8395 parse_start = RExC_parse;
8397 nextchar(pRExC_state);
8399 *flagp = (op != '+') ? (WORST|SPSTART|HASWIDTH) : (WORST|HASWIDTH);
8401 if (op == '*' && (flags&SIMPLE)) {
8402 reginsert(pRExC_state, STAR, ret, depth+1);
8406 else if (op == '*') {
8410 else if (op == '+' && (flags&SIMPLE)) {
8411 reginsert(pRExC_state, PLUS, ret, depth+1);
8415 else if (op == '+') {
8419 else if (op == '?') {
8424 if (!SIZE_ONLY && !(flags&(HASWIDTH|POSTPONED)) && max > REG_INFTY/3) {
8425 ckWARN3reg(RExC_parse,
8426 "%.*s matches null string many times",
8427 (int)(RExC_parse >= origparse ? RExC_parse - origparse : 0),
8431 if (RExC_parse < RExC_end && *RExC_parse == '?') {
8432 nextchar(pRExC_state);
8433 reginsert(pRExC_state, MINMOD, ret, depth+1);
8434 REGTAIL(pRExC_state, ret, ret + NODE_STEP_REGNODE);
8436 #ifndef REG_ALLOW_MINMOD_SUSPEND
8439 if (RExC_parse < RExC_end && *RExC_parse == '+') {
8441 nextchar(pRExC_state);
8442 ender = reg_node(pRExC_state, SUCCEED);
8443 REGTAIL(pRExC_state, ret, ender);
8444 reginsert(pRExC_state, SUSPEND, ret, depth+1);
8446 ender = reg_node(pRExC_state, TAIL);
8447 REGTAIL(pRExC_state, ret, ender);
8451 if (RExC_parse < RExC_end && ISMULT2(RExC_parse)) {
8453 vFAIL("Nested quantifiers");
8460 /* reg_namedseq(pRExC_state,UVp, UV depth)
8462 This is expected to be called by a parser routine that has
8463 recognized '\N' and needs to handle the rest. RExC_parse is
8464 expected to point at the first char following the N at the time
8467 The \N may be inside (indicated by valuep not being NULL) or outside a
8470 \N may begin either a named sequence, or if outside a character class, mean
8471 to match a non-newline. For non single-quoted regexes, the tokenizer has
8472 attempted to decide which, and in the case of a named sequence converted it
8473 into one of the forms: \N{} (if the sequence is null), or \N{U+c1.c2...},
8474 where c1... are the characters in the sequence. For single-quoted regexes,
8475 the tokenizer passes the \N sequence through unchanged; this code will not
8476 attempt to determine this nor expand those. The net effect is that if the
8477 beginning of the passed-in pattern isn't '{U+' or there is no '}', it
8478 signals that this \N occurrence means to match a non-newline.
8480 Only the \N{U+...} form should occur in a character class, for the same
8481 reason that '.' inside a character class means to just match a period: it
8482 just doesn't make sense.
8484 If valuep is non-null then it is assumed that we are parsing inside
8485 of a charclass definition and the first codepoint in the resolved
8486 string is returned via *valuep and the routine will return NULL.
8487 In this mode if a multichar string is returned from the charnames
8488 handler, a warning will be issued, and only the first char in the
8489 sequence will be examined. If the string returned is zero length
8490 then the value of *valuep is undefined and NON-NULL will
8491 be returned to indicate failure. (This will NOT be a valid pointer
8494 If valuep is null then it is assumed that we are parsing normal text and a
8495 new EXACT node is inserted into the program containing the resolved string,
8496 and a pointer to the new node is returned. But if the string is zero length
8497 a NOTHING node is emitted instead.
8499 On success RExC_parse is set to the char following the endbrace.
8500 Parsing failures will generate a fatal error via vFAIL(...)
8503 S_reg_namedseq(pTHX_ RExC_state_t *pRExC_state, UV *valuep, I32 *flagp, U32 depth)
8505 char * endbrace; /* '}' following the name */
8506 regnode *ret = NULL;
8509 GET_RE_DEBUG_FLAGS_DECL;
8511 PERL_ARGS_ASSERT_REG_NAMEDSEQ;
8515 /* The [^\n] meaning of \N ignores spaces and comments under the /x
8516 * modifier. The other meaning does not */
8517 p = (RExC_flags & RXf_PMf_EXTENDED)
8518 ? regwhite( pRExC_state, RExC_parse )
8521 /* Disambiguate between \N meaning a named character versus \N meaning
8522 * [^\n]. The former is assumed when it can't be the latter. */
8523 if (*p != '{' || regcurly(p)) {
8526 /* no bare \N in a charclass */
8527 vFAIL("\\N in a character class must be a named character: \\N{...}");
8529 nextchar(pRExC_state);
8530 ret = reg_node(pRExC_state, REG_ANY);
8531 *flagp |= HASWIDTH|SIMPLE;
8534 Set_Node_Length(ret, 1); /* MJD */
8538 /* Here, we have decided it should be a named sequence */
8540 /* The test above made sure that the next real character is a '{', but
8541 * under the /x modifier, it could be separated by space (or a comment and
8542 * \n) and this is not allowed (for consistency with \x{...} and the
8543 * tokenizer handling of \N{NAME}). */
8544 if (*RExC_parse != '{') {
8545 vFAIL("Missing braces on \\N{}");
8548 RExC_parse++; /* Skip past the '{' */
8550 if (! (endbrace = strchr(RExC_parse, '}')) /* no trailing brace */
8551 || ! (endbrace == RExC_parse /* nothing between the {} */
8552 || (endbrace - RExC_parse >= 2 /* U+ (bad hex is checked below */
8553 && strnEQ(RExC_parse, "U+", 2)))) /* for a better error msg) */
8555 if (endbrace) RExC_parse = endbrace; /* position msg's '<--HERE' */
8556 vFAIL("\\N{NAME} must be resolved by the lexer");
8559 if (endbrace == RExC_parse) { /* empty: \N{} */
8561 RExC_parse = endbrace + 1;
8562 return reg_node(pRExC_state,NOTHING);
8566 ckWARNreg(RExC_parse,
8567 "Ignoring zero length \\N{} in character class"
8569 RExC_parse = endbrace + 1;
8572 return (regnode *) &RExC_parse; /* Invalid regnode pointer */
8575 REQUIRE_UTF8; /* named sequences imply Unicode semantics */
8576 RExC_parse += 2; /* Skip past the 'U+' */
8578 if (valuep) { /* In a bracketed char class */
8579 /* We only pay attention to the first char of
8580 multichar strings being returned. I kinda wonder
8581 if this makes sense as it does change the behaviour
8582 from earlier versions, OTOH that behaviour was broken
8583 as well. XXX Solution is to recharacterize as
8584 [rest-of-class]|multi1|multi2... */
8586 STRLEN length_of_hex;
8587 I32 flags = PERL_SCAN_ALLOW_UNDERSCORES
8588 | PERL_SCAN_DISALLOW_PREFIX
8589 | (SIZE_ONLY ? PERL_SCAN_SILENT_ILLDIGIT : 0);
8591 char * endchar = RExC_parse + strcspn(RExC_parse, ".}");
8592 if (endchar < endbrace) {
8593 ckWARNreg(endchar, "Using just the first character returned by \\N{} in character class");
8596 length_of_hex = (STRLEN)(endchar - RExC_parse);
8597 *valuep = grok_hex(RExC_parse, &length_of_hex, &flags, NULL);
8599 /* The tokenizer should have guaranteed validity, but it's possible to
8600 * bypass it by using single quoting, so check */
8601 if (length_of_hex == 0
8602 || length_of_hex != (STRLEN)(endchar - RExC_parse) )
8604 RExC_parse += length_of_hex; /* Includes all the valid */
8605 RExC_parse += (RExC_orig_utf8) /* point to after 1st invalid */
8606 ? UTF8SKIP(RExC_parse)
8608 /* Guard against malformed utf8 */
8609 if (RExC_parse >= endchar) RExC_parse = endchar;
8610 vFAIL("Invalid hexadecimal number in \\N{U+...}");
8613 RExC_parse = endbrace + 1;
8614 if (endchar == endbrace) return NULL;
8616 ret = (regnode *) &RExC_parse; /* Invalid regnode pointer */
8618 else { /* Not a char class */
8620 /* What is done here is to convert this to a sub-pattern of the form
8621 * (?:\x{char1}\x{char2}...)
8622 * and then call reg recursively. That way, it retains its atomicness,
8623 * while not having to worry about special handling that some code
8624 * points may have. toke.c has converted the original Unicode values
8625 * to native, so that we can just pass on the hex values unchanged. We
8626 * do have to set a flag to keep recoding from happening in the
8629 SV * substitute_parse = newSVpvn_flags("?:", 2, SVf_UTF8|SVs_TEMP);
8631 char *endchar; /* Points to '.' or '}' ending cur char in the input
8633 char *orig_end = RExC_end;
8635 while (RExC_parse < endbrace) {
8637 /* Code points are separated by dots. If none, there is only one
8638 * code point, and is terminated by the brace */
8639 endchar = RExC_parse + strcspn(RExC_parse, ".}");
8641 /* Convert to notation the rest of the code understands */
8642 sv_catpv(substitute_parse, "\\x{");
8643 sv_catpvn(substitute_parse, RExC_parse, endchar - RExC_parse);
8644 sv_catpv(substitute_parse, "}");
8646 /* Point to the beginning of the next character in the sequence. */
8647 RExC_parse = endchar + 1;
8649 sv_catpv(substitute_parse, ")");
8651 RExC_parse = SvPV(substitute_parse, len);
8653 /* Don't allow empty number */
8655 vFAIL("Invalid hexadecimal number in \\N{U+...}");
8657 RExC_end = RExC_parse + len;
8659 /* The values are Unicode, and therefore not subject to recoding */
8660 RExC_override_recoding = 1;
8662 ret = reg(pRExC_state, 1, flagp, depth+1);
8664 RExC_parse = endbrace;
8665 RExC_end = orig_end;
8666 RExC_override_recoding = 0;
8668 nextchar(pRExC_state);
8678 * It returns the code point in utf8 for the value in *encp.
8679 * value: a code value in the source encoding
8680 * encp: a pointer to an Encode object
8682 * If the result from Encode is not a single character,
8683 * it returns U+FFFD (Replacement character) and sets *encp to NULL.
8686 S_reg_recode(pTHX_ const char value, SV **encp)
8689 SV * const sv = newSVpvn_flags(&value, numlen, SVs_TEMP);
8690 const char * const s = *encp ? sv_recode_to_utf8(sv, *encp) : SvPVX(sv);
8691 const STRLEN newlen = SvCUR(sv);
8692 UV uv = UNICODE_REPLACEMENT;
8694 PERL_ARGS_ASSERT_REG_RECODE;
8698 ? utf8n_to_uvchr((U8*)s, newlen, &numlen, UTF8_ALLOW_DEFAULT)
8701 if (!newlen || numlen != newlen) {
8702 uv = UNICODE_REPLACEMENT;
8710 - regatom - the lowest level
8712 Try to identify anything special at the start of the pattern. If there
8713 is, then handle it as required. This may involve generating a single regop,
8714 such as for an assertion; or it may involve recursing, such as to
8715 handle a () structure.
8717 If the string doesn't start with something special then we gobble up
8718 as much literal text as we can.
8720 Once we have been able to handle whatever type of thing started the
8721 sequence, we return.
8723 Note: we have to be careful with escapes, as they can be both literal
8724 and special, and in the case of \10 and friends can either, depending
8725 on context. Specifically there are two separate switches for handling
8726 escape sequences, with the one for handling literal escapes requiring
8727 a dummy entry for all of the special escapes that are actually handled
8732 S_regatom(pTHX_ RExC_state_t *pRExC_state, I32 *flagp, U32 depth)
8735 register regnode *ret = NULL;
8737 char *parse_start = RExC_parse;
8739 GET_RE_DEBUG_FLAGS_DECL;
8740 DEBUG_PARSE("atom");
8741 *flagp = WORST; /* Tentatively. */
8743 PERL_ARGS_ASSERT_REGATOM;
8746 switch ((U8)*RExC_parse) {
8748 RExC_seen_zerolen++;
8749 nextchar(pRExC_state);
8750 if (RExC_flags & RXf_PMf_MULTILINE)
8751 ret = reg_node(pRExC_state, MBOL);
8752 else if (RExC_flags & RXf_PMf_SINGLELINE)
8753 ret = reg_node(pRExC_state, SBOL);
8755 ret = reg_node(pRExC_state, BOL);
8756 Set_Node_Length(ret, 1); /* MJD */
8759 nextchar(pRExC_state);
8761 RExC_seen_zerolen++;
8762 if (RExC_flags & RXf_PMf_MULTILINE)
8763 ret = reg_node(pRExC_state, MEOL);
8764 else if (RExC_flags & RXf_PMf_SINGLELINE)
8765 ret = reg_node(pRExC_state, SEOL);
8767 ret = reg_node(pRExC_state, EOL);
8768 Set_Node_Length(ret, 1); /* MJD */
8771 nextchar(pRExC_state);
8772 if (RExC_flags & RXf_PMf_SINGLELINE)
8773 ret = reg_node(pRExC_state, SANY);
8775 ret = reg_node(pRExC_state, REG_ANY);
8776 *flagp |= HASWIDTH|SIMPLE;
8778 Set_Node_Length(ret, 1); /* MJD */
8782 char * const oregcomp_parse = ++RExC_parse;
8783 ret = regclass(pRExC_state,depth+1);
8784 if (*RExC_parse != ']') {
8785 RExC_parse = oregcomp_parse;
8786 vFAIL("Unmatched [");
8788 nextchar(pRExC_state);
8789 *flagp |= HASWIDTH|SIMPLE;
8790 Set_Node_Length(ret, RExC_parse - oregcomp_parse + 1); /* MJD */
8794 nextchar(pRExC_state);
8795 ret = reg(pRExC_state, 1, &flags,depth+1);
8797 if (flags & TRYAGAIN) {
8798 if (RExC_parse == RExC_end) {
8799 /* Make parent create an empty node if needed. */
8807 *flagp |= flags&(HASWIDTH|SPSTART|SIMPLE|POSTPONED);
8811 if (flags & TRYAGAIN) {
8815 vFAIL("Internal urp");
8816 /* Supposed to be caught earlier. */
8819 if (!regcurly(RExC_parse)) {
8828 vFAIL("Quantifier follows nothing");
8833 This switch handles escape sequences that resolve to some kind
8834 of special regop and not to literal text. Escape sequnces that
8835 resolve to literal text are handled below in the switch marked
8838 Every entry in this switch *must* have a corresponding entry
8839 in the literal escape switch. However, the opposite is not
8840 required, as the default for this switch is to jump to the
8841 literal text handling code.
