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
84 EXTERN_C const struct regexp_engine my_reg_engine;
89 #include "dquote_static.c"
90 #include "charclass_invlists.h"
91 #include "inline_invlist.c"
92 #include "unicode_constants.h"
94 #define HAS_NONLATIN1_FOLD_CLOSURE(i) \
95 _HAS_NONLATIN1_FOLD_CLOSURE_ONLY_FOR_USE_BY_REGCOMP_DOT_C_AND_REGEXEC_DOT_C(i)
96 #define IS_NON_FINAL_FOLD(c) _IS_NON_FINAL_FOLD_ONLY_FOR_USE_BY_REGCOMP_DOT_C(c)
97 #define IS_IN_SOME_FOLD_L1(c) _IS_IN_SOME_FOLD_ONLY_FOR_USE_BY_REGCOMP_DOT_C(c)
100 #define STATIC static
104 struct RExC_state_t {
105 U32 flags; /* RXf_* are we folding, multilining? */
106 U32 pm_flags; /* PMf_* stuff from the calling PMOP */
107 char *precomp; /* uncompiled string. */
108 REGEXP *rx_sv; /* The SV that is the regexp. */
109 regexp *rx; /* perl core regexp structure */
110 regexp_internal *rxi; /* internal data for regexp object
112 char *start; /* Start of input for compile */
113 char *end; /* End of input for compile */
114 char *parse; /* Input-scan pointer. */
115 SSize_t whilem_seen; /* number of WHILEM in this expr */
116 regnode *emit_start; /* Start of emitted-code area */
117 regnode *emit_bound; /* First regnode outside of the
119 regnode *emit; /* Code-emit pointer; if = &emit_dummy,
120 implies compiling, so don't emit */
121 regnode_ssc emit_dummy; /* placeholder for emit to point to;
122 large enough for the largest
123 non-EXACTish node, so can use it as
125 I32 naughty; /* How bad is this pattern? */
126 I32 sawback; /* Did we see \1, ...? */
128 SSize_t size; /* Code size. */
129 I32 npar; /* Capture buffer count, (OPEN) plus
130 one. ("par" 0 is the whole
132 I32 nestroot; /* root parens we are in - used by
136 regnode **open_parens; /* pointers to open parens */
137 regnode **close_parens; /* pointers to close parens */
138 regnode *opend; /* END node in program */
139 I32 utf8; /* whether the pattern is utf8 or not */
140 I32 orig_utf8; /* whether the pattern was originally in utf8 */
141 /* XXX use this for future optimisation of case
142 * where pattern must be upgraded to utf8. */
143 I32 uni_semantics; /* If a d charset modifier should use unicode
144 rules, even if the pattern is not in
146 HV *paren_names; /* Paren names */
148 regnode **recurse; /* Recurse regops */
149 I32 recurse_count; /* Number of recurse regops */
150 U8 *study_chunk_recursed; /* bitmap of which parens we have moved
152 U32 study_chunk_recursed_bytes; /* bytes in bitmap */
156 I32 override_recoding;
157 I32 in_multi_char_class;
158 struct reg_code_block *code_blocks; /* positions of literal (?{})
160 int num_code_blocks; /* size of code_blocks[] */
161 int code_index; /* next code_blocks[] slot */
162 SSize_t maxlen; /* mininum possible number of chars in string to match */
163 #ifdef ADD_TO_REGEXEC
164 char *starttry; /* -Dr: where regtry was called. */
165 #define RExC_starttry (pRExC_state->starttry)
167 SV *runtime_code_qr; /* qr with the runtime code blocks */
169 const char *lastparse;
171 AV *paren_name_list; /* idx -> name */
172 #define RExC_lastparse (pRExC_state->lastparse)
173 #define RExC_lastnum (pRExC_state->lastnum)
174 #define RExC_paren_name_list (pRExC_state->paren_name_list)
178 #define RExC_flags (pRExC_state->flags)
179 #define RExC_pm_flags (pRExC_state->pm_flags)
180 #define RExC_precomp (pRExC_state->precomp)
181 #define RExC_rx_sv (pRExC_state->rx_sv)
182 #define RExC_rx (pRExC_state->rx)
183 #define RExC_rxi (pRExC_state->rxi)
184 #define RExC_start (pRExC_state->start)
185 #define RExC_end (pRExC_state->end)
186 #define RExC_parse (pRExC_state->parse)
187 #define RExC_whilem_seen (pRExC_state->whilem_seen)
188 #ifdef RE_TRACK_PATTERN_OFFSETS
189 #define RExC_offsets (pRExC_state->rxi->u.offsets) /* I am not like the
192 #define RExC_emit (pRExC_state->emit)
193 #define RExC_emit_dummy (pRExC_state->emit_dummy)
194 #define RExC_emit_start (pRExC_state->emit_start)
195 #define RExC_emit_bound (pRExC_state->emit_bound)
196 #define RExC_naughty (pRExC_state->naughty)
197 #define RExC_sawback (pRExC_state->sawback)
198 #define RExC_seen (pRExC_state->seen)
199 #define RExC_size (pRExC_state->size)
200 #define RExC_maxlen (pRExC_state->maxlen)
201 #define RExC_npar (pRExC_state->npar)
202 #define RExC_nestroot (pRExC_state->nestroot)
203 #define RExC_extralen (pRExC_state->extralen)
204 #define RExC_seen_zerolen (pRExC_state->seen_zerolen)
205 #define RExC_utf8 (pRExC_state->utf8)
206 #define RExC_uni_semantics (pRExC_state->uni_semantics)
207 #define RExC_orig_utf8 (pRExC_state->orig_utf8)
208 #define RExC_open_parens (pRExC_state->open_parens)
209 #define RExC_close_parens (pRExC_state->close_parens)
210 #define RExC_opend (pRExC_state->opend)
211 #define RExC_paren_names (pRExC_state->paren_names)
212 #define RExC_recurse (pRExC_state->recurse)
213 #define RExC_recurse_count (pRExC_state->recurse_count)
214 #define RExC_study_chunk_recursed (pRExC_state->study_chunk_recursed)
215 #define RExC_study_chunk_recursed_bytes \
216 (pRExC_state->study_chunk_recursed_bytes)
217 #define RExC_in_lookbehind (pRExC_state->in_lookbehind)
218 #define RExC_contains_locale (pRExC_state->contains_locale)
219 #define RExC_contains_i (pRExC_state->contains_i)
220 #define RExC_override_recoding (pRExC_state->override_recoding)
221 #define RExC_in_multi_char_class (pRExC_state->in_multi_char_class)
224 #define ISMULT1(c) ((c) == '*' || (c) == '+' || (c) == '?')
225 #define ISMULT2(s) ((*s) == '*' || (*s) == '+' || (*s) == '?' || \
226 ((*s) == '{' && regcurly(s, FALSE)))
229 * Flags to be passed up and down.
231 #define WORST 0 /* Worst case. */
232 #define HASWIDTH 0x01 /* Known to match non-null strings. */
234 /* Simple enough to be STAR/PLUS operand; in an EXACTish node must be a single
235 * character. (There needs to be a case: in the switch statement in regexec.c
236 * for any node marked SIMPLE.) Note that this is not the same thing as
239 #define SPSTART 0x04 /* Starts with * or + */
240 #define POSTPONED 0x08 /* (?1),(?&name), (??{...}) or similar */
241 #define TRYAGAIN 0x10 /* Weeded out a declaration. */
242 #define RESTART_UTF8 0x20 /* Restart, need to calcuate sizes as UTF-8 */
244 #define REG_NODE_NUM(x) ((x) ? (int)((x)-RExC_emit_start) : -1)
246 /* whether trie related optimizations are enabled */
247 #if PERL_ENABLE_EXTENDED_TRIE_OPTIMISATION
248 #define TRIE_STUDY_OPT
249 #define FULL_TRIE_STUDY
255 #define PBYTE(u8str,paren) ((U8*)(u8str))[(paren) >> 3]
256 #define PBITVAL(paren) (1 << ((paren) & 7))
257 #define PAREN_TEST(u8str,paren) ( PBYTE(u8str,paren) & PBITVAL(paren))
258 #define PAREN_SET(u8str,paren) PBYTE(u8str,paren) |= PBITVAL(paren)
259 #define PAREN_UNSET(u8str,paren) PBYTE(u8str,paren) &= (~PBITVAL(paren))
261 #define REQUIRE_UTF8 STMT_START { \
263 *flagp = RESTART_UTF8; \
268 /* This converts the named class defined in regcomp.h to its equivalent class
269 * number defined in handy.h. */
270 #define namedclass_to_classnum(class) ((int) ((class) / 2))
271 #define classnum_to_namedclass(classnum) ((classnum) * 2)
273 #define _invlist_union_complement_2nd(a, b, output) \
274 _invlist_union_maybe_complement_2nd(a, b, TRUE, output)
275 #define _invlist_intersection_complement_2nd(a, b, output) \
276 _invlist_intersection_maybe_complement_2nd(a, b, TRUE, output)
278 /* About scan_data_t.
280 During optimisation we recurse through the regexp program performing
281 various inplace (keyhole style) optimisations. In addition study_chunk
282 and scan_commit populate this data structure with information about
283 what strings MUST appear in the pattern. We look for the longest
284 string that must appear at a fixed location, and we look for the
285 longest string that may appear at a floating location. So for instance
290 Both 'FOO' and 'A' are fixed strings. Both 'B' and 'BAR' are floating
291 strings (because they follow a .* construct). study_chunk will identify
292 both FOO and BAR as being the longest fixed and floating strings respectively.
294 The strings can be composites, for instance
298 will result in a composite fixed substring 'foo'.
300 For each string some basic information is maintained:
302 - offset or min_offset
303 This is the position the string must appear at, or not before.
304 It also implicitly (when combined with minlenp) tells us how many
305 characters must match before the string we are searching for.
306 Likewise when combined with minlenp and the length of the string it
307 tells us how many characters must appear after the string we have
311 Only used for floating strings. This is the rightmost point that
312 the string can appear at. If set to SSize_t_MAX it indicates that the
313 string can occur infinitely far to the right.
316 A pointer to the minimum number of characters of the pattern that the
317 string was found inside. This is important as in the case of positive
318 lookahead or positive lookbehind we can have multiple patterns
323 The minimum length of the pattern overall is 3, the minimum length
324 of the lookahead part is 3, but the minimum length of the part that
325 will actually match is 1. So 'FOO's minimum length is 3, but the
326 minimum length for the F is 1. This is important as the minimum length
327 is used to determine offsets in front of and behind the string being
328 looked for. Since strings can be composites this is the length of the
329 pattern at the time it was committed with a scan_commit. Note that
330 the length is calculated by study_chunk, so that the minimum lengths
331 are not known until the full pattern has been compiled, thus the
332 pointer to the value.
336 In the case of lookbehind the string being searched for can be
337 offset past the start point of the final matching string.
338 If this value was just blithely removed from the min_offset it would
339 invalidate some of the calculations for how many chars must match
340 before or after (as they are derived from min_offset and minlen and
341 the length of the string being searched for).
342 When the final pattern is compiled and the data is moved from the
343 scan_data_t structure into the regexp structure the information
344 about lookbehind is factored in, with the information that would
345 have been lost precalculated in the end_shift field for the
348 The fields pos_min and pos_delta are used to store the minimum offset
349 and the delta to the maximum offset at the current point in the pattern.
353 typedef struct scan_data_t {
354 /*I32 len_min; unused */
355 /*I32 len_delta; unused */
359 SSize_t last_end; /* min value, <0 unless valid. */
360 SSize_t last_start_min;
361 SSize_t last_start_max;
362 SV **longest; /* Either &l_fixed, or &l_float. */
363 SV *longest_fixed; /* longest fixed string found in pattern */
364 SSize_t offset_fixed; /* offset where it starts */
365 SSize_t *minlen_fixed; /* pointer to the minlen relevant to the string */
366 I32 lookbehind_fixed; /* is the position of the string modfied by LB */
367 SV *longest_float; /* longest floating string found in pattern */
368 SSize_t offset_float_min; /* earliest point in string it can appear */
369 SSize_t offset_float_max; /* latest point in string it can appear */
370 SSize_t *minlen_float; /* pointer to the minlen relevant to the string */
371 SSize_t lookbehind_float; /* is the pos of the string modified by LB */
374 SSize_t *last_closep;
375 regnode_ssc *start_class;
378 /* The below is perhaps overboard, but this allows us to save a test at the
379 * expense of a mask. This is because on both EBCDIC and ASCII machines, 'A'
380 * and 'a' differ by a single bit; the same with the upper and lower case of
381 * all other ASCII-range alphabetics. On ASCII platforms, they are 32 apart;
382 * on EBCDIC, they are 64. This uses an exclusive 'or' to find that bit and
383 * then inverts it to form a mask, with just a single 0, in the bit position
384 * where the upper- and lowercase differ. XXX There are about 40 other
385 * instances in the Perl core where this micro-optimization could be used.
386 * Should decide if maintenance cost is worse, before changing those
388 * Returns a boolean as to whether or not 'v' is either a lowercase or
389 * uppercase instance of 'c', where 'c' is in [A-Za-z]. If 'c' is a
390 * compile-time constant, the generated code is better than some optimizing
391 * compilers figure out, amounting to a mask and test. The results are
392 * meaningless if 'c' is not one of [A-Za-z] */
393 #define isARG2_lower_or_UPPER_ARG1(c, v) \
394 (((v) & ~('A' ^ 'a')) == ((c) & ~('A' ^ 'a')))
397 * Forward declarations for pregcomp()'s friends.
400 static const scan_data_t zero_scan_data =
401 { 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0 ,0};
403 #define SF_BEFORE_EOL (SF_BEFORE_SEOL|SF_BEFORE_MEOL)
404 #define SF_BEFORE_SEOL 0x0001
405 #define SF_BEFORE_MEOL 0x0002
406 #define SF_FIX_BEFORE_EOL (SF_FIX_BEFORE_SEOL|SF_FIX_BEFORE_MEOL)
407 #define SF_FL_BEFORE_EOL (SF_FL_BEFORE_SEOL|SF_FL_BEFORE_MEOL)
409 #define SF_FIX_SHIFT_EOL (+2)
410 #define SF_FL_SHIFT_EOL (+4)
412 #define SF_FIX_BEFORE_SEOL (SF_BEFORE_SEOL << SF_FIX_SHIFT_EOL)
413 #define SF_FIX_BEFORE_MEOL (SF_BEFORE_MEOL << SF_FIX_SHIFT_EOL)
415 #define SF_FL_BEFORE_SEOL (SF_BEFORE_SEOL << SF_FL_SHIFT_EOL)
416 #define SF_FL_BEFORE_MEOL (SF_BEFORE_MEOL << SF_FL_SHIFT_EOL) /* 0x20 */
417 #define SF_IS_INF 0x0040
418 #define SF_HAS_PAR 0x0080
419 #define SF_IN_PAR 0x0100
420 #define SF_HAS_EVAL 0x0200
421 #define SCF_DO_SUBSTR 0x0400
422 #define SCF_DO_STCLASS_AND 0x0800
423 #define SCF_DO_STCLASS_OR 0x1000
424 #define SCF_DO_STCLASS (SCF_DO_STCLASS_AND|SCF_DO_STCLASS_OR)
425 #define SCF_WHILEM_VISITED_POS 0x2000
427 #define SCF_TRIE_RESTUDY 0x4000 /* Do restudy? */
428 #define SCF_SEEN_ACCEPT 0x8000
429 #define SCF_TRIE_DOING_RESTUDY 0x10000
431 #define UTF cBOOL(RExC_utf8)
433 /* The enums for all these are ordered so things work out correctly */
434 #define LOC (get_regex_charset(RExC_flags) == REGEX_LOCALE_CHARSET)
435 #define DEPENDS_SEMANTICS (get_regex_charset(RExC_flags) \
436 == REGEX_DEPENDS_CHARSET)
437 #define UNI_SEMANTICS (get_regex_charset(RExC_flags) == REGEX_UNICODE_CHARSET)
438 #define AT_LEAST_UNI_SEMANTICS (get_regex_charset(RExC_flags) \
439 >= REGEX_UNICODE_CHARSET)
440 #define ASCII_RESTRICTED (get_regex_charset(RExC_flags) \
441 == REGEX_ASCII_RESTRICTED_CHARSET)
442 #define AT_LEAST_ASCII_RESTRICTED (get_regex_charset(RExC_flags) \
443 >= REGEX_ASCII_RESTRICTED_CHARSET)
444 #define ASCII_FOLD_RESTRICTED (get_regex_charset(RExC_flags) \
445 == REGEX_ASCII_MORE_RESTRICTED_CHARSET)
447 #define FOLD cBOOL(RExC_flags & RXf_PMf_FOLD)
449 /* For programs that want to be strictly Unicode compatible by dying if any
450 * attempt is made to match a non-Unicode code point against a Unicode
452 #define ALWAYS_WARN_SUPER ckDEAD(packWARN(WARN_NON_UNICODE))
454 #define OOB_NAMEDCLASS -1
456 /* There is no code point that is out-of-bounds, so this is problematic. But
457 * its only current use is to initialize a variable that is always set before
459 #define OOB_UNICODE 0xDEADBEEF
461 #define CHR_SVLEN(sv) (UTF ? sv_len_utf8(sv) : SvCUR(sv))
462 #define CHR_DIST(a,b) (UTF ? utf8_distance(a,b) : a - b)
465 /* length of regex to show in messages that don't mark a position within */
466 #define RegexLengthToShowInErrorMessages 127
469 * If MARKER[12] are adjusted, be sure to adjust the constants at the top
470 * of t/op/regmesg.t, the tests in t/op/re_tests, and those in
471 * op/pragma/warn/regcomp.
473 #define MARKER1 "<-- HERE" /* marker as it appears in the description */
474 #define MARKER2 " <-- HERE " /* marker as it appears within the regex */
476 #define REPORT_LOCATION " in regex; marked by " MARKER1 \
477 " in m/%"UTF8f MARKER2 "%"UTF8f"/"
479 #define REPORT_LOCATION_ARGS(offset) \
480 UTF8fARG(UTF, offset, RExC_precomp), \
481 UTF8fARG(UTF, RExC_end - RExC_precomp - offset, RExC_precomp + offset)
484 * Calls SAVEDESTRUCTOR_X if needed, then calls Perl_croak with the given
485 * arg. Show regex, up to a maximum length. If it's too long, chop and add
488 #define _FAIL(code) STMT_START { \
489 const char *ellipses = ""; \
490 IV len = RExC_end - RExC_precomp; \
493 SAVEFREESV(RExC_rx_sv); \
494 if (len > RegexLengthToShowInErrorMessages) { \
495 /* chop 10 shorter than the max, to ensure meaning of "..." */ \
496 len = RegexLengthToShowInErrorMessages - 10; \
502 #define FAIL(msg) _FAIL( \
503 Perl_croak(aTHX_ "%s in regex m/%"UTF8f"%s/", \
504 msg, UTF8fARG(UTF, len, RExC_precomp), ellipses))
506 #define FAIL2(msg,arg) _FAIL( \
507 Perl_croak(aTHX_ msg " in regex m/%"UTF8f"%s/", \
508 arg, UTF8fARG(UTF, len, RExC_precomp), ellipses))
511 * Simple_vFAIL -- like FAIL, but marks the current location in the scan
513 #define Simple_vFAIL(m) STMT_START { \
514 const IV offset = RExC_parse - RExC_precomp; \
515 Perl_croak(aTHX_ "%s" REPORT_LOCATION, \
516 m, REPORT_LOCATION_ARGS(offset)); \
520 * Calls SAVEDESTRUCTOR_X if needed, then Simple_vFAIL()
522 #define vFAIL(m) STMT_START { \
524 SAVEFREESV(RExC_rx_sv); \
529 * Like Simple_vFAIL(), but accepts two arguments.
531 #define Simple_vFAIL2(m,a1) STMT_START { \
532 const IV offset = RExC_parse - RExC_precomp; \
533 S_re_croak2(aTHX_ UTF, m, REPORT_LOCATION, a1, \
534 REPORT_LOCATION_ARGS(offset)); \
538 * Calls SAVEDESTRUCTOR_X if needed, then Simple_vFAIL2().
540 #define vFAIL2(m,a1) STMT_START { \
542 SAVEFREESV(RExC_rx_sv); \
543 Simple_vFAIL2(m, a1); \
548 * Like Simple_vFAIL(), but accepts three arguments.
550 #define Simple_vFAIL3(m, a1, a2) STMT_START { \
551 const IV offset = RExC_parse - RExC_precomp; \
552 S_re_croak2(aTHX_ UTF, m, REPORT_LOCATION, a1, a2, \
553 REPORT_LOCATION_ARGS(offset)); \
557 * Calls SAVEDESTRUCTOR_X if needed, then Simple_vFAIL3().
559 #define vFAIL3(m,a1,a2) STMT_START { \
561 SAVEFREESV(RExC_rx_sv); \
562 Simple_vFAIL3(m, a1, a2); \
566 * Like Simple_vFAIL(), but accepts four arguments.