8843 switch ((U8)*++RExC_parse) {
8844 /* Special Escapes */
8846 RExC_seen_zerolen++;
8847 ret = reg_node(pRExC_state, SBOL);
8849 goto finish_meta_pat;
8851 ret = reg_node(pRExC_state, GPOS);
8852 RExC_seen |= REG_SEEN_GPOS;
8854 goto finish_meta_pat;
8856 RExC_seen_zerolen++;
8857 ret = reg_node(pRExC_state, KEEPS);
8859 /* XXX:dmq : disabling in-place substitution seems to
8860 * be necessary here to avoid cases of memory corruption, as
8861 * with: C<$_="x" x 80; s/x\K/y/> -- rgs
8863 RExC_seen |= REG_SEEN_LOOKBEHIND;
8864 goto finish_meta_pat;
8866 ret = reg_node(pRExC_state, SEOL);
8868 RExC_seen_zerolen++; /* Do not optimize RE away */
8869 goto finish_meta_pat;
8871 ret = reg_node(pRExC_state, EOS);
8873 RExC_seen_zerolen++; /* Do not optimize RE away */
8874 goto finish_meta_pat;
8876 ret = reg_node(pRExC_state, CANY);
8877 RExC_seen |= REG_SEEN_CANY;
8878 *flagp |= HASWIDTH|SIMPLE;
8879 goto finish_meta_pat;
8881 ret = reg_node(pRExC_state, CLUMP);
8883 goto finish_meta_pat;
8885 switch (get_regex_charset(RExC_flags)) {
8886 case REGEX_LOCALE_CHARSET:
8889 case REGEX_UNICODE_CHARSET:
8892 case REGEX_ASCII_RESTRICTED_CHARSET:
8893 case REGEX_ASCII_MORE_RESTRICTED_CHARSET:
8896 case REGEX_DEPENDS_CHARSET:
8902 ret = reg_node(pRExC_state, op);
8903 *flagp |= HASWIDTH|SIMPLE;
8904 goto finish_meta_pat;
8906 switch (get_regex_charset(RExC_flags)) {
8907 case REGEX_LOCALE_CHARSET:
8910 case REGEX_UNICODE_CHARSET:
8913 case REGEX_ASCII_RESTRICTED_CHARSET:
8914 case REGEX_ASCII_MORE_RESTRICTED_CHARSET:
8917 case REGEX_DEPENDS_CHARSET:
8923 ret = reg_node(pRExC_state, op);
8924 *flagp |= HASWIDTH|SIMPLE;
8925 goto finish_meta_pat;
8927 RExC_seen_zerolen++;
8928 RExC_seen |= REG_SEEN_LOOKBEHIND;
8929 switch (get_regex_charset(RExC_flags)) {
8930 case REGEX_LOCALE_CHARSET:
8933 case REGEX_UNICODE_CHARSET:
8936 case REGEX_ASCII_RESTRICTED_CHARSET:
8937 case REGEX_ASCII_MORE_RESTRICTED_CHARSET:
8940 case REGEX_DEPENDS_CHARSET:
8946 ret = reg_node(pRExC_state, op);
8947 FLAGS(ret) = get_regex_charset(RExC_flags);
8949 if (! SIZE_ONLY && (U8) *(RExC_parse + 1) == '{') {
8950 ckWARNregdep(RExC_parse, "\"\\b{\" is deprecated; use \"\\b\\{\" instead");
8952 goto finish_meta_pat;
8954 RExC_seen_zerolen++;
8955 RExC_seen |= REG_SEEN_LOOKBEHIND;
8956 switch (get_regex_charset(RExC_flags)) {
8957 case REGEX_LOCALE_CHARSET:
8960 case REGEX_UNICODE_CHARSET:
8963 case REGEX_ASCII_RESTRICTED_CHARSET:
8964 case REGEX_ASCII_MORE_RESTRICTED_CHARSET:
8967 case REGEX_DEPENDS_CHARSET:
8973 ret = reg_node(pRExC_state, op);
8974 FLAGS(ret) = get_regex_charset(RExC_flags);
8976 if (! SIZE_ONLY && (U8) *(RExC_parse + 1) == '{') {
8977 ckWARNregdep(RExC_parse, "\"\\B{\" is deprecated; use \"\\B\\{\" instead");
8979 goto finish_meta_pat;
8981 switch (get_regex_charset(RExC_flags)) {
8982 case REGEX_LOCALE_CHARSET:
8985 case REGEX_UNICODE_CHARSET:
8988 case REGEX_ASCII_RESTRICTED_CHARSET:
8989 case REGEX_ASCII_MORE_RESTRICTED_CHARSET:
8992 case REGEX_DEPENDS_CHARSET:
8998 ret = reg_node(pRExC_state, op);
8999 *flagp |= HASWIDTH|SIMPLE;
9000 goto finish_meta_pat;
9002 switch (get_regex_charset(RExC_flags)) {
9003 case REGEX_LOCALE_CHARSET:
9006 case REGEX_UNICODE_CHARSET:
9009 case REGEX_ASCII_RESTRICTED_CHARSET:
9010 case REGEX_ASCII_MORE_RESTRICTED_CHARSET:
9013 case REGEX_DEPENDS_CHARSET:
9019 ret = reg_node(pRExC_state, op);
9020 *flagp |= HASWIDTH|SIMPLE;
9021 goto finish_meta_pat;
9023 switch (get_regex_charset(RExC_flags)) {
9024 case REGEX_LOCALE_CHARSET:
9027 case REGEX_ASCII_RESTRICTED_CHARSET:
9028 case REGEX_ASCII_MORE_RESTRICTED_CHARSET:
9031 case REGEX_DEPENDS_CHARSET: /* No difference between these */
9032 case REGEX_UNICODE_CHARSET:
9038 ret = reg_node(pRExC_state, op);
9039 *flagp |= HASWIDTH|SIMPLE;
9040 goto finish_meta_pat;
9042 switch (get_regex_charset(RExC_flags)) {
9043 case REGEX_LOCALE_CHARSET:
9046 case REGEX_ASCII_RESTRICTED_CHARSET:
9047 case REGEX_ASCII_MORE_RESTRICTED_CHARSET:
9050 case REGEX_DEPENDS_CHARSET: /* No difference between these */
9051 case REGEX_UNICODE_CHARSET:
9057 ret = reg_node(pRExC_state, op);
9058 *flagp |= HASWIDTH|SIMPLE;
9059 goto finish_meta_pat;
9061 ret = reg_node(pRExC_state, LNBREAK);
9062 *flagp |= HASWIDTH|SIMPLE;
9063 goto finish_meta_pat;
9065 ret = reg_node(pRExC_state, HORIZWS);
9066 *flagp |= HASWIDTH|SIMPLE;
9067 goto finish_meta_pat;
9069 ret = reg_node(pRExC_state, NHORIZWS);
9070 *flagp |= HASWIDTH|SIMPLE;
9071 goto finish_meta_pat;
9073 ret = reg_node(pRExC_state, VERTWS);
9074 *flagp |= HASWIDTH|SIMPLE;
9075 goto finish_meta_pat;
9077 ret = reg_node(pRExC_state, NVERTWS);
9078 *flagp |= HASWIDTH|SIMPLE;
9080 nextchar(pRExC_state);
9081 Set_Node_Length(ret, 2); /* MJD */
9086 char* const oldregxend = RExC_end;
9088 char* parse_start = RExC_parse - 2;
9091 if (RExC_parse[1] == '{') {
9092 /* a lovely hack--pretend we saw [\pX] instead */
9093 RExC_end = strchr(RExC_parse, '}');
9095 const U8 c = (U8)*RExC_parse;
9097 RExC_end = oldregxend;
9098 vFAIL2("Missing right brace on \\%c{}", c);
9103 RExC_end = RExC_parse + 2;
9104 if (RExC_end > oldregxend)
9105 RExC_end = oldregxend;
9109 ret = regclass(pRExC_state,depth+1);
9111 RExC_end = oldregxend;
9114 Set_Node_Offset(ret, parse_start + 2);
9115 Set_Node_Cur_Length(ret);
9116 nextchar(pRExC_state);
9117 *flagp |= HASWIDTH|SIMPLE;
9121 /* Handle \N and \N{NAME} here and not below because it can be
9122 multicharacter. join_exact() will join them up later on.
9123 Also this makes sure that things like /\N{BLAH}+/ and
9124 \N{BLAH} being multi char Just Happen. dmq*/
9126 ret= reg_namedseq(pRExC_state, NULL, flagp, depth);
9128 case 'k': /* Handle \k<NAME> and \k'NAME' */
9131 char ch= RExC_parse[1];
9132 if (ch != '<' && ch != '\'' && ch != '{') {
9134 vFAIL2("Sequence %.2s... not terminated",parse_start);
9136 /* this pretty much dupes the code for (?P=...) in reg(), if
9137 you change this make sure you change that */
9138 char* name_start = (RExC_parse += 2);
9140 SV *sv_dat = reg_scan_name(pRExC_state,
9141 SIZE_ONLY ? REG_RSN_RETURN_NULL : REG_RSN_RETURN_DATA);
9142 ch= (ch == '<') ? '>' : (ch == '{') ? '}' : '\'';
9143 if (RExC_parse == name_start || *RExC_parse != ch)
9144 vFAIL2("Sequence %.3s... not terminated",parse_start);
9147 num = add_data( pRExC_state, 1, "S" );
9148 RExC_rxi->data->data[num]=(void*)sv_dat;
9149 SvREFCNT_inc_simple_void(sv_dat);
9153 ret = reganode(pRExC_state,
9156 : (MORE_ASCII_RESTRICTED)
9158 : (AT_LEAST_UNI_SEMANTICS)
9166 /* override incorrect value set in reganode MJD */
9167 Set_Node_Offset(ret, parse_start+1);
9168 Set_Node_Cur_Length(ret); /* MJD */
9169 nextchar(pRExC_state);
9175 case '1': case '2': case '3': case '4':
9176 case '5': case '6': case '7': case '8': case '9':
9179 bool isg = *RExC_parse == 'g';
9184 if (*RExC_parse == '{') {
9188 if (*RExC_parse == '-') {
9192 if (hasbrace && !isDIGIT(*RExC_parse)) {
9193 if (isrel) RExC_parse--;
9195 goto parse_named_seq;
9197 num = atoi(RExC_parse);
9198 if (isg && num == 0)
9199 vFAIL("Reference to invalid group 0");
9201 num = RExC_npar - num;
9203 vFAIL("Reference to nonexistent or unclosed group");
9205 if (!isg && num > 9 && num >= RExC_npar)
9208 char * const parse_start = RExC_parse - 1; /* MJD */
9209 while (isDIGIT(*RExC_parse))
9211 if (parse_start == RExC_parse - 1)
9212 vFAIL("Unterminated \\g... pattern");
9214 if (*RExC_parse != '}')
9215 vFAIL("Unterminated \\g{...} pattern");
9219 if (num > (I32)RExC_rx->nparens)
9220 vFAIL("Reference to nonexistent group");
9223 ret = reganode(pRExC_state,
9226 : (MORE_ASCII_RESTRICTED)
9228 : (AT_LEAST_UNI_SEMANTICS)
9236 /* override incorrect value set in reganode MJD */
9237 Set_Node_Offset(ret, parse_start+1);
9238 Set_Node_Cur_Length(ret); /* MJD */
9240 nextchar(pRExC_state);
9245 if (RExC_parse >= RExC_end)
9246 FAIL("Trailing \\");
9249 /* Do not generate "unrecognized" warnings here, we fall
9250 back into the quick-grab loop below */
9257 if (RExC_flags & RXf_PMf_EXTENDED) {
9258 if ( reg_skipcomment( pRExC_state ) )
9265 parse_start = RExC_parse - 1;
9270 register STRLEN len;
9275 U8 tmpbuf[UTF8_MAXBYTES_CASE+1], *foldbuf;
9276 regnode * orig_emit;
9279 /* Is this a LATIN LOWER CASE SHARP S in an EXACTFU node? If so,
9280 * it is folded to 'ss' even if not utf8 */
9281 bool is_exactfu_sharp_s;
9284 orig_emit = RExC_emit; /* Save the original output node position in
9285 case we need to output a different node
9287 node_type = ((! FOLD) ? EXACT
9290 : (MORE_ASCII_RESTRICTED)
9292 : (AT_LEAST_UNI_SEMANTICS)
9295 ret = reg_node(pRExC_state, node_type);
9298 /* XXX The node can hold up to 255 bytes, yet this only goes to
9299 * 127. I (khw) do not know why. Keeping it somewhat less than
9300 * 255 allows us to not have to worry about overflow due to
9301 * converting to utf8 and fold expansion, but that value is
9302 * 255-UTF8_MAXBYTES_CASE. join_exact() may join adjacent nodes
9303 * split up by this limit into a single one using the real max of
9304 * 255. Even at 127, this breaks under rare circumstances. If
9305 * folding, we do not want to split a node at a character that is a
9306 * non-final in a multi-char fold, as an input string could just
9307 * happen to want to match across the node boundary. The join
9308 * would solve that problem if the join actually happens. But a
9309 * series of more than two nodes in a row each of 127 would cause
9310 * the first join to succeed to get to 254, but then there wouldn't
9311 * be room for the next one, which could at be one of those split
9312 * multi-char folds. I don't know of any fool-proof solution. One
9313 * could back off to end with only a code point that isn't such a
9314 * non-final, but it is possible for there not to be any in the
9316 for (len = 0, p = RExC_parse - 1;
9317 len < 127 && p < RExC_end;
9320 char * const oldp = p;
9322 if (RExC_flags & RXf_PMf_EXTENDED)
9323 p = regwhite( pRExC_state, p );
9334 /* Literal Escapes Switch
9336 This switch is meant to handle escape sequences that
9337 resolve to a literal character.
9339 Every escape sequence that represents something
9340 else, like an assertion or a char class, is handled
9341 in the switch marked 'Special Escapes' above in this
9342 routine, but also has an entry here as anything that
9343 isn't explicitly mentioned here will be treated as
9344 an unescaped equivalent literal.
9348 /* These are all the special escapes. */
9349 case 'A': /* Start assertion */
9350 case 'b': case 'B': /* Word-boundary assertion*/
9351 case 'C': /* Single char !DANGEROUS! */
9352 case 'd': case 'D': /* digit class */
9353 case 'g': case 'G': /* generic-backref, pos assertion */
9354 case 'h': case 'H': /* HORIZWS */
9355 case 'k': case 'K': /* named backref, keep marker */
9356 case 'N': /* named char sequence */
9357 case 'p': case 'P': /* Unicode property */
9358 case 'R': /* LNBREAK */
9359 case 's': case 'S': /* space class */
9360 case 'v': case 'V': /* VERTWS */
9361 case 'w': case 'W': /* word class */
9362 case 'X': /* eXtended Unicode "combining character sequence" */
9363 case 'z': case 'Z': /* End of line/string assertion */
9367 /* Anything after here is an escape that resolves to a
9368 literal. (Except digits, which may or may not)
9387 ender = ASCII_TO_NATIVE('\033');
9391 ender = ASCII_TO_NATIVE('\007');
9396 STRLEN brace_len = len;
9398 const char* error_msg;
9400 bool valid = grok_bslash_o(p,
9407 RExC_parse = p; /* going to die anyway; point
9408 to exact spot of failure */
9415 if (PL_encoding && ender < 0x100) {
9416 goto recode_encoding;
9425 char* const e = strchr(p, '}');
9429 vFAIL("Missing right brace on \\x{}");
9432 I32 flags = PERL_SCAN_ALLOW_UNDERSCORES
9433 | PERL_SCAN_DISALLOW_PREFIX;
9434 STRLEN numlen = e - p - 1;
9435 ender = grok_hex(p + 1, &numlen, &flags, NULL);
9442 I32 flags = PERL_SCAN_DISALLOW_PREFIX;
9444 ender = grok_hex(p, &numlen, &flags, NULL);
9447 if (PL_encoding && ender < 0x100)
9448 goto recode_encoding;
9452 ender = grok_bslash_c(*p++, UTF, SIZE_ONLY);
9454 case '0': case '1': case '2': case '3':case '4':
9455 case '5': case '6': case '7': case '8':case '9':
9457 (isDIGIT(p[1]) && atoi(p) >= RExC_npar))
9459 I32 flags = PERL_SCAN_SILENT_ILLDIGIT;
9461 ender = grok_oct(p, &numlen, &flags, NULL);
9471 if (PL_encoding && ender < 0x100)
9472 goto recode_encoding;
9475 if (! RExC_override_recoding) {
9476 SV* enc = PL_encoding;
9477 ender = reg_recode((const char)(U8)ender, &enc);
9478 if (!enc && SIZE_ONLY)
9479 ckWARNreg(p, "Invalid escape in the specified encoding");
9485 FAIL("Trailing \\");
9488 if (!SIZE_ONLY&& isALPHA(*p)) {
9489 /* Include any { following the alpha to emphasize
9490 * that it could be part of an escape at some point
9492 int len = (*(p + 1) == '{') ? 2 : 1;
9493 ckWARN3reg(p + len, "Unrecognized escape \\%.*s passed through", len, p);
9495 goto normal_default;
9500 if (UTF8_IS_START(*p) && UTF) {
9502 ender = utf8n_to_uvchr((U8*)p, RExC_end - p,
9503 &numlen, UTF8_ALLOW_DEFAULT);
9509 } /* End of switch on the literal */
9511 is_exactfu_sharp_s = (node_type == EXACTFU
9512 && ender == LATIN_SMALL_LETTER_SHARP_S);
9513 if ( RExC_flags & RXf_PMf_EXTENDED)
9514 p = regwhite( pRExC_state, p );
9515 if ((UTF && FOLD) || is_exactfu_sharp_s) {
9516 /* Prime the casefolded buffer. Locale rules, which apply
9517 * only to code points < 256, aren't known until execution,
9518 * so for them, just output the original character using
9519 * utf8. If we start to fold non-UTF patterns, be sure to
9520 * update join_exact() */
9521 if (LOC && ender < 256) {
9522 if (UNI_IS_INVARIANT(ender)) {
9523 *tmpbuf = (U8) ender;
9526 *tmpbuf = UTF8_TWO_BYTE_HI(ender);
9527 *(tmpbuf + 1) = UTF8_TWO_BYTE_LO(ender);
9531 else if (isASCII(ender)) { /* Note: Here can't also be LOC
9533 ender = toLOWER(ender);
9534 *tmpbuf = (U8) ender;
9537 else if (! MORE_ASCII_RESTRICTED && ! LOC) {
9539 /* Locale and /aa require more selectivity about the
9540 * fold, so are handled below. Otherwise, here, just
9542 ender = toFOLD_uni(ender, tmpbuf, &foldlen);
9545 /* Under locale rules or /aa we are not to mix,
9546 * respectively, ords < 256 or ASCII with non-. So
9547 * reject folds that mix them, using only the
9548 * non-folded code point. So do the fold to a
9549 * temporary, and inspect each character in it. */
9550 U8 trialbuf[UTF8_MAXBYTES_CASE+1];
9552 UV tmpender = toFOLD_uni(ender, trialbuf, &foldlen);
9553 U8* e = s + foldlen;
9554 bool fold_ok = TRUE;
9558 || (LOC && (UTF8_IS_INVARIANT(*s)
9559 || UTF8_IS_DOWNGRADEABLE_START(*s))))
9567 Copy(trialbuf, tmpbuf, foldlen, U8);
9571 uvuni_to_utf8(tmpbuf, ender);
9572 foldlen = UNISKIP(ender);
9576 if (p < RExC_end && ISMULT2(p)) { /* Back off on ?+*. */
9579 else if (UTF || is_exactfu_sharp_s) {
9581 /* Emit all the Unicode characters. */
9583 for (foldbuf = tmpbuf;
9585 foldlen -= numlen) {
9586 ender = utf8_to_uvchr(foldbuf, &numlen);
9588 const STRLEN unilen = reguni(pRExC_state, ender, s);
9591 /* In EBCDIC the numlen
9592 * and unilen can differ. */
9594 if (numlen >= foldlen)
9598 break; /* "Can't happen." */
9602 const STRLEN unilen = reguni(pRExC_state, ender, s);
9611 REGC((char)ender, s++);
9615 if (UTF || is_exactfu_sharp_s) {
9617 /* Emit all the Unicode characters. */
9619 for (foldbuf = tmpbuf;
9621 foldlen -= numlen) {
9622 ender = utf8_to_uvchr(foldbuf, &numlen);
9624 const STRLEN unilen = reguni(pRExC_state, ender, s);
9627 /* In EBCDIC the numlen
9628 * and unilen can differ. */
9630 if (numlen >= foldlen)
9638 const STRLEN unilen = reguni(pRExC_state, ender, s);
9647 REGC((char)ender, s++);
9650 loopdone: /* Jumped to when encounters something that shouldn't be in
9653 Set_Node_Cur_Length(ret); /* MJD */
9654 nextchar(pRExC_state);
9656 /* len is STRLEN which is unsigned, need to copy to signed */
9659 vFAIL("Internal disaster");
9663 if (len == 1 && UNI_IS_INVARIANT(ender))
9667 RExC_size += STR_SZ(len);
9670 RExC_emit += STR_SZ(len);
9678 /* Jumped to when an unrecognized character set is encountered */
9680 Perl_croak(aTHX_ "panic: Unknown regex character set encoding: %u", get_regex_charset(RExC_flags));
9685 S_regwhite( RExC_state_t *pRExC_state, char *p )
9687 const char *e = RExC_end;
9689 PERL_ARGS_ASSERT_REGWHITE;
9694 else if (*p == '#') {
9703 RExC_seen |= REG_SEEN_RUN_ON_COMMENT;
9711 /* Parse POSIX character classes: [[:foo:]], [[=foo=]], [[.foo.]].
9712 Character classes ([:foo:]) can also be negated ([:^foo:]).
9713 Returns a named class id (ANYOF_XXX) if successful, -1 otherwise.
9714 Equivalence classes ([=foo=]) and composites ([.foo.]) are parsed,
9715 but trigger failures because they are currently unimplemented. */
9717 #define POSIXCC_DONE(c) ((c) == ':')
9718 #define POSIXCC_NOTYET(c) ((c) == '=' || (c) == '.')