568 #define Simple_vFAIL4(m, a1, a2, a3) STMT_START { \
569 const IV offset = RExC_parse - RExC_precomp; \
570 S_re_croak2(aTHX_ UTF, m, REPORT_LOCATION, a1, a2, a3, \
571 REPORT_LOCATION_ARGS(offset)); \
574 #define vFAIL4(m,a1,a2,a3) STMT_START { \
576 SAVEFREESV(RExC_rx_sv); \
577 Simple_vFAIL4(m, a1, a2, a3); \
580 /* A specialized version of vFAIL2 that works with UTF8f */
581 #define vFAIL2utf8f(m, a1) STMT_START { \
582 const IV offset = RExC_parse - RExC_precomp; \
584 SAVEFREESV(RExC_rx_sv); \
585 S_re_croak2(aTHX_ UTF, m, REPORT_LOCATION, a1, \
586 REPORT_LOCATION_ARGS(offset)); \
590 /* m is not necessarily a "literal string", in this macro */
591 #define reg_warn_non_literal_string(loc, m) STMT_START { \
592 const IV offset = loc - RExC_precomp; \
593 Perl_warner(aTHX_ packWARN(WARN_REGEXP), "%s" REPORT_LOCATION, \
594 m, REPORT_LOCATION_ARGS(offset)); \
597 #define ckWARNreg(loc,m) STMT_START { \
598 const IV offset = loc - RExC_precomp; \
599 Perl_ck_warner(aTHX_ packWARN(WARN_REGEXP), m REPORT_LOCATION, \
600 REPORT_LOCATION_ARGS(offset)); \
603 #define vWARN_dep(loc, m) STMT_START { \
604 const IV offset = loc - RExC_precomp; \
605 Perl_warner(aTHX_ packWARN(WARN_DEPRECATED), m REPORT_LOCATION, \
606 REPORT_LOCATION_ARGS(offset)); \
609 #define ckWARNdep(loc,m) STMT_START { \
610 const IV offset = loc - RExC_precomp; \
611 Perl_ck_warner_d(aTHX_ packWARN(WARN_DEPRECATED), \
613 REPORT_LOCATION_ARGS(offset)); \
616 #define ckWARNregdep(loc,m) STMT_START { \
617 const IV offset = loc - RExC_precomp; \
618 Perl_ck_warner_d(aTHX_ packWARN2(WARN_DEPRECATED, WARN_REGEXP), \
620 REPORT_LOCATION_ARGS(offset)); \
623 #define ckWARN2reg_d(loc,m, a1) STMT_START { \
624 const IV offset = loc - RExC_precomp; \
625 Perl_ck_warner_d(aTHX_ packWARN(WARN_REGEXP), \
627 a1, REPORT_LOCATION_ARGS(offset)); \
630 #define ckWARN2reg(loc, m, a1) STMT_START { \
631 const IV offset = loc - RExC_precomp; \
632 Perl_ck_warner(aTHX_ packWARN(WARN_REGEXP), m REPORT_LOCATION, \
633 a1, REPORT_LOCATION_ARGS(offset)); \
636 #define vWARN3(loc, m, a1, a2) STMT_START { \
637 const IV offset = loc - RExC_precomp; \
638 Perl_warner(aTHX_ packWARN(WARN_REGEXP), m REPORT_LOCATION, \
639 a1, a2, REPORT_LOCATION_ARGS(offset)); \
642 #define ckWARN3reg(loc, m, a1, a2) STMT_START { \
643 const IV offset = loc - RExC_precomp; \
644 Perl_ck_warner(aTHX_ packWARN(WARN_REGEXP), m REPORT_LOCATION, \
645 a1, a2, REPORT_LOCATION_ARGS(offset)); \
648 #define vWARN4(loc, m, a1, a2, a3) STMT_START { \
649 const IV offset = loc - RExC_precomp; \
650 Perl_warner(aTHX_ packWARN(WARN_REGEXP), m REPORT_LOCATION, \
651 a1, a2, a3, REPORT_LOCATION_ARGS(offset)); \
654 #define ckWARN4reg(loc, m, a1, a2, a3) STMT_START { \
655 const IV offset = loc - RExC_precomp; \
656 Perl_ck_warner(aTHX_ packWARN(WARN_REGEXP), m REPORT_LOCATION, \
657 a1, a2, a3, REPORT_LOCATION_ARGS(offset)); \
660 #define vWARN5(loc, m, a1, a2, a3, a4) STMT_START { \
661 const IV offset = loc - RExC_precomp; \
662 Perl_warner(aTHX_ packWARN(WARN_REGEXP), m REPORT_LOCATION, \
663 a1, a2, a3, a4, REPORT_LOCATION_ARGS(offset)); \
667 /* Allow for side effects in s */
668 #define REGC(c,s) STMT_START { \
669 if (!SIZE_ONLY) *(s) = (c); else (void)(s); \
672 /* Macros for recording node offsets. 20001227 mjd@plover.com
673 * Nodes are numbered 1, 2, 3, 4. Node #n's position is recorded in
674 * element 2*n-1 of the array. Element #2n holds the byte length node #n.
675 * Element 0 holds the number n.
676 * Position is 1 indexed.
678 #ifndef RE_TRACK_PATTERN_OFFSETS
679 #define Set_Node_Offset_To_R(node,byte)
680 #define Set_Node_Offset(node,byte)
681 #define Set_Cur_Node_Offset
682 #define Set_Node_Length_To_R(node,len)
683 #define Set_Node_Length(node,len)
684 #define Set_Node_Cur_Length(node,start)
685 #define Node_Offset(n)
686 #define Node_Length(n)
687 #define Set_Node_Offset_Length(node,offset,len)
688 #define ProgLen(ri) ri->u.proglen
689 #define SetProgLen(ri,x) ri->u.proglen = x
691 #define ProgLen(ri) ri->u.offsets[0]
692 #define SetProgLen(ri,x) ri->u.offsets[0] = x
693 #define Set_Node_Offset_To_R(node,byte) STMT_START { \
695 MJD_OFFSET_DEBUG(("** (%d) offset of node %d is %d.\n", \
696 __LINE__, (int)(node), (int)(byte))); \
698 Perl_croak(aTHX_ "value of node is %d in Offset macro", \
701 RExC_offsets[2*(node)-1] = (byte); \
706 #define Set_Node_Offset(node,byte) \
707 Set_Node_Offset_To_R((node)-RExC_emit_start, (byte)-RExC_start)
708 #define Set_Cur_Node_Offset Set_Node_Offset(RExC_emit, RExC_parse)
710 #define Set_Node_Length_To_R(node,len) STMT_START { \
712 MJD_OFFSET_DEBUG(("** (%d) size of node %d is %d.\n", \
713 __LINE__, (int)(node), (int)(len))); \
715 Perl_croak(aTHX_ "value of node is %d in Length macro", \
718 RExC_offsets[2*(node)] = (len); \
723 #define Set_Node_Length(node,len) \
724 Set_Node_Length_To_R((node)-RExC_emit_start, len)
725 #define Set_Node_Cur_Length(node, start) \
726 Set_Node_Length(node, RExC_parse - start)
728 /* Get offsets and lengths */
729 #define Node_Offset(n) (RExC_offsets[2*((n)-RExC_emit_start)-1])
730 #define Node_Length(n) (RExC_offsets[2*((n)-RExC_emit_start)])
732 #define Set_Node_Offset_Length(node,offset,len) STMT_START { \
733 Set_Node_Offset_To_R((node)-RExC_emit_start, (offset)); \
734 Set_Node_Length_To_R((node)-RExC_emit_start, (len)); \
738 #if PERL_ENABLE_EXPERIMENTAL_REGEX_OPTIMISATIONS
739 #define EXPERIMENTAL_INPLACESCAN
740 #endif /*PERL_ENABLE_EXPERIMENTAL_REGEX_OPTIMISATIONS*/
742 #define DEBUG_RExC_seen() \
743 DEBUG_OPTIMISE_MORE_r({ \
744 PerlIO_printf(Perl_debug_log,"RExC_seen: "); \
746 if (RExC_seen & REG_ZERO_LEN_SEEN) \
747 PerlIO_printf(Perl_debug_log,"REG_ZERO_LEN_SEEN "); \
749 if (RExC_seen & REG_LOOKBEHIND_SEEN) \
750 PerlIO_printf(Perl_debug_log,"REG_LOOKBEHIND_SEEN "); \
752 if (RExC_seen & REG_GPOS_SEEN) \
753 PerlIO_printf(Perl_debug_log,"REG_GPOS_SEEN "); \
755 if (RExC_seen & REG_CANY_SEEN) \
756 PerlIO_printf(Perl_debug_log,"REG_CANY_SEEN "); \
758 if (RExC_seen & REG_RECURSE_SEEN) \
759 PerlIO_printf(Perl_debug_log,"REG_RECURSE_SEEN "); \
761 if (RExC_seen & REG_TOP_LEVEL_BRANCHES_SEEN) \
762 PerlIO_printf(Perl_debug_log,"REG_TOP_LEVEL_BRANCHES_SEEN "); \
764 if (RExC_seen & REG_VERBARG_SEEN) \
765 PerlIO_printf(Perl_debug_log,"REG_VERBARG_SEEN "); \
767 if (RExC_seen & REG_CUTGROUP_SEEN) \
768 PerlIO_printf(Perl_debug_log,"REG_CUTGROUP_SEEN "); \
770 if (RExC_seen & REG_RUN_ON_COMMENT_SEEN) \
771 PerlIO_printf(Perl_debug_log,"REG_RUN_ON_COMMENT_SEEN "); \
773 if (RExC_seen & REG_UNFOLDED_MULTI_SEEN) \
774 PerlIO_printf(Perl_debug_log,"REG_UNFOLDED_MULTI_SEEN "); \
776 if (RExC_seen & REG_GOSTART_SEEN) \
777 PerlIO_printf(Perl_debug_log,"REG_GOSTART_SEEN "); \
779 if (RExC_seen & REG_UNBOUNDED_QUANTIFIER_SEEN) \
780 PerlIO_printf(Perl_debug_log,"REG_UNBOUNDED_QUANTIFIER_SEEN "); \
782 PerlIO_printf(Perl_debug_log,"\n"); \
785 #define DEBUG_STUDYDATA(str,data,depth) \
786 DEBUG_OPTIMISE_MORE_r(if(data){ \
787 PerlIO_printf(Perl_debug_log, \
788 "%*s" str "Pos:%"IVdf"/%"IVdf \
789 " Flags: 0x%"UVXf" Whilem_c: %"IVdf" Lcp: %"IVdf" %s", \
790 (int)(depth)*2, "", \
791 (IV)((data)->pos_min), \
792 (IV)((data)->pos_delta), \
793 (UV)((data)->flags), \
794 (IV)((data)->whilem_c), \
795 (IV)((data)->last_closep ? *((data)->last_closep) : -1), \
796 is_inf ? "INF " : "" \
798 if ((data)->last_found) \
799 PerlIO_printf(Perl_debug_log, \
800 "Last:'%s' %"IVdf":%"IVdf"/%"IVdf" %sFixed:'%s' @ %"IVdf \
801 " %sFloat: '%s' @ %"IVdf"/%"IVdf"", \
802 SvPVX_const((data)->last_found), \
803 (IV)((data)->last_end), \
804 (IV)((data)->last_start_min), \
805 (IV)((data)->last_start_max), \
806 ((data)->longest && \
807 (data)->longest==&((data)->longest_fixed)) ? "*" : "", \
808 SvPVX_const((data)->longest_fixed), \
809 (IV)((data)->offset_fixed), \
810 ((data)->longest && \
811 (data)->longest==&((data)->longest_float)) ? "*" : "", \
812 SvPVX_const((data)->longest_float), \
813 (IV)((data)->offset_float_min), \
814 (IV)((data)->offset_float_max) \
816 PerlIO_printf(Perl_debug_log,"\n"); \
819 /* Mark that we cannot extend a found fixed substring at this point.
820 Update the longest found anchored substring and the longest found
821 floating substrings if needed. */
824 S_scan_commit(pTHX_ const RExC_state_t *pRExC_state, scan_data_t *data,
825 SSize_t *minlenp, int is_inf)
827 const STRLEN l = CHR_SVLEN(data->last_found);
828 const STRLEN old_l = CHR_SVLEN(*data->longest);
829 GET_RE_DEBUG_FLAGS_DECL;
831 PERL_ARGS_ASSERT_SCAN_COMMIT;
833 if ((l >= old_l) && ((l > old_l) || (data->flags & SF_BEFORE_EOL))) {
834 SvSetMagicSV(*data->longest, data->last_found);
835 if (*data->longest == data->longest_fixed) {
836 data->offset_fixed = l ? data->last_start_min : data->pos_min;
837 if (data->flags & SF_BEFORE_EOL)
839 |= ((data->flags & SF_BEFORE_EOL) << SF_FIX_SHIFT_EOL);
841 data->flags &= ~SF_FIX_BEFORE_EOL;
842 data->minlen_fixed=minlenp;
843 data->lookbehind_fixed=0;
845 else { /* *data->longest == data->longest_float */
846 data->offset_float_min = l ? data->last_start_min : data->pos_min;
847 data->offset_float_max = (l
848 ? data->last_start_max
849 : (data->pos_delta == SSize_t_MAX
851 : data->pos_min + data->pos_delta));
853 || (STRLEN)data->offset_float_max > (STRLEN)SSize_t_MAX)
854 data->offset_float_max = SSize_t_MAX;
855 if (data->flags & SF_BEFORE_EOL)
857 |= ((data->flags & SF_BEFORE_EOL) << SF_FL_SHIFT_EOL);
859 data->flags &= ~SF_FL_BEFORE_EOL;
860 data->minlen_float=minlenp;
861 data->lookbehind_float=0;
864 SvCUR_set(data->last_found, 0);
866 SV * const sv = data->last_found;
867 if (SvUTF8(sv) && SvMAGICAL(sv)) {
868 MAGIC * const mg = mg_find(sv, PERL_MAGIC_utf8);
874 data->flags &= ~SF_BEFORE_EOL;
875 DEBUG_STUDYDATA("commit: ",data,0);
878 /* An SSC is just a regnode_charclass_posix with an extra field: the inversion
879 * list that describes which code points it matches */
882 S_ssc_anything(pTHX_ regnode_ssc *ssc)
884 /* Set the SSC 'ssc' to match an empty string or any code point */
886 PERL_ARGS_ASSERT_SSC_ANYTHING;
888 assert(is_ANYOF_SYNTHETIC(ssc));
890 ssc->invlist = sv_2mortal(_new_invlist(2)); /* mortalize so won't leak */
891 _append_range_to_invlist(ssc->invlist, 0, UV_MAX);
892 ANYOF_FLAGS(ssc) |= ANYOF_EMPTY_STRING; /* Plus match empty string */
896 S_ssc_is_anything(pTHX_ const regnode_ssc *ssc)
898 /* Returns TRUE if the SSC 'ssc' can match the empty string and any code
899 * point; FALSE otherwise. Thus, this is used to see if using 'ssc' buys
900 * us anything: if the function returns TRUE, 'ssc' hasn't been restricted
901 * in any way, so there's no point in using it */
906 PERL_ARGS_ASSERT_SSC_IS_ANYTHING;
908 assert(is_ANYOF_SYNTHETIC(ssc));
910 if (! (ANYOF_FLAGS(ssc) & ANYOF_EMPTY_STRING)) {
914 /* See if the list consists solely of the range 0 - Infinity */
915 invlist_iterinit(ssc->invlist);
916 ret = invlist_iternext(ssc->invlist, &start, &end)
920 invlist_iterfinish(ssc->invlist);
926 /* If e.g., both \w and \W are set, matches everything */
927 if (ANYOF_POSIXL_SSC_TEST_ANY_SET(ssc)) {
929 for (i = 0; i < ANYOF_POSIXL_MAX; i += 2) {
930 if (ANYOF_POSIXL_TEST(ssc, i) && ANYOF_POSIXL_TEST(ssc, i+1)) {
940 S_ssc_init(pTHX_ const RExC_state_t *pRExC_state, regnode_ssc *ssc)
942 /* Initializes the SSC 'ssc'. This includes setting it to match an empty
943 * string, any code point, or any posix class under locale */
945 PERL_ARGS_ASSERT_SSC_INIT;
947 Zero(ssc, 1, regnode_ssc);
948 set_ANYOF_SYNTHETIC(ssc);
949 ARG_SET(ssc, ANYOF_NONBITMAP_EMPTY);
952 /* If any portion of the regex is to operate under locale rules,
953 * initialization includes it. The reason this isn't done for all regexes
954 * is that the optimizer was written under the assumption that locale was
955 * all-or-nothing. Given the complexity and lack of documentation in the
956 * optimizer, and that there are inadequate test cases for locale, many
957 * parts of it may not work properly, it is safest to avoid locale unless
959 if (RExC_contains_locale) {
960 ANYOF_POSIXL_SETALL(ssc);
963 ANYOF_POSIXL_ZERO(ssc);
968 S_ssc_is_cp_posixl_init(pTHX_ const RExC_state_t *pRExC_state,
969 const regnode_ssc *ssc)
971 /* Returns TRUE if the SSC 'ssc' is in its initial state with regard only
972 * to the list of code points matched, and locale posix classes; hence does
973 * not check its flags) */
978 PERL_ARGS_ASSERT_SSC_IS_CP_POSIXL_INIT;
980 assert(is_ANYOF_SYNTHETIC(ssc));
982 invlist_iterinit(ssc->invlist);
983 ret = invlist_iternext(ssc->invlist, &start, &end)
987 invlist_iterfinish(ssc->invlist);
993 if (RExC_contains_locale && ! ANYOF_POSIXL_SSC_TEST_ALL_SET(ssc)) {
1001 S_get_ANYOF_cp_list_for_ssc(pTHX_ const RExC_state_t *pRExC_state,
1002 const regnode_charclass* const node)
1004 /* Returns a mortal inversion list defining which code points are matched
1005 * by 'node', which is of type ANYOF. Handles complementing the result if
1006 * appropriate. If some code points aren't knowable at this time, the
1007 * returned list must, and will, contain every code point that is a
1010 SV* invlist = sv_2mortal(_new_invlist(0));
1011 SV* only_utf8_locale_invlist = NULL;
1013 const U32 n = ARG(node);
1014 bool new_node_has_latin1 = FALSE;
1016 PERL_ARGS_ASSERT_GET_ANYOF_CP_LIST_FOR_SSC;
1018 /* Look at the data structure created by S_set_ANYOF_arg() */
1019 if (n != ANYOF_NONBITMAP_EMPTY) {
1020 SV * const rv = MUTABLE_SV(RExC_rxi->data->data[n]);
1021 AV * const av = MUTABLE_AV(SvRV(rv));
1022 SV **const ary = AvARRAY(av);
1023 assert(RExC_rxi->data->what[n] == 's');
1025 if (ary[1] && ary[1] != &PL_sv_undef) { /* Has compile-time swash */
1026 invlist = sv_2mortal(invlist_clone(_get_swash_invlist(ary[1])));
1028 else if (ary[0] && ary[0] != &PL_sv_undef) {
1030 /* Here, no compile-time swash, and there are things that won't be
1031 * known until runtime -- we have to assume it could be anything */
1032 return _add_range_to_invlist(invlist, 0, UV_MAX);
1034 else if (ary[3] && ary[3] != &PL_sv_undef) {
1036 /* Here no compile-time swash, and no run-time only data. Use the
1037 * node's inversion list */
1038 invlist = sv_2mortal(invlist_clone(ary[3]));
1041 /* Get the code points valid only under UTF-8 locales */
1042 if ((ANYOF_FLAGS(node) & ANYOF_LOC_FOLD)
1043 && ary[2] && ary[2] != &PL_sv_undef)
1045 only_utf8_locale_invlist = ary[2];
1049 /* An ANYOF node contains a bitmap for the first 256 code points, and an
1050 * inversion list for the others, but if there are code points that should
1051 * match only conditionally on the target string being UTF-8, those are
1052 * placed in the inversion list, and not the bitmap. Since there are
1053 * circumstances under which they could match, they are included in the
1054 * SSC. But if the ANYOF node is to be inverted, we have to exclude them
1055 * here, so that when we invert below, the end result actually does include
1056 * them. (Think about "\xe0" =~ /[^\xc0]/di;). We have to do this here
1057 * before we add the unconditionally matched code points */
1058 if (ANYOF_FLAGS(node) & ANYOF_INVERT) {
1059 _invlist_intersection_complement_2nd(invlist,
1064 /* Add in the points from the bit map */
1065 for (i = 0; i < 256; i++) {
1066 if (ANYOF_BITMAP_TEST(node, i)) {
1067 invlist = add_cp_to_invlist(invlist, i);
1068 new_node_has_latin1 = TRUE;
1072 /* If this can match all upper Latin1 code points, have to add them
1074 if (ANYOF_FLAGS(node) & ANYOF_NON_UTF8_NON_ASCII_ALL) {
1075 _invlist_union(invlist, PL_UpperLatin1, &invlist);
1078 /* Similarly for these */
1079 if (ANYOF_FLAGS(node) & ANYOF_ABOVE_LATIN1_ALL) {
1080 invlist = _add_range_to_invlist(invlist, 256, UV_MAX);
1083 if (ANYOF_FLAGS(node) & ANYOF_INVERT) {
1084 _invlist_invert(invlist);
1086 else if (new_node_has_latin1 && ANYOF_FLAGS(node) & ANYOF_LOC_FOLD) {
1088 /* Under /li, any 0-255 could fold to any other 0-255, depending on the
1089 * locale. We can skip this if there are no 0-255 at all. */
1090 _invlist_union(invlist, PL_Latin1, &invlist);
1093 /* Similarly add the UTF-8 locale possible matches. These have to be
1094 * deferred until after the non-UTF-8 locale ones are taken care of just
1095 * above, or it leads to wrong results under ANYOF_INVERT */
1096 if (only_utf8_locale_invlist) {
1097 _invlist_union_maybe_complement_2nd(invlist,
1098 only_utf8_locale_invlist,
1099 ANYOF_FLAGS(node) & ANYOF_INVERT,
1106 /* These two functions currently do the exact same thing */
1107 #define ssc_init_zero ssc_init
1109 #define ssc_add_cp(ssc, cp) ssc_add_range((ssc), (cp), (cp))
1110 #define ssc_match_all_cp(ssc) ssc_add_range(ssc, 0, UV_MAX)
1112 /* 'AND' a given class with another one. Can create false positives. 'ssc'
1113 * should not be inverted. 'and_with->flags & ANYOF_POSIXL' should be 0 if
1114 * 'and_with' is a regnode_charclass instead of a regnode_ssc. */
1117 S_ssc_and(pTHX_ const RExC_state_t *pRExC_state, regnode_ssc *ssc,
1118 const regnode_charclass *and_with)
1120 /* Accumulate into SSC 'ssc' its 'AND' with 'and_with', which is either
1121 * another SSC or a regular ANYOF class. Can create false positives. */
1126 PERL_ARGS_ASSERT_SSC_AND;
1128 assert(is_ANYOF_SYNTHETIC(ssc));
1130 /* 'and_with' is used as-is if it too is an SSC; otherwise have to extract
1131 * the code point inversion list and just the relevant flags */
1132 if (is_ANYOF_SYNTHETIC(and_with)) {
1133 anded_cp_list = ((regnode_ssc *)and_with)->invlist;
1134 anded_flags = ANYOF_FLAGS(and_with);
1136 /* XXX This is a kludge around what appears to be deficiencies in the
1137 * optimizer. If we make S_ssc_anything() add in the WARN_SUPER flag,
1138 * there are paths through the optimizer where it doesn't get weeded
1139 * out when it should. And if we don't make some extra provision for
1140 * it like the code just below, it doesn't get added when it should.