9719 #define POSIXCC(c) (POSIXCC_DONE(c) || POSIXCC_NOTYET(c))
9722 S_regpposixcc(pTHX_ RExC_state_t *pRExC_state, I32 value)
9725 I32 namedclass = OOB_NAMEDCLASS;
9727 PERL_ARGS_ASSERT_REGPPOSIXCC;
9729 if (value == '[' && RExC_parse + 1 < RExC_end &&
9730 /* I smell either [: or [= or [. -- POSIX has been here, right? */
9731 POSIXCC(UCHARAT(RExC_parse))) {
9732 const char c = UCHARAT(RExC_parse);
9733 char* const s = RExC_parse++;
9735 while (RExC_parse < RExC_end && UCHARAT(RExC_parse) != c)
9737 if (RExC_parse == RExC_end)
9738 /* Grandfather lone [:, [=, [. */
9741 const char* const t = RExC_parse++; /* skip over the c */
9744 if (UCHARAT(RExC_parse) == ']') {
9745 const char *posixcc = s + 1;
9746 RExC_parse++; /* skip over the ending ] */
9749 const I32 complement = *posixcc == '^' ? *posixcc++ : 0;
9750 const I32 skip = t - posixcc;
9752 /* Initially switch on the length of the name. */
9755 if (memEQ(posixcc, "word", 4)) /* this is not POSIX, this is the Perl \w */
9756 namedclass = complement ? ANYOF_NALNUM : ANYOF_ALNUM;
9759 /* Names all of length 5. */
9760 /* alnum alpha ascii blank cntrl digit graph lower
9761 print punct space upper */
9762 /* Offset 4 gives the best switch position. */
9763 switch (posixcc[4]) {
9765 if (memEQ(posixcc, "alph", 4)) /* alpha */
9766 namedclass = complement ? ANYOF_NALPHA : ANYOF_ALPHA;
9769 if (memEQ(posixcc, "spac", 4)) /* space */
9770 namedclass = complement ? ANYOF_NPSXSPC : ANYOF_PSXSPC;
9773 if (memEQ(posixcc, "grap", 4)) /* graph */
9774 namedclass = complement ? ANYOF_NGRAPH : ANYOF_GRAPH;
9777 if (memEQ(posixcc, "asci", 4)) /* ascii */
9778 namedclass = complement ? ANYOF_NASCII : ANYOF_ASCII;
9781 if (memEQ(posixcc, "blan", 4)) /* blank */
9782 namedclass = complement ? ANYOF_NBLANK : ANYOF_BLANK;
9785 if (memEQ(posixcc, "cntr", 4)) /* cntrl */
9786 namedclass = complement ? ANYOF_NCNTRL : ANYOF_CNTRL;
9789 if (memEQ(posixcc, "alnu", 4)) /* alnum */
9790 namedclass = complement ? ANYOF_NALNUMC : ANYOF_ALNUMC;
9793 if (memEQ(posixcc, "lowe", 4)) /* lower */
9794 namedclass = complement ? ANYOF_NLOWER : ANYOF_LOWER;
9795 else if (memEQ(posixcc, "uppe", 4)) /* upper */
9796 namedclass = complement ? ANYOF_NUPPER : ANYOF_UPPER;
9799 if (memEQ(posixcc, "digi", 4)) /* digit */
9800 namedclass = complement ? ANYOF_NDIGIT : ANYOF_DIGIT;
9801 else if (memEQ(posixcc, "prin", 4)) /* print */
9802 namedclass = complement ? ANYOF_NPRINT : ANYOF_PRINT;
9803 else if (memEQ(posixcc, "punc", 4)) /* punct */
9804 namedclass = complement ? ANYOF_NPUNCT : ANYOF_PUNCT;
9809 if (memEQ(posixcc, "xdigit", 6))
9810 namedclass = complement ? ANYOF_NXDIGIT : ANYOF_XDIGIT;
9814 if (namedclass == OOB_NAMEDCLASS)
9815 Simple_vFAIL3("POSIX class [:%.*s:] unknown",
9817 assert (posixcc[skip] == ':');
9818 assert (posixcc[skip+1] == ']');
9819 } else if (!SIZE_ONLY) {
9820 /* [[=foo=]] and [[.foo.]] are still future. */
9822 /* adjust RExC_parse so the warning shows after
9824 while (UCHARAT(RExC_parse) && UCHARAT(RExC_parse) != ']')
9826 Simple_vFAIL3("POSIX syntax [%c %c] is reserved for future extensions", c, c);
9829 /* Maternal grandfather:
9830 * "[:" ending in ":" but not in ":]" */
9840 S_checkposixcc(pTHX_ RExC_state_t *pRExC_state)
9844 PERL_ARGS_ASSERT_CHECKPOSIXCC;
9846 if (POSIXCC(UCHARAT(RExC_parse))) {
9847 const char *s = RExC_parse;
9848 const char c = *s++;
9852 if (*s && c == *s && s[1] == ']') {
9854 "POSIX syntax [%c %c] belongs inside character classes",
9857 /* [[=foo=]] and [[.foo.]] are still future. */
9858 if (POSIXCC_NOTYET(c)) {
9859 /* adjust RExC_parse so the error shows after
9861 while (UCHARAT(RExC_parse) && UCHARAT(RExC_parse++) != ']')
9863 Simple_vFAIL3("POSIX syntax [%c %c] is reserved for future extensions", c, c);
9869 /* No locale test, and always Unicode semantics, no ignore-case differences */
9870 #define _C_C_T_NOLOC_(NAME,TEST,WORD) \
9872 for (value = 0; value < 256; value++) \
9874 stored += set_regclass_bit(pRExC_state, ret, (U8) value, &l1_fold_invlist, &unicode_alternate); \
9878 case ANYOF_N##NAME: \
9879 for (value = 0; value < 256; value++) \
9881 stored += set_regclass_bit(pRExC_state, ret, (U8) value, &l1_fold_invlist, &unicode_alternate); \
9886 /* Like the above, but there are differences if we are in uni-8-bit or not, so
9887 * there are two tests passed in, to use depending on that. There aren't any
9888 * cases where the label is different from the name, so no need for that
9890 * Sets 'what' to WORD which is the property name for non-bitmap code points;
9891 * But, uses FOLD_WORD instead if /i has been selected, to allow a different
9893 #define _C_C_T_(NAME, TEST_8, TEST_7, WORD, FOLD_WORD) \
9895 if (LOC) ANYOF_CLASS_SET(ret, ANYOF_##NAME); \
9896 else if (UNI_SEMANTICS) { \
9897 for (value = 0; value < 256; value++) { \
9898 if (TEST_8(value)) stored += \
9899 set_regclass_bit(pRExC_state, ret, (U8) value, &l1_fold_invlist, &unicode_alternate); \
9903 for (value = 0; value < 128; value++) { \
9904 if (TEST_7(UNI_TO_NATIVE(value))) stored += \
9905 set_regclass_bit(pRExC_state, ret, \
9906 (U8) UNI_TO_NATIVE(value), &l1_fold_invlist, &unicode_alternate); \
9917 case ANYOF_N##NAME: \
9918 if (LOC) ANYOF_CLASS_SET(ret, ANYOF_N##NAME); \
9919 else if (UNI_SEMANTICS) { \
9920 for (value = 0; value < 256; value++) { \
9921 if (! TEST_8(value)) stored += \
9922 set_regclass_bit(pRExC_state, ret, (U8) value, &l1_fold_invlist, &unicode_alternate); \
9926 for (value = 0; value < 128; value++) { \
9927 if (! TEST_7(UNI_TO_NATIVE(value))) stored += set_regclass_bit( \
9928 pRExC_state, ret, (U8) UNI_TO_NATIVE(value), &l1_fold_invlist, &unicode_alternate); \
9930 if (AT_LEAST_ASCII_RESTRICTED) { \
9931 for (value = 128; value < 256; value++) { \
9932 stored += set_regclass_bit( \
9933 pRExC_state, ret, (U8) UNI_TO_NATIVE(value), &l1_fold_invlist, &unicode_alternate); \
9935 ANYOF_FLAGS(ret) |= ANYOF_UNICODE_ALL; \
9938 /* For a non-ut8 target string with DEPENDS semantics, all above \
9939 * ASCII Latin1 code points match the complement of any of the \
9940 * classes. But in utf8, they have their Unicode semantics, so \
9941 * can't just set them in the bitmap, or else regexec.c will think \
9942 * they matched when they shouldn't. */ \
9943 ANYOF_FLAGS(ret) |= ANYOF_NON_UTF8_LATIN1_ALL; \
9956 S_set_regclass_bit_fold(pTHX_ RExC_state_t *pRExC_state, regnode* node, const U8 value, SV** invlist_ptr, AV** alternate_ptr)
9959 /* Handle the setting of folds in the bitmap for non-locale ANYOF nodes.
9960 * Locale folding is done at run-time, so this function should not be
9961 * called for nodes that are for locales.
9963 * This function sets the bit corresponding to the fold of the input
9964 * 'value', if not already set. The fold of 'f' is 'F', and the fold of
9967 * It also knows about the characters that are in the bitmap that have
9968 * folds that are matchable only outside it, and sets the appropriate lists
9971 * It returns the number of bits that actually changed from 0 to 1 */
9976 PERL_ARGS_ASSERT_SET_REGCLASS_BIT_FOLD;
9978 fold = (AT_LEAST_UNI_SEMANTICS) ? PL_fold_latin1[value]
9981 /* It assumes the bit for 'value' has already been set */
9982 if (fold != value && ! ANYOF_BITMAP_TEST(node, fold)) {
9983 ANYOF_BITMAP_SET(node, fold);
9986 if (_HAS_NONLATIN1_FOLD_CLOSURE_ONLY_FOR_USE_BY_REGCOMP_DOT_C_AND_REGEXEC_DOT_C(value) && (! isASCII(value) || ! MORE_ASCII_RESTRICTED)) {
9987 /* Certain Latin1 characters have matches outside the bitmap. To get
9988 * here, 'value' is one of those characters. None of these matches is
9989 * valid for ASCII characters under /aa, which have been excluded by
9990 * the 'if' above. The matches fall into three categories:
9991 * 1) They are singly folded-to or -from an above 255 character, as
9992 * LATIN SMALL LETTER Y WITH DIAERESIS and LATIN CAPITAL LETTER Y
9994 * 2) They are part of a multi-char fold with another character in the
9995 * bitmap, only LATIN SMALL LETTER SHARP S => "ss" fits that bill;
9996 * 3) They are part of a multi-char fold with a character not in the
9997 * bitmap, such as various ligatures.
9998 * We aren't dealing fully with multi-char folds, except we do deal
9999 * with the pattern containing a character that has a multi-char fold
10000 * (not so much the inverse).
10001 * For types 1) and 3), the matches only happen when the target string
10002 * is utf8; that's not true for 2), and we set a flag for it.
10004 * The code below adds to the passed in inversion list the single fold
10005 * closures for 'value'. The values are hard-coded here so that an
10006 * innocent-looking character class, like /[ks]/i won't have to go out
10007 * to disk to find the possible matches. XXX It would be better to
10008 * generate these via regen, in case a new version of the Unicode
10009 * standard adds new mappings, though that is not really likely. */
10014 *invlist_ptr = add_cp_to_invlist(*invlist_ptr, 0x212A);
10018 /* LATIN SMALL LETTER LONG S */
10019 *invlist_ptr = add_cp_to_invlist(*invlist_ptr, 0x017F);
10022 *invlist_ptr = add_cp_to_invlist(*invlist_ptr,
10023 GREEK_SMALL_LETTER_MU);
10024 *invlist_ptr = add_cp_to_invlist(*invlist_ptr,
10025 GREEK_CAPITAL_LETTER_MU);
10027 case LATIN_CAPITAL_LETTER_A_WITH_RING_ABOVE:
10028 case LATIN_SMALL_LETTER_A_WITH_RING_ABOVE:
10029 /* ANGSTROM SIGN */
10030 *invlist_ptr = add_cp_to_invlist(*invlist_ptr, 0x212B);
10031 if (DEPENDS_SEMANTICS) { /* See DEPENDS comment below */
10032 *invlist_ptr = add_cp_to_invlist(*invlist_ptr,
10033 PL_fold_latin1[value]);
10036 case LATIN_SMALL_LETTER_Y_WITH_DIAERESIS:
10037 *invlist_ptr = add_cp_to_invlist(*invlist_ptr,
10038 LATIN_CAPITAL_LETTER_Y_WITH_DIAERESIS);
10040 case LATIN_SMALL_LETTER_SHARP_S:
10041 *invlist_ptr = add_cp_to_invlist(*invlist_ptr,
10042 LATIN_CAPITAL_LETTER_SHARP_S);
10044 /* Under /a, /d, and /u, this can match the two chars "ss" */
10045 if (! MORE_ASCII_RESTRICTED) {
10046 add_alternate(alternate_ptr, (U8 *) "ss", 2);
10048 /* And under /u or /a, it can match even if the target is
10050 if (AT_LEAST_UNI_SEMANTICS) {
10051 ANYOF_FLAGS(node) |= ANYOF_NONBITMAP_NON_UTF8;
10055 case 'F': case 'f':
10056 case 'I': case 'i':
10057 case 'L': case 'l':
10058 case 'T': case 't':
10059 case 'A': case 'a':
10060 case 'H': case 'h':
10061 case 'J': case 'j':
10062 case 'N': case 'n':
10063 case 'W': case 'w':
10064 case 'Y': case 'y':
10065 /* These all are targets of multi-character folds from code
10066 * points that require UTF8 to express, so they can't match
10067 * unless the target string is in UTF-8, so no action here is
10068 * necessary, as regexec.c properly handles the general case
10069 * for UTF-8 matching */
10072 /* Use deprecated warning to increase the chances of this
10074 ckWARN2regdep(RExC_parse, "Perl folding rules are not up-to-date for 0x%x; please use the perlbug utility to report;", value);
10078 else if (DEPENDS_SEMANTICS
10079 && ! isASCII(value)
10080 && PL_fold_latin1[value] != value)
10082 /* Under DEPENDS rules, non-ASCII Latin1 characters match their
10083 * folds only when the target string is in UTF-8. We add the fold
10084 * here to the list of things to match outside the bitmap, which
10085 * won't be looked at unless it is UTF8 (or else if something else
10086 * says to look even if not utf8, but those things better not happen
10087 * under DEPENDS semantics. */
10088 *invlist_ptr = add_cp_to_invlist(*invlist_ptr, PL_fold_latin1[value]);
10095 PERL_STATIC_INLINE U8
10096 S_set_regclass_bit(pTHX_ RExC_state_t *pRExC_state, regnode* node, const U8 value, SV** invlist_ptr, AV** alternate_ptr)
10098 /* This inline function sets a bit in the bitmap if not already set, and if
10099 * appropriate, its fold, returning the number of bits that actually
10100 * changed from 0 to 1 */
10104 PERL_ARGS_ASSERT_SET_REGCLASS_BIT;
10106 if (ANYOF_BITMAP_TEST(node, value)) { /* Already set */
10110 ANYOF_BITMAP_SET(node, value);
10113 if (FOLD && ! LOC) { /* Locale folds aren't known until runtime */
10114 stored += set_regclass_bit_fold(pRExC_state, node, value, invlist_ptr, alternate_ptr);
10121 S_add_alternate(pTHX_ AV** alternate_ptr, U8* string, STRLEN len)
10123 /* Adds input 'string' with length 'len' to the ANYOF node's unicode
10124 * alternate list, pointed to by 'alternate_ptr'. This is an array of
10125 * the multi-character folds of characters in the node */
10128 PERL_ARGS_ASSERT_ADD_ALTERNATE;
10130 if (! *alternate_ptr) {
10131 *alternate_ptr = newAV();
10133 sv = newSVpvn_utf8((char*)string, len, TRUE);
10134 av_push(*alternate_ptr, sv);
10139 parse a class specification and produce either an ANYOF node that
10140 matches the pattern or perhaps will be optimized into an EXACTish node
10141 instead. The node contains a bit map for the first 256 characters, with the
10142 corresponding bit set if that character is in the list. For characters
10143 above 255, a range list is used */
10146 S_regclass(pTHX_ RExC_state_t *pRExC_state, U32 depth)
10149 register UV nextvalue;
10150 register IV prevvalue = OOB_UNICODE;
10151 register IV range = 0;
10152 UV value = 0; /* XXX:dmq: needs to be referenceable (unfortunately) */
10153 register regnode *ret;
10156 char *rangebegin = NULL;
10157 bool need_class = 0;
10158 bool allow_full_fold = TRUE; /* Assume wants multi-char folding */
10160 STRLEN initial_listsv_len = 0; /* Kind of a kludge to see if it is more
10161 than just initialized. */
10162 SV* properties = NULL; /* Code points that match \p{} \P{} */
10163 UV element_count = 0; /* Number of distinct elements in the class.
10164 Optimizations may be possible if this is tiny */
10167 /* Unicode properties are stored in a swash; this holds the current one
10168 * being parsed. If this swash is the only above-latin1 component of the
10169 * character class, an optimization is to pass it directly on to the
10170 * execution engine. Otherwise, it is set to NULL to indicate that there
10171 * are other things in the class that have to be dealt with at execution
10173 SV* swash = NULL; /* Code points that match \p{} \P{} */
10175 /* Set if a component of this character class is user-defined; just passed
10176 * on to the engine */
10177 UV has_user_defined_property = 0;
10179 /* code points this node matches that can't be stored in the bitmap */
10180 SV* nonbitmap = NULL;
10182 /* The items that are to match that aren't stored in the bitmap, but are a
10183 * result of things that are stored there. This is the fold closure of
10184 * such a character, either because it has DEPENDS semantics and shouldn't
10185 * be matched unless the target string is utf8, or is a code point that is
10186 * too large for the bit map, as for example, the fold of the MICRO SIGN is
10187 * above 255. This all is solely for performance reasons. By having this
10188 * code know the outside-the-bitmap folds that the bitmapped characters are
10189 * involved with, we don't have to go out to disk to find the list of
10190 * matches, unless the character class includes code points that aren't
10191 * storable in the bit map. That means that a character class with an 's'
10192 * in it, for example, doesn't need to go out to disk to find everything
10193 * that matches. A 2nd list is used so that the 'nonbitmap' list is kept
10194 * empty unless there is something whose fold we don't know about, and will
10195 * have to go out to the disk to find. */
10196 SV* l1_fold_invlist = NULL;
10198 /* List of multi-character folds that are matched by this node */
10199 AV* unicode_alternate = NULL;
10201 UV literal_endpoint = 0;
10203 UV stored = 0; /* how many chars stored in the bitmap */
10205 regnode * const orig_emit = RExC_emit; /* Save the original RExC_emit in
10206 case we need to change the emitted regop to an EXACT. */
10207 const char * orig_parse = RExC_parse;
10208 GET_RE_DEBUG_FLAGS_DECL;
10210 PERL_ARGS_ASSERT_REGCLASS;
10212 PERL_UNUSED_ARG(depth);
10215 DEBUG_PARSE("clas");
10217 /* Assume we are going to generate an ANYOF node. */
10218 ret = reganode(pRExC_state, ANYOF, 0);
10222 ANYOF_FLAGS(ret) = 0;
10225 if (UCHARAT(RExC_parse) == '^') { /* Complement of range. */
10229 ANYOF_FLAGS(ret) |= ANYOF_INVERT;
10231 /* We have decided to not allow multi-char folds in inverted character
10232 * classes, due to the confusion that can happen, especially with
10233 * classes that are designed for a non-Unicode world: You have the
10234 * peculiar case that:
10235 "s s" =~ /^[^\xDF]+$/i => Y
10236 "ss" =~ /^[^\xDF]+$/i => N
10238 * See [perl #89750] */
10239 allow_full_fold = FALSE;
10243 RExC_size += ANYOF_SKIP;
10244 listsv = &PL_sv_undef; /* For code scanners: listsv always non-NULL. */
10247 RExC_emit += ANYOF_SKIP;
10249 ANYOF_FLAGS(ret) |= ANYOF_LOCALE;
10251 ANYOF_BITMAP_ZERO(ret);
10252 listsv = newSVpvs("# comment\n");
10253 initial_listsv_len = SvCUR(listsv);
10256 nextvalue = RExC_parse < RExC_end ? UCHARAT(RExC_parse) : 0;
10258 if (!SIZE_ONLY && POSIXCC(nextvalue))
10259 checkposixcc(pRExC_state);
10261 /* allow 1st char to be ] (allowing it to be - is dealt with later) */
10262 if (UCHARAT(RExC_parse) == ']')
10263 goto charclassloop;
10266 while (RExC_parse < RExC_end && UCHARAT(RExC_parse) != ']') {
10270 namedclass = OOB_NAMEDCLASS; /* initialize as illegal */
10273 rangebegin = RExC_parse;
10277 value = utf8n_to_uvchr((U8*)RExC_parse,
10278 RExC_end - RExC_parse,
10279 &numlen, UTF8_ALLOW_DEFAULT);
10280 RExC_parse += numlen;
10283 value = UCHARAT(RExC_parse++);
10285 nextvalue = RExC_parse < RExC_end ? UCHARAT(RExC_parse) : 0;
10286 if (value == '[' && POSIXCC(nextvalue))
10287 namedclass = regpposixcc(pRExC_state, value);
10288 else if (value == '\\') {
10290 value = utf8n_to_uvchr((U8*)RExC_parse,
10291 RExC_end - RExC_parse,
10292 &numlen, UTF8_ALLOW_DEFAULT);
10293 RExC_parse += numlen;
10296 value = UCHARAT(RExC_parse++);
10297 /* Some compilers cannot handle switching on 64-bit integer
10298 * values, therefore value cannot be an UV. Yes, this will
10299 * be a problem later if we want switch on Unicode.
10300 * A similar issue a little bit later when switching on
10301 * namedclass. --jhi */
10302 switch ((I32)value) {
10303 case 'w': namedclass = ANYOF_ALNUM; break;
10304 case 'W': namedclass = ANYOF_NALNUM; break;
10305 case 's': namedclass = ANYOF_SPACE; break;
10306 case 'S': namedclass = ANYOF_NSPACE; break;
10307 case 'd': namedclass = ANYOF_DIGIT; break;
10308 case 'D': namedclass = ANYOF_NDIGIT; break;
10309 case 'v': namedclass = ANYOF_VERTWS; break;
10310 case 'V': namedclass = ANYOF_NVERTWS; break;
10311 case 'h': namedclass = ANYOF_HORIZWS; break;
10312 case 'H': namedclass = ANYOF_NHORIZWS; break;
10313 case 'N': /* Handle \N{NAME} in class */
10315 /* We only pay attention to the first char of
10316 multichar strings being returned. I kinda wonder
10317 if this makes sense as it does change the behaviour
10318 from earlier versions, OTOH that behaviour was broken
10320 UV v; /* value is register so we cant & it /grrr */
10321 if (reg_namedseq(pRExC_state, &v, NULL, depth)) {
10331 if (RExC_parse >= RExC_end)
10332 vFAIL2("Empty \\%c{}", (U8)value);
10333 if (*RExC_parse == '{') {
10334 const U8 c = (U8)value;
10335 e = strchr(RExC_parse++, '}');
10337 vFAIL2("Missing right brace on \\%c{}", c);
10338 while (isSPACE(UCHARAT(RExC_parse)))
10340 if (e == RExC_parse)
10341 vFAIL2("Empty \\%c{}", c);
10342 n = e - RExC_parse;
10343 while (isSPACE(UCHARAT(RExC_parse + n - 1)))
10354 if (UCHARAT(RExC_parse) == '^') {
10357 value = value == 'p' ? 'P' : 'p'; /* toggle */
10358 while (isSPACE(UCHARAT(RExC_parse))) {
10363 /* Try to get the definition of the property into
10364 * <invlist>. If /i is in effect, the effective property
10365 * will have its name be <__NAME_i>. The design is
10366 * discussed in commit
10367 * 2f833f5208e26b208886e51e09e2c072b5eabb46 */
10368 Newx(name, n + sizeof("_i__\n"), char);
10370 sprintf(name, "%s%.*s%s\n",
10371 (FOLD) ? "__" : "",
10377 /* Look up the property name, and get its swash and
10378 * inversion list, if the property is found */
10380 SvREFCNT_dec(swash);
10382 swash = _core_swash_init("utf8", name, &PL_sv_undef,
10385 TRUE, /* this routine will handle
10386 undefined properties */
10387 NULL, FALSE /* No inversion list */
10391 || ! SvTYPE(SvRV(swash)) == SVt_PVHV
10393 hv_fetchs(MUTABLE_HV(SvRV(swash)),
10395 || ! (invlist = *invlistsvp))
10398 SvREFCNT_dec(swash);
10402 /* Here didn't find it. It could be a user-defined
10403 * property that will be available at run-time. Add it
10404 * to the list to look up then */
10405 Perl_sv_catpvf(aTHX_ listsv, "%cutf8::%s\n",
10406 (value == 'p' ? '+' : '!'),
10408 has_user_defined_property = 1;
10410 /* We don't know yet, so have to assume that the
10411 * property could match something in the Latin1 range,
10412 * hence something that isn't utf8 */
10413 ANYOF_FLAGS(ret) |= ANYOF_NONBITMAP_NON_UTF8;
10417 /* Here, did get the swash and its inversion list. If
10418 * the swash is from a user-defined property, then this
10419 * whole character class should be regarded as such */
10420 SV** user_defined_svp =
10421 hv_fetchs(MUTABLE_HV(SvRV(swash)),
10422 "USER_DEFINED", FALSE);
10423 if (user_defined_svp) {
10424 has_user_defined_property
10425 |= SvUV(*user_defined_svp);
10428 /* Invert if asking for the complement */
10429 if (value == 'P') {
10431 /* Add to any existing list */
10432 if (! properties) {
10433 properties = invlist_clone(invlist);
10434 _invlist_invert(properties);
10437 invlist = invlist_clone(invlist);
10438 _invlist_invert(invlist);
10439 _invlist_union(properties, invlist, &properties);
10440 SvREFCNT_dec(invlist);
10443 /* The swash can't be used as-is, because we've
10444 * inverted things; delay removing it to here after
10445 * have copied its invlist above */
10446 SvREFCNT_dec(swash);
10450 if (! properties) {
10451 properties = invlist_clone(invlist);
10454 _invlist_union(properties, invlist, &properties);
10460 RExC_parse = e + 1;
10461 namedclass = ANYOF_MAX; /* no official name, but it's named */
10463 /* \p means they want Unicode semantics */
10464 RExC_uni_semantics = 1;
10467 case 'n': value = '\n'; break;
10468 case 'r': value = '\r'; break;
10469 case 't': value = '\t'; break;
10470 case 'f': value = '\f'; break;
10471 case 'b': value = '\b'; break;
10472 case 'e': value = ASCII_TO_NATIVE('\033');break;
10473 case 'a': value = ASCII_TO_NATIVE('\007');break;
10475 RExC_parse--; /* function expects to be pointed at the 'o' */
10477 const char* error_msg;
10478 bool valid = grok_bslash_o(RExC_parse,
10483 RExC_parse += numlen;
10488 if (PL_encoding && value < 0x100) {
10489 goto recode_encoding;
10493 if (*RExC_parse == '{') {
10494 I32 flags = PERL_SCAN_ALLOW_UNDERSCORES
10495 | PERL_SCAN_DISALLOW_PREFIX;
10496 char * const e = strchr(RExC_parse++, '}');
10498 vFAIL("Missing right brace on \\x{}");
10500 numlen = e - RExC_parse;
10501 value = grok_hex(RExC_parse, &numlen, &flags, NULL);
10502 RExC_parse = e + 1;
10505 I32 flags = PERL_SCAN_DISALLOW_PREFIX;
10507 value = grok_hex(RExC_parse, &numlen, &flags, NULL);
10508 RExC_parse += numlen;
10510 if (PL_encoding && value < 0x100)
10511 goto recode_encoding;
10514 value = grok_bslash_c(*RExC_parse++, UTF, SIZE_ONLY);
10516 case '0': case '1': case '2': case '3': case '4':
10517 case '5': case '6': case '7':
10519 /* Take 1-3 octal digits */
10520 I32 flags = PERL_SCAN_SILENT_ILLDIGIT;
10522 value = grok_oct(--RExC_parse, &numlen, &flags, NULL);
10523 RExC_parse += numlen;
10524 if (PL_encoding && value < 0x100)
10525 goto recode_encoding;
10529 if (! RExC_override_recoding) {
10530 SV* enc = PL_encoding;
10531 value = reg_recode((const char)(U8)value, &enc);
10532 if (!enc && SIZE_ONLY)
10533 ckWARNreg(RExC_parse,
10534 "Invalid escape in the specified encoding");
10538 /* Allow \_ to not give an error */
10539 if (!SIZE_ONLY && isALNUM(value) && value != '_') {
10540 ckWARN2reg(RExC_parse,
10541 "Unrecognized escape \\%c in character class passed through",
10546 } /* end of \blah */
10549 literal_endpoint++;
10552 if (namedclass > OOB_NAMEDCLASS) { /* this is a named class \blah */
10554 /* What matches in a locale is not known until runtime, so need to
10555 * (one time per class) allocate extra space to pass to regexec.
10556 * The space will contain a bit for each named class that is to be
10557 * matched against. This isn't needed for \p{} and pseudo-classes,
10558 * as they are not affected by locale, and hence are dealt with
10560 if (LOC && namedclass < ANYOF_MAX && ! need_class) {
10563 RExC_size += ANYOF_CLASS_SKIP - ANYOF_SKIP;
10566 RExC_emit += ANYOF_CLASS_SKIP - ANYOF_SKIP;
10567 ANYOF_CLASS_ZERO(ret);
10569 ANYOF_FLAGS(ret) |= ANYOF_CLASS;
10572 /* a bad range like a-\d, a-[:digit:]. The '-' is taken as a
10573 * literal, as is the character that began the false range, i.e.
10574 * the 'a' in the examples */
10578 RExC_parse >= rangebegin ?
10579 RExC_parse - rangebegin : 0;
10580 ckWARN4reg(RExC_parse,
10581 "False [] range \"%*.*s\"",
10585 set_regclass_bit(pRExC_state, ret, '-', &l1_fold_invlist, &unicode_alternate);
10586 if (prevvalue < 256) {
10588 set_regclass_bit(pRExC_state, ret, (U8) prevvalue, &l1_fold_invlist, &unicode_alternate);
10591 nonbitmap = add_cp_to_invlist(nonbitmap, prevvalue);
10595 range = 0; /* this was not a true range */
10599 const char *what = NULL;
10602 /* Possible truncation here but in some 64-bit environments
10603 * the compiler gets heartburn about switch on 64-bit values.
10604 * A similar issue a little earlier when switching on value.
10606 switch ((I32)namedclass) {
10608 case _C_C_T_(ALNUMC, isALNUMC_L1, isALNUMC, "XPosixAlnum", "XPosixAlnum");
10609 case _C_C_T_(ALPHA, isALPHA_L1, isALPHA, "XPosixAlpha", "XPosixAlpha");
10610 case _C_C_T_(BLANK, isBLANK_L1, isBLANK, "XPosixBlank", "XPosixBlank");
10611 case _C_C_T_(CNTRL, isCNTRL_L1, isCNTRL, "XPosixCntrl", "XPosixCntrl");
10612 case _C_C_T_(GRAPH, isGRAPH_L1, isGRAPH, "XPosixGraph", "XPosixGraph");
10613 case _C_C_T_(LOWER, isLOWER_L1, isLOWER, "XPosixLower", "__XPosixLower_i");
10614 case _C_C_T_(PRINT, isPRINT_L1, isPRINT, "XPosixPrint", "XPosixPrint");
10615 case _C_C_T_(PSXSPC, isPSXSPC_L1, isPSXSPC, "XPosixSpace", "XPosixSpace");
10616 case _C_C_T_(PUNCT, isPUNCT_L1, isPUNCT, "XPosixPunct", "XPosixPunct");
10617 case _C_C_T_(UPPER, isUPPER_L1, isUPPER, "XPosixUpper", "__XPosixUpper_i");
10618 /* \s, \w match all unicode if utf8. */
10619 case _C_C_T_(SPACE, isSPACE_L1, isSPACE, "SpacePerl", "SpacePerl");
10620 case _C_C_T_(ALNUM, isWORDCHAR_L1, isALNUM, "Word", "Word");
10621 case _C_C_T_(XDIGIT, isXDIGIT_L1, isXDIGIT, "XPosixXDigit", "XPosixXDigit");
10622 case _C_C_T_NOLOC_(VERTWS, is_VERTWS_latin1(&value), "VertSpace");
10623 case _C_C_T_NOLOC_(HORIZWS, is_HORIZWS_latin1(&value), "HorizSpace");
10626 ANYOF_CLASS_SET(ret, ANYOF_ASCII);
10628 for (value = 0; value < 128; value++)
10630 set_regclass_bit(pRExC_state, ret, (U8) ASCII_TO_NATIVE(value), &l1_fold_invlist, &unicode_alternate);
10633 what = NULL; /* Doesn't match outside ascii, so
10634 don't want to add +utf8:: */
10638 ANYOF_CLASS_SET(ret, ANYOF_NASCII);
10640 for (value = 128; value < 256; value++)
10642 set_regclass_bit(pRExC_state, ret, (U8) ASCII_TO_NATIVE(value), &l1_fold_invlist, &unicode_alternate);
10644 ANYOF_FLAGS(ret) |= ANYOF_UNICODE_ALL;
10650 ANYOF_CLASS_SET(ret, ANYOF_DIGIT);
10652 /* consecutive digits assumed */
10653 for (value = '0'; value <= '9'; value++)
10655 set_regclass_bit(pRExC_state, ret, (U8) value, &l1_fold_invlist, &unicode_alternate);
10662 ANYOF_CLASS_SET(ret, ANYOF_NDIGIT);
10664 /* consecutive digits assumed */
10665 for (value = 0; value < '0'; value++)
10667 set_regclass_bit(pRExC_state, ret, (U8) value, &l1_fold_invlist, &unicode_alternate);
10668 for (value = '9' + 1; value < 256; value++)
10670 set_regclass_bit(pRExC_state, ret, (U8) value, &l1_fold_invlist, &unicode_alternate);
10674 if (AT_LEAST_ASCII_RESTRICTED ) {
10675 ANYOF_FLAGS(ret) |= ANYOF_UNICODE_ALL;
10679 /* this is to handle \p and \P */
10682 vFAIL("Invalid [::] class");
10685 if (what && ! (AT_LEAST_ASCII_RESTRICTED)) {
10686 /* Strings such as "+utf8::isWord\n" */
10687 Perl_sv_catpvf(aTHX_ listsv, "%cutf8::%s\n", yesno, what);
10692 } /* end of namedclass \blah */
10695 if (prevvalue > (IV)value) /* b-a */ {
10696 const int w = RExC_parse - rangebegin;
10697 Simple_vFAIL4("Invalid [] range \"%*.*s\"", w, w, rangebegin);
10698 range = 0; /* not a valid range */
10702 prevvalue = value; /* save the beginning of the range */
10703 if (RExC_parse+1 < RExC_end
10704 && *RExC_parse == '-'
10705 && RExC_parse[1] != ']')
10709 /* a bad range like \w-, [:word:]- ? */
10710 if (namedclass > OOB_NAMEDCLASS) {
10711 if (ckWARN(WARN_REGEXP)) {
10713 RExC_parse >= rangebegin ?
10714 RExC_parse - rangebegin : 0;
10716 "False [] range \"%*.*s\"",
10721 set_regclass_bit(pRExC_state, ret, '-', &l1_fold_invlist, &unicode_alternate);
10723 range = 1; /* yeah, it's a range! */
10724 continue; /* but do it the next time */
10728 /* non-Latin1 code point implies unicode semantics. Must be set in
10729 * pass1 so is there for the whole of pass 2 */
10731 RExC_uni_semantics = 1;
10734 /* now is the next time */
10736 if (prevvalue < 256) {
10737 const IV ceilvalue = value < 256 ? value : 255;
10740 /* In EBCDIC [\x89-\x91] should include
10741 * the \x8e but [i-j] should not. */
10742 if (literal_endpoint == 2 &&
10743 ((isLOWER(prevvalue) && isLOWER(ceilvalue)) ||
10744 (isUPPER(prevvalue) && isUPPER(ceilvalue))))
10746 if (isLOWER(prevvalue)) {
10747 for (i = prevvalue; i <= ceilvalue; i++)
10748 if (isLOWER(i) && !ANYOF_BITMAP_TEST(ret,i)) {
10750 set_regclass_bit(pRExC_state, ret, (U8) i, &l1_fold_invlist, &unicode_alternate);
10753 for (i = prevvalue; i <= ceilvalue; i++)
10754 if (isUPPER(i) && !ANYOF_BITMAP_TEST(ret,i)) {
10756 set_regclass_bit(pRExC_state, ret, (U8) i, &l1_fold_invlist, &unicode_alternate);
10762 for (i = prevvalue; i <= ceilvalue; i++) {
10763 stored += set_regclass_bit(pRExC_state, ret, (U8) i, &l1_fold_invlist, &unicode_alternate);
10767 const UV prevnatvalue = NATIVE_TO_UNI(prevvalue);
10768 const UV natvalue = NATIVE_TO_UNI(value);
10769 nonbitmap = add_range_to_invlist(nonbitmap, prevnatvalue, natvalue);
10772 literal_endpoint = 0;
10776 range = 0; /* this range (if it was one) is done now */
10783 /****** !SIZE_ONLY AFTER HERE *********/
10785 /* If folding and there are code points above 255, we calculate all
10786 * characters that could fold to or from the ones already on the list */
10787 if (FOLD && nonbitmap) {
10788 UV start, end; /* End points of code point ranges */
10790 SV* fold_intersection = NULL;
10792 /* This is a list of all the characters that participate in folds
10793 * (except marks, etc in multi-char folds */
10794 if (! PL_utf8_foldable) {
10795 SV* swash = swash_init("utf8", "Cased", &PL_sv_undef, 1, 0);
10796 PL_utf8_foldable = _swash_to_invlist(swash);
10797 SvREFCNT_dec(swash);
10800 /* This is a hash that for a particular fold gives all characters
10801 * that are involved in it */
10802 if (! PL_utf8_foldclosures) {
10804 /* If we were unable to find any folds, then we likely won't be
10805 * able to find the closures. So just create an empty list.