1141 * This solution is to add it only when AND'ing, which is here, and
1142 * only when what is being AND'ed is the pristine, original node
1143 * matching anything. Thus it is like adding it to ssc_anything() but
1144 * only when the result is to be AND'ed. Probably the same solution
1145 * could be adopted for the same problem we have with /l matching,
1146 * which is solved differently in S_ssc_init(), and that would lead to
1147 * fewer false positives than that solution has. But if this solution
1148 * creates bugs, the consequences are only that a warning isn't raised
1149 * that should be; while the consequences for having /l bugs is
1150 * incorrect matches */
1151 if (ssc_is_anything((regnode_ssc *)and_with)) {
1152 anded_flags |= ANYOF_WARN_SUPER;
1156 anded_cp_list = get_ANYOF_cp_list_for_ssc(pRExC_state, and_with);
1157 anded_flags = ANYOF_FLAGS(and_with) & ANYOF_COMMON_FLAGS;
1160 ANYOF_FLAGS(ssc) &= anded_flags;
1162 /* Below, C1 is the list of code points in 'ssc'; P1, its posix classes.
1163 * C2 is the list of code points in 'and-with'; P2, its posix classes.
1164 * 'and_with' may be inverted. When not inverted, we have the situation of
1166 * (C1 | P1) & (C2 | P2)
1167 * = (C1 & (C2 | P2)) | (P1 & (C2 | P2))
1168 * = ((C1 & C2) | (C1 & P2)) | ((P1 & C2) | (P1 & P2))
1169 * <= ((C1 & C2) | P2)) | ( P1 | (P1 & P2))
1170 * <= ((C1 & C2) | P1 | P2)
1171 * Alternatively, the last few steps could be:
1172 * = ((C1 & C2) | (C1 & P2)) | ((P1 & C2) | (P1 & P2))
1173 * <= ((C1 & C2) | C1 ) | ( C2 | (P1 & P2))
1174 * <= (C1 | C2 | (P1 & P2))
1175 * We favor the second approach if either P1 or P2 is non-empty. This is
1176 * because these components are a barrier to doing optimizations, as what
1177 * they match cannot be known until the moment of matching as they are
1178 * dependent on the current locale, 'AND"ing them likely will reduce or
1180 * But we can do better if we know that C1,P1 are in their initial state (a
1181 * frequent occurrence), each matching everything:
1182 * (<everything>) & (C2 | P2) = C2 | P2
1183 * Similarly, if C2,P2 are in their initial state (again a frequent
1184 * occurrence), the result is a no-op
1185 * (C1 | P1) & (<everything>) = C1 | P1
1188 * (C1 | P1) & ~(C2 | P2) = (C1 | P1) & (~C2 & ~P2)
1189 * = (C1 & (~C2 & ~P2)) | (P1 & (~C2 & ~P2))
1190 * <= (C1 & ~C2) | (P1 & ~P2)
1193 if ((ANYOF_FLAGS(and_with) & ANYOF_INVERT)
1194 && ! is_ANYOF_SYNTHETIC(and_with))
1198 ssc_intersection(ssc,
1200 FALSE /* Has already been inverted */
1203 /* If either P1 or P2 is empty, the intersection will be also; can skip
1205 if (! (ANYOF_FLAGS(and_with) & ANYOF_POSIXL)) {
1206 ANYOF_POSIXL_ZERO(ssc);
1208 else if (ANYOF_POSIXL_SSC_TEST_ANY_SET(ssc)) {
1210 /* Note that the Posix class component P from 'and_with' actually
1212 * P = Pa | Pb | ... | Pn
1213 * where each component is one posix class, such as in [\w\s].
1215 * ~P = ~(Pa | Pb | ... | Pn)
1216 * = ~Pa & ~Pb & ... & ~Pn
1217 * <= ~Pa | ~Pb | ... | ~Pn
1218 * The last is something we can easily calculate, but unfortunately
1219 * is likely to have many false positives. We could do better
1220 * in some (but certainly not all) instances if two classes in
1221 * P have known relationships. For example
1222 * :lower: <= :alpha: <= :alnum: <= \w <= :graph: <= :print:
1224 * :lower: & :print: = :lower:
1225 * And similarly for classes that must be disjoint. For example,
1226 * since \s and \w can have no elements in common based on rules in
1227 * the POSIX standard,
1228 * \w & ^\S = nothing
1229 * Unfortunately, some vendor locales do not meet the Posix
1230 * standard, in particular almost everything by Microsoft.
1231 * The loop below just changes e.g., \w into \W and vice versa */
1233 regnode_charclass_posixl temp;
1234 int add = 1; /* To calculate the index of the complement */
1236 ANYOF_POSIXL_ZERO(&temp);
1237 for (i = 0; i < ANYOF_MAX; i++) {
1239 || ! ANYOF_POSIXL_TEST((regnode_charclass_posixl*) and_with, i)
1240 || ! ANYOF_POSIXL_TEST((regnode_charclass_posixl*) and_with, i + 1));
1242 if (ANYOF_POSIXL_TEST((regnode_charclass_posixl*) and_with, i)) {
1243 ANYOF_POSIXL_SET(&temp, i + add);
1245 add = 0 - add; /* 1 goes to -1; -1 goes to 1 */
1247 ANYOF_POSIXL_AND(&temp, ssc);
1249 } /* else ssc already has no posixes */
1250 } /* else: Not inverted. This routine is a no-op if 'and_with' is an SSC
1251 in its initial state */
1252 else if (! is_ANYOF_SYNTHETIC(and_with)
1253 || ! ssc_is_cp_posixl_init(pRExC_state, (regnode_ssc *)and_with))
1255 /* But if 'ssc' is in its initial state, the result is just 'and_with';
1256 * copy it over 'ssc' */
1257 if (ssc_is_cp_posixl_init(pRExC_state, ssc)) {
1258 if (is_ANYOF_SYNTHETIC(and_with)) {
1259 StructCopy(and_with, ssc, regnode_ssc);
1262 ssc->invlist = anded_cp_list;
1263 ANYOF_POSIXL_ZERO(ssc);
1264 if (ANYOF_FLAGS(and_with) & ANYOF_POSIXL) {
1265 ANYOF_POSIXL_OR((regnode_charclass_posixl*) and_with, ssc);
1269 else if (ANYOF_POSIXL_SSC_TEST_ANY_SET(ssc)
1270 || (ANYOF_FLAGS(and_with) & ANYOF_POSIXL))
1272 /* One or the other of P1, P2 is non-empty. */
1273 if (ANYOF_FLAGS(and_with) & ANYOF_POSIXL) {
1274 ANYOF_POSIXL_AND((regnode_charclass_posixl*) and_with, ssc);
1276 ssc_union(ssc, anded_cp_list, FALSE);
1278 else { /* P1 = P2 = empty */
1279 ssc_intersection(ssc, anded_cp_list, FALSE);
1285 S_ssc_or(pTHX_ const RExC_state_t *pRExC_state, regnode_ssc *ssc,
1286 const regnode_charclass *or_with)
1288 /* Accumulate into SSC 'ssc' its 'OR' with 'or_with', which is either
1289 * another SSC or a regular ANYOF class. Can create false positives if
1290 * 'or_with' is to be inverted. */
1295 PERL_ARGS_ASSERT_SSC_OR;
1297 assert(is_ANYOF_SYNTHETIC(ssc));
1299 /* 'or_with' is used as-is if it too is an SSC; otherwise have to extract
1300 * the code point inversion list and just the relevant flags */
1301 if (is_ANYOF_SYNTHETIC(or_with)) {
1302 ored_cp_list = ((regnode_ssc*) or_with)->invlist;
1303 ored_flags = ANYOF_FLAGS(or_with);
1306 ored_cp_list = get_ANYOF_cp_list_for_ssc(pRExC_state, or_with);
1307 ored_flags = ANYOF_FLAGS(or_with) & ANYOF_COMMON_FLAGS;
1310 ANYOF_FLAGS(ssc) |= ored_flags;
1312 /* Below, C1 is the list of code points in 'ssc'; P1, its posix classes.
1313 * C2 is the list of code points in 'or-with'; P2, its posix classes.
1314 * 'or_with' may be inverted. When not inverted, we have the simple
1315 * situation of computing:
1316 * (C1 | P1) | (C2 | P2) = (C1 | C2) | (P1 | P2)
1317 * If P1|P2 yields a situation with both a class and its complement are
1318 * set, like having both \w and \W, this matches all code points, and we
1319 * can delete these from the P component of the ssc going forward. XXX We
1320 * might be able to delete all the P components, but I (khw) am not certain
1321 * about this, and it is better to be safe.
1324 * (C1 | P1) | ~(C2 | P2) = (C1 | P1) | (~C2 & ~P2)
1325 * <= (C1 | P1) | ~C2
1326 * <= (C1 | ~C2) | P1
1327 * (which results in actually simpler code than the non-inverted case)
1330 if ((ANYOF_FLAGS(or_with) & ANYOF_INVERT)
1331 && ! is_ANYOF_SYNTHETIC(or_with))
1333 /* We ignore P2, leaving P1 going forward */
1334 } /* else Not inverted */
1335 else if (ANYOF_FLAGS(or_with) & ANYOF_POSIXL) {
1336 ANYOF_POSIXL_OR((regnode_charclass_posixl*)or_with, ssc);
1337 if (ANYOF_POSIXL_SSC_TEST_ANY_SET(ssc)) {
1339 for (i = 0; i < ANYOF_MAX; i += 2) {
1340 if (ANYOF_POSIXL_TEST(ssc, i) && ANYOF_POSIXL_TEST(ssc, i + 1))
1342 ssc_match_all_cp(ssc);
1343 ANYOF_POSIXL_CLEAR(ssc, i);
1344 ANYOF_POSIXL_CLEAR(ssc, i+1);
1352 FALSE /* Already has been inverted */
1356 PERL_STATIC_INLINE void
1357 S_ssc_union(pTHX_ regnode_ssc *ssc, SV* const invlist, const bool invert2nd)
1359 PERL_ARGS_ASSERT_SSC_UNION;
1361 assert(is_ANYOF_SYNTHETIC(ssc));
1363 _invlist_union_maybe_complement_2nd(ssc->invlist,
1369 PERL_STATIC_INLINE void
1370 S_ssc_intersection(pTHX_ regnode_ssc *ssc,
1372 const bool invert2nd)
1374 PERL_ARGS_ASSERT_SSC_INTERSECTION;
1376 assert(is_ANYOF_SYNTHETIC(ssc));
1378 _invlist_intersection_maybe_complement_2nd(ssc->invlist,
1384 PERL_STATIC_INLINE void
1385 S_ssc_add_range(pTHX_ regnode_ssc *ssc, const UV start, const UV end)
1387 PERL_ARGS_ASSERT_SSC_ADD_RANGE;
1389 assert(is_ANYOF_SYNTHETIC(ssc));
1391 ssc->invlist = _add_range_to_invlist(ssc->invlist, start, end);
1394 PERL_STATIC_INLINE void
1395 S_ssc_cp_and(pTHX_ regnode_ssc *ssc, const UV cp)
1397 /* AND just the single code point 'cp' into the SSC 'ssc' */
1399 SV* cp_list = _new_invlist(2);
1401 PERL_ARGS_ASSERT_SSC_CP_AND;
1403 assert(is_ANYOF_SYNTHETIC(ssc));
1405 cp_list = add_cp_to_invlist(cp_list, cp);
1406 ssc_intersection(ssc, cp_list,
1407 FALSE /* Not inverted */
1409 SvREFCNT_dec_NN(cp_list);
1412 PERL_STATIC_INLINE void
1413 S_ssc_clear_locale(pTHX_ regnode_ssc *ssc)
1415 /* Set the SSC 'ssc' to not match any locale things */
1417 PERL_ARGS_ASSERT_SSC_CLEAR_LOCALE;
1419 assert(is_ANYOF_SYNTHETIC(ssc));
1421 ANYOF_POSIXL_ZERO(ssc);
1422 ANYOF_FLAGS(ssc) &= ~ANYOF_LOCALE_FLAGS;
1426 S_ssc_finalize(pTHX_ RExC_state_t *pRExC_state, regnode_ssc *ssc)
1428 /* The inversion list in the SSC is marked mortal; now we need a more
1429 * permanent copy, which is stored the same way that is done in a regular
1430 * ANYOF node, with the first 256 code points in a bit map */
1432 SV* invlist = invlist_clone(ssc->invlist);
1434 PERL_ARGS_ASSERT_SSC_FINALIZE;
1436 assert(is_ANYOF_SYNTHETIC(ssc));
1438 /* The code in this file assumes that all but these flags aren't relevant
1439 * to the SSC, except ANYOF_EMPTY_STRING, which should be cleared by the
1440 * time we reach here */
1441 assert(! (ANYOF_FLAGS(ssc) & ~ANYOF_COMMON_FLAGS));
1443 populate_ANYOF_from_invlist( (regnode *) ssc, &invlist);
1445 set_ANYOF_arg(pRExC_state, (regnode *) ssc, invlist,
1446 NULL, NULL, NULL, FALSE);
1448 /* Make sure is clone-safe */
1449 ssc->invlist = NULL;
1451 if (ANYOF_POSIXL_SSC_TEST_ANY_SET(ssc)) {
1452 ANYOF_FLAGS(ssc) |= ANYOF_POSIXL;
1455 assert(! (ANYOF_FLAGS(ssc) & ANYOF_LOCALE_FLAGS) || RExC_contains_locale);
1458 #define TRIE_LIST_ITEM(state,idx) (trie->states[state].trans.list)[ idx ]
1459 #define TRIE_LIST_CUR(state) ( TRIE_LIST_ITEM( state, 0 ).forid )
1460 #define TRIE_LIST_LEN(state) ( TRIE_LIST_ITEM( state, 0 ).newstate )
1461 #define TRIE_LIST_USED(idx) ( trie->states[state].trans.list \
1462 ? (TRIE_LIST_CUR( idx ) - 1) \
1468 dump_trie(trie,widecharmap,revcharmap)
1469 dump_trie_interim_list(trie,widecharmap,revcharmap,next_alloc)
1470 dump_trie_interim_table(trie,widecharmap,revcharmap,next_alloc)
1472 These routines dump out a trie in a somewhat readable format.
1473 The _interim_ variants are used for debugging the interim
1474 tables that are used to generate the final compressed
1475 representation which is what dump_trie expects.
1477 Part of the reason for their existence is to provide a form
1478 of documentation as to how the different representations function.
1483 Dumps the final compressed table form of the trie to Perl_debug_log.
1484 Used for debugging make_trie().
1488 S_dump_trie(pTHX_ const struct _reg_trie_data *trie, HV *widecharmap,
1489 AV *revcharmap, U32 depth)
1492 SV *sv=sv_newmortal();
1493 int colwidth= widecharmap ? 6 : 4;
1495 GET_RE_DEBUG_FLAGS_DECL;
1497 PERL_ARGS_ASSERT_DUMP_TRIE;
1499 PerlIO_printf( Perl_debug_log, "%*sChar : %-6s%-6s%-4s ",
1500 (int)depth * 2 + 2,"",
1501 "Match","Base","Ofs" );
1503 for( state = 0 ; state < trie->uniquecharcount ; state++ ) {
1504 SV ** const tmp = av_fetch( revcharmap, state, 0);
1506 PerlIO_printf( Perl_debug_log, "%*s",
1508 pv_pretty(sv, SvPV_nolen_const(*tmp), SvCUR(*tmp), colwidth,
1509 PL_colors[0], PL_colors[1],
1510 (SvUTF8(*tmp) ? PERL_PV_ESCAPE_UNI : 0) |
1511 PERL_PV_ESCAPE_FIRSTCHAR
1516 PerlIO_printf( Perl_debug_log, "\n%*sState|-----------------------",
1517 (int)depth * 2 + 2,"");
1519 for( state = 0 ; state < trie->uniquecharcount ; state++ )
1520 PerlIO_printf( Perl_debug_log, "%.*s", colwidth, "--------");
1521 PerlIO_printf( Perl_debug_log, "\n");
1523 for( state = 1 ; state < trie->statecount ; state++ ) {
1524 const U32 base = trie->states[ state ].trans.base;
1526 PerlIO_printf( Perl_debug_log, "%*s#%4"UVXf"|",
1527 (int)depth * 2 + 2,"", (UV)state);
1529 if ( trie->states[ state ].wordnum ) {
1530 PerlIO_printf( Perl_debug_log, " W%4X",
1531 trie->states[ state ].wordnum );
1533 PerlIO_printf( Perl_debug_log, "%6s", "" );
1536 PerlIO_printf( Perl_debug_log, " @%4"UVXf" ", (UV)base );
1541 while( ( base + ofs < trie->uniquecharcount ) ||
1542 ( base + ofs - trie->uniquecharcount < trie->lasttrans
1543 && trie->trans[ base + ofs - trie->uniquecharcount ].check
1547 PerlIO_printf( Perl_debug_log, "+%2"UVXf"[ ", (UV)ofs);
1549 for ( ofs = 0 ; ofs < trie->uniquecharcount ; ofs++ ) {
1550 if ( ( base + ofs >= trie->uniquecharcount )
1551 && ( base + ofs - trie->uniquecharcount
1553 && trie->trans[ base + ofs
1554 - trie->uniquecharcount ].check == state )
1556 PerlIO_printf( Perl_debug_log, "%*"UVXf,
1558 (UV)trie->trans[ base + ofs
1559 - trie->uniquecharcount ].next );
1561 PerlIO_printf( Perl_debug_log, "%*s",colwidth," ." );
1565 PerlIO_printf( Perl_debug_log, "]");
1568 PerlIO_printf( Perl_debug_log, "\n" );
1570 PerlIO_printf(Perl_debug_log, "%*sword_info N:(prev,len)=",
1572 for (word=1; word <= trie->wordcount; word++) {
1573 PerlIO_printf(Perl_debug_log, " %d:(%d,%d)",
1574 (int)word, (int)(trie->wordinfo[word].prev),
1575 (int)(trie->wordinfo[word].len));
1577 PerlIO_printf(Perl_debug_log, "\n" );
1580 Dumps a fully constructed but uncompressed trie in list form.
1581 List tries normally only are used for construction when the number of
1582 possible chars (trie->uniquecharcount) is very high.
1583 Used for debugging make_trie().
1586 S_dump_trie_interim_list(pTHX_ const struct _reg_trie_data *trie,
1587 HV *widecharmap, AV *revcharmap, U32 next_alloc,
1591 SV *sv=sv_newmortal();
1592 int colwidth= widecharmap ? 6 : 4;
1593 GET_RE_DEBUG_FLAGS_DECL;
1595 PERL_ARGS_ASSERT_DUMP_TRIE_INTERIM_LIST;
1597 /* print out the table precompression. */
1598 PerlIO_printf( Perl_debug_log, "%*sState :Word | Transition Data\n%*s%s",
1599 (int)depth * 2 + 2,"", (int)depth * 2 + 2,"",
1600 "------:-----+-----------------\n" );
1602 for( state=1 ; state < next_alloc ; state ++ ) {
1605 PerlIO_printf( Perl_debug_log, "%*s %4"UVXf" :",
1606 (int)depth * 2 + 2,"", (UV)state );
1607 if ( ! trie->states[ state ].wordnum ) {
1608 PerlIO_printf( Perl_debug_log, "%5s| ","");
1610 PerlIO_printf( Perl_debug_log, "W%4x| ",
1611 trie->states[ state ].wordnum
1614 for( charid = 1 ; charid <= TRIE_LIST_USED( state ) ; charid++ ) {
1615 SV ** const tmp = av_fetch( revcharmap,
1616 TRIE_LIST_ITEM(state,charid).forid, 0);
1618 PerlIO_printf( Perl_debug_log, "%*s:%3X=%4"UVXf" | ",
1620 pv_pretty(sv, SvPV_nolen_const(*tmp), SvCUR(*tmp),
1622 PL_colors[0], PL_colors[1],
1623 (SvUTF8(*tmp) ? PERL_PV_ESCAPE_UNI : 0)
1624 | PERL_PV_ESCAPE_FIRSTCHAR
1626 TRIE_LIST_ITEM(state,charid).forid,
1627 (UV)TRIE_LIST_ITEM(state,charid).newstate
1630 PerlIO_printf(Perl_debug_log, "\n%*s| ",
1631 (int)((depth * 2) + 14), "");
1634 PerlIO_printf( Perl_debug_log, "\n");
1639 Dumps a fully constructed but uncompressed trie in table form.
1640 This is the normal DFA style state transition table, with a few
1641 twists to facilitate compression later.
1642 Used for debugging make_trie().
1645 S_dump_trie_interim_table(pTHX_ const struct _reg_trie_data *trie,
1646 HV *widecharmap, AV *revcharmap, U32 next_alloc,
1651 SV *sv=sv_newmortal();
1652 int colwidth= widecharmap ? 6 : 4;
1653 GET_RE_DEBUG_FLAGS_DECL;
1655 PERL_ARGS_ASSERT_DUMP_TRIE_INTERIM_TABLE;
1658 print out the table precompression so that we can do a visual check
1659 that they are identical.
1662 PerlIO_printf( Perl_debug_log, "%*sChar : ",(int)depth * 2 + 2,"" );
1664 for( charid = 0 ; charid < trie->uniquecharcount ; charid++ ) {
1665 SV ** const tmp = av_fetch( revcharmap, charid, 0);
1667 PerlIO_printf( Perl_debug_log, "%*s",
1669 pv_pretty(sv, SvPV_nolen_const(*tmp), SvCUR(*tmp), colwidth,
1670 PL_colors[0], PL_colors[1],
1671 (SvUTF8(*tmp) ? PERL_PV_ESCAPE_UNI : 0) |
1672 PERL_PV_ESCAPE_FIRSTCHAR
1678 PerlIO_printf( Perl_debug_log, "\n%*sState+-",(int)depth * 2 + 2,"" );
1680 for( charid=0 ; charid < trie->uniquecharcount ; charid++ ) {
1681 PerlIO_printf( Perl_debug_log, "%.*s", colwidth,"--------");
1684 PerlIO_printf( Perl_debug_log, "\n" );
1686 for( state=1 ; state < next_alloc ; state += trie->uniquecharcount ) {
1688 PerlIO_printf( Perl_debug_log, "%*s%4"UVXf" : ",
1689 (int)depth * 2 + 2,"",
1690 (UV)TRIE_NODENUM( state ) );
1692 for( charid = 0 ; charid < trie->uniquecharcount ; charid++ ) {
1693 UV v=(UV)SAFE_TRIE_NODENUM( trie->trans[ state + charid ].next );
1695 PerlIO_printf( Perl_debug_log, "%*"UVXf, colwidth, v );
1697 PerlIO_printf( Perl_debug_log, "%*s", colwidth, "." );
1699 if ( ! trie->states[ TRIE_NODENUM( state ) ].wordnum ) {
1700 PerlIO_printf( Perl_debug_log, " (%4"UVXf")\n",
1701 (UV)trie->trans[ state ].check );
1703 PerlIO_printf( Perl_debug_log, " (%4"UVXf") W%4X\n",
1704 (UV)trie->trans[ state ].check,
1705 trie->states[ TRIE_NODENUM( state ) ].wordnum );
1713 /* make_trie(startbranch,first,last,tail,word_count,flags,depth)
1714 startbranch: the first branch in the whole branch sequence
1715 first : start branch of sequence of branch-exact nodes.
1716 May be the same as startbranch
1717 last : Thing following the last branch.
1718 May be the same as tail.
1719 tail : item following the branch sequence
1720 count : words in the sequence
1721 flags : currently the OP() type we will be building one of /EXACT(|F|Fl)/
1722 depth : indent depth
1724 Inplace optimizes a sequence of 2 or more Branch-Exact nodes into a TRIE node.
1726 A trie is an N'ary tree where the branches are determined by digital
1727 decomposition of the key. IE, at the root node you look up the 1st character and
1728 follow that branch repeat until you find the end of the branches. Nodes can be
1729 marked as "accepting" meaning they represent a complete word. Eg:
1733 would convert into the following structure. Numbers represent states, letters
1734 following numbers represent valid transitions on the letter from that state, if
1735 the number is in square brackets it represents an accepting state, otherwise it
1736 will be in parenthesis.
1738 +-h->+-e->[3]-+-r->(8)-+-s->[9]
1742 (1) +-i->(6)-+-s->[7]
1744 +-s->(3)-+-h->(4)-+-e->[5]
1746 Accept Word Mapping: 3=>1 (he),5=>2 (she), 7=>3 (his), 9=>4 (hers)
1748 This shows that when matching against the string 'hers' we will begin at state 1
1749 read 'h' and move to state 2, read 'e' and move to state 3 which is accepting,
1750 then read 'r' and go to state 8 followed by 's' which takes us to state 9 which
1751 is also accepting. Thus we know that we can match both 'he' and 'hers' with a
1752 single traverse. We store a mapping from accepting to state to which word was
1753 matched, and then when we have multiple possibilities we try to complete the
1754 rest of the regex in the order in which they occured in the alternation.
1756 The only prior NFA like behaviour that would be changed by the TRIE support is
1757 the silent ignoring of duplicate alternations which are of the form:
1759 / (DUPE|DUPE) X? (?{ ... }) Y /x
1761 Thus EVAL blocks following a trie may be called a different number of times with
1762 and without the optimisation. With the optimisations dupes will be silently
1763 ignored. This inconsistent behaviour of EVAL type nodes is well established as
1764 the following demonstrates:
1766 'words'=~/(word|word|word)(?{ print $1 })[xyz]/
1768 which prints out 'word' three times, but
1770 'words'=~/(word|word|word)(?{ print $1 })S/
1772 which doesnt print it out at all. This is due to other optimisations kicking in.
1774 Example of what happens on a structural level:
1776 The regexp /(ac|ad|ab)+/ will produce the following debug output:
1778 1: CURLYM[1] {1,32767}(18)
1789 This would be optimizable with startbranch=5, first=5, last=16, tail=16
1790 and should turn into:
1792 1: CURLYM[1] {1,32767}(18)
1794 [Words:3 Chars Stored:6 Unique Chars:4 States:5 NCP:1]
1802 Cases where tail != last would be like /(?foo|bar)baz/:
1812 which would be optimizable with startbranch=1, first=1, last=7, tail=8
1813 and would end up looking like:
1816 [Words:2 Chars Stored:6 Unique Chars:5 States:7 NCP:1]
1823 d = uvchr_to_utf8_flags(d, uv, 0);
1825 is the recommended Unicode-aware way of saying
1830 #define TRIE_STORE_REVCHAR(val) \
1833 SV *zlopp = newSV(7); /* XXX: optimize me */ \
1834 unsigned char *flrbbbbb = (unsigned char *) SvPVX(zlopp); \
1835 unsigned const char *const kapow = uvchr_to_utf8(flrbbbbb, val); \
1836 SvCUR_set(zlopp, kapow - flrbbbbb); \
1839 av_push(revcharmap, zlopp); \
1841 char ooooff = (char)val; \
1842 av_push(revcharmap, newSVpvn(&ooooff, 1)); \
1846 /* This gets the next character from the input, folding it if not already
1848 #define TRIE_READ_CHAR STMT_START { \
1851 /* if it is UTF then it is either already folded, or does not need \
1853 uvc = valid_utf8_to_uvchr( (const U8*) uc, &len); \
1855 else if (folder == PL_fold_latin1) { \
1856 /* This folder implies Unicode rules, which in the range expressible \
1857 * by not UTF is the lower case, with the two exceptions, one of \
1858 * which should have been taken care of before calling this */ \
1859 assert(*uc != LATIN_SMALL_LETTER_SHARP_S); \
1860 uvc = toLOWER_L1(*uc); \
1861 if (UNLIKELY(uvc == MICRO_SIGN)) uvc = GREEK_SMALL_LETTER_MU; \
1864 /* raw data, will be folded later if needed */ \
1872 #define TRIE_LIST_PUSH(state,fid,ns) STMT_START { \
1873 if ( TRIE_LIST_CUR( state ) >=TRIE_LIST_LEN( state ) ) { \
1874 U32 ging = TRIE_LIST_LEN( state ) *= 2; \
1875 Renew( trie->states[ state ].trans.list, ging, reg_trie_trans_le ); \
1877 TRIE_LIST_ITEM( state, TRIE_LIST_CUR( state ) ).forid = fid; \
1878 TRIE_LIST_ITEM( state, TRIE_LIST_CUR( state ) ).newstate = ns; \
1879 TRIE_LIST_CUR( state )++; \
1882 #define TRIE_LIST_NEW(state) STMT_START { \
1883 Newxz( trie->states[ state ].trans.list, \
1884 4, reg_trie_trans_le ); \
1885 TRIE_LIST_CUR( state ) = 1; \
1886 TRIE_LIST_LEN( state ) = 4; \
1889 #define TRIE_HANDLE_WORD(state) STMT_START { \
1890 U16 dupe= trie->states[ state ].wordnum; \
1891 regnode * const noper_next = regnext( noper ); \
1894 /* store the word for dumping */ \
1896 if (OP(noper) != NOTHING) \
1897 tmp = newSVpvn_utf8(STRING(noper), STR_LEN(noper), UTF); \
1899 tmp = newSVpvn_utf8( "", 0, UTF ); \
1900 av_push( trie_words, tmp ); \
1904 trie->wordinfo[curword].prev = 0; \
1905 trie->wordinfo[curword].len = wordlen; \
1906 trie->wordinfo[curword].accept = state; \
1908 if ( noper_next < tail ) { \
1910 trie->jump = (U16 *) PerlMemShared_calloc( word_count + 1, \
1912 trie->jump[curword] = (U16)(noper_next - convert); \
1914 jumper = noper_next; \
1916 nextbranch= regnext(cur); \
1920 /* It's a dupe. Pre-insert into the wordinfo[].prev */\
1921 /* chain, so that when the bits of chain are later */\
1922 /* linked together, the dups appear in the chain */\
1923 trie->wordinfo[curword].prev = trie->wordinfo[dupe].prev; \
1924 trie->wordinfo[dupe].prev = curword; \
1926 /* we haven't inserted this word yet. */ \
1927 trie->states[ state ].wordnum = curword; \
1932 #define TRIE_TRANS_STATE(state,base,ucharcount,charid,special) \
1933 ( ( base + charid >= ucharcount \
1934 && base + charid < ubound \
1935 && state == trie->trans[ base - ucharcount + charid ].check \
1936 && trie->trans[ base - ucharcount + charid ].next ) \
1937 ? trie->trans[ base - ucharcount + charid ].next \
1938 : ( state==1 ? special : 0 ) \
1942 #define MADE_JUMP_TRIE 2
1943 #define MADE_EXACT_TRIE 4
1946 S_make_trie(pTHX_ RExC_state_t *pRExC_state, regnode *startbranch,
1947 regnode *first, regnode *last, regnode *tail,
1948 U32 word_count, U32 flags, U32 depth)
1951 /* first pass, loop through and scan words */
1952 reg_trie_data *trie;
1953 HV *widecharmap = NULL;
1954 AV *revcharmap = newAV();
1960 regnode *jumper = NULL;
1961 regnode *nextbranch = NULL;
1962 regnode *convert = NULL;
1963 U32 *prev_states; /* temp array mapping each state to previous one */
1964 /* we just use folder as a flag in utf8 */
1965 const U8 * folder = NULL;
1968 const U32 data_slot = add_data( pRExC_state, STR_WITH_LEN("tuuu"));
1969 AV *trie_words = NULL;
1970 /* along with revcharmap, this only used during construction but both are
1971 * useful during debugging so we store them in the struct when debugging.
1974 const U32 data_slot = add_data( pRExC_state, STR_WITH_LEN("tu"));
1975 STRLEN trie_charcount=0;
1977 SV *re_trie_maxbuff;
1978 GET_RE_DEBUG_FLAGS_DECL;
1980 PERL_ARGS_ASSERT_MAKE_TRIE;
1982 PERL_UNUSED_ARG(depth);
1989 case EXACTFU: folder = PL_fold_latin1; break;
1990 case EXACTF: folder = PL_fold; break;
1991 default: Perl_croak( aTHX_ "panic! In trie construction, unknown node type %u %s", (unsigned) flags, PL_reg_name[flags] );
1994 trie = (reg_trie_data *) PerlMemShared_calloc( 1, sizeof(reg_trie_data) );
1996 trie->startstate = 1;
1997 trie->wordcount = word_count;
1998 RExC_rxi->data->data[ data_slot ] = (void*)trie;
1999 trie->charmap = (U16 *) PerlMemShared_calloc( 256, sizeof(U16) );
2001 trie->bitmap = (char *) PerlMemShared_calloc( ANYOF_BITMAP_SIZE, 1 );
2002 trie->wordinfo = (reg_trie_wordinfo *) PerlMemShared_calloc(
2003 trie->wordcount+1, sizeof(reg_trie_wordinfo));
2006 trie_words = newAV();
2009 re_trie_maxbuff = get_sv(RE_TRIE_MAXBUF_NAME, 1);
2010 if (!SvIOK(re_trie_maxbuff)) {
2011 sv_setiv(re_trie_maxbuff, RE_TRIE_MAXBUF_INIT);
2013 DEBUG_TRIE_COMPILE_r({
2014 PerlIO_printf( Perl_debug_log,
2015 "%*smake_trie start==%d, first==%d, last==%d, tail==%d depth=%d\n",
2016 (int)depth * 2 + 2, "",
2017 REG_NODE_NUM(startbranch),REG_NODE_NUM(first),
2018 REG_NODE_NUM(last), REG_NODE_NUM(tail), (int)depth);
2021 /* Find the node we are going to overwrite */
2022 if ( first == startbranch && OP( last ) != BRANCH ) {
2023 /* whole branch chain */
2026 /* branch sub-chain */
2027 convert = NEXTOPER( first );
2030 /* -- First loop and Setup --
2032 We first traverse the branches and scan each word to determine if it
2033 contains widechars, and how many unique chars there are, this is
2034 important as we have to build a table with at least as many columns as we
2037 We use an array of integers to represent the character codes 0..255
2038 (trie->charmap) and we use a an HV* to store Unicode characters. We use
2039 the native representation of the character value as the key and IV's for
2042 *TODO* If we keep track of how many times each character is used we can
2043 remap the columns so that the table compression later on is more
2044 efficient in terms of memory by ensuring the most common value is in the
2045 middle and the least common are on the outside. IMO this would be better
2046 than a most to least common mapping as theres a decent chance the most
2047 common letter will share a node with the least common, meaning the node
2048 will not be compressible. With a middle is most common approach the worst
2049 case is when we have the least common nodes twice.
2053 for ( cur = first ; cur < last ; cur = regnext( cur ) ) {
2054 regnode *noper = NEXTOPER( cur );
2055 const U8 *uc = (U8*)STRING( noper );
2056 const U8 *e = uc + STR_LEN( noper );
2058 U32 wordlen = 0; /* required init */
2059 STRLEN minchars = 0;
2060 STRLEN maxchars = 0;
2061 bool set_bit = trie->bitmap ? 1 : 0; /*store the first char in the
2064 if (OP(noper) == NOTHING) {
2065 regnode *noper_next= regnext(noper);
2066 if (noper_next != tail && OP(noper_next) == flags) {
2068 uc= (U8*)STRING(noper);
2069 e= uc + STR_LEN(noper);
2070 trie->minlen= STR_LEN(noper);
2077 if ( set_bit ) { /* bitmap only alloced when !(UTF&&Folding) */
2078 TRIE_BITMAP_SET(trie,*uc); /* store the raw first byte
2079 regardless of encoding */
2080 if (OP( noper ) == EXACTFU_SS) {
2081 /* false positives are ok, so just set this */
2082 TRIE_BITMAP_SET(trie, LATIN_SMALL_LETTER_SHARP_S);
2085 for ( ; uc < e ; uc += len ) { /* Look at each char in the current
2087 TRIE_CHARCOUNT(trie)++;
2090 /* TRIE_READ_CHAR returns the current character, or its fold if /i
2091 * is in effect. Under /i, this character can match itself, or
2092 * anything that folds to it. If not under /i, it can match just
2093 * itself. Most folds are 1-1, for example k, K, and KELVIN SIGN
2094 * all fold to k, and all are single characters. But some folds
2095 * expand to more than one character, so for example LATIN SMALL
2096 * LIGATURE FFI folds to the three character sequence 'ffi'. If
2097 * the string beginning at 'uc' is 'ffi', it could be matched by
2098 * three characters, or just by the one ligature character. (It
2099 * could also be matched by two characters: LATIN SMALL LIGATURE FF
2100 * followed by 'i', or by 'f' followed by LATIN SMALL LIGATURE FI).
2101 * (Of course 'I' and/or 'F' instead of 'i' and 'f' can also
2102 * match.) The trie needs to know the minimum and maximum number
2103 * of characters that could match so that it can use size alone to
2104 * quickly reject many match attempts. The max is simple: it is
2105 * the number of folded characters in this branch (since a fold is
2106 * never shorter than what folds to it. */
2110 /* And the min is equal to the max if not under /i (indicated by
2111 * 'folder' being NULL), or there are no multi-character folds. If
2112 * there is a multi-character fold, the min is incremented just
2113 * once, for the character that folds to the sequence. Each
2114 * character in the sequence needs to be added to the list below of
2115 * characters in the trie, but we count only the first towards the
2116 * min number of characters needed. This is done through the
2117 * variable 'foldlen', which is returned by the macros that look
2118 * for these sequences as the number of bytes the sequence
2119 * occupies. Each time through the loop, we decrement 'foldlen' by
2120 * how many bytes the current char occupies. Only when it reaches
2121 * 0 do we increment 'minchars' or look for another multi-character
2123 if (folder == NULL) {
2126 else if (foldlen > 0) {
2127 foldlen -= (UTF) ? UTF8SKIP(uc) : 1;
2132 /* See if *uc is the beginning of a multi-character fold. If
2133 * so, we decrement the length remaining to look at, to account
2134 * for the current character this iteration. (We can use 'uc'
2135 * instead of the fold returned by TRIE_READ_CHAR because for
2136 * non-UTF, the latin1_safe macro is smart enough to account
2137 * for all the unfolded characters, and because for UTF, the
2138 * string will already have been folded earlier in the
2139 * compilation process */
2141 if ((foldlen = is_MULTI_CHAR_FOLD_utf8_safe(uc, e))) {
2142 foldlen -= UTF8SKIP(uc);
2145 else if ((foldlen = is_MULTI_CHAR_FOLD_latin1_safe(uc, e))) {
2150 /* The current character (and any potential folds) should be added
2151 * to the possible matching characters for this position in this
2155 U8 folded= folder[ (U8) uvc ];
2156 if ( !trie->charmap[ folded ] ) {
2157 trie->charmap[ folded ]=( ++trie->uniquecharcount );
2158 TRIE_STORE_REVCHAR( folded );
2161 if ( !trie->charmap[ uvc ] ) {
2162 trie->charmap[ uvc ]=( ++trie->uniquecharcount );
2163 TRIE_STORE_REVCHAR( uvc );
2166 /* store the codepoint in the bitmap, and its folded
2168 TRIE_BITMAP_SET(trie, uvc);
2170 /* store the folded codepoint */
2171 if ( folder ) TRIE_BITMAP_SET(trie, folder[(U8) uvc ]);
2174 /* store first byte of utf8 representation of
2175 variant codepoints */
2176 if (! UVCHR_IS_INVARIANT(uvc)) {
2177 TRIE_BITMAP_SET(trie, UTF8_TWO_BYTE_HI(uvc));
2180 set_bit = 0; /* We've done our bit :-) */
2184 /* XXX We could come up with the list of code points that fold
2185 * to this using PL_utf8_foldclosures, except not for
2186 * multi-char folds, as there may be multiple combinations
2187 * there that could work, which needs to wait until runtime to
2188 * resolve (The comment about LIGATURE FFI above is such an
2193 widecharmap = newHV();
2195 svpp = hv_fetch( widecharmap, (char*)&uvc, sizeof( UV ), 1 );
2198 Perl_croak( aTHX_ "error creating/fetching widecharmap entry for 0x%"UVXf, uvc );
2200 if ( !SvTRUE( *svpp ) ) {
2201 sv_setiv( *svpp, ++trie->uniquecharcount );
2202 TRIE_STORE_REVCHAR(uvc);
2205 } /* end loop through characters in this branch of the trie */
2207 /* We take the min and max for this branch and combine to find the min
2208 * and max for all branches processed so far */
2209 if( cur == first ) {
2210 trie->minlen = minchars;
2211 trie->maxlen = maxchars;
2212 } else if (minchars < trie->minlen) {
2213 trie->minlen = minchars;
2214 } else if (maxchars > trie->maxlen) {
2215 trie->maxlen = maxchars;
2217 } /* end first pass */
2218 DEBUG_TRIE_COMPILE_r(
2219 PerlIO_printf( Perl_debug_log,
2220 "%*sTRIE(%s): W:%d C:%d Uq:%d Min:%d Max:%d\n",
2221 (int)depth * 2 + 2,"",
2222 ( widecharmap ? "UTF8" : "NATIVE" ), (int)word_count,
2223 (int)TRIE_CHARCOUNT(trie), trie->uniquecharcount,
2224 (int)trie->minlen, (int)trie->maxlen )
2228 We now know what we are dealing with in terms of unique chars and
2229 string sizes so we can calculate how much memory a naive
2230 representation using a flat table will take. If it's over a reasonable
2231 limit (as specified by ${^RE_TRIE_MAXBUF}) we use a more memory
2232 conservative but potentially much slower representation using an array
2235 At the end we convert both representations into the same compressed
2236 form that will be used in regexec.c for matching with. The latter
2237 is a form that cannot be used to construct with but has memory
2238 properties similar to the list form and access properties similar
2239 to the table form making it both suitable for fast searches and
2240 small enough that its feasable to store for the duration of a program.