10806 * Folding will effectively be restricted to the non-Unicode rules
10807 * hard-coded into Perl. (This case happens legitimately during
10808 * compilation of Perl itself before the Unicode tables are
10810 if (invlist_len(PL_utf8_foldable) == 0) {
10811 PL_utf8_foldclosures = newHV();
10813 /* If the folds haven't been read in, call a fold function
10815 if (! PL_utf8_tofold) {
10816 U8 dummy[UTF8_MAXBYTES+1];
10819 /* This particular string is above \xff in both UTF-8 and
10821 to_utf8_fold((U8*) "\xC8\x80", dummy, &dummy_len);
10822 assert(PL_utf8_tofold); /* Verify that worked */
10824 PL_utf8_foldclosures = _swash_inversion_hash(PL_utf8_tofold);
10828 /* Only the characters in this class that participate in folds need be
10829 * checked. Get the intersection of this class and all the possible
10830 * characters that are foldable. This can quickly narrow down a large
10832 _invlist_intersection(PL_utf8_foldable, nonbitmap, &fold_intersection);
10834 /* Now look at the foldable characters in this class individually */
10835 invlist_iterinit(fold_intersection);
10836 while (invlist_iternext(fold_intersection, &start, &end)) {
10839 /* Look at every character in the range */
10840 for (j = start; j <= end; j++) {
10843 U8 foldbuf[UTF8_MAXBYTES_CASE+1];
10846 _to_uni_fold_flags(j, foldbuf, &foldlen, allow_full_fold);
10848 if (foldlen > (STRLEN)UNISKIP(f)) {
10850 /* Any multicharacter foldings (disallowed in lookbehind
10851 * patterns) require the following transform: [ABCDEF] ->
10852 * (?:[ABCabcDEFd]|pq|rst) where E folds into "pq" and F
10853 * folds into "rst", all other characters fold to single
10854 * characters. We save away these multicharacter foldings,
10855 * to be later saved as part of the additional "s" data. */
10856 if (! RExC_in_lookbehind) {
10858 U8* e = foldbuf + foldlen;
10860 /* If any of the folded characters of this are in the
10861 * Latin1 range, tell the regex engine that this can
10862 * match a non-utf8 target string. The only multi-byte
10863 * fold whose source is in the Latin1 range (U+00DF)
10864 * applies only when the target string is utf8, or
10865 * under unicode rules */
10866 if (j > 255 || AT_LEAST_UNI_SEMANTICS) {
10869 /* Can't mix ascii with non- under /aa */
10870 if (MORE_ASCII_RESTRICTED
10871 && (isASCII(*loc) != isASCII(j)))
10873 goto end_multi_fold;
10875 if (UTF8_IS_INVARIANT(*loc)
10876 || UTF8_IS_DOWNGRADEABLE_START(*loc))
10878 /* Can't mix above and below 256 under LOC
10881 goto end_multi_fold;
10884 |= ANYOF_NONBITMAP_NON_UTF8;
10887 loc += UTF8SKIP(loc);
10891 add_alternate(&unicode_alternate, foldbuf, foldlen);
10895 /* This is special-cased, as it is the only letter which
10896 * has both a multi-fold and single-fold in Latin1. All
10897 * the other chars that have single and multi-folds are
10898 * always in utf8, and the utf8 folding algorithm catches
10900 if (! LOC && j == LATIN_CAPITAL_LETTER_SHARP_S) {
10901 stored += set_regclass_bit(pRExC_state,
10903 LATIN_SMALL_LETTER_SHARP_S,
10904 &l1_fold_invlist, &unicode_alternate);
10908 /* Single character fold. Add everything in its fold
10909 * closure to the list that this node should match */
10912 /* The fold closures data structure is a hash with the keys
10913 * being every character that is folded to, like 'k', and
10914 * the values each an array of everything that folds to its
10915 * key. e.g. [ 'k', 'K', KELVIN_SIGN ] */
10916 if ((listp = hv_fetch(PL_utf8_foldclosures,
10917 (char *) foldbuf, foldlen, FALSE)))
10919 AV* list = (AV*) *listp;
10921 for (k = 0; k <= av_len(list); k++) {
10922 SV** c_p = av_fetch(list, k, FALSE);
10925 Perl_croak(aTHX_ "panic: invalid PL_utf8_foldclosures structure");
10929 /* /aa doesn't allow folds between ASCII and non-;
10930 * /l doesn't allow them between above and below
10932 if ((MORE_ASCII_RESTRICTED
10933 && (isASCII(c) != isASCII(j)))
10934 || (LOC && ((c < 256) != (j < 256))))
10939 if (c < 256 && AT_LEAST_UNI_SEMANTICS) {
10940 stored += set_regclass_bit(pRExC_state,
10943 &l1_fold_invlist, &unicode_alternate);
10945 /* It may be that the code point is already in
10946 * this range or already in the bitmap, in
10947 * which case we need do nothing */
10948 else if ((c < start || c > end)
10950 || ! ANYOF_BITMAP_TEST(ret, c)))
10952 nonbitmap = add_cp_to_invlist(nonbitmap, c);
10959 SvREFCNT_dec(fold_intersection);
10962 /* Combine the two lists into one. */
10963 if (l1_fold_invlist) {
10965 _invlist_union(nonbitmap, l1_fold_invlist, &nonbitmap);
10966 SvREFCNT_dec(l1_fold_invlist);
10969 nonbitmap = l1_fold_invlist;
10973 /* And combine the result (if any) with any inversion list from properties.
10974 * The lists are kept separate up to now because we don't want to fold the
10978 _invlist_union(nonbitmap, properties, &nonbitmap);
10979 SvREFCNT_dec(properties);
10982 nonbitmap = properties;
10986 /* Here, <nonbitmap> contains all the code points we can determine at
10987 * compile time that we haven't put into the bitmap. Go through it, and
10988 * for things that belong in the bitmap, put them there, and delete from
10992 /* Above-ASCII code points in /d have to stay in <nonbitmap>, as they
10993 * possibly only should match when the target string is UTF-8 */
10994 UV max_cp_to_set = (DEPENDS_SEMANTICS) ? 127 : 255;
10996 /* This gets set if we actually need to modify things */
10997 bool change_invlist = FALSE;
11001 /* Start looking through <nonbitmap> */
11002 invlist_iterinit(nonbitmap);
11003 while (invlist_iternext(nonbitmap, &start, &end)) {
11007 /* Quit if are above what we should change */
11008 if (start > max_cp_to_set) {
11012 change_invlist = TRUE;
11014 /* Set all the bits in the range, up to the max that we are doing */
11015 high = (end < max_cp_to_set) ? end : max_cp_to_set;
11016 for (i = start; i <= (int) high; i++) {
11017 if (! ANYOF_BITMAP_TEST(ret, i)) {
11018 ANYOF_BITMAP_SET(ret, i);
11026 /* Done with loop; set <nonbitmap> to not include any code points that
11027 * are in the bitmap */
11028 if (change_invlist) {
11029 SV* keep_list = _new_invlist(2);
11030 _append_range_to_invlist(keep_list, max_cp_to_set + 1, UV_MAX);
11031 _invlist_intersection(nonbitmap, keep_list, &nonbitmap);
11032 SvREFCNT_dec(keep_list);
11035 /* If have completely emptied it, remove it completely */
11036 if (invlist_len(nonbitmap) == 0) {
11037 SvREFCNT_dec(nonbitmap);
11042 /* Here, we have calculated what code points should be in the character
11043 * class. <nonbitmap> does not overlap the bitmap except possibly in the
11044 * case of DEPENDS rules.
11046 * Now we can see about various optimizations. Fold calculation (which we
11047 * did above) needs to take place before inversion. Otherwise /[^k]/i
11048 * would invert to include K, which under /i would match k, which it
11051 /* Optimize inverted simple patterns (e.g. [^a-z]). Note that we haven't
11052 * set the FOLD flag yet, so this does optimize those. It doesn't
11053 * optimize locale. Doing so perhaps could be done as long as there is
11054 * nothing like \w in it; some thought also would have to be given to the
11055 * interaction with above 0x100 chars */
11056 if ((ANYOF_FLAGS(ret) & ANYOF_INVERT)
11058 && ! unicode_alternate
11059 /* In case of /d, there are some things that should match only when in
11060 * not in the bitmap, i.e., they require UTF8 to match. These are
11061 * listed in nonbitmap, but if ANYOF_NONBITMAP_NON_UTF8 is set in this
11062 * case, they don't require UTF8, so can invert here */
11064 || ! DEPENDS_SEMANTICS
11065 || (ANYOF_FLAGS(ret) & ANYOF_NONBITMAP_NON_UTF8))
11066 && SvCUR(listsv) == initial_listsv_len)
11070 for (i = 0; i < 256; ++i) {
11071 if (ANYOF_BITMAP_TEST(ret, i)) {
11072 ANYOF_BITMAP_CLEAR(ret, i);
11075 ANYOF_BITMAP_SET(ret, i);
11080 /* The inversion means that everything above 255 is matched */
11081 ANYOF_FLAGS(ret) |= ANYOF_UNICODE_ALL;
11084 /* Here, also has things outside the bitmap that may overlap with
11085 * the bitmap. We have to sync them up, so that they get inverted
11086 * in both places. Earlier, we removed all overlaps except in the
11087 * case of /d rules, so no syncing is needed except for this case
11089 SV *remove_list = NULL;
11091 if (DEPENDS_SEMANTICS) {
11094 /* Set the bits that correspond to the ones that aren't in the
11095 * bitmap. Otherwise, when we invert, we'll miss these.
11096 * Earlier, we removed from the nonbitmap all code points
11097 * < 128, so there is no extra work here */
11098 invlist_iterinit(nonbitmap);
11099 while (invlist_iternext(nonbitmap, &start, &end)) {
11100 if (start > 255) { /* The bit map goes to 255 */
11106 for (i = start; i <= (int) end; ++i) {
11107 ANYOF_BITMAP_SET(ret, i);
11114 /* Now invert both the bitmap and the nonbitmap. Anything in the
11115 * bitmap has to also be removed from the non-bitmap, but again,
11116 * there should not be overlap unless is /d rules. */
11117 _invlist_invert(nonbitmap);
11119 for (i = 0; i < 256; ++i) {
11120 if (ANYOF_BITMAP_TEST(ret, i)) {
11121 ANYOF_BITMAP_CLEAR(ret, i);
11122 if (DEPENDS_SEMANTICS) {
11123 if (! remove_list) {
11124 remove_list = _new_invlist(2);
11126 remove_list = add_cp_to_invlist(remove_list, i);
11130 ANYOF_BITMAP_SET(ret, i);
11136 /* And do the removal */
11137 if (DEPENDS_SEMANTICS) {
11139 _invlist_subtract(nonbitmap, remove_list, &nonbitmap);
11140 SvREFCNT_dec(remove_list);
11144 /* There is no overlap for non-/d, so just delete anything
11146 SV* keep_list = _new_invlist(2);
11147 _append_range_to_invlist(keep_list, 256, UV_MAX);
11148 _invlist_intersection(nonbitmap, keep_list, &nonbitmap);
11149 SvREFCNT_dec(keep_list);
11153 stored = 256 - stored;
11155 /* Clear the invert flag since have just done it here */
11156 ANYOF_FLAGS(ret) &= ~ANYOF_INVERT;
11159 /* Folding in the bitmap is taken care of above, but not for locale (for
11160 * which we have to wait to see what folding is in effect at runtime), and
11161 * for some things not in the bitmap (only the upper latin folds in this
11162 * case, as all other single-char folding has been set above). Set
11163 * run-time fold flag for these */
11165 || (DEPENDS_SEMANTICS
11167 && ! (ANYOF_FLAGS(ret) & ANYOF_NONBITMAP_NON_UTF8))
11168 || unicode_alternate))
11170 ANYOF_FLAGS(ret) |= ANYOF_LOC_NONBITMAP_FOLD;
11173 /* A single character class can be "optimized" into an EXACTish node.
11174 * Note that since we don't currently count how many characters there are
11175 * outside the bitmap, we are XXX missing optimization possibilities for
11176 * them. This optimization can't happen unless this is a truly single
11177 * character class, which means that it can't be an inversion into a
11178 * many-character class, and there must be no possibility of there being
11179 * things outside the bitmap. 'stored' (only) for locales doesn't include
11180 * \w, etc, so have to make a special test that they aren't present
11182 * Similarly A 2-character class of the very special form like [bB] can be
11183 * optimized into an EXACTFish node, but only for non-locales, and for
11184 * characters which only have the two folds; so things like 'fF' and 'Ii'
11185 * wouldn't work because they are part of the fold of 'LATIN SMALL LIGATURE
11188 && ! unicode_alternate
11189 && SvCUR(listsv) == initial_listsv_len
11190 && ! (ANYOF_FLAGS(ret) & (ANYOF_INVERT|ANYOF_UNICODE_ALL))
11191 && (((stored == 1 && ((! (ANYOF_FLAGS(ret) & ANYOF_LOCALE))
11192 || (! ANYOF_CLASS_TEST_ANY_SET(ret)))))
11193 || (stored == 2 && ((! (ANYOF_FLAGS(ret) & ANYOF_LOCALE))
11194 && (! _HAS_NONLATIN1_FOLD_CLOSURE_ONLY_FOR_USE_BY_REGCOMP_DOT_C_AND_REGEXEC_DOT_C(value))
11195 /* If the latest code point has a fold whose
11196 * bit is set, it must be the only other one */
11197 && ((prevvalue = PL_fold_latin1[value]) != (IV)value)
11198 && ANYOF_BITMAP_TEST(ret, prevvalue)))))
11200 /* Note that the information needed to decide to do this optimization
11201 * is not currently available until the 2nd pass, and that the actually
11202 * used EXACTish node takes less space than the calculated ANYOF node,
11203 * and hence the amount of space calculated in the first pass is larger
11204 * than actually used, so this optimization doesn't gain us any space.
11205 * But an EXACT node is faster than an ANYOF node, and can be combined
11206 * with any adjacent EXACT nodes later by the optimizer for further
11207 * gains. The speed of executing an EXACTF is similar to an ANYOF
11208 * node, so the optimization advantage comes from the ability to join
11209 * it to adjacent EXACT nodes */
11211 const char * cur_parse= RExC_parse;
11213 RExC_emit = (regnode *)orig_emit;
11214 RExC_parse = (char *)orig_parse;
11218 /* A locale node with one point can be folded; all the other cases
11219 * with folding will have two points, since we calculate them above
11221 if (ANYOF_FLAGS(ret) & ANYOF_LOC_NONBITMAP_FOLD) {
11228 else { /* else 2 chars in the bit map: the folds of each other */
11230 /* Use the folded value, which for the cases where we get here,
11231 * is just the lower case of the current one (which may resolve to
11232 * itself, or to the other one */
11233 value = toLOWER_LATIN1(value);
11235 /* To join adjacent nodes, they must be the exact EXACTish type.
11236 * Try to use the most likely type, by using EXACTFA if possible,
11237 * then EXACTFU if the regex calls for it, or is required because
11238 * the character is non-ASCII. (If <value> is ASCII, its fold is
11239 * also ASCII for the cases where we get here.) */
11240 if (MORE_ASCII_RESTRICTED && isASCII(value)) {
11243 else if (AT_LEAST_UNI_SEMANTICS || !isASCII(value)) {
11246 else { /* Otherwise, more likely to be EXACTF type */
11251 ret = reg_node(pRExC_state, op);
11252 RExC_parse = (char *)cur_parse;
11253 if (UTF && ! NATIVE_IS_INVARIANT(value)) {
11254 *STRING(ret)= UTF8_EIGHT_BIT_HI((U8) value);
11255 *(STRING(ret) + 1)= UTF8_EIGHT_BIT_LO((U8) value);
11257 RExC_emit += STR_SZ(2);
11260 *STRING(ret)= (char)value;
11262 RExC_emit += STR_SZ(1);
11264 SvREFCNT_dec(listsv);
11268 /* If there is a swash and more than one element, we can't use the swash in
11269 * the optimization below. */
11270 if (swash && element_count > 1) {
11271 SvREFCNT_dec(swash);
11275 && SvCUR(listsv) == initial_listsv_len
11276 && ! unicode_alternate)
11278 ARG_SET(ret, ANYOF_NONBITMAP_EMPTY);
11279 SvREFCNT_dec(listsv);
11280 SvREFCNT_dec(unicode_alternate);
11283 /* av[0] stores the character class description in its textual form:
11284 * used later (regexec.c:Perl_regclass_swash()) to initialize the
11285 * appropriate swash, and is also useful for dumping the regnode.
11286 * av[1] if NULL, is a placeholder to later contain the swash computed
11287 * from av[0]. But if no further computation need be done, the
11288 * swash is stored there now.
11289 * av[2] stores the multicharacter foldings, used later in
11290 * regexec.c:S_reginclass().
11291 * av[3] stores the nonbitmap inversion list for use in addition or
11292 * instead of av[0]; not used if av[1] isn't NULL
11293 * av[4] is set if any component of the class is from a user-defined
11294 * property; not used if av[1] isn't NULL */
11295 AV * const av = newAV();
11298 av_store(av, 0, (SvCUR(listsv) == initial_listsv_len)
11302 av_store(av, 1, swash);
11303 SvREFCNT_dec(nonbitmap);
11306 av_store(av, 1, NULL);
11308 av_store(av, 3, nonbitmap);
11309 av_store(av, 4, newSVuv(has_user_defined_property));
11313 /* Store any computed multi-char folds only if we are allowing
11315 if (allow_full_fold) {
11316 av_store(av, 2, MUTABLE_SV(unicode_alternate));
11317 if (unicode_alternate) { /* This node is variable length */
11322 av_store(av, 2, NULL);
11324 rv = newRV_noinc(MUTABLE_SV(av));
11325 n = add_data(pRExC_state, 1, "s");
11326 RExC_rxi->data->data[n] = (void*)rv;
11334 /* reg_skipcomment()
11336 Absorbs an /x style # comments from the input stream.
11337 Returns true if there is more text remaining in the stream.
11338 Will set the REG_SEEN_RUN_ON_COMMENT flag if the comment
11339 terminates the pattern without including a newline.
11341 Note its the callers responsibility to ensure that we are
11342 actually in /x mode
11347 S_reg_skipcomment(pTHX_ RExC_state_t *pRExC_state)
11351 PERL_ARGS_ASSERT_REG_SKIPCOMMENT;
11353 while (RExC_parse < RExC_end)
11354 if (*RExC_parse++ == '\n') {
11359 /* we ran off the end of the pattern without ending
11360 the comment, so we have to add an \n when wrapping */
11361 RExC_seen |= REG_SEEN_RUN_ON_COMMENT;
11369 Advances the parse position, and optionally absorbs
11370 "whitespace" from the inputstream.
11372 Without /x "whitespace" means (?#...) style comments only,
11373 with /x this means (?#...) and # comments and whitespace proper.
11375 Returns the RExC_parse point from BEFORE the scan occurs.
11377 This is the /x friendly way of saying RExC_parse++.
11381 S_nextchar(pTHX_ RExC_state_t *pRExC_state)
11383 char* const retval = RExC_parse++;
11385 PERL_ARGS_ASSERT_NEXTCHAR;
11388 if (RExC_end - RExC_parse >= 3
11389 && *RExC_parse == '('
11390 && RExC_parse[1] == '?'
11391 && RExC_parse[2] == '#')
11393 while (*RExC_parse != ')') {
11394 if (RExC_parse == RExC_end)
11395 FAIL("Sequence (?#... not terminated");
11401 if (RExC_flags & RXf_PMf_EXTENDED) {
11402 if (isSPACE(*RExC_parse)) {
11406 else if (*RExC_parse == '#') {
11407 if ( reg_skipcomment( pRExC_state ) )
11416 - reg_node - emit a node
11418 STATIC regnode * /* Location. */
11419 S_reg_node(pTHX_ RExC_state_t *pRExC_state, U8 op)
11422 register regnode *ptr;
11423 regnode * const ret = RExC_emit;
11424 GET_RE_DEBUG_FLAGS_DECL;
11426 PERL_ARGS_ASSERT_REG_NODE;
11429 SIZE_ALIGN(RExC_size);
11433 if (RExC_emit >= RExC_emit_bound)
11434 Perl_croak(aTHX_ "panic: reg_node overrun trying to emit %d, %p>=%p",
11435 op, RExC_emit, RExC_emit_bound);
11437 NODE_ALIGN_FILL(ret);
11439 FILL_ADVANCE_NODE(ptr, op);
11440 REH_CALL_COMP_NODE_HOOK(pRExC_state->rx, (ptr) - 1);
11441 #ifdef RE_TRACK_PATTERN_OFFSETS
11442 if (RExC_offsets) { /* MJD */
11443 MJD_OFFSET_DEBUG(("%s:%d: (op %s) %s %"UVuf" (len %"UVuf") (max %"UVuf").\n",
11444 "reg_node", __LINE__,
11446 (UV)(RExC_emit - RExC_emit_start) > RExC_offsets[0]
11447 ? "Overwriting end of array!\n" : "OK",
11448 (UV)(RExC_emit - RExC_emit_start),
11449 (UV)(RExC_parse - RExC_start),
11450 (UV)RExC_offsets[0]));
11451 Set_Node_Offset(RExC_emit, RExC_parse + (op == END));
11459 - reganode - emit a node with an argument
11461 STATIC regnode * /* Location. */
11462 S_reganode(pTHX_ RExC_state_t *pRExC_state, U8 op, U32 arg)
11465 register regnode *ptr;
11466 regnode * const ret = RExC_emit;
11467 GET_RE_DEBUG_FLAGS_DECL;
11469 PERL_ARGS_ASSERT_REGANODE;
11472 SIZE_ALIGN(RExC_size);
11477 assert(2==regarglen[op]+1);
11479 Anything larger than this has to allocate the extra amount.
11480 If we changed this to be:
11482 RExC_size += (1 + regarglen[op]);
11484 then it wouldn't matter. Its not clear what side effect
11485 might come from that so its not done so far.
11490 if (RExC_emit >= RExC_emit_bound)
11491 Perl_croak(aTHX_ "panic: reg_node overrun trying to emit %d, %p>=%p",
11492 op, RExC_emit, RExC_emit_bound);
11494 NODE_ALIGN_FILL(ret);
11496 FILL_ADVANCE_NODE_ARG(ptr, op, arg);
11497 REH_CALL_COMP_NODE_HOOK(pRExC_state->rx, (ptr) - 2);
11498 #ifdef RE_TRACK_PATTERN_OFFSETS
11499 if (RExC_offsets) { /* MJD */
11500 MJD_OFFSET_DEBUG(("%s(%d): (op %s) %s %"UVuf" <- %"UVuf" (max %"UVuf").\n",
11504 (UV)(RExC_emit - RExC_emit_start) > RExC_offsets[0] ?
11505 "Overwriting end of array!\n" : "OK",
11506 (UV)(RExC_emit - RExC_emit_start),
11507 (UV)(RExC_parse - RExC_start),
11508 (UV)RExC_offsets[0]));
11509 Set_Cur_Node_Offset;
11517 - reguni - emit (if appropriate) a Unicode character
11520 S_reguni(pTHX_ const RExC_state_t *pRExC_state, UV uv, char* s)
11524 PERL_ARGS_ASSERT_REGUNI;
11526 return SIZE_ONLY ? UNISKIP(uv) : (uvchr_to_utf8((U8*)s, uv) - (U8*)s);
11530 - reginsert - insert an operator in front of already-emitted operand
11532 * Means relocating the operand.
11535 S_reginsert(pTHX_ RExC_state_t *pRExC_state, U8 op, regnode *opnd, U32 depth)
11538 register regnode *src;
11539 register regnode *dst;
11540 register regnode *place;
11541 const int offset = regarglen[(U8)op];
11542 const int size = NODE_STEP_REGNODE + offset;
11543 GET_RE_DEBUG_FLAGS_DECL;
11545 PERL_ARGS_ASSERT_REGINSERT;
11546 PERL_UNUSED_ARG(depth);
11547 /* (PL_regkind[(U8)op] == CURLY ? EXTRA_STEP_2ARGS : 0); */
11548 DEBUG_PARSE_FMT("inst"," - %s",PL_reg_name[op]);
11557 if (RExC_open_parens) {
11559 /*DEBUG_PARSE_FMT("inst"," - %"IVdf, (IV)RExC_npar);*/
11560 for ( paren=0 ; paren < RExC_npar ; paren++ ) {
11561 if ( RExC_open_parens[paren] >= opnd ) {
11562 /*DEBUG_PARSE_FMT("open"," - %d",size);*/
11563 RExC_open_parens[paren] += size;
11565 /*DEBUG_PARSE_FMT("open"," - %s","ok");*/
11567 if ( RExC_close_parens[paren] >= opnd ) {
11568 /*DEBUG_PARSE_FMT("close"," - %d",size);*/
11569 RExC_close_parens[paren] += size;
11571 /*DEBUG_PARSE_FMT("close"," - %s","ok");*/
11576 while (src > opnd) {
11577 StructCopy(--src, --dst, regnode);
11578 #ifdef RE_TRACK_PATTERN_OFFSETS
11579 if (RExC_offsets) { /* MJD 20010112 */
11580 MJD_OFFSET_DEBUG(("%s(%d): (op %s) %s copy %"UVuf" -> %"UVuf" (max %"UVuf").\n",
11584 (UV)(dst - RExC_emit_start) > RExC_offsets[0]
11585 ? "Overwriting end of array!\n" : "OK",
11586 (UV)(src - RExC_emit_start),
11587 (UV)(dst - RExC_emit_start),
11588 (UV)RExC_offsets[0]));
11589 Set_Node_Offset_To_R(dst-RExC_emit_start, Node_Offset(src));
11590 Set_Node_Length_To_R(dst-RExC_emit_start, Node_Length(src));
11596 place = opnd; /* Op node, where operand used to be. */
11597 #ifdef RE_TRACK_PATTERN_OFFSETS
11598 if (RExC_offsets) { /* MJD */
11599 MJD_OFFSET_DEBUG(("%s(%d): (op %s) %s %"UVuf" <- %"UVuf" (max %"UVuf").\n",
11603 (UV)(place - RExC_emit_start) > RExC_offsets[0]
11604 ? "Overwriting end of array!\n" : "OK",
11605 (UV)(place - RExC_emit_start),
11606 (UV)(RExC_parse - RExC_start),
11607 (UV)RExC_offsets[0]));
11608 Set_Node_Offset(place, RExC_parse);
11609 Set_Node_Length(place, 1);
11612 src = NEXTOPER(place);
11613 FILL_ADVANCE_NODE(place, op);
11614 REH_CALL_COMP_NODE_HOOK(pRExC_state->rx, (place) - 1);
11615 Zero(src, offset, regnode);
11619 - regtail - set the next-pointer at the end of a node chain of p to val.
11620 - SEE ALSO: regtail_study
11622 /* TODO: All three parms should be const */
11624 S_regtail(pTHX_ RExC_state_t *pRExC_state, regnode *p, const regnode *val,U32 depth)
11627 register regnode *scan;
11628 GET_RE_DEBUG_FLAGS_DECL;
11630 PERL_ARGS_ASSERT_REGTAIL;
11632 PERL_UNUSED_ARG(depth);
11638 /* Find last node. */
11641 regnode * const temp = regnext(scan);
11643 SV * const mysv=sv_newmortal();
11644 DEBUG_PARSE_MSG((scan==p ? "tail" : ""));
11645 regprop(RExC_rx, mysv, scan);
11646 PerlIO_printf(Perl_debug_log, "~ %s (%d) %s %s\n",
11647 SvPV_nolen_const(mysv), REG_NODE_NUM(scan),
11648 (temp == NULL ? "->" : ""),
11649 (temp == NULL ? PL_reg_name[OP(val)] : "")
11657 if (reg_off_by_arg[OP(scan)]) {
11658 ARG_SET(scan, val - scan);
11661 NEXT_OFF(scan) = val - scan;
11667 - regtail_study - set the next-pointer at the end of a node chain of p to val.
11668 - Look for optimizable sequences at the same time.
11669 - currently only looks for EXACT chains.
11671 This is experimental code. The idea is to use this routine to perform
11672 in place optimizations on branches and groups as they are constructed,
11673 with the long term intention of removing optimization from study_chunk so
11674 that it is purely analytical.
11676 Currently only used when in DEBUG mode. The macro REGTAIL_STUDY() is used
11677 to control which is which.
11680 /* TODO: All four parms should be const */
11683 S_regtail_study(pTHX_ RExC_state_t *pRExC_state, regnode *p, const regnode *val,U32 depth)
11686 register regnode *scan;
11688 #ifdef EXPERIMENTAL_INPLACESCAN
11691 GET_RE_DEBUG_FLAGS_DECL;
11693 PERL_ARGS_ASSERT_REGTAIL_STUDY;
11699 /* Find last node. */
11703 regnode * const temp = regnext(scan);
11704 #ifdef EXPERIMENTAL_INPLACESCAN
11705 if (PL_regkind[OP(scan)] == EXACT) {
11706 bool has_exactf_sharp_s; /* Unexamined in this routine */
11707 if (join_exact(pRExC_state,scan,&min, &has_exactf_sharp_s, 1,val,depth+1))
11712 switch (OP(scan)) {
11718 case EXACTFU_NO_TRIE:
11720 if( exact == PSEUDO )
11722 else if ( exact != OP(scan) )
11731 SV * const mysv=sv_newmortal();
11732 DEBUG_PARSE_MSG((scan==p ? "tsdy" : ""));
11733 regprop(RExC_rx, mysv, scan);
11734 PerlIO_printf(Perl_debug_log, "~ %s (%d) -> %s\n",
11735 SvPV_nolen_const(mysv),
11736 REG_NODE_NUM(scan),
11737 PL_reg_name[exact]);
11744 SV * const mysv_val=sv_newmortal();
11745 DEBUG_PARSE_MSG("");
11746 regprop(RExC_rx, mysv_val, val);
11747 PerlIO_printf(Perl_debug_log, "~ attach to %s (%"IVdf") offset to %"IVdf"\n",
11748 SvPV_nolen_const(mysv_val),
11749 (IV)REG_NODE_NUM(val),
11753 if (reg_off_by_arg[OP(scan)]) {
11754 ARG_SET(scan, val - scan);
11757 NEXT_OFF(scan) = val - scan;
11765 - regdump - dump a regexp onto Perl_debug_log in vaguely comprehensible form
11769 S_regdump_extflags(pTHX_ const char *lead, const U32 flags)
11775 for (bit=0; bit<32; bit++) {
11776 if (flags & (1<<bit)) {
11777 if ((1<<bit) & RXf_PMf_CHARSET) { /* Output separately, below */
11780 if (!set++ && lead)
11781 PerlIO_printf(Perl_debug_log, "%s",lead);
11782 PerlIO_printf(Perl_debug_log, "%s ",PL_reg_extflags_name[bit]);
11785 if ((cs = get_regex_charset(flags)) != REGEX_DEPENDS_CHARSET) {
11786 if (!set++ && lead) {
11787 PerlIO_printf(Perl_debug_log, "%s",lead);
11790 case REGEX_UNICODE_CHARSET:
11791 PerlIO_printf(Perl_debug_log, "UNICODE");
11793 case REGEX_LOCALE_CHARSET:
11794 PerlIO_printf(Perl_debug_log, "LOCALE");
11796 case REGEX_ASCII_RESTRICTED_CHARSET:
11797 PerlIO_printf(Perl_debug_log, "ASCII-RESTRICTED");
11799 case REGEX_ASCII_MORE_RESTRICTED_CHARSET:
11800 PerlIO_printf(Perl_debug_log, "ASCII-MORE_RESTRICTED");
11803 PerlIO_printf(Perl_debug_log, "UNKNOWN CHARACTER SET");
11809 PerlIO_printf(Perl_debug_log, "\n");
11811 PerlIO_printf(Perl_debug_log, "%s[none-set]\n",lead);
11817 Perl_regdump(pTHX_ const regexp *r)
11821 SV * const sv = sv_newmortal();
11822 SV *dsv= sv_newmortal();
11823 RXi_GET_DECL(r,ri);
11824 GET_RE_DEBUG_FLAGS_DECL;
11826 PERL_ARGS_ASSERT_REGDUMP;
11828 (void)dumpuntil(r, ri->program, ri->program + 1, NULL, NULL, sv, 0, 0);
11830 /* Header fields of interest. */
11831 if (r->anchored_substr) {
11832 RE_PV_QUOTED_DECL(s, 0, dsv, SvPVX_const(r->anchored_substr),
11833 RE_SV_DUMPLEN(r->anchored_substr), 30);
11834 PerlIO_printf(Perl_debug_log,
11835 "anchored %s%s at %"IVdf" ",
11836 s, RE_SV_TAIL(r->anchored_substr),
11837 (IV)r->anchored_offset);
11838 } else if (r->anchored_utf8) {
11839 RE_PV_QUOTED_DECL(s, 1, dsv, SvPVX_const(r->anchored_utf8),
11840 RE_SV_DUMPLEN(r->anchored_utf8), 30);
11841 PerlIO_printf(Perl_debug_log,
11842 "anchored utf8 %s%s at %"IVdf" ",
11843 s, RE_SV_TAIL(r->anchored_utf8),
11844 (IV)r->anchored_offset);
11846 if (r->float_substr) {
11847 RE_PV_QUOTED_DECL(s, 0, dsv, SvPVX_const(r->float_substr),
11848 RE_SV_DUMPLEN(r->float_substr), 30);
11849 PerlIO_printf(Perl_debug_log,
11850 "floating %s%s at %"IVdf"..%"UVuf" ",
11851 s, RE_SV_TAIL(r->float_substr),
11852 (IV)r->float_min_offset, (UV)r->float_max_offset);
11853 } else if (r->float_utf8) {
11854 RE_PV_QUOTED_DECL(s, 1, dsv, SvPVX_const(r->float_utf8),
11855 RE_SV_DUMPLEN(r->float_utf8), 30);
11856 PerlIO_printf(Perl_debug_log,
11857 "floating utf8 %s%s at %"IVdf"..%"UVuf" ",
11858 s, RE_SV_TAIL(r->float_utf8),
11859 (IV)r->float_min_offset, (UV)r->float_max_offset);
11861 if (r->check_substr || r->check_utf8)
11862 PerlIO_printf(Perl_debug_log,
11864 (r->check_substr == r->float_substr
11865 && r->check_utf8 == r->float_utf8
11866 ? "(checking floating" : "(checking anchored"));
11867 if (r->extflags & RXf_NOSCAN)
11868 PerlIO_printf(Perl_debug_log, " noscan");
11869 if (r->extflags & RXf_CHECK_ALL)
11870 PerlIO_printf(Perl_debug_log, " isall");
11871 if (r->check_substr || r->check_utf8)
11872 PerlIO_printf(Perl_debug_log, ") ");
11874 if (ri->regstclass) {
11875 regprop(r, sv, ri->regstclass);
11876 PerlIO_printf(Perl_debug_log, "stclass %s ", SvPVX_const(sv));
11878 if (r->extflags & RXf_ANCH) {
11879 PerlIO_printf(Perl_debug_log, "anchored");
11880 if (r->extflags & RXf_ANCH_BOL)
11881 PerlIO_printf(Perl_debug_log, "(BOL)");
11882 if (r->extflags & RXf_ANCH_MBOL)
11883 PerlIO_printf(Perl_debug_log, "(MBOL)");
11884 if (r->extflags & RXf_ANCH_SBOL)
11885 PerlIO_printf(Perl_debug_log, "(SBOL)");
11886 if (r->extflags & RXf_ANCH_GPOS)
11887 PerlIO_printf(Perl_debug_log, "(GPOS)");
11888 PerlIO_putc(Perl_debug_log, ' ');
11890 if (r->extflags & RXf_GPOS_SEEN)
11891 PerlIO_printf(Perl_debug_log, "GPOS:%"UVuf" ", (UV)r->gofs);
11892 if (r->intflags & PREGf_SKIP)
11893 PerlIO_printf(Perl_debug_log, "plus ");
11894 if (r->intflags & PREGf_IMPLICIT)
11895 PerlIO_printf(Perl_debug_log, "implicit ");
11896 PerlIO_printf(Perl_debug_log, "minlen %"IVdf" ", (IV)r->minlen);
11897 if (r->extflags & RXf_EVAL_SEEN)
11898 PerlIO_printf(Perl_debug_log, "with eval ");
11899 PerlIO_printf(Perl_debug_log, "\n");
11900 DEBUG_FLAGS_r(regdump_extflags("r->extflags: ",r->extflags));
11902 PERL_ARGS_ASSERT_REGDUMP;
11903 PERL_UNUSED_CONTEXT;
11904 PERL_UNUSED_ARG(r);
11905 #endif /* DEBUGGING */
11909 - regprop - printable representation of opcode
11911 #define EMIT_ANYOF_TEST_SEPARATOR(do_sep,sv,flags) \
11914 Perl_sv_catpvf(aTHX_ sv,"%s][%s",PL_colors[1],PL_colors[0]); \
11915 if (flags & ANYOF_INVERT) \
11916 /*make sure the invert info is in each */ \
11917 sv_catpvs(sv, "^"); \
11923 Perl_regprop(pTHX_ const regexp *prog, SV *sv, const regnode *o)
11928 RXi_GET_DECL(prog,progi);
11929 GET_RE_DEBUG_FLAGS_DECL;
11931 PERL_ARGS_ASSERT_REGPROP;
11935 if (OP(o) > REGNODE_MAX) /* regnode.type is unsigned */
11936 /* It would be nice to FAIL() here, but this may be called from
11937 regexec.c, and it would be hard to supply pRExC_state. */
11938 Perl_croak(aTHX_ "Corrupted regexp opcode %d > %d", (int)OP(o), (int)REGNODE_MAX);
11939 sv_catpv(sv, PL_reg_name[OP(o)]); /* Take off const! */
11941 k = PL_regkind[OP(o)];
11944 sv_catpvs(sv, " ");
11945 /* Using is_utf8_string() (via PERL_PV_UNI_DETECT)
11946 * is a crude hack but it may be the best for now since
11947 * we have no flag "this EXACTish node was UTF-8"
11949 pv_pretty(sv, STRING(o), STR_LEN(o), 60, PL_colors[0], PL_colors[1],
11950 PERL_PV_ESCAPE_UNI_DETECT |
11951 PERL_PV_ESCAPE_NONASCII |
11952 PERL_PV_PRETTY_ELLIPSES |
11953 PERL_PV_PRETTY_LTGT |
11954 PERL_PV_PRETTY_NOCLEAR
11956 } else if (k == TRIE) {
11957 /* print the details of the trie in dumpuntil instead, as
11958 * progi->data isn't available here */
11959 const char op = OP(o);
11960 const U32 n = ARG(o);
11961 const reg_ac_data * const ac = IS_TRIE_AC(op) ?