2242 See the comment in the code where the compressed table is produced
2243 inplace from the flat tabe representation for an explanation of how
2244 the compression works.
2249 Newx(prev_states, TRIE_CHARCOUNT(trie) + 2, U32);
2252 if ( (IV)( ( TRIE_CHARCOUNT(trie) + 1 ) * trie->uniquecharcount + 1)
2253 > SvIV(re_trie_maxbuff) )
2256 Second Pass -- Array Of Lists Representation
2258 Each state will be represented by a list of charid:state records
2259 (reg_trie_trans_le) the first such element holds the CUR and LEN
2260 points of the allocated array. (See defines above).
2262 We build the initial structure using the lists, and then convert
2263 it into the compressed table form which allows faster lookups
2264 (but cant be modified once converted).
2267 STRLEN transcount = 1;
2269 DEBUG_TRIE_COMPILE_MORE_r( PerlIO_printf( Perl_debug_log,
2270 "%*sCompiling trie using list compiler\n",
2271 (int)depth * 2 + 2, ""));
2273 trie->states = (reg_trie_state *)
2274 PerlMemShared_calloc( TRIE_CHARCOUNT(trie) + 2,
2275 sizeof(reg_trie_state) );
2279 for ( cur = first ; cur < last ; cur = regnext( cur ) ) {
2281 regnode *noper = NEXTOPER( cur );
2282 U8 *uc = (U8*)STRING( noper );
2283 const U8 *e = uc + STR_LEN( noper );
2284 U32 state = 1; /* required init */
2285 U16 charid = 0; /* sanity init */
2286 U32 wordlen = 0; /* required init */
2288 if (OP(noper) == NOTHING) {
2289 regnode *noper_next= regnext(noper);
2290 if (noper_next != tail && OP(noper_next) == flags) {
2292 uc= (U8*)STRING(noper);
2293 e= uc + STR_LEN(noper);
2297 if (OP(noper) != NOTHING) {
2298 for ( ; uc < e ; uc += len ) {
2303 charid = trie->charmap[ uvc ];
2305 SV** const svpp = hv_fetch( widecharmap,
2312 charid=(U16)SvIV( *svpp );
2315 /* charid is now 0 if we dont know the char read, or
2316 * nonzero if we do */
2323 if ( !trie->states[ state ].trans.list ) {
2324 TRIE_LIST_NEW( state );
2327 check <= TRIE_LIST_USED( state );
2330 if ( TRIE_LIST_ITEM( state, check ).forid
2333 newstate = TRIE_LIST_ITEM( state, check ).newstate;
2338 newstate = next_alloc++;
2339 prev_states[newstate] = state;
2340 TRIE_LIST_PUSH( state, charid, newstate );
2345 Perl_croak( aTHX_ "panic! In trie construction, no char mapping for %"IVdf, uvc );
2349 TRIE_HANDLE_WORD(state);
2351 } /* end second pass */
2353 /* next alloc is the NEXT state to be allocated */
2354 trie->statecount = next_alloc;
2355 trie->states = (reg_trie_state *)
2356 PerlMemShared_realloc( trie->states,
2358 * sizeof(reg_trie_state) );
2360 /* and now dump it out before we compress it */
2361 DEBUG_TRIE_COMPILE_MORE_r(dump_trie_interim_list(trie, widecharmap,
2362 revcharmap, next_alloc,
2366 trie->trans = (reg_trie_trans *)
2367 PerlMemShared_calloc( transcount, sizeof(reg_trie_trans) );
2374 for( state=1 ; state < next_alloc ; state ++ ) {
2378 DEBUG_TRIE_COMPILE_MORE_r(
2379 PerlIO_printf( Perl_debug_log, "tp: %d zp: %d ",tp,zp)
2383 if (trie->states[state].trans.list) {
2384 U16 minid=TRIE_LIST_ITEM( state, 1).forid;
2388 for( idx = 2 ; idx <= TRIE_LIST_USED( state ) ; idx++ ) {
2389 const U16 forid = TRIE_LIST_ITEM( state, idx).forid;
2390 if ( forid < minid ) {
2392 } else if ( forid > maxid ) {
2396 if ( transcount < tp + maxid - minid + 1) {
2398 trie->trans = (reg_trie_trans *)
2399 PerlMemShared_realloc( trie->trans,
2401 * sizeof(reg_trie_trans) );
2402 Zero( trie->trans + (transcount / 2),
2406 base = trie->uniquecharcount + tp - minid;
2407 if ( maxid == minid ) {
2409 for ( ; zp < tp ; zp++ ) {
2410 if ( ! trie->trans[ zp ].next ) {
2411 base = trie->uniquecharcount + zp - minid;
2412 trie->trans[ zp ].next = TRIE_LIST_ITEM( state,
2414 trie->trans[ zp ].check = state;
2420 trie->trans[ tp ].next = TRIE_LIST_ITEM( state,
2422 trie->trans[ tp ].check = state;
2427 for ( idx=1; idx <= TRIE_LIST_USED( state ) ; idx++ ) {
2428 const U32 tid = base
2429 - trie->uniquecharcount
2430 + TRIE_LIST_ITEM( state, idx ).forid;
2431 trie->trans[ tid ].next = TRIE_LIST_ITEM( state,
2433 trie->trans[ tid ].check = state;
2435 tp += ( maxid - minid + 1 );
2437 Safefree(trie->states[ state ].trans.list);
2440 DEBUG_TRIE_COMPILE_MORE_r(
2441 PerlIO_printf( Perl_debug_log, " base: %d\n",base);
2444 trie->states[ state ].trans.base=base;
2446 trie->lasttrans = tp + 1;
2450 Second Pass -- Flat Table Representation.
2452 we dont use the 0 slot of either trans[] or states[] so we add 1 to
2453 each. We know that we will need Charcount+1 trans at most to store
2454 the data (one row per char at worst case) So we preallocate both
2455 structures assuming worst case.
2457 We then construct the trie using only the .next slots of the entry
2460 We use the .check field of the first entry of the node temporarily
2461 to make compression both faster and easier by keeping track of how
2462 many non zero fields are in the node.
2464 Since trans are numbered from 1 any 0 pointer in the table is a FAIL
2467 There are two terms at use here: state as a TRIE_NODEIDX() which is
2468 a number representing the first entry of the node, and state as a
2469 TRIE_NODENUM() which is the trans number. state 1 is TRIE_NODEIDX(1)
2470 and TRIE_NODENUM(1), state 2 is TRIE_NODEIDX(2) and TRIE_NODENUM(3)
2471 if there are 2 entrys per node. eg:
2479 The table is internally in the right hand, idx form. However as we
2480 also have to deal with the states array which is indexed by nodenum
2481 we have to use TRIE_NODENUM() to convert.
2484 DEBUG_TRIE_COMPILE_MORE_r( PerlIO_printf( Perl_debug_log,
2485 "%*sCompiling trie using table compiler\n",
2486 (int)depth * 2 + 2, ""));
2488 trie->trans = (reg_trie_trans *)
2489 PerlMemShared_calloc( ( TRIE_CHARCOUNT(trie) + 1 )
2490 * trie->uniquecharcount + 1,
2491 sizeof(reg_trie_trans) );
2492 trie->states = (reg_trie_state *)
2493 PerlMemShared_calloc( TRIE_CHARCOUNT(trie) + 2,
2494 sizeof(reg_trie_state) );
2495 next_alloc = trie->uniquecharcount + 1;
2498 for ( cur = first ; cur < last ; cur = regnext( cur ) ) {
2500 regnode *noper = NEXTOPER( cur );
2501 const U8 *uc = (U8*)STRING( noper );
2502 const U8 *e = uc + STR_LEN( noper );
2504 U32 state = 1; /* required init */
2506 U16 charid = 0; /* sanity init */
2507 U32 accept_state = 0; /* sanity init */
2509 U32 wordlen = 0; /* required init */
2511 if (OP(noper) == NOTHING) {
2512 regnode *noper_next= regnext(noper);
2513 if (noper_next != tail && OP(noper_next) == flags) {
2515 uc= (U8*)STRING(noper);
2516 e= uc + STR_LEN(noper);
2520 if ( OP(noper) != NOTHING ) {
2521 for ( ; uc < e ; uc += len ) {
2526 charid = trie->charmap[ uvc ];
2528 SV* const * const svpp = hv_fetch( widecharmap,
2532 charid = svpp ? (U16)SvIV(*svpp) : 0;
2536 if ( !trie->trans[ state + charid ].next ) {
2537 trie->trans[ state + charid ].next = next_alloc;
2538 trie->trans[ state ].check++;
2539 prev_states[TRIE_NODENUM(next_alloc)]
2540 = TRIE_NODENUM(state);
2541 next_alloc += trie->uniquecharcount;
2543 state = trie->trans[ state + charid ].next;
2545 Perl_croak( aTHX_ "panic! In trie construction, no char mapping for %"IVdf, uvc );
2547 /* charid is now 0 if we dont know the char read, or
2548 * nonzero if we do */
2551 accept_state = TRIE_NODENUM( state );
2552 TRIE_HANDLE_WORD(accept_state);
2554 } /* end second pass */
2556 /* and now dump it out before we compress it */
2557 DEBUG_TRIE_COMPILE_MORE_r(dump_trie_interim_table(trie, widecharmap,
2559 next_alloc, depth+1));
2563 * Inplace compress the table.*
2565 For sparse data sets the table constructed by the trie algorithm will
2566 be mostly 0/FAIL transitions or to put it another way mostly empty.
2567 (Note that leaf nodes will not contain any transitions.)
2569 This algorithm compresses the tables by eliminating most such
2570 transitions, at the cost of a modest bit of extra work during lookup:
2572 - Each states[] entry contains a .base field which indicates the
2573 index in the state[] array wheres its transition data is stored.
2575 - If .base is 0 there are no valid transitions from that node.
2577 - If .base is nonzero then charid is added to it to find an entry in
2580 -If trans[states[state].base+charid].check!=state then the
2581 transition is taken to be a 0/Fail transition. Thus if there are fail
2582 transitions at the front of the node then the .base offset will point
2583 somewhere inside the previous nodes data (or maybe even into a node
2584 even earlier), but the .check field determines if the transition is
2588 The following process inplace converts the table to the compressed
2589 table: We first do not compress the root node 1,and mark all its
2590 .check pointers as 1 and set its .base pointer as 1 as well. This
2591 allows us to do a DFA construction from the compressed table later,
2592 and ensures that any .base pointers we calculate later are greater
2595 - We set 'pos' to indicate the first entry of the second node.
2597 - We then iterate over the columns of the node, finding the first and
2598 last used entry at l and m. We then copy l..m into pos..(pos+m-l),
2599 and set the .check pointers accordingly, and advance pos
2600 appropriately and repreat for the next node. Note that when we copy
2601 the next pointers we have to convert them from the original
2602 NODEIDX form to NODENUM form as the former is not valid post
2605 - If a node has no transitions used we mark its base as 0 and do not
2606 advance the pos pointer.
2608 - If a node only has one transition we use a second pointer into the
2609 structure to fill in allocated fail transitions from other states.
2610 This pointer is independent of the main pointer and scans forward
2611 looking for null transitions that are allocated to a state. When it
2612 finds one it writes the single transition into the "hole". If the
2613 pointer doesnt find one the single transition is appended as normal.
2615 - Once compressed we can Renew/realloc the structures to release the
2618 See "Table-Compression Methods" in sec 3.9 of the Red Dragon,
2619 specifically Fig 3.47 and the associated pseudocode.
2623 const U32 laststate = TRIE_NODENUM( next_alloc );
2626 trie->statecount = laststate;
2628 for ( state = 1 ; state < laststate ; state++ ) {
2630 const U32 stateidx = TRIE_NODEIDX( state );
2631 const U32 o_used = trie->trans[ stateidx ].check;
2632 U32 used = trie->trans[ stateidx ].check;
2633 trie->trans[ stateidx ].check = 0;
2636 used && charid < trie->uniquecharcount;
2639 if ( flag || trie->trans[ stateidx + charid ].next ) {
2640 if ( trie->trans[ stateidx + charid ].next ) {
2642 for ( ; zp < pos ; zp++ ) {
2643 if ( ! trie->trans[ zp ].next ) {
2647 trie->states[ state ].trans.base
2649 + trie->uniquecharcount
2651 trie->trans[ zp ].next
2652 = SAFE_TRIE_NODENUM( trie->trans[ stateidx
2654 trie->trans[ zp ].check = state;
2655 if ( ++zp > pos ) pos = zp;
2662 trie->states[ state ].trans.base
2663 = pos + trie->uniquecharcount - charid ;
2665 trie->trans[ pos ].next
2666 = SAFE_TRIE_NODENUM(
2667 trie->trans[ stateidx + charid ].next );
2668 trie->trans[ pos ].check = state;
2673 trie->lasttrans = pos + 1;
2674 trie->states = (reg_trie_state *)
2675 PerlMemShared_realloc( trie->states, laststate
2676 * sizeof(reg_trie_state) );
2677 DEBUG_TRIE_COMPILE_MORE_r(
2678 PerlIO_printf( Perl_debug_log,
2679 "%*sAlloc: %d Orig: %"IVdf" elements, Final:%"IVdf". Savings of %%%5.2f\n",
2680 (int)depth * 2 + 2,"",
2681 (int)( ( TRIE_CHARCOUNT(trie) + 1 ) * trie->uniquecharcount
2685 ( ( next_alloc - pos ) * 100 ) / (double)next_alloc );
2688 } /* end table compress */
2690 DEBUG_TRIE_COMPILE_MORE_r(
2691 PerlIO_printf(Perl_debug_log,
2692 "%*sStatecount:%"UVxf" Lasttrans:%"UVxf"\n",
2693 (int)depth * 2 + 2, "",
2694 (UV)trie->statecount,
2695 (UV)trie->lasttrans)
2697 /* resize the trans array to remove unused space */
2698 trie->trans = (reg_trie_trans *)
2699 PerlMemShared_realloc( trie->trans, trie->lasttrans
2700 * sizeof(reg_trie_trans) );
2702 { /* Modify the program and insert the new TRIE node */
2703 U8 nodetype =(U8)(flags & 0xFF);
2707 regnode *optimize = NULL;
2708 #ifdef RE_TRACK_PATTERN_OFFSETS
2711 U32 mjd_nodelen = 0;
2712 #endif /* RE_TRACK_PATTERN_OFFSETS */
2713 #endif /* DEBUGGING */
2715 This means we convert either the first branch or the first Exact,
2716 depending on whether the thing following (in 'last') is a branch
2717 or not and whther first is the startbranch (ie is it a sub part of
2718 the alternation or is it the whole thing.)
2719 Assuming its a sub part we convert the EXACT otherwise we convert
2720 the whole branch sequence, including the first.
2722 /* Find the node we are going to overwrite */
2723 if ( first != startbranch || OP( last ) == BRANCH ) {
2724 /* branch sub-chain */
2725 NEXT_OFF( first ) = (U16)(last - first);
2726 #ifdef RE_TRACK_PATTERN_OFFSETS
2728 mjd_offset= Node_Offset((convert));
2729 mjd_nodelen= Node_Length((convert));
2732 /* whole branch chain */
2734 #ifdef RE_TRACK_PATTERN_OFFSETS
2737 const regnode *nop = NEXTOPER( convert );
2738 mjd_offset= Node_Offset((nop));
2739 mjd_nodelen= Node_Length((nop));
2743 PerlIO_printf(Perl_debug_log,
2744 "%*sMJD offset:%"UVuf" MJD length:%"UVuf"\n",
2745 (int)depth * 2 + 2, "",
2746 (UV)mjd_offset, (UV)mjd_nodelen)
2749 /* But first we check to see if there is a common prefix we can
2750 split out as an EXACT and put in front of the TRIE node. */
2751 trie->startstate= 1;
2752 if ( trie->bitmap && !widecharmap && !trie->jump ) {
2754 for ( state = 1 ; state < trie->statecount-1 ; state++ ) {
2758 const U32 base = trie->states[ state ].trans.base;
2760 if ( trie->states[state].wordnum )
2763 for ( ofs = 0 ; ofs < trie->uniquecharcount ; ofs++ ) {
2764 if ( ( base + ofs >= trie->uniquecharcount ) &&
2765 ( base + ofs - trie->uniquecharcount < trie->lasttrans ) &&
2766 trie->trans[ base + ofs - trie->uniquecharcount ].check == state )
2768 if ( ++count > 1 ) {
2769 SV **tmp = av_fetch( revcharmap, ofs, 0);
2770 const U8 *ch = (U8*)SvPV_nolen_const( *tmp );
2771 if ( state == 1 ) break;
2773 Zero(trie->bitmap, ANYOF_BITMAP_SIZE, char);
2775 PerlIO_printf(Perl_debug_log,
2776 "%*sNew Start State=%"UVuf" Class: [",
2777 (int)depth * 2 + 2, "",
2780 SV ** const tmp = av_fetch( revcharmap, idx, 0);
2781 const U8 * const ch = (U8*)SvPV_nolen_const( *tmp );
2783 TRIE_BITMAP_SET(trie,*ch);
2785 TRIE_BITMAP_SET(trie, folder[ *ch ]);
2787 PerlIO_printf(Perl_debug_log, "%s", (char*)ch)
2791 TRIE_BITMAP_SET(trie,*ch);
2793 TRIE_BITMAP_SET(trie,folder[ *ch ]);
2794 DEBUG_OPTIMISE_r(PerlIO_printf( Perl_debug_log,"%s", ch));
2800 SV **tmp = av_fetch( revcharmap, idx, 0);
2802 char *ch = SvPV( *tmp, len );
2804 SV *sv=sv_newmortal();
2805 PerlIO_printf( Perl_debug_log,
2806 "%*sPrefix State: %"UVuf" Idx:%"UVuf" Char='%s'\n",
2807 (int)depth * 2 + 2, "",
2809 pv_pretty(sv, SvPV_nolen_const(*tmp), SvCUR(*tmp), 6,
2810 PL_colors[0], PL_colors[1],
2811 (SvUTF8(*tmp) ? PERL_PV_ESCAPE_UNI : 0) |
2812 PERL_PV_ESCAPE_FIRSTCHAR
2817 OP( convert ) = nodetype;
2818 str=STRING(convert);
2821 STR_LEN(convert) += len;
2827 DEBUG_OPTIMISE_r(PerlIO_printf( Perl_debug_log,"]\n"));
2832 trie->prefixlen = (state-1);
2834 regnode *n = convert+NODE_SZ_STR(convert);
2835 NEXT_OFF(convert) = NODE_SZ_STR(convert);
2836 trie->startstate = state;
2837 trie->minlen -= (state - 1);
2838 trie->maxlen -= (state - 1);
2840 /* At least the UNICOS C compiler choked on this
2841 * being argument to DEBUG_r(), so let's just have
2844 #ifdef PERL_EXT_RE_BUILD
2850 regnode *fix = convert;
2851 U32 word = trie->wordcount;
2853 Set_Node_Offset_Length(convert, mjd_offset, state - 1);
2854 while( ++fix < n ) {
2855 Set_Node_Offset_Length(fix, 0, 0);
2858 SV ** const tmp = av_fetch( trie_words, word, 0 );
2860 if ( STR_LEN(convert) <= SvCUR(*tmp) )
2861 sv_chop(*tmp, SvPV_nolen(*tmp) + STR_LEN(convert));
2863 sv_chop(*tmp, SvPV_nolen(*tmp) + SvCUR(*tmp));
2871 NEXT_OFF(convert) = (U16)(tail - convert);
2872 DEBUG_r(optimize= n);
2878 if ( trie->maxlen ) {
2879 NEXT_OFF( convert ) = (U16)(tail - convert);
2880 ARG_SET( convert, data_slot );
2881 /* Store the offset to the first unabsorbed branch in
2882 jump[0], which is otherwise unused by the jump logic.
2883 We use this when dumping a trie and during optimisation. */
2885 trie->jump[0] = (U16)(nextbranch - convert);
2887 /* If the start state is not accepting (meaning there is no empty string/NOTHING)
2888 * and there is a bitmap
2889 * and the first "jump target" node we found leaves enough room
2890 * then convert the TRIE node into a TRIEC node, with the bitmap
2891 * embedded inline in the opcode - this is hypothetically faster.
2893 if ( !trie->states[trie->startstate].wordnum
2895 && ( (char *)jumper - (char *)convert) >= (int)sizeof(struct regnode_charclass) )
2897 OP( convert ) = TRIEC;
2898 Copy(trie->bitmap, ((struct regnode_charclass *)convert)->bitmap, ANYOF_BITMAP_SIZE, char);
2899 PerlMemShared_free(trie->bitmap);
2902 OP( convert ) = TRIE;
2904 /* store the type in the flags */
2905 convert->flags = nodetype;
2909 + regarglen[ OP( convert ) ];
2911 /* XXX We really should free up the resource in trie now,
2912 as we won't use them - (which resources?) dmq */
2914 /* needed for dumping*/
2915 DEBUG_r(if (optimize) {
2916 regnode *opt = convert;
2918 while ( ++opt < optimize) {
2919 Set_Node_Offset_Length(opt,0,0);
2922 Try to clean up some of the debris left after the
2925 while( optimize < jumper ) {
2926 mjd_nodelen += Node_Length((optimize));
2927 OP( optimize ) = OPTIMIZED;
2928 Set_Node_Offset_Length(optimize,0,0);
2931 Set_Node_Offset_Length(convert,mjd_offset,mjd_nodelen);
2933 } /* end node insert */
2935 /* Finish populating the prev field of the wordinfo array. Walk back
2936 * from each accept state until we find another accept state, and if
2937 * so, point the first word's .prev field at the second word. If the
2938 * second already has a .prev field set, stop now. This will be the
2939 * case either if we've already processed that word's accept state,
2940 * or that state had multiple words, and the overspill words were
2941 * already linked up earlier.
2948 for (word=1; word <= trie->wordcount; word++) {
2950 if (trie->wordinfo[word].prev)
2952 state = trie->wordinfo[word].accept;
2954 state = prev_states[state];
2957 prev = trie->states[state].wordnum;
2961 trie->wordinfo[word].prev = prev;
2963 Safefree(prev_states);
2967 /* and now dump out the compressed format */
2968 DEBUG_TRIE_COMPILE_r(dump_trie(trie, widecharmap, revcharmap, depth+1));
2970 RExC_rxi->data->data[ data_slot + 1 ] = (void*)widecharmap;
2972 RExC_rxi->data->data[ data_slot + TRIE_WORDS_OFFSET ] = (void*)trie_words;
2973 RExC_rxi->data->data[ data_slot + 3 ] = (void*)revcharmap;
2975 SvREFCNT_dec_NN(revcharmap);
2979 : trie->startstate>1
2985 S_make_trie_failtable(pTHX_ RExC_state_t *pRExC_state, regnode *source, regnode *stclass, U32 depth)
2987 /* The Trie is constructed and compressed now so we can build a fail array if
2990 This is basically the Aho-Corasick algorithm. Its from exercise 3.31 and
2992 "Red Dragon" -- Compilers, principles, techniques, and tools. Aho, Sethi,
2996 We find the fail state for each state in the trie, this state is the longest
2997 proper suffix of the current state's 'word' that is also a proper prefix of
2998 another word in our trie. State 1 represents the word '' and is thus the
2999 default fail state. This allows the DFA not to have to restart after its
3000 tried and failed a word at a given point, it simply continues as though it
3001 had been matching the other word in the first place.
3003 'abcdgu'=~/abcdefg|cdgu/
3004 When we get to 'd' we are still matching the first word, we would encounter
3005 'g' which would fail, which would bring us to the state representing 'd' in
3006 the second word where we would try 'g' and succeed, proceeding to match
3009 /* add a fail transition */
3010 const U32 trie_offset = ARG(source);
3011 reg_trie_data *trie=(reg_trie_data *)RExC_rxi->data->data[trie_offset];
3013 const U32 ucharcount = trie->uniquecharcount;
3014 const U32 numstates = trie->statecount;
3015 const U32 ubound = trie->lasttrans + ucharcount;
3019 U32 base = trie->states[ 1 ].trans.base;
3022 const U32 data_slot = add_data( pRExC_state, STR_WITH_LEN("T"));
3023 GET_RE_DEBUG_FLAGS_DECL;
3025 PERL_ARGS_ASSERT_MAKE_TRIE_FAILTABLE;
3027 PERL_UNUSED_ARG(depth);
3031 ARG_SET( stclass, data_slot );
3032 aho = (reg_ac_data *) PerlMemShared_calloc( 1, sizeof(reg_ac_data) );
3033 RExC_rxi->data->data[ data_slot ] = (void*)aho;
3034 aho->trie=trie_offset;
3035 aho->states=(reg_trie_state *)PerlMemShared_malloc( numstates * sizeof(reg_trie_state) );
3036 Copy( trie->states, aho->states, numstates, reg_trie_state );
3037 Newxz( q, numstates, U32);
3038 aho->fail = (U32 *) PerlMemShared_calloc( numstates, sizeof(U32) );
3041 /* initialize fail[0..1] to be 1 so that we always have
3042 a valid final fail state */
3043 fail[ 0 ] = fail[ 1 ] = 1;
3045 for ( charid = 0; charid < ucharcount ; charid++ ) {
3046 const U32 newstate = TRIE_TRANS_STATE( 1, base, ucharcount, charid, 0 );
3048 q[ q_write ] = newstate;
3049 /* set to point at the root */
3050 fail[ q[ q_write++ ] ]=1;
3053 while ( q_read < q_write) {
3054 const U32 cur = q[ q_read++ % numstates ];
3055 base = trie->states[ cur ].trans.base;
3057 for ( charid = 0 ; charid < ucharcount ; charid++ ) {
3058 const U32 ch_state = TRIE_TRANS_STATE( cur, base, ucharcount, charid, 1 );
3060 U32 fail_state = cur;
3063 fail_state = fail[ fail_state ];
3064 fail_base = aho->states[ fail_state ].trans.base;
3065 } while ( !TRIE_TRANS_STATE( fail_state, fail_base, ucharcount, charid, 1 ) );
3067 fail_state = TRIE_TRANS_STATE( fail_state, fail_base, ucharcount, charid, 1 );
3068 fail[ ch_state ] = fail_state;
3069 if ( !aho->states[ ch_state ].wordnum && aho->states[ fail_state ].wordnum )
3071 aho->states[ ch_state ].wordnum = aho->states[ fail_state ].wordnum;
3073 q[ q_write++ % numstates] = ch_state;
3077 /* restore fail[0..1] to 0 so that we "fall out" of the AC loop
3078 when we fail in state 1, this allows us to use the
3079 charclass scan to find a valid start char. This is based on the principle
3080 that theres a good chance the string being searched contains lots of stuff
3081 that cant be a start char.
3083 fail[ 0 ] = fail[ 1 ] = 0;
3084 DEBUG_TRIE_COMPILE_r({
3085 PerlIO_printf(Perl_debug_log,
3086 "%*sStclass Failtable (%"UVuf" states): 0",
3087 (int)(depth * 2), "", (UV)numstates
3089 for( q_read=1; q_read<numstates; q_read++ ) {
3090 PerlIO_printf(Perl_debug_log, ", %"UVuf, (UV)fail[q_read]);
3092 PerlIO_printf(Perl_debug_log, "\n");
3095 /*RExC_seen |= REG_TRIEDFA_SEEN;*/
3099 #define DEBUG_PEEP(str,scan,depth) \
3100 DEBUG_OPTIMISE_r({if (scan){ \
3101 SV * const mysv=sv_newmortal(); \
3102 regnode *Next = regnext(scan); \
3103 regprop(RExC_rx, mysv, scan, NULL); \
3104 PerlIO_printf(Perl_debug_log, "%*s" str ">%3d: %s (%d)\n", \
3105 (int)depth*2, "", REG_NODE_NUM(scan), SvPV_nolen_const(mysv),\
3106 Next ? (REG_NODE_NUM(Next)) : 0 ); \
3110 /* The below joins as many adjacent EXACTish nodes as possible into a single
3111 * one. The regop may be changed if the node(s) contain certain sequences that
3112 * require special handling. The joining is only done if:
3113 * 1) there is room in the current conglomerated node to entirely contain the
3115 * 2) they are the exact same node type
3117 * The adjacent nodes actually may be separated by NOTHING-kind nodes, and
3118 * these get optimized out
3120 * If a node is to match under /i (folded), the number of characters it matches
3121 * can be different than its character length if it contains a multi-character
3122 * fold. *min_subtract is set to the total delta number of characters of the
3125 * And *unfolded_multi_char is set to indicate whether or not the node contains
3126 * an unfolded multi-char fold. This happens when whether the fold is valid or
3127 * not won't be known until runtime; namely for EXACTF nodes that contain LATIN
3128 * SMALL LETTER SHARP S, as only if the target string being matched against
3129 * turns out to be UTF-8 is that fold valid; and also for EXACTFL nodes whose
3130 * folding rules depend on the locale in force at runtime. (Multi-char folds
3131 * whose components are all above the Latin1 range are not run-time locale
3132 * dependent, and have already been folded by the time this function is
3135 * This is as good a place as any to discuss the design of handling these
3136 * multi-character fold sequences. It's been wrong in Perl for a very long
3137 * time. There are three code points in Unicode whose multi-character folds
3138 * were long ago discovered to mess things up. The previous designs for
3139 * dealing with these involved assigning a special node for them. This
3140 * approach doesn't always work, as evidenced by this example:
3141 * "\xDFs" =~ /s\xDF/ui # Used to fail before these patches
3142 * Both sides fold to "sss", but if the pattern is parsed to create a node that
3143 * would match just the \xDF, it won't be able to handle the case where a
3144 * successful match would have to cross the node's boundary. The new approach
3145 * that hopefully generally solves the problem generates an EXACTFU_SS node
3146 * that is "sss" in this case.
3148 * It turns out that there are problems with all multi-character folds, and not
3149 * just these three. Now the code is general, for all such cases. The
3150 * approach taken is:
3151 * 1) This routine examines each EXACTFish node that could contain multi-
3152 * character folded sequences. Since a single character can fold into
3153 * such a sequence, the minimum match length for this node is less than
3154 * the number of characters in the node. This routine returns in
3155 * *min_subtract how many characters to subtract from the the actual
3156 * length of the string to get a real minimum match length; it is 0 if
3157 * there are no multi-char foldeds. This delta is used by the caller to
3158 * adjust the min length of the match, and the delta between min and max,
3159 * so that the optimizer doesn't reject these possibilities based on size
3161 * 2) For the sequence involving the Sharp s (\xDF), the node type EXACTFU_SS
3162 * is used for an EXACTFU node that contains at least one "ss" sequence in
3163 * it. For non-UTF-8 patterns and strings, this is the only case where
3164 * there is a possible fold length change. That means that a regular
3165 * EXACTFU node without UTF-8 involvement doesn't have to concern itself
3166 * with length changes, and so can be processed faster. regexec.c takes
3167 * advantage of this. Generally, an EXACTFish node that is in UTF-8 is
3168 * pre-folded by regcomp.c (except EXACTFL, some of whose folds aren't
3169 * known until runtime). This saves effort in regex matching. However,
3170 * the pre-folding isn't done for non-UTF8 patterns because the fold of
3171 * the MICRO SIGN requires UTF-8, and we don't want to slow things down by
3172 * forcing the pattern into UTF8 unless necessary. Also what EXACTF (and,
3173 * again, EXACTFL) nodes fold to isn't known until runtime. The fold
3174 * possibilities for the non-UTF8 patterns are quite simple, except for
3175 * the sharp s. All the ones that don't involve a UTF-8 target string are
3176 * members of a fold-pair, and arrays are set up for all of them so that
3177 * the other member of the pair can be found quickly. Code elsewhere in
3178 * this file makes sure that in EXACTFU nodes, the sharp s gets folded to
3179 * 'ss', even if the pattern isn't UTF-8. This avoids the issues
3180 * described in the next item.
3181 * 3) A problem remains for unfolded multi-char folds. (These occur when the
3182 * validity of the fold won't be known until runtime, and so must remain
3183 * unfolded for now. This happens for the sharp s in EXACTF and EXACTFA
3184 * nodes when the pattern isn't in UTF-8. (Note, BTW, that there cannot
3185 * be an EXACTF node with a UTF-8 pattern.) They also occur for various
3186 * folds in EXACTFL nodes, regardless of the UTF-ness of the pattern.)
3187 * The reason this is a problem is that the optimizer part of regexec.c
3188 * (probably unwittingly, in Perl_regexec_flags()) makes an assumption
3189 * that a character in the pattern corresponds to at most a single
3190 * character in the target string. (And I do mean character, and not byte
3191 * here, unlike other parts of the documentation that have never been
3192 * updated to account for multibyte Unicode.) sharp s in EXACTF and
3193 * EXACTFL nodes can match the two character string 'ss'; in EXACTFA nodes
3194 * it can match "\x{17F}\x{17F}". These, along with other ones in EXACTFL
3195 * nodes, violate the assumption, and they are the only instances where it
3196 * is violated. I'm reluctant to try to change the assumption, as the
3197 * code involved is impenetrable to me (khw), so instead the code here
3198 * punts. This routine examines EXACTFL nodes, and (when the pattern
3199 * isn't UTF-8) EXACTF and EXACTFA for such unfolded folds, and returns a
3200 * boolean indicating whether or not the node contains such a fold. When
3201 * it is true, the caller sets a flag that later causes the optimizer in
3202 * this file to not set values for the floating and fixed string lengths,
3203 * and thus avoids the optimizer code in regexec.c that makes the invalid
3204 * assumption. Thus, there is no optimization based on string lengths for
3205 * EXACTFL nodes that contain these few folds, nor for non-UTF8-pattern
3206 * EXACTF and EXACTFA nodes that contain the sharp s. (The reason the
3207 * assumption is wrong only in these cases is that all other non-UTF-8
3208 * folds are 1-1; and, for UTF-8 patterns, we pre-fold all other folds to
3209 * their expanded versions. (Again, we can't prefold sharp s to 'ss' in
3210 * EXACTF nodes because we don't know at compile time if it actually
3211 * matches 'ss' or not. For EXACTF nodes it will match iff the target
3212 * string is in UTF-8. This is in contrast to EXACTFU nodes, where it
3213 * always matches; and EXACTFA where it never does. In an EXACTFA node in
3214 * a UTF-8 pattern, sharp s is folded to "\x{17F}\x{17F}, avoiding the
3215 * problem; but in a non-UTF8 pattern, folding it to that above-Latin1
3216 * string would require the pattern to be forced into UTF-8, the overhead
3217 * of which we want to avoid. Similarly the unfolded multi-char folds in
3218 * EXACTFL nodes will match iff the locale at the time of match is a UTF-8
3221 * Similarly, the code that generates tries doesn't currently handle
3222 * not-already-folded multi-char folds, and it looks like a pain to change
3223 * that. Therefore, trie generation of EXACTFA nodes with the sharp s
3224 * doesn't work. Instead, such an EXACTFA is turned into a new regnode,
3225 * EXACTFA_NO_TRIE, which the trie code knows not to handle. Most people
3226 * using /iaa matching will be doing so almost entirely with ASCII
3227 * strings, so this should rarely be encountered in practice */
3229 #define JOIN_EXACT(scan,min_subtract,unfolded_multi_char, flags) \
3230 if (PL_regkind[OP(scan)] == EXACT) \
3231 join_exact(pRExC_state,(scan),(min_subtract),unfolded_multi_char, (flags),NULL,depth+1)
3234 S_join_exact(pTHX_ RExC_state_t *pRExC_state, regnode *scan,
3235 UV *min_subtract, bool *unfolded_multi_char,
3236 U32 flags,regnode *val, U32 depth)
3238 /* Merge several consecutive EXACTish nodes into one. */
3239 regnode *n = regnext(scan);
3241 regnode *next = scan + NODE_SZ_STR(scan);
3245 regnode *stop = scan;
3246 GET_RE_DEBUG_FLAGS_DECL;
3248 PERL_UNUSED_ARG(depth);
3251 PERL_ARGS_ASSERT_JOIN_EXACT;
3252 #ifndef EXPERIMENTAL_INPLACESCAN
3253 PERL_UNUSED_ARG(flags);
3254 PERL_UNUSED_ARG(val);
3256 DEBUG_PEEP("join",scan,depth);
3258 /* Look through the subsequent nodes in the chain. Skip NOTHING, merge
3259 * EXACT ones that are mergeable to the current one. */
3261 && (PL_regkind[OP(n)] == NOTHING
3262 || (stringok && OP(n) == OP(scan)))
3264 && NEXT_OFF(scan) + NEXT_OFF(n) < I16_MAX)
3267 if (OP(n) == TAIL || n > next)
3269 if (PL_regkind[OP(n)] == NOTHING) {
3270 DEBUG_PEEP("skip:",n,depth);
3271 NEXT_OFF(scan) += NEXT_OFF(n);
3272 next = n + NODE_STEP_REGNODE;
3279 else if (stringok) {
3280 const unsigned int oldl = STR_LEN(scan);
3281 regnode * const nnext = regnext(n);
3283 /* XXX I (khw) kind of doubt that this works on platforms (should
3284 * Perl ever run on one) where U8_MAX is above 255 because of lots
3285 * of other assumptions */
3286 /* Don't join if the sum can't fit into a single node */
3287 if (oldl + STR_LEN(n) > U8_MAX)
3290 DEBUG_PEEP("merg",n,depth);
3293 NEXT_OFF(scan) += NEXT_OFF(n);
3294 STR_LEN(scan) += STR_LEN(n);
3295 next = n + NODE_SZ_STR(n);
3296 /* Now we can overwrite *n : */
3297 Move(STRING(n), STRING(scan) + oldl, STR_LEN(n), char);
3305 #ifdef EXPERIMENTAL_INPLACESCAN
3306 if (flags && !NEXT_OFF(n)) {
3307 DEBUG_PEEP("atch", val, depth);
3308 if (reg_off_by_arg[OP(n)]) {
3309 ARG_SET(n, val - n);
3312 NEXT_OFF(n) = val - n;
3320 *unfolded_multi_char = FALSE;
3322 /* Here, all the adjacent mergeable EXACTish nodes have been merged. We
3323 * can now analyze for sequences of problematic code points. (Prior to
3324 * this final joining, sequences could have been split over boundaries, and
3325 * hence missed). The sequences only happen in folding, hence for any
3326 * non-EXACT EXACTish node */
3327 if (OP(scan) != EXACT) {
3328 U8* s0 = (U8*) STRING(scan);
3330 U8* s_end = s0 + STR_LEN(scan);
3332 int total_count_delta = 0; /* Total delta number of characters that
3333 multi-char folds expand to */
3335 /* One pass is made over the node's string looking for all the
3336 * possibilities. To avoid some tests in the loop, there are two main
3337 * cases, for UTF-8 patterns (which can't have EXACTF nodes) and
3342 if (OP(scan) == EXACTFL) {
3345 /* An EXACTFL node would already have been changed to another
3346 * node type unless there is at least one character in it that
3347 * is problematic; likely a character whose fold definition
3348 * won't be known until runtime, and so has yet to be folded.
3349 * For all but the UTF-8 locale, folds are 1-1 in length, but
3350 * to handle the UTF-8 case, we need to create a temporary
3351 * folded copy using UTF-8 locale rules in order to analyze it.
3352 * This is because our macros that look to see if a sequence is
3353 * a multi-char fold assume everything is folded (otherwise the
3354 * tests in those macros would be too complicated and slow).
3355 * Note that here, the non-problematic folds will have already
3356 * been done, so we can just copy such characters. We actually
3357 * don't completely fold the EXACTFL string. We skip the
3358 * unfolded multi-char folds, as that would just create work
3359 * below to figure out the size they already are */
3361 Newx(folded, UTF8_MAX_FOLD_CHAR_EXPAND * STR_LEN(scan) + 1, U8);
3364 STRLEN s_len = UTF8SKIP(s);
3365 if (! is_PROBLEMATIC_LOCALE_FOLD_utf8(s)) {
3366 Copy(s, d, s_len, U8);
3369 else if (is_FOLDS_TO_MULTI_utf8(s)) {
3370 *unfolded_multi_char = TRUE;
3371 Copy(s, d, s_len, U8);
3374 else if (isASCII(*s)) {
3375 *(d++) = toFOLD(*s);
3379 _to_utf8_fold_flags(s, d, &len, FOLD_FLAGS_FULL);
3385 /* Point the remainder of the routine to look at our temporary
3389 } /* End of creating folded copy of EXACTFL string */
3391 /* Examine the string for a multi-character fold sequence. UTF-8
3392 * patterns have all characters pre-folded by the time this code is
3394 while (s < s_end - 1) /* Can stop 1 before the end, as minimum
3395 length sequence we are looking for is 2 */
3397 int count = 0; /* How many characters in a multi-char fold */
3398 int len = is_MULTI_CHAR_FOLD_utf8_safe(s, s_end);
3399 if (! len) { /* Not a multi-char fold: get next char */
3404 /* Nodes with 'ss' require special handling, except for
3405 * EXACTFA-ish for which there is no multi-char fold to this */
3406 if (len == 2 && *s == 's' && *(s+1) == 's'
3407 && OP(scan) != EXACTFA
3408 && OP(scan) != EXACTFA_NO_TRIE)
3411 if (OP(scan) != EXACTFL) {
3412 OP(scan) = EXACTFU_SS;
3416 else { /* Here is a generic multi-char fold. */
3417 U8* multi_end = s + len;
3419 /* Count how many characters in it. In the case of /aa, no
3420 * folds which contain ASCII code points are allowed, so
3421 * check for those, and skip if found. */
3422 if (OP(scan) != EXACTFA && OP(scan) != EXACTFA_NO_TRIE) {
3423 count = utf8_length(s, multi_end);
3427 while (s < multi_end) {
3430 goto next_iteration;
3440 /* The delta is how long the sequence is minus 1 (1 is how long
3441 * the character that folds to the sequence is) */
3442 total_count_delta += count - 1;
3446 /* We created a temporary folded copy of the string in EXACTFL
3447 * nodes. Therefore we need to be sure it doesn't go below zero,
3448 * as the real string could be shorter */
3449 if (OP(scan) == EXACTFL) {
3450 int total_chars = utf8_length((U8*) STRING(scan),
3451 (U8*) STRING(scan) + STR_LEN(scan));
3452 if (total_count_delta > total_chars) {
3453 total_count_delta = total_chars;
3457 *min_subtract += total_count_delta;
3460 else if (OP(scan) == EXACTFA) {
3462 /* Non-UTF-8 pattern, EXACTFA node. There can't be a multi-char
3463 * fold to the ASCII range (and there are no existing ones in the
3464 * upper latin1 range). But, as outlined in the comments preceding
3465 * this function, we need to flag any occurrences of the sharp s.
3466 * This character forbids trie formation (because of added
3469 if (*s == LATIN_SMALL_LETTER_SHARP_S) {
3470 OP(scan) = EXACTFA_NO_TRIE;
3471 *unfolded_multi_char = TRUE;
3480 /* Non-UTF-8 pattern, not EXACTFA node. Look for the multi-char
3481 * folds that are all Latin1. As explained in the comments
3482 * preceding this function, we look also for the sharp s in EXACTF
3483 * and EXACTFL nodes; it can be in the final position. Otherwise
3484 * we can stop looking 1 byte earlier because have to find at least
3485 * two characters for a multi-fold */
3486 const U8* upper = (OP(scan) == EXACTF || OP(scan) == EXACTFL)
3491 int len = is_MULTI_CHAR_FOLD_latin1_safe(s, s_end);
3492 if (! len) { /* Not a multi-char fold. */
3493 if (*s == LATIN_SMALL_LETTER_SHARP_S
3494 && (OP(scan) == EXACTF || OP(scan) == EXACTFL))
3496 *unfolded_multi_char = TRUE;
3503 && isARG2_lower_or_UPPER_ARG1('s', *s)
3504 && isARG2_lower_or_UPPER_ARG1('s', *(s+1)))
3507 /* EXACTF nodes need to know that the minimum length
3508 * changed so that a sharp s in the string can match this
3509 * ss in the pattern, but they remain EXACTF nodes, as they
3510 * won't match this unless the target string is is UTF-8,
3511 * which we don't know until runtime. EXACTFL nodes can't
3512 * transform into EXACTFU nodes */
3513 if (OP(scan) != EXACTF && OP(scan) != EXACTFL) {
3514 OP(scan) = EXACTFU_SS;
3518 *min_subtract += len - 1;
3525 /* Allow dumping but overwriting the collection of skipped
3526 * ops and/or strings with fake optimized ops */
3527 n = scan + NODE_SZ_STR(scan);
3535 DEBUG_OPTIMISE_r(if (merged){DEBUG_PEEP("finl",scan,depth)});
3539 /* REx optimizer. Converts nodes into quicker variants "in place".
3540 Finds fixed substrings. */
3542 /* Stops at toplevel WHILEM as well as at "last". At end *scanp is set
3543 to the position after last scanned or to NULL. */
3545 #define INIT_AND_WITHP \
3546 assert(!and_withp); \
3547 Newx(and_withp,1, regnode_ssc); \
3548 SAVEFREEPV(and_withp)
3550 /* this is a chain of data about sub patterns we are processing that
3551 need to be handled separately/specially in study_chunk. Its so
3552 we can simulate recursion without losing state. */
3554 typedef struct scan_frame {
3555 regnode *last; /* last node to process in this frame */
3556 regnode *next; /* next node to process when last is reached */
3557 struct scan_frame *prev; /*previous frame*/
3558 U32 prev_recursed_depth;
3559 I32 stop; /* what stopparen do we use */
3564 S_study_chunk(pTHX_ RExC_state_t *pRExC_state, regnode **scanp,
3565 SSize_t *minlenp, SSize_t *deltap,
3570 regnode_ssc *and_withp,
3571 U32 flags, U32 depth)
3572 /* scanp: Start here (read-write). */
3573 /* deltap: Write maxlen-minlen here. */
3574 /* last: Stop before this one. */
3575 /* data: string data about the pattern */
3576 /* stopparen: treat close N as END */
3577 /* recursed: which subroutines have we recursed into */
3578 /* and_withp: Valid if flags & SCF_DO_STCLASS_OR */
3581 /* There must be at least this number of characters to match */
3584 regnode *scan = *scanp, *next;
3586 int is_inf = (flags & SCF_DO_SUBSTR) && (data->flags & SF_IS_INF);
3587 int is_inf_internal = 0; /* The studied chunk is infinite */
3588 I32 is_par = OP(scan) == OPEN ? ARG(scan) : 0;
3589 scan_data_t data_fake;
3590 SV *re_trie_maxbuff = NULL;
3591 regnode *first_non_open = scan;
3592 SSize_t stopmin = SSize_t_MAX;
3593 scan_frame *frame = NULL;
3594 GET_RE_DEBUG_FLAGS_DECL;
3596 PERL_ARGS_ASSERT_STUDY_CHUNK;
3599 StructCopy(&zero_scan_data, &data_fake, scan_data_t);
3602 while (first_non_open && OP(first_non_open) == OPEN)
3603 first_non_open=regnext(first_non_open);
3608 while ( scan && OP(scan) != END && scan < last ){
3609 UV min_subtract = 0; /* How mmany chars to subtract from the minimum
3610 node length to get a real minimum (because
3611 the folded version may be shorter) */
3612 bool unfolded_multi_char = FALSE;
3613 /* Peephole optimizer: */
3614 DEBUG_OPTIMISE_MORE_r(
3616 PerlIO_printf(Perl_debug_log,
3617 "%*sstudy_chunk stopparen=%ld depth=%lu recursed_depth=%lu ",
3618 ((int) depth*2), "", (long)stopparen,
3619 (unsigned long)depth, (unsigned long)recursed_depth);
3620 if (recursed_depth) {
3623 for ( j = 0 ; j < recursed_depth ; j++ ) {
3624 PerlIO_printf(Perl_debug_log,"[");
3625 for ( i = 0 ; i < (U32)RExC_npar ; i++ )
3626 PerlIO_printf(Perl_debug_log,"%d",
3627 PAREN_TEST(RExC_study_chunk_recursed +
3628 (j * RExC_study_chunk_recursed_bytes), i)
3631 PerlIO_printf(Perl_debug_log,"]");
3634 PerlIO_printf(Perl_debug_log,"\n");
3637 DEBUG_STUDYDATA("Peep:", data, depth);
3638 DEBUG_PEEP("Peep", scan, depth);
3641 /* The reason we do this here we need to deal with things like /(?:f)(?:o)(?:o)/
3642 * which cant be dealt with by the normal EXACT parsing code, as each (?:..) is handled
3643 * by a different invocation of reg() -- Yves
3645 JOIN_EXACT(scan,&min_subtract, &unfolded_multi_char, 0);
3647 /* Follow the next-chain of the current node and optimize
3648 away all the NOTHINGs from it. */
3649 if (OP(scan) != CURLYX) {
3650 const int max = (reg_off_by_arg[OP(scan)]
3652 /* I32 may be smaller than U16 on CRAYs! */
3653 : (I32_MAX < U16_MAX ? I32_MAX : U16_MAX));
3654 int off = (reg_off_by_arg[OP(scan)] ? ARG(scan) : NEXT_OFF(scan));
3658 /* Skip NOTHING and LONGJMP. */
3659 while ((n = regnext(n))
3660 && ((PL_regkind[OP(n)] == NOTHING && (noff = NEXT_OFF(n)))
3661 || ((OP(n) == LONGJMP) && (noff = ARG(n))))
3662 && off + noff < max)
3664 if (reg_off_by_arg[OP(scan)])
3667 NEXT_OFF(scan) = off;
3672 /* The principal pseudo-switch. Cannot be a switch, since we
3673 look into several different things. */
3674 if (OP(scan) == BRANCH || OP(scan) == BRANCHJ
3675 || OP(scan) == IFTHEN) {
3676 next = regnext(scan);
3678 /* demq: the op(next)==code check is to see if we have
3679 * "branch-branch" AFAICT */
3681 if (OP(next) == code || code == IFTHEN) {
3682 /* NOTE - There is similar code to this block below for
3683 * handling TRIE nodes on a re-study. If you change stuff here
3684 * check there too. */
3685 SSize_t max1 = 0, min1 = SSize_t_MAX, num = 0;
3687 regnode * const startbranch=scan;
3689 if (flags & SCF_DO_SUBSTR) {
3690 /* Cannot merge strings after this. */
3691 scan_commit(pRExC_state, data, minlenp, is_inf);
3694 if (flags & SCF_DO_STCLASS)
3695 ssc_init_zero(pRExC_state, &accum);
3697 while (OP(scan) == code) {
3698 SSize_t deltanext, minnext, fake;
3700 regnode_ssc this_class;
3703 data_fake.flags = 0;
3705 data_fake.whilem_c = data->whilem_c;
3706 data_fake.last_closep = data->last_closep;
3709 data_fake.last_closep = &fake;
3711 data_fake.pos_delta = delta;
3712 next = regnext(scan);
3713 scan = NEXTOPER(scan);
3715 scan = NEXTOPER(scan);
3716 if (flags & SCF_DO_STCLASS) {
3717 ssc_init(pRExC_state, &this_class);
3718 data_fake.start_class = &this_class;
3719 f = SCF_DO_STCLASS_AND;
3721 if (flags & SCF_WHILEM_VISITED_POS)
3722 f |= SCF_WHILEM_VISITED_POS;
3724 /* we suppose the run is continuous, last=next...*/
3725 minnext = study_chunk(pRExC_state, &scan, minlenp,
3726 &deltanext, next, &data_fake, stopparen,
3727 recursed_depth, NULL, f,depth+1);
3730 if (deltanext == SSize_t_MAX) {
3731 is_inf = is_inf_internal = 1;
3733 } else if (max1 < minnext + deltanext)
3734 max1 = minnext + deltanext;
3736 if (data_fake.flags & (SF_HAS_PAR|SF_IN_PAR))
3738 if (data_fake.flags & SCF_SEEN_ACCEPT) {
3739 if ( stopmin > minnext)
3740 stopmin = min + min1;
3741 flags &= ~SCF_DO_SUBSTR;
3743 data->flags |= SCF_SEEN_ACCEPT;
3746 if (data_fake.flags & SF_HAS_EVAL)
3747 data->flags |= SF_HAS_EVAL;
3748 data->whilem_c = data_fake.whilem_c;
3750 if (flags & SCF_DO_STCLASS)
3751 ssc_or(pRExC_state, &accum, (regnode_charclass*)&this_class);
3753 if (code == IFTHEN && num < 2) /* Empty ELSE branch */
3755 if (flags & SCF_DO_SUBSTR) {
3756 data->pos_min += min1;
3757 if (data->pos_delta >= SSize_t_MAX - (max1 - min1))
3758 data->pos_delta = SSize_t_MAX;
3760 data->pos_delta += max1 - min1;
3761 if (max1 != min1 || is_inf)
3762 data->longest = &(data->longest_float);
3765 if (delta == SSize_t_MAX
3766 || SSize_t_MAX - delta - (max1 - min1) < 0)
3767 delta = SSize_t_MAX;
3769 delta += max1 - min1;
3770 if (flags & SCF_DO_STCLASS_OR) {
3771 ssc_or(pRExC_state, data->start_class, (regnode_charclass*) &accum);
3773 ssc_and(pRExC_state, data->start_class, (regnode_charclass *) and_withp);
3774 flags &= ~SCF_DO_STCLASS;
3777 else if (flags & SCF_DO_STCLASS_AND) {
3779 ssc_and(pRExC_state, data->start_class, (regnode_charclass *) &accum);
3780 flags &= ~SCF_DO_STCLASS;
3783 /* Switch to OR mode: cache the old value of
3784 * data->start_class */
3786 StructCopy(data->start_class, and_withp, regnode_ssc);
3787 flags &= ~SCF_DO_STCLASS_AND;
3788 StructCopy(&accum, data->start_class, regnode_ssc);
3789 flags |= SCF_DO_STCLASS_OR;
3793 if (PERL_ENABLE_TRIE_OPTIMISATION &&
3794 OP( startbranch ) == BRANCH )
3798 Assuming this was/is a branch we are dealing with: 'scan'
3799 now points at the item that follows the branch sequence,
3800 whatever it is. We now start at the beginning of the
3801 sequence and look for subsequences of
3807 which would be constructed from a pattern like
3810 If we can find such a subsequence we need to turn the first
3811 element into a trie and then add the subsequent branch exact
3812 strings to the trie.
3816 1. patterns where the whole set of branches can be
3819 2. patterns where only a subset can be converted.
3821 In case 1 we can replace the whole set with a single regop
3822 for the trie. In case 2 we need to keep the start and end
3825 'BRANCH EXACT; BRANCH EXACT; BRANCH X'
3826 becomes BRANCH TRIE; BRANCH X;
3828 There is an additional case, that being where there is a
3829 common prefix, which gets split out into an EXACT like node
3830 preceding the TRIE node.
3832 If x(1..n)==tail then we can do a simple trie, if not we make
3833 a "jump" trie, such that when we match the appropriate word
3834 we "jump" to the appropriate tail node. Essentially we turn
3835 a nested if into a case structure of sorts.
3840 if (!re_trie_maxbuff) {
3841 re_trie_maxbuff = get_sv(RE_TRIE_MAXBUF_NAME, 1);
3842 if (!SvIOK(re_trie_maxbuff))
3843 sv_setiv(re_trie_maxbuff, RE_TRIE_MAXBUF_INIT);
3845 if ( SvIV(re_trie_maxbuff)>=0 ) {
3847 regnode *first = (regnode *)NULL;
3848 regnode *last = (regnode *)NULL;
3849 regnode *tail = scan;
3854 SV * const mysv = sv_newmortal(); /* for dumping */
3856 /* var tail is used because there may be a TAIL
3857 regop in the way. Ie, the exacts will point to the
3858 thing following the TAIL, but the last branch will
3859 point at the TAIL. So we advance tail. If we
3860 have nested (?:) we may have to move through several
3864 while ( OP( tail ) == TAIL ) {
3865 /* this is the TAIL generated by (?:) */
3866 tail = regnext( tail );
3870 DEBUG_TRIE_COMPILE_r({
3871 regprop(RExC_rx, mysv, tail, NULL);
3872 PerlIO_printf( Perl_debug_log, "%*s%s%s\n",
3873 (int)depth * 2 + 2, "",
3874 "Looking for TRIE'able sequences. Tail node is: ",
3875 SvPV_nolen_const( mysv )
3881 Step through the branches
3882 cur represents each branch,
3883 noper is the first thing to be matched as part
3885 noper_next is the regnext() of that node.
3887 We normally handle a case like this
3888 /FOO[xyz]|BAR[pqr]/ via a "jump trie" but we also
3889 support building with NOJUMPTRIE, which restricts
3890 the trie logic to structures like /FOO|BAR/.
3892 If noper is a trieable nodetype then the branch is
3893 a possible optimization target. If we are building
3894 under NOJUMPTRIE then we require that noper_next is
3895 the same as scan (our current position in the regex
3898 Once we have two or more consecutive such branches
3899 we can create a trie of the EXACT's contents and
3900 stitch it in place into the program.
3902 If the sequence represents all of the branches in
3903 the alternation we replace the entire thing with a
3906 Otherwise when it is a subsequence we need to
3907 stitch it in place and replace only the relevant
3908 branches. This means the first branch has to remain
3909 as it is used by the alternation logic, and its
3910 next pointer, and needs to be repointed at the item
3911 on the branch chain following the last branch we
3912 have optimized away.
3914 This could be either a BRANCH, in which case the
3915 subsequence is internal, or it could be the item
3916 following the branch sequence in which case the
3917 subsequence is at the end (which does not
3918 necessarily mean the first node is the start of the
3921 TRIE_TYPE(X) is a define which maps the optype to a
3925 ----------------+-----------
3929 EXACTFU_SS | EXACTFU
3934 #define TRIE_TYPE(X) ( ( NOTHING == (X) ) ? NOTHING : \
3935 ( EXACT == (X) ) ? EXACT : \
3936 ( EXACTFU == (X) || EXACTFU_SS == (X) ) ? EXACTFU : \
3937 ( EXACTFA == (X) ) ? EXACTFA : \
3940 /* dont use tail as the end marker for this traverse */
3941 for ( cur = startbranch ; cur != scan ; cur = regnext( cur ) ) {
3942 regnode * const noper = NEXTOPER( cur );
3943 U8 noper_type = OP( noper );
3944 U8 noper_trietype = TRIE_TYPE( noper_type );
3945 #if defined(DEBUGGING) || defined(NOJUMPTRIE)
3946 regnode * const noper_next = regnext( noper );
3947 U8 noper_next_type = (noper_next && noper_next != tail) ? OP(noper_next) : 0;
3948 U8 noper_next_trietype = (noper_next && noper_next != tail) ? TRIE_TYPE( noper_next_type ) :0;
3951 DEBUG_TRIE_COMPILE_r({
3952 regprop(RExC_rx, mysv, cur, NULL);
3953 PerlIO_printf( Perl_debug_log, "%*s- %s (%d)",
3954 (int)depth * 2 + 2,"", SvPV_nolen_const( mysv ), REG_NODE_NUM(cur) );
3956 regprop(RExC_rx, mysv, noper, NULL);
3957 PerlIO_printf( Perl_debug_log, " -> %s",
3958 SvPV_nolen_const(mysv));
3961 regprop(RExC_rx, mysv, noper_next, NULL);
3962 PerlIO_printf( Perl_debug_log,"\t=> %s\t",
3963 SvPV_nolen_const(mysv));
3965 PerlIO_printf( Perl_debug_log, "(First==%d,Last==%d,Cur==%d,tt==%s,nt==%s,nnt==%s)\n",
3966 REG_NODE_NUM(first), REG_NODE_NUM(last), REG_NODE_NUM(cur),
3967 PL_reg_name[trietype], PL_reg_name[noper_trietype], PL_reg_name[noper_next_trietype]
3971 /* Is noper a trieable nodetype that can be merged
3972 * with the current trie (if there is one)? */
3976 ( noper_trietype == NOTHING)
3977 || ( trietype == NOTHING )
3978 || ( trietype == noper_trietype )
3981 && noper_next == tail
3985 /* Handle mergable triable node Either we are
3986 * the first node in a new trieable sequence,
3987 * in which case we do some bookkeeping,
3988 * otherwise we update the end pointer. */
3991 if ( noper_trietype == NOTHING ) {
3992 #if !defined(DEBUGGING) && !defined(NOJUMPTRIE)
3993 regnode * const noper_next = regnext( noper );
3994 U8 noper_next_type = (noper_next && noper_next!=tail) ? OP(noper_next) : 0;
3995 U8 noper_next_trietype = noper_next_type ? TRIE_TYPE( noper_next_type ) :0;
3998 if ( noper_next_trietype ) {
3999 trietype = noper_next_trietype;
4000 } else if (noper_next_type) {
4001 /* a NOTHING regop is 1 regop wide.