11962 (reg_ac_data *)progi->data->data[n] :
11964 const reg_trie_data * const trie
11965 = (reg_trie_data*)progi->data->data[!IS_TRIE_AC(op) ? n : ac->trie];
11967 Perl_sv_catpvf(aTHX_ sv, "-%s",PL_reg_name[o->flags]);
11968 DEBUG_TRIE_COMPILE_r(
11969 Perl_sv_catpvf(aTHX_ sv,
11970 "<S:%"UVuf"/%"IVdf" W:%"UVuf" L:%"UVuf"/%"UVuf" C:%"UVuf"/%"UVuf">",
11971 (UV)trie->startstate,
11972 (IV)trie->statecount-1, /* -1 because of the unused 0 element */
11973 (UV)trie->wordcount,
11976 (UV)TRIE_CHARCOUNT(trie),
11977 (UV)trie->uniquecharcount
11980 if ( IS_ANYOF_TRIE(op) || trie->bitmap ) {
11982 int rangestart = -1;
11983 U8* bitmap = IS_ANYOF_TRIE(op) ? (U8*)ANYOF_BITMAP(o) : (U8*)TRIE_BITMAP(trie);
11984 sv_catpvs(sv, "[");
11985 for (i = 0; i <= 256; i++) {
11986 if (i < 256 && BITMAP_TEST(bitmap,i)) {
11987 if (rangestart == -1)
11989 } else if (rangestart != -1) {
11990 if (i <= rangestart + 3)
11991 for (; rangestart < i; rangestart++)
11992 put_byte(sv, rangestart);
11994 put_byte(sv, rangestart);
11995 sv_catpvs(sv, "-");
11996 put_byte(sv, i - 1);
12001 sv_catpvs(sv, "]");
12004 } else if (k == CURLY) {
12005 if (OP(o) == CURLYM || OP(o) == CURLYN || OP(o) == CURLYX)
12006 Perl_sv_catpvf(aTHX_ sv, "[%d]", o->flags); /* Parenth number */
12007 Perl_sv_catpvf(aTHX_ sv, " {%d,%d}", ARG1(o), ARG2(o));
12009 else if (k == WHILEM && o->flags) /* Ordinal/of */
12010 Perl_sv_catpvf(aTHX_ sv, "[%d/%d]", o->flags & 0xf, o->flags>>4);
12011 else if (k == REF || k == OPEN || k == CLOSE || k == GROUPP || OP(o)==ACCEPT) {
12012 Perl_sv_catpvf(aTHX_ sv, "%d", (int)ARG(o)); /* Parenth number */
12013 if ( RXp_PAREN_NAMES(prog) ) {
12014 if ( k != REF || (OP(o) < NREF)) {
12015 AV *list= MUTABLE_AV(progi->data->data[progi->name_list_idx]);
12016 SV **name= av_fetch(list, ARG(o), 0 );
12018 Perl_sv_catpvf(aTHX_ sv, " '%"SVf"'", SVfARG(*name));
12021 AV *list= MUTABLE_AV(progi->data->data[ progi->name_list_idx ]);
12022 SV *sv_dat= MUTABLE_SV(progi->data->data[ ARG( o ) ]);
12023 I32 *nums=(I32*)SvPVX(sv_dat);
12024 SV **name= av_fetch(list, nums[0], 0 );
12027 for ( n=0; n<SvIVX(sv_dat); n++ ) {
12028 Perl_sv_catpvf(aTHX_ sv, "%s%"IVdf,
12029 (n ? "," : ""), (IV)nums[n]);
12031 Perl_sv_catpvf(aTHX_ sv, " '%"SVf"'", SVfARG(*name));
12035 } else if (k == GOSUB)
12036 Perl_sv_catpvf(aTHX_ sv, "%d[%+d]", (int)ARG(o),(int)ARG2L(o)); /* Paren and offset */
12037 else if (k == VERB) {
12039 Perl_sv_catpvf(aTHX_ sv, ":%"SVf,
12040 SVfARG((MUTABLE_SV(progi->data->data[ ARG( o ) ]))));
12041 } else if (k == LOGICAL)
12042 Perl_sv_catpvf(aTHX_ sv, "[%d]", o->flags); /* 2: embedded, otherwise 1 */
12043 else if (k == ANYOF) {
12044 int i, rangestart = -1;
12045 const U8 flags = ANYOF_FLAGS(o);
12048 /* Should be synchronized with * ANYOF_ #xdefines in regcomp.h */
12049 static const char * const anyofs[] = {
12082 if (flags & ANYOF_LOCALE)
12083 sv_catpvs(sv, "{loc}");
12084 if (flags & ANYOF_LOC_NONBITMAP_FOLD)
12085 sv_catpvs(sv, "{i}");
12086 Perl_sv_catpvf(aTHX_ sv, "[%s", PL_colors[0]);
12087 if (flags & ANYOF_INVERT)
12088 sv_catpvs(sv, "^");
12090 /* output what the standard cp 0-255 bitmap matches */
12091 for (i = 0; i <= 256; i++) {
12092 if (i < 256 && ANYOF_BITMAP_TEST(o,i)) {
12093 if (rangestart == -1)
12095 } else if (rangestart != -1) {
12096 if (i <= rangestart + 3)
12097 for (; rangestart < i; rangestart++)
12098 put_byte(sv, rangestart);
12100 put_byte(sv, rangestart);
12101 sv_catpvs(sv, "-");
12102 put_byte(sv, i - 1);
12109 EMIT_ANYOF_TEST_SEPARATOR(do_sep,sv,flags);
12110 /* output any special charclass tests (used entirely under use locale) */
12111 if (ANYOF_CLASS_TEST_ANY_SET(o))
12112 for (i = 0; i < (int)(sizeof(anyofs)/sizeof(char*)); i++)
12113 if (ANYOF_CLASS_TEST(o,i)) {
12114 sv_catpv(sv, anyofs[i]);
12118 EMIT_ANYOF_TEST_SEPARATOR(do_sep,sv,flags);
12120 if (flags & ANYOF_NON_UTF8_LATIN1_ALL) {
12121 sv_catpvs(sv, "{non-utf8-latin1-all}");
12124 /* output information about the unicode matching */
12125 if (flags & ANYOF_UNICODE_ALL)
12126 sv_catpvs(sv, "{unicode_all}");
12127 else if (ANYOF_NONBITMAP(o))
12128 sv_catpvs(sv, "{unicode}");
12129 if (flags & ANYOF_NONBITMAP_NON_UTF8)
12130 sv_catpvs(sv, "{outside bitmap}");
12132 if (ANYOF_NONBITMAP(o)) {
12133 SV *lv; /* Set if there is something outside the bit map */
12134 SV * const sw = regclass_swash(prog, o, FALSE, &lv, 0);
12135 bool byte_output = FALSE; /* If something in the bitmap has been
12138 if (lv && lv != &PL_sv_undef) {
12140 U8 s[UTF8_MAXBYTES_CASE+1];
12142 for (i = 0; i <= 256; i++) { /* Look at chars in bitmap */
12143 uvchr_to_utf8(s, i);
12146 && ! ANYOF_BITMAP_TEST(o, i) /* Don't duplicate
12150 && swash_fetch(sw, s, TRUE))
12152 if (rangestart == -1)
12154 } else if (rangestart != -1) {
12155 byte_output = TRUE;
12156 if (i <= rangestart + 3)
12157 for (; rangestart < i; rangestart++) {
12158 put_byte(sv, rangestart);
12161 put_byte(sv, rangestart);
12162 sv_catpvs(sv, "-");
12171 char *s = savesvpv(lv);
12172 char * const origs = s;
12174 while (*s && *s != '\n')
12178 const char * const t = ++s;
12181 sv_catpvs(sv, " ");
12187 /* Truncate very long output */
12188 if (s - origs > 256) {
12189 Perl_sv_catpvf(aTHX_ sv,
12191 (int) (s - origs - 1),
12197 else if (*s == '\t') {
12216 Perl_sv_catpvf(aTHX_ sv, "%s]", PL_colors[1]);
12218 else if (k == BRANCHJ && (OP(o) == UNLESSM || OP(o) == IFMATCH))
12219 Perl_sv_catpvf(aTHX_ sv, "[%d]", -(o->flags));
12221 PERL_UNUSED_CONTEXT;
12222 PERL_UNUSED_ARG(sv);
12223 PERL_UNUSED_ARG(o);
12224 PERL_UNUSED_ARG(prog);
12225 #endif /* DEBUGGING */
12229 Perl_re_intuit_string(pTHX_ REGEXP * const r)
12230 { /* Assume that RE_INTUIT is set */
12232 struct regexp *const prog = (struct regexp *)SvANY(r);
12233 GET_RE_DEBUG_FLAGS_DECL;
12235 PERL_ARGS_ASSERT_RE_INTUIT_STRING;
12236 PERL_UNUSED_CONTEXT;
12240 const char * const s = SvPV_nolen_const(prog->check_substr
12241 ? prog->check_substr : prog->check_utf8);
12243 if (!PL_colorset) reginitcolors();
12244 PerlIO_printf(Perl_debug_log,
12245 "%sUsing REx %ssubstr:%s \"%s%.60s%s%s\"\n",
12247 prog->check_substr ? "" : "utf8 ",
12248 PL_colors[5],PL_colors[0],
12251 (strlen(s) > 60 ? "..." : ""));
12254 return prog->check_substr ? prog->check_substr : prog->check_utf8;
12260 handles refcounting and freeing the perl core regexp structure. When
12261 it is necessary to actually free the structure the first thing it
12262 does is call the 'free' method of the regexp_engine associated to
12263 the regexp, allowing the handling of the void *pprivate; member
12264 first. (This routine is not overridable by extensions, which is why
12265 the extensions free is called first.)
12267 See regdupe and regdupe_internal if you change anything here.
12269 #ifndef PERL_IN_XSUB_RE
12271 Perl_pregfree(pTHX_ REGEXP *r)
12277 Perl_pregfree2(pTHX_ REGEXP *rx)
12280 struct regexp *const r = (struct regexp *)SvANY(rx);
12281 GET_RE_DEBUG_FLAGS_DECL;
12283 PERL_ARGS_ASSERT_PREGFREE2;
12285 if (r->mother_re) {
12286 ReREFCNT_dec(r->mother_re);
12288 CALLREGFREE_PVT(rx); /* free the private data */
12289 SvREFCNT_dec(RXp_PAREN_NAMES(r));
12292 SvREFCNT_dec(r->anchored_substr);
12293 SvREFCNT_dec(r->anchored_utf8);
12294 SvREFCNT_dec(r->float_substr);
12295 SvREFCNT_dec(r->float_utf8);
12296 Safefree(r->substrs);
12298 RX_MATCH_COPY_FREE(rx);
12299 #ifdef PERL_OLD_COPY_ON_WRITE
12300 SvREFCNT_dec(r->saved_copy);
12307 This is a hacky workaround to the structural issue of match results
12308 being stored in the regexp structure which is in turn stored in
12309 PL_curpm/PL_reg_curpm. The problem is that due to qr// the pattern
12310 could be PL_curpm in multiple contexts, and could require multiple
12311 result sets being associated with the pattern simultaneously, such
12312 as when doing a recursive match with (??{$qr})
12314 The solution is to make a lightweight copy of the regexp structure
12315 when a qr// is returned from the code executed by (??{$qr}) this
12316 lightweight copy doesn't actually own any of its data except for
12317 the starp/end and the actual regexp structure itself.
12323 Perl_reg_temp_copy (pTHX_ REGEXP *ret_x, REGEXP *rx)
12325 struct regexp *ret;
12326 struct regexp *const r = (struct regexp *)SvANY(rx);
12327 register const I32 npar = r->nparens+1;
12329 PERL_ARGS_ASSERT_REG_TEMP_COPY;
12332 ret_x = (REGEXP*) newSV_type(SVt_REGEXP);
12333 ret = (struct regexp *)SvANY(ret_x);
12335 (void)ReREFCNT_inc(rx);
12336 /* We can take advantage of the existing "copied buffer" mechanism in SVs
12337 by pointing directly at the buffer, but flagging that the allocated
12338 space in the copy is zero. As we've just done a struct copy, it's now
12339 a case of zero-ing that, rather than copying the current length. */
12340 SvPV_set(ret_x, RX_WRAPPED(rx));
12341 SvFLAGS(ret_x) |= SvFLAGS(rx) & (SVf_POK|SVp_POK|SVf_UTF8);
12342 memcpy(&(ret->xpv_cur), &(r->xpv_cur),
12343 sizeof(regexp) - STRUCT_OFFSET(regexp, xpv_cur));
12344 SvLEN_set(ret_x, 0);
12345 SvSTASH_set(ret_x, NULL);
12346 SvMAGIC_set(ret_x, NULL);
12347 Newx(ret->offs, npar, regexp_paren_pair);
12348 Copy(r->offs, ret->offs, npar, regexp_paren_pair);
12350 Newx(ret->substrs, 1, struct reg_substr_data);
12351 StructCopy(r->substrs, ret->substrs, struct reg_substr_data);
12353 SvREFCNT_inc_void(ret->anchored_substr);
12354 SvREFCNT_inc_void(ret->anchored_utf8);
12355 SvREFCNT_inc_void(ret->float_substr);
12356 SvREFCNT_inc_void(ret->float_utf8);
12358 /* check_substr and check_utf8, if non-NULL, point to either their
12359 anchored or float namesakes, and don't hold a second reference. */
12361 RX_MATCH_COPIED_off(ret_x);
12362 #ifdef PERL_OLD_COPY_ON_WRITE
12363 ret->saved_copy = NULL;
12365 ret->mother_re = rx;
12371 /* regfree_internal()
12373 Free the private data in a regexp. This is overloadable by
12374 extensions. Perl takes care of the regexp structure in pregfree(),
12375 this covers the *pprivate pointer which technically perl doesn't
12376 know about, however of course we have to handle the
12377 regexp_internal structure when no extension is in use.
12379 Note this is called before freeing anything in the regexp
12384 Perl_regfree_internal(pTHX_ REGEXP * const rx)
12387 struct regexp *const r = (struct regexp *)SvANY(rx);
12388 RXi_GET_DECL(r,ri);
12389 GET_RE_DEBUG_FLAGS_DECL;
12391 PERL_ARGS_ASSERT_REGFREE_INTERNAL;
12397 SV *dsv= sv_newmortal();
12398 RE_PV_QUOTED_DECL(s, RX_UTF8(rx),
12399 dsv, RX_PRECOMP(rx), RX_PRELEN(rx), 60);
12400 PerlIO_printf(Perl_debug_log,"%sFreeing REx:%s %s\n",
12401 PL_colors[4],PL_colors[5],s);
12404 #ifdef RE_TRACK_PATTERN_OFFSETS
12406 Safefree(ri->u.offsets); /* 20010421 MJD */
12409 int n = ri->data->count;
12410 PAD* new_comppad = NULL;
12415 /* If you add a ->what type here, update the comment in regcomp.h */
12416 switch (ri->data->what[n]) {
12421 SvREFCNT_dec(MUTABLE_SV(ri->data->data[n]));
12424 Safefree(ri->data->data[n]);
12427 new_comppad = MUTABLE_AV(ri->data->data[n]);
12430 if (new_comppad == NULL)
12431 Perl_croak(aTHX_ "panic: pregfree comppad");
12432 PAD_SAVE_LOCAL(old_comppad,
12433 /* Watch out for global destruction's random ordering. */
12434 (SvTYPE(new_comppad) == SVt_PVAV) ? new_comppad : NULL
12437 refcnt = OpREFCNT_dec((OP_4tree*)ri->data->data[n]);
12440 op_free((OP_4tree*)ri->data->data[n]);
12442 PAD_RESTORE_LOCAL(old_comppad);
12443 SvREFCNT_dec(MUTABLE_SV(new_comppad));
12444 new_comppad = NULL;
12449 { /* Aho Corasick add-on structure for a trie node.