4002 * We need at least two for a trie
4003 * so we can't merge this in */
4007 trietype = noper_trietype;
4010 if ( trietype == NOTHING )
4011 trietype = noper_trietype;
4016 } /* end handle mergable triable node */
4018 /* handle unmergable node -
4019 * noper may either be a triable node which can
4020 * not be tried together with the current trie,
4021 * or a non triable node */
4023 /* If last is set and trietype is not
4024 * NOTHING then we have found at least two
4025 * triable branch sequences in a row of a
4026 * similar trietype so we can turn them
4027 * into a trie. If/when we allow NOTHING to
4028 * start a trie sequence this condition
4029 * will be required, and it isn't expensive
4030 * so we leave it in for now. */
4031 if ( trietype && trietype != NOTHING )
4032 make_trie( pRExC_state,
4033 startbranch, first, cur, tail,
4034 count, trietype, depth+1 );
4035 last = NULL; /* note: we clear/update
4036 first, trietype etc below,
4037 so we dont do it here */
4041 && noper_next == tail
4044 /* noper is triable, so we can start a new
4048 trietype = noper_trietype;
4050 /* if we already saw a first but the
4051 * current node is not triable then we have
4052 * to reset the first information. */
4057 } /* end handle unmergable node */
4058 } /* loop over branches */
4059 DEBUG_TRIE_COMPILE_r({
4060 regprop(RExC_rx, mysv, cur, NULL);
4061 PerlIO_printf( Perl_debug_log,
4062 "%*s- %s (%d) <SCAN FINISHED>\n",
4064 "", SvPV_nolen_const( mysv ),REG_NODE_NUM(cur));
4067 if ( last && trietype ) {
4068 if ( trietype != NOTHING ) {
4069 /* the last branch of the sequence was part of
4070 * a trie, so we have to construct it here
4071 * outside of the loop */
4072 made= make_trie( pRExC_state, startbranch,
4073 first, scan, tail, count,
4074 trietype, depth+1 );
4075 #ifdef TRIE_STUDY_OPT
4076 if ( ((made == MADE_EXACT_TRIE &&
4077 startbranch == first)
4078 || ( first_non_open == first )) &&
4080 flags |= SCF_TRIE_RESTUDY;
4081 if ( startbranch == first
4084 RExC_seen &=~REG_TOP_LEVEL_BRANCHES_SEEN;
4089 /* at this point we know whatever we have is a
4090 * NOTHING sequence/branch AND if 'startbranch'
4091 * is 'first' then we can turn the whole thing
4094 if ( startbranch == first ) {
4096 /* the entire thing is a NOTHING sequence,
4097 * something like this: (?:|) So we can
4098 * turn it into a plain NOTHING op. */
4099 DEBUG_TRIE_COMPILE_r({
4100 regprop(RExC_rx, mysv, cur, NULL);
4101 PerlIO_printf( Perl_debug_log,
4102 "%*s- %s (%d) <NOTHING BRANCH SEQUENCE>\n", (int)depth * 2 + 2,
4103 "", SvPV_nolen_const( mysv ),REG_NODE_NUM(cur));
4106 OP(startbranch)= NOTHING;
4107 NEXT_OFF(startbranch)= tail - startbranch;
4108 for ( opt= startbranch + 1; opt < tail ; opt++ )
4112 } /* end if ( last) */
4113 } /* TRIE_MAXBUF is non zero */
4118 else if ( code == BRANCHJ ) { /* single branch is optimized. */
4119 scan = NEXTOPER(NEXTOPER(scan));
4120 } else /* single branch is optimized. */
4121 scan = NEXTOPER(scan);
4123 } else if (OP(scan) == SUSPEND || OP(scan) == GOSUB || OP(scan) == GOSTART) {
4124 scan_frame *newframe = NULL;
4128 U32 my_recursed_depth= recursed_depth;
4130 if (OP(scan) != SUSPEND) {
4131 /* set the pointer */
4132 if (OP(scan) == GOSUB) {
4134 RExC_recurse[ARG2L(scan)] = scan;
4135 start = RExC_open_parens[paren-1];
4136 end = RExC_close_parens[paren-1];
4139 start = RExC_rxi->program + 1;
4144 !PAREN_TEST(RExC_study_chunk_recursed + ((recursed_depth-1) * RExC_study_chunk_recursed_bytes), paren)
4146 if (!recursed_depth) {
4147 Zero(RExC_study_chunk_recursed, RExC_study_chunk_recursed_bytes, U8);
4149 Copy(RExC_study_chunk_recursed + ((recursed_depth-1) * RExC_study_chunk_recursed_bytes),
4150 RExC_study_chunk_recursed + (recursed_depth * RExC_study_chunk_recursed_bytes),
4151 RExC_study_chunk_recursed_bytes, U8);
4153 /* we havent recursed into this paren yet, so recurse into it */
4154 DEBUG_STUDYDATA("set:", data,depth);
4155 PAREN_SET(RExC_study_chunk_recursed + (recursed_depth * RExC_study_chunk_recursed_bytes), paren);
4156 my_recursed_depth= recursed_depth + 1;
4157 Newx(newframe,1,scan_frame);
4159 DEBUG_STUDYDATA("inf:", data,depth);
4160 /* some form of infinite recursion, assume infinite length
4162 if (flags & SCF_DO_SUBSTR) {
4163 scan_commit(pRExC_state, data, minlenp, is_inf);
4164 data->longest = &(data->longest_float);
4166 is_inf = is_inf_internal = 1;
4167 if (flags & SCF_DO_STCLASS_OR) /* Allow everything */
4168 ssc_anything(data->start_class);
4169 flags &= ~SCF_DO_STCLASS;
4172 Newx(newframe,1,scan_frame);
4175 end = regnext(scan);
4180 SAVEFREEPV(newframe);
4181 newframe->next = regnext(scan);
4182 newframe->last = last;
4183 newframe->stop = stopparen;
4184 newframe->prev = frame;
4185 newframe->prev_recursed_depth = recursed_depth;
4187 DEBUG_STUDYDATA("frame-new:",data,depth);
4188 DEBUG_PEEP("fnew", scan, depth);
4195 recursed_depth= my_recursed_depth;
4200 else if (OP(scan) == EXACT) {
4201 SSize_t l = STR_LEN(scan);
4204 const U8 * const s = (U8*)STRING(scan);
4205 uc = utf8_to_uvchr_buf(s, s + l, NULL);
4206 l = utf8_length(s, s + l);
4208 uc = *((U8*)STRING(scan));
4211 if (flags & SCF_DO_SUBSTR) { /* Update longest substr. */
4212 /* The code below prefers earlier match for fixed
4213 offset, later match for variable offset. */
4214 if (data->last_end == -1) { /* Update the start info. */
4215 data->last_start_min = data->pos_min;
4216 data->last_start_max = is_inf
4217 ? SSize_t_MAX : data->pos_min + data->pos_delta;
4219 sv_catpvn(data->last_found, STRING(scan), STR_LEN(scan));
4221 SvUTF8_on(data->last_found);
4223 SV * const sv = data->last_found;
4224 MAGIC * const mg = SvUTF8(sv) && SvMAGICAL(sv) ?
4225 mg_find(sv, PERL_MAGIC_utf8) : NULL;
4226 if (mg && mg->mg_len >= 0)
4227 mg->mg_len += utf8_length((U8*)STRING(scan),
4228 (U8*)STRING(scan)+STR_LEN(scan));
4230 data->last_end = data->pos_min + l;
4231 data->pos_min += l; /* As in the first entry. */
4232 data->flags &= ~SF_BEFORE_EOL;
4235 /* ANDing the code point leaves at most it, and not in locale, and
4236 * can't match null string */
4237 if (flags & SCF_DO_STCLASS_AND) {
4238 ssc_cp_and(data->start_class, uc);
4239 ANYOF_FLAGS(data->start_class) &= ~ANYOF_EMPTY_STRING;
4240 ssc_clear_locale(data->start_class);
4242 else if (flags & SCF_DO_STCLASS_OR) {
4243 ssc_add_cp(data->start_class, uc);
4244 ssc_and(pRExC_state, data->start_class, (regnode_charclass *) and_withp);
4246 /* See commit msg 749e076fceedeb708a624933726e7989f2302f6a */
4247 ANYOF_FLAGS(data->start_class) &= ~ANYOF_EMPTY_STRING;
4249 flags &= ~SCF_DO_STCLASS;
4251 else if (PL_regkind[OP(scan)] == EXACT) { /* But OP != EXACT! */
4252 SSize_t l = STR_LEN(scan);
4253 UV uc = *((U8*)STRING(scan));
4254 SV* EXACTF_invlist = _new_invlist(4); /* Start out big enough for 2
4255 separate code points */
4257 /* Search for fixed substrings supports EXACT only. */
4258 if (flags & SCF_DO_SUBSTR) {
4260 scan_commit(pRExC_state, data, minlenp, is_inf);
4263 const U8 * const s = (U8 *)STRING(scan);
4264 uc = utf8_to_uvchr_buf(s, s + l, NULL);
4265 l = utf8_length(s, s + l);
4267 if (unfolded_multi_char) {
4268 RExC_seen |= REG_UNFOLDED_MULTI_SEEN;
4270 min += l - min_subtract;
4272 delta += min_subtract;
4273 if (flags & SCF_DO_SUBSTR) {
4274 data->pos_min += l - min_subtract;
4275 if (data->pos_min < 0) {
4278 data->pos_delta += min_subtract;
4280 data->longest = &(data->longest_float);
4283 if (OP(scan) == EXACTFL) {
4285 /* We don't know what the folds are; it could be anything. XXX
4286 * Actually, we only support UTF-8 encoding for code points
4287 * above Latin1, so we could know what those folds are. */
4288 EXACTF_invlist = _add_range_to_invlist(EXACTF_invlist,
4292 else { /* Non-locale EXACTFish */
4293 EXACTF_invlist = add_cp_to_invlist(EXACTF_invlist, uc);
4294 if (flags & SCF_DO_STCLASS_AND) {
4295 ssc_clear_locale(data->start_class);
4297 if (uc < 256) { /* We know what the Latin1 folds are ... */
4298 if (IS_IN_SOME_FOLD_L1(uc)) { /* For instance, we
4299 know if anything folds
4301 EXACTF_invlist = add_cp_to_invlist(EXACTF_invlist,
4302 PL_fold_latin1[uc]);
4303 if (OP(scan) != EXACTFA) { /* The folds below aren't
4305 if (isARG2_lower_or_UPPER_ARG1('s', uc)) {
4307 = add_cp_to_invlist(EXACTF_invlist,
4308 LATIN_SMALL_LETTER_SHARP_S);
4310 else if (uc == LATIN_SMALL_LETTER_SHARP_S) {
4312 = add_cp_to_invlist(EXACTF_invlist, 's');
4314 = add_cp_to_invlist(EXACTF_invlist, 'S');
4318 /* We also know if there are above-Latin1 code points
4319 * that fold to this (none legal for ASCII and /iaa) */
4320 if ((! isASCII(uc) || OP(scan) != EXACTFA)
4321 && HAS_NONLATIN1_FOLD_CLOSURE(uc))
4323 /* XXX We could know exactly what does fold to this
4324 * if the reverse folds are loaded, as currently in
4326 _invlist_union(EXACTF_invlist,
4332 else { /* Non-locale, above Latin1. XXX We don't currently
4333 know what participates in folds with this, so have
4334 to assume anything could */
4336 /* XXX We could know exactly what does fold to this if the
4337 * reverse folds are loaded, as currently in S_regclass().
4338 * But we do know that under /iaa nothing in the ASCII
4339 * range can participate */
4340 if (OP(scan) == EXACTFA) {
4341 _invlist_union_complement_2nd(EXACTF_invlist,
4342 PL_XPosix_ptrs[_CC_ASCII],
4346 EXACTF_invlist = _add_range_to_invlist(EXACTF_invlist,
4351 if (flags & SCF_DO_STCLASS_AND) {
4352 ANYOF_FLAGS(data->start_class) &= ~ANYOF_EMPTY_STRING;
4353 ANYOF_POSIXL_ZERO(data->start_class);
4354 ssc_intersection(data->start_class, EXACTF_invlist, FALSE);
4356 else if (flags & SCF_DO_STCLASS_OR) {
4357 ssc_union(data->start_class, EXACTF_invlist, FALSE);
4358 ssc_and(pRExC_state, data->start_class, (regnode_charclass *) and_withp);
4360 /* See commit msg 749e076fceedeb708a624933726e7989f2302f6a */
4361 ANYOF_FLAGS(data->start_class) &= ~ANYOF_EMPTY_STRING;
4363 flags &= ~SCF_DO_STCLASS;
4364 SvREFCNT_dec(EXACTF_invlist);
4366 else if (REGNODE_VARIES(OP(scan))) {
4367 SSize_t mincount, maxcount, minnext, deltanext, pos_before = 0;
4368 I32 fl = 0, f = flags;
4369 regnode * const oscan = scan;
4370 regnode_ssc this_class;
4371 regnode_ssc *oclass = NULL;
4372 I32 next_is_eval = 0;
4374 switch (PL_regkind[OP(scan)]) {
4375 case WHILEM: /* End of (?:...)* . */
4376 scan = NEXTOPER(scan);
4379 if (flags & (SCF_DO_SUBSTR | SCF_DO_STCLASS)) {
4380 next = NEXTOPER(scan);
4381 if (OP(next) == EXACT || (flags & SCF_DO_STCLASS)) {
4383 maxcount = REG_INFTY;
4384 next = regnext(scan);
4385 scan = NEXTOPER(scan);
4389 if (flags & SCF_DO_SUBSTR)
4394 if (flags & SCF_DO_STCLASS) {
4396 maxcount = REG_INFTY;
4397 next = regnext(scan);
4398 scan = NEXTOPER(scan);
4401 if (flags & SCF_DO_SUBSTR) {
4402 scan_commit(pRExC_state, data, minlenp, is_inf);
4403 /* Cannot extend fixed substrings */
4404 data->longest = &(data->longest_float);
4406 is_inf = is_inf_internal = 1;
4407 scan = regnext(scan);
4408 goto optimize_curly_tail;
4410 if (stopparen>0 && (OP(scan)==CURLYN || OP(scan)==CURLYM)
4411 && (scan->flags == stopparen))
4416 mincount = ARG1(scan);
4417 maxcount = ARG2(scan);
4419 next = regnext(scan);
4420 if (OP(scan) == CURLYX) {
4421 I32 lp = (data ? *(data->last_closep) : 0);
4422 scan->flags = ((lp <= (I32)U8_MAX) ? (U8)lp : U8_MAX);
4424 scan = NEXTOPER(scan) + EXTRA_STEP_2ARGS;
4425 next_is_eval = (OP(scan) == EVAL);
4427 if (flags & SCF_DO_SUBSTR) {
4429 scan_commit(pRExC_state, data, minlenp, is_inf);
4430 /* Cannot extend fixed substrings */
4431 pos_before = data->pos_min;
4435 data->flags &= ~(SF_HAS_PAR|SF_IN_PAR|SF_HAS_EVAL);
4437 data->flags |= SF_IS_INF;
4439 if (flags & SCF_DO_STCLASS) {
4440 ssc_init(pRExC_state, &this_class);
4441 oclass = data->start_class;
4442 data->start_class = &this_class;
4443 f |= SCF_DO_STCLASS_AND;
4444 f &= ~SCF_DO_STCLASS_OR;
4446 /* Exclude from super-linear cache processing any {n,m}
4447 regops for which the combination of input pos and regex
4448 pos is not enough information to determine if a match
4451 For example, in the regex /foo(bar\s*){4,8}baz/ with the
4452 regex pos at the \s*, the prospects for a match depend not
4453 only on the input position but also on how many (bar\s*)
4454 repeats into the {4,8} we are. */
4455 if ((mincount > 1) || (maxcount > 1 && maxcount != REG_INFTY))
4456 f &= ~SCF_WHILEM_VISITED_POS;
4458 /* This will finish on WHILEM, setting scan, or on NULL: */
4459 minnext = study_chunk(pRExC_state, &scan, minlenp, &deltanext,
4460 last, data, stopparen, recursed_depth, NULL,
4462 ? (f & ~SCF_DO_SUBSTR)
4466 if (flags & SCF_DO_STCLASS)
4467 data->start_class = oclass;
4468 if (mincount == 0 || minnext == 0) {
4469 if (flags & SCF_DO_STCLASS_OR) {
4470 ssc_or(pRExC_state, data->start_class, (regnode_charclass *) &this_class);
4472 else if (flags & SCF_DO_STCLASS_AND) {
4473 /* Switch to OR mode: cache the old value of
4474 * data->start_class */
4476 StructCopy(data->start_class, and_withp, regnode_ssc);
4477 flags &= ~SCF_DO_STCLASS_AND;
4478 StructCopy(&this_class, data->start_class, regnode_ssc);
4479 flags |= SCF_DO_STCLASS_OR;
4480 ANYOF_FLAGS(data->start_class) |= ANYOF_EMPTY_STRING;
4482 } else { /* Non-zero len */
4483 if (flags & SCF_DO_STCLASS_OR) {
4484 ssc_or(pRExC_state, data->start_class, (regnode_charclass *) &this_class);
4485 ssc_and(pRExC_state, data->start_class, (regnode_charclass *) and_withp);
4487 else if (flags & SCF_DO_STCLASS_AND)
4488 ssc_and(pRExC_state, data->start_class, (regnode_charclass *) &this_class);
4489 flags &= ~SCF_DO_STCLASS;
4491 if (!scan) /* It was not CURLYX, but CURLY. */
4493 if (!(flags & SCF_TRIE_DOING_RESTUDY)
4494 /* ? quantifier ok, except for (?{ ... }) */
4495 && (next_is_eval || !(mincount == 0 && maxcount == 1))
4496 && (minnext == 0) && (deltanext == 0)
4497 && data && !(data->flags & (SF_HAS_PAR|SF_IN_PAR))
4498 && maxcount <= REG_INFTY/3) /* Complement check for big
4501 /* Fatal warnings may leak the regexp without this: */
4502 SAVEFREESV(RExC_rx_sv);
4503 ckWARNreg(RExC_parse,
4504 "Quantifier unexpected on zero-length expression");
4505 (void)ReREFCNT_inc(RExC_rx_sv);
4508 min += minnext * mincount;
4509 is_inf_internal |= deltanext == SSize_t_MAX
4510 || (maxcount == REG_INFTY && minnext + deltanext > 0);
4511 is_inf |= is_inf_internal;
4513 delta = SSize_t_MAX;
4515 delta += (minnext + deltanext) * maxcount
4516 - minnext * mincount;
4518 /* Try powerful optimization CURLYX => CURLYN. */
4519 if ( OP(oscan) == CURLYX && data
4520 && data->flags & SF_IN_PAR
4521 && !(data->flags & SF_HAS_EVAL)
4522 && !deltanext && minnext == 1 ) {
4523 /* Try to optimize to CURLYN. */
4524 regnode *nxt = NEXTOPER(oscan) + EXTRA_STEP_2ARGS;
4525 regnode * const nxt1 = nxt;
4532 if (!REGNODE_SIMPLE(OP(nxt))
4533 && !(PL_regkind[OP(nxt)] == EXACT
4534 && STR_LEN(nxt) == 1))
4540 if (OP(nxt) != CLOSE)
4542 if (RExC_open_parens) {
4543 RExC_open_parens[ARG(nxt1)-1]=oscan; /*open->CURLYM*/
4544 RExC_close_parens[ARG(nxt1)-1]=nxt+2; /*close->while*/
4546 /* Now we know that nxt2 is the only contents: */
4547 oscan->flags = (U8)ARG(nxt);
4549 OP(nxt1) = NOTHING; /* was OPEN. */
4552 OP(nxt1 + 1) = OPTIMIZED; /* was count. */
4553 NEXT_OFF(nxt1+ 1) = 0; /* just for consistency. */
4554 NEXT_OFF(nxt2) = 0; /* just for consistency with CURLY. */
4555 OP(nxt) = OPTIMIZED; /* was CLOSE. */
4556 OP(nxt + 1) = OPTIMIZED; /* was count. */
4557 NEXT_OFF(nxt+ 1) = 0; /* just for consistency. */
4562 /* Try optimization CURLYX => CURLYM. */
4563 if ( OP(oscan) == CURLYX && data
4564 && !(data->flags & SF_HAS_PAR)
4565 && !(data->flags & SF_HAS_EVAL)
4566 && !deltanext /* atom is fixed width */
4567 && minnext != 0 /* CURLYM can't handle zero width */
4569 /* Nor characters whose fold at run-time may be
4570 * multi-character */
4571 && ! (RExC_seen & REG_UNFOLDED_MULTI_SEEN)
4573 /* XXXX How to optimize if data == 0? */
4574 /* Optimize to a simpler form. */
4575 regnode *nxt = NEXTOPER(oscan) + EXTRA_STEP_2ARGS; /* OPEN */
4579 while ( (nxt2 = regnext(nxt)) /* skip over embedded stuff*/
4580 && (OP(nxt2) != WHILEM))
4582 OP(nxt2) = SUCCEED; /* Whas WHILEM */
4583 /* Need to optimize away parenths. */
4584 if ((data->flags & SF_IN_PAR) && OP(nxt) == CLOSE) {
4585 /* Set the parenth number. */
4586 regnode *nxt1 = NEXTOPER(oscan) + EXTRA_STEP_2ARGS; /* OPEN*/
4588 oscan->flags = (U8)ARG(nxt);
4589 if (RExC_open_parens) {
4590 RExC_open_parens[ARG(nxt1)-1]=oscan; /*open->CURLYM*/
4591 RExC_close_parens[ARG(nxt1)-1]=nxt2+1; /*close->NOTHING*/
4593 OP(nxt1) = OPTIMIZED; /* was OPEN. */
4594 OP(nxt) = OPTIMIZED; /* was CLOSE. */
4597 OP(nxt1 + 1) = OPTIMIZED; /* was count. */
4598 OP(nxt + 1) = OPTIMIZED; /* was count. */
4599 NEXT