12450 Used in stclass optimization only */
12452 reg_ac_data *aho=(reg_ac_data*)ri->data->data[n];
12454 refcount = --aho->refcount;
12457 PerlMemShared_free(aho->states);
12458 PerlMemShared_free(aho->fail);
12459 /* do this last!!!! */
12460 PerlMemShared_free(ri->data->data[n]);
12461 PerlMemShared_free(ri->regstclass);
12467 /* trie structure. */
12469 reg_trie_data *trie=(reg_trie_data*)ri->data->data[n];
12471 refcount = --trie->refcount;
12474 PerlMemShared_free(trie->charmap);
12475 PerlMemShared_free(trie->states);
12476 PerlMemShared_free(trie->trans);
12478 PerlMemShared_free(trie->bitmap);
12480 PerlMemShared_free(trie->jump);
12481 PerlMemShared_free(trie->wordinfo);
12482 /* do this last!!!! */
12483 PerlMemShared_free(ri->data->data[n]);
12488 Perl_croak(aTHX_ "panic: regfree data code '%c'", ri->data->what[n]);
12491 Safefree(ri->data->what);
12492 Safefree(ri->data);
12498 #define av_dup_inc(s,t) MUTABLE_AV(sv_dup_inc((const SV *)s,t))
12499 #define hv_dup_inc(s,t) MUTABLE_HV(sv_dup_inc((const SV *)s,t))
12500 #define SAVEPVN(p,n) ((p) ? savepvn(p,n) : NULL)
12503 re_dup - duplicate a regexp.
12505 This routine is expected to clone a given regexp structure. It is only
12506 compiled under USE_ITHREADS.
12508 After all of the core data stored in struct regexp is duplicated
12509 the regexp_engine.dupe method is used to copy any private data
12510 stored in the *pprivate pointer. This allows extensions to handle
12511 any duplication it needs to do.
12513 See pregfree() and regfree_internal() if you change anything here.
12515 #if defined(USE_ITHREADS)
12516 #ifndef PERL_IN_XSUB_RE
12518 Perl_re_dup_guts(pTHX_ const REGEXP *sstr, REGEXP *dstr, CLONE_PARAMS *param)
12522 const struct regexp *r = (const struct regexp *)SvANY(sstr);
12523 struct regexp *ret = (struct regexp *)SvANY(dstr);
12525 PERL_ARGS_ASSERT_RE_DUP_GUTS;
12527 npar = r->nparens+1;
12528 Newx(ret->offs, npar, regexp_paren_pair);
12529 Copy(r->offs, ret->offs, npar, regexp_paren_pair);
12531 /* no need to copy these */
12532 Newx(ret->swap, npar, regexp_paren_pair);
12535 if (ret->substrs) {
12536 /* Do it this way to avoid reading from *r after the StructCopy().
12537 That way, if any of the sv_dup_inc()s dislodge *r from the L1
12538 cache, it doesn't matter. */
12539 const bool anchored = r->check_substr
12540 ? r->check_substr == r->anchored_substr
12541 : r->check_utf8 == r->anchored_utf8;
12542 Newx(ret->substrs, 1, struct reg_substr_data);
12543 StructCopy(r->substrs, ret->substrs, struct reg_substr_data);
12545 ret->anchored_substr = sv_dup_inc(ret->anchored_substr, param);
12546 ret->anchored_utf8 = sv_dup_inc(ret->anchored_utf8, param);
12547 ret->float_substr = sv_dup_inc(ret->float_substr, param);
12548 ret->float_utf8 = sv_dup_inc(ret->float_utf8, param);
12550 /* check_substr and check_utf8, if non-NULL, point to either their
12551 anchored or float namesakes, and don't hold a second reference. */
12553 if (ret->check_substr) {
12555 assert(r->check_utf8 == r->anchored_utf8);
12556 ret->check_substr = ret->anchored_substr;
12557 ret->check_utf8 = ret->anchored_utf8;
12559 assert(r->check_substr == r->float_substr);
12560 assert(r->check_utf8 == r->float_utf8);
12561 ret->check_substr = ret->float_substr;
12562 ret->check_utf8 = ret->float_utf8;
12564 } else if (ret->check_utf8) {
12566 ret->check_utf8 = ret->anchored_utf8;
12568 ret->check_utf8 = ret->float_utf8;
12573 RXp_PAREN_NAMES(ret) = hv_dup_inc(RXp_PAREN_NAMES(ret), param);
12576 RXi_SET(ret,CALLREGDUPE_PVT(dstr,param));
12578 if (RX_MATCH_COPIED(dstr))
12579 ret->subbeg = SAVEPVN(ret->subbeg, ret->sublen);
12581 ret->subbeg = NULL;
12582 #ifdef PERL_OLD_COPY_ON_WRITE
12583 ret->saved_copy = NULL;
12586 if (ret->mother_re) {
12587 if (SvPVX_const(dstr) == SvPVX_const(ret->mother_re)) {
12588 /* Our storage points directly to our mother regexp, but that's
12589 1: a buffer in a different thread
12590 2: something we no longer hold a reference on
12591 so we need to copy it locally. */
12592 /* Note we need to use SvCUR(), rather than
12593 SvLEN(), on our mother_re, because it, in
12594 turn, may well be pointing to its own mother_re. */
12595 SvPV_set(dstr, SAVEPVN(SvPVX_const(ret->mother_re),
12596 SvCUR(ret->mother_re)+1));
12597 SvLEN_set(dstr, SvCUR(ret->mother_re)+1);
12599 ret->mother_re = NULL;
12603 #endif /* PERL_IN_XSUB_RE */
12608 This is the internal complement to regdupe() which is used to copy
12609 the structure pointed to by the *pprivate pointer in the regexp.
12610 This is the core version of the extension overridable cloning hook.
12611 The regexp structure being duplicated will be copied by perl prior
12612 to this and will be provided as the regexp *r argument, however
12613 with the /old/ structures pprivate pointer value. Thus this routine
12614 may override any copying normally done by perl.
12616 It returns a pointer to the new regexp_internal structure.
12620 Perl_regdupe_internal(pTHX_ REGEXP * const rx, CLONE_PARAMS *param)
12623 struct regexp *const r = (struct regexp *)SvANY(rx);
12624 regexp_internal *reti;
12626 RXi_GET_DECL(r,ri);
12628 PERL_ARGS_ASSERT_REGDUPE_INTERNAL;
12632 Newxc(reti, sizeof(regexp_internal) + len*sizeof(regnode), char, regexp_internal);
12633 Copy(ri->program, reti->program, len+1, regnode);
12636 reti->regstclass = NULL;
12639 struct reg_data *d;
12640 const int count = ri->data->count;
12643 Newxc(d, sizeof(struct reg_data) + count*sizeof(void *),
12644 char, struct reg_data);
12645 Newx(d->what, count, U8);
12648 for (i = 0; i < count; i++) {
12649 d->what[i] = ri->data->what[i];
12650 switch (d->what[i]) {
12651 /* legal options are one of: sSfpontTua
12652 see also regcomp.h and pregfree() */
12653 case 'a': /* actually an AV, but the dup function is identical. */
12656 case 'p': /* actually an AV, but the dup function is identical. */
12657 case 'u': /* actually an HV, but the dup function is identical. */
12658 d->data[i] = sv_dup_inc((const SV *)ri->data->data[i], param);
12661 /* This is cheating. */
12662 Newx(d->data[i], 1, struct regnode_charclass_class);
12663 StructCopy(ri->data->data[i], d->data[i],
12664 struct regnode_charclass_class);
12665 reti->regstclass = (regnode*)d->data[i];
12668 /* Compiled op trees are readonly and in shared memory,
12669 and can thus be shared without duplication. */
12671 d->data[i] = (void*)OpREFCNT_inc((OP*)ri->data->data[i]);
12675 /* Trie stclasses are readonly and can thus be shared
12676 * without duplication. We free the stclass in pregfree
12677 * when the corresponding reg_ac_data struct is freed.
12679 reti->regstclass= ri->regstclass;
12683 ((reg_trie_data*)ri->data->data[i])->refcount++;
12687 d->data[i] = ri->data->data[i];
12690 Perl_croak(aTHX_ "panic: re_dup unknown data code '%c'", ri->data->what[i]);
12699 reti->name_list_idx = ri->name_list_idx;
12701 #ifdef RE_TRACK_PATTERN_OFFSETS
12702 if (ri->u.offsets) {
12703 Newx(reti->u.offsets, 2*len+1, U32);
12704 Copy(ri->u.offsets, reti->u.offsets, 2*len+1, U32);
12707 SetProgLen(reti,len);
12710 return (void*)reti;
12713 #endif /* USE_ITHREADS */
12715 #ifndef PERL_IN_XSUB_RE
12718 - regnext - dig the "next" pointer out of a node
12721 Perl_regnext(pTHX_ register regnode *p)
12724 register I32 offset;
12729 if (OP(p) > REGNODE_MAX) { /* regnode.type is unsigned */
12730 Perl_croak(aTHX_ "Corrupted regexp opcode %d > %d", (int)OP(p), (int)REGNODE_MAX);
12733 offset = (reg_off_by_arg[OP(p)] ? ARG(p) : NEXT_OFF(p));
12742 S_re_croak2(pTHX_ const char* pat1,const char* pat2,...)
12745 STRLEN l1 = strlen(pat1);
12746 STRLEN l2 = strlen(pat2);
12749 const char *message;
12751 PERL_ARGS_ASSERT_RE_CROAK2;
12757 Copy(pat1, buf, l1 , char);
12758 Copy(pat2, buf + l1, l2 , char);
12759 buf[l1 + l2] = '\n';
12760 buf[l1 + l2 + 1] = '\0';
12762 /* ANSI variant takes additional second argument */
12763 va_start(args, pat2);
12767 msv = vmess(buf, &args);
12769 message = SvPV_const(msv,l1);
12772 Copy(message, buf, l1 , char);
12773 buf[l1-1] = '\0'; /* Overwrite \n */
12774 Perl_croak(aTHX_ "%s", buf);
12777 /* XXX Here's a total kludge. But we need to re-enter for swash routines. */
12779 #ifndef PERL_IN_XSUB_RE
12781 Perl_save_re_context(pTHX)
12785 struct re_save_state *state;
12787 SAVEVPTR(PL_curcop);
12788 SSGROW(SAVESTACK_ALLOC_FOR_RE_SAVE_STATE + 1);
12790 state = (struct re_save_state *)(PL_savestack + PL_savestack_ix);
12791 PL_savestack_ix += SAVESTACK_ALLOC_FOR_RE_SAVE_STATE;
12792 SSPUSHUV(SAVEt_RE_STATE);
12794 Copy(&PL_reg_state, state, 1, struct re_save_state);
12796 PL_reg_start_tmp = 0;
12797 PL_reg_start_tmpl = 0;
12798 PL_reg_oldsaved = NULL;
12799 PL_reg_oldsavedlen = 0;
12800 PL_reg_maxiter = 0;
12801 PL_reg_leftiter = 0;
12802 PL_reg_poscache = NULL;
12803 PL_reg_poscache_size = 0;
12804 #ifdef PERL_OLD_COPY_ON_WRITE
12808 /* Save $1..$n (#18107: UTF-8 s/(\w+)/uc($1)/e); AMS 20021106. */
12810 const REGEXP * const rx = PM_GETRE(PL_curpm);
12813 for (i = 1; i <= RX_NPARENS(rx); i++) {
12814 char digits[TYPE_CHARS(long)];
12815 const STRLEN len = my_snprintf(digits, sizeof(digits), "%lu", (long)i);
12816 GV *const *const gvp
12817 = (GV**)hv_fetch(PL_defstash, digits, len, 0);
12820 GV * const gv = *gvp;
12821 if (SvTYPE(gv) == SVt_PVGV && GvSV(gv))
12831 clear_re(pTHX_ void *r)
12834 ReREFCNT_dec((REGEXP *)r);
12840 S_put_byte(pTHX_ SV *sv, int c)
12842 PERL_ARGS_ASSERT_PUT_BYTE;
12844 /* Our definition of isPRINT() ignores locales, so only bytes that are
12845 not part of UTF-8 are considered printable. I assume that the same
12846 holds for UTF-EBCDIC.
12847 Also, code point 255 is not printable in either (it's E0 in EBCDIC,
12848 which Wikipedia says:
12850 EO, or Eight Ones, is an 8-bit EBCDIC character code represented as all
12851 ones (binary 1111 1111, hexadecimal FF). It is similar, but not
12852 identical, to the ASCII delete (DEL) or rubout control character.
12853 ) So the old condition can be simplified to !isPRINT(c) */
12856 Perl_sv_catpvf(aTHX_ sv, "\\x%02x", c);
12859 Perl_sv_catpvf(aTHX_ sv, "\\x{%x}", c);
12863 const char string = c;
12864 if (c == '-' || c == ']' || c == '\\' || c == '^')
12865 sv_catpvs(sv, "\\");
12866 sv_catpvn(sv, &string, 1);
12871 #define CLEAR_OPTSTART \
12872 if (optstart) STMT_START { \
12873 DEBUG_OPTIMISE_r(PerlIO_printf(Perl_debug_log, " (%"IVdf" nodes)\n", (IV)(node - optstart))); \
12877 #define DUMPUNTIL(b,e) CLEAR_OPTSTART; node=dumpuntil(r,start,(b),(e),last,sv,indent+1,depth+1);
12879 STATIC const regnode *
12880 S_dumpuntil(pTHX_ const regexp *r, const regnode *start, const regnode *node,
12881 const regnode *last, const regnode *plast,
12882 SV* sv, I32 indent, U32 depth)
12885 register U8 op = PSEUDO; /* Arbitrary non-END op. */
12886 register const regnode *next;
12887 const regnode *optstart= NULL;
12889 RXi_GET_DECL(r,ri);
12890 GET_RE_DEBUG_FLAGS_DECL;
12892 PERL_ARGS_ASSERT_DUMPUNTIL;
12894 #ifdef DEBUG_DUMPUNTIL
12895 PerlIO_printf(Perl_debug_log, "--- %d : %d - %d - %d\n",indent,node-start,
12896 last ? last-start : 0,plast ? plast-start : 0);
12899 if (plast && plast < last)
12902 while (PL_regkind[op] != END && (!last || node < last)) {
12903 /* While that wasn't END last time... */
12906 if (op == CLOSE || op == WHILEM)
12908 next = regnext((regnode *)node);
12911 if (OP(node) == OPTIMIZED) {
12912 if (!optstart && RE_DEBUG_FLAG(RE_DEBUG_COMPILE_OPTIMISE))
12919 regprop(r, sv, node);
12920 PerlIO_printf(Perl_debug_log, "%4"IVdf":%*s%s", (IV)(node - start),
12921 (int)(2*indent + 1), "", SvPVX_const(sv));
12923 if (OP(node) != OPTIMIZED) {
12924 if (next == NULL) /* Next ptr. */
12925 PerlIO_printf(Perl_debug_log, " (0)");
12926 else if (PL_regkind[(U8)op] == BRANCH && PL_regkind[OP(next)] != BRANCH )
12927 PerlIO_printf(Perl_debug_log, " (FAIL)");
12929 PerlIO_printf(Perl_debug_log, " (%"IVdf")", (IV)(next - start));
12930 (void)PerlIO_putc(Perl_debug_log, '\n');
12934 if (PL_regkind[(U8)op] == BRANCHJ) {
12937 register const regnode *nnode = (OP(next) == LONGJMP
12938 ? regnext((regnode *)next)
12940 if (last && nnode > last)
12942 DUMPUNTIL(NEXTOPER(NEXTOPER(node)), nnode);
12945 else if (PL_regkind[(U8)op] == BRANCH) {
12947 DUMPUNTIL(NEXTOPER(node), next);
12949 else if ( PL_regkind[(U8)op] == TRIE ) {
12950 const regnode *this_trie = node;
12951 const char op = OP(node);
12952 const U32 n = ARG(node);
12953 const reg_ac_data * const ac = op>=AHOCORASICK ?
12954 (reg_ac_data *)ri->data->data[n] :
12956 const reg_trie_data * const trie =
12957 (reg_trie_data*)ri->data->data[op<AHOCORASICK ? n : ac->trie];
12959 AV *const trie_words = MUTABLE_AV(ri->data->data[n + TRIE_WORDS_OFFSET]);
12961 const regnode *nextbranch= NULL;
12964 for (word_idx= 0; word_idx < (I32)trie->wordcount; word_idx++) {
12965 SV ** const elem_ptr = av_fetch(trie_words,word_idx,0);
12967 PerlIO_printf(Perl_debug_log, "%*s%s ",
12968 (int)(2*(indent+3)), "",
12969 elem_ptr ? pv_pretty(sv, SvPV_nolen_const(*elem_ptr), SvCUR(*elem_ptr), 60,
12970 PL_colors[0], PL_colors[1],
12971 (SvUTF8(*elem_ptr) ? PERL_PV_ESCAPE_UNI : 0) |
12972 PERL_PV_PRETTY_ELLIPSES |
12973 PERL_PV_PRETTY_LTGT
12978 U16 dist= trie->jump[word_idx+1];
12979 PerlIO_printf(Perl_debug_log, "(%"UVuf")\n",
12980 (UV)((dist ? this_trie + dist : next) - start));
12983 nextbranch= this_trie + trie->jump[0];
12984 DUMPUNTIL(this_trie + dist, nextbranch);
12986 if (nextbranch && PL_regkind[OP(nextbranch)]==BRANCH)
12987 nextbranch= regnext((regnode *)nextbranch);
12989 PerlIO_printf(Perl_debug_log, "\n");
12992 if (last && next > last)
12997 else if ( op == CURLY ) { /* "next" might be very big: optimizer */
12998 DUMPUNTIL(NEXTOPER(node) + EXTRA_STEP_2ARGS,
12999 NEXTOPER(node) + EXTRA_STEP_2ARGS + 1);
13001 else if (PL_regkind[(U8)op] == CURLY && op != CURLYX) {
13003 DUMPUNTIL(NEXTOPER(node) + EXTRA_STEP_2ARGS, next);
13005 else if ( op == PLUS || op == STAR) {
13006 DUMPUNTIL(NEXTOPER(node), NEXTOPER(node) + 1);
13008 else if (PL_regkind[(U8)op] == ANYOF) {
13009 /* arglen 1 + class block */
13010 node += 1 + ((ANYOF_FLAGS(node) & ANYOF_CLASS)
13011 ? ANYOF_CLASS_SKIP : ANYOF_SKIP);
13012 node = NEXTOPER(node);
13014 else if (PL_regkind[(U8)op] == EXACT) {
13015 /* Literal string, where present. */
13016 node += NODE_SZ_STR(node) - 1;
13017 node = NEXTOPER(node);
13020 node = NEXTOPER(node);
13021 node += regarglen[(U8)op];
13023 if (op == CURLYX || op == OPEN)
13027 #ifdef DEBUG_DUMPUNTIL
13028 PerlIO_printf(Perl_debug_log, "--- %d\n", (int)indent);
13033 #endif /* DEBUGGING */
13037 * c-indentation-style: bsd
13038 * c-basic-offset: 4
13039 * indent-tabs-mode: t
13042 * ex: set ts=8 sts=4 sw=4 noet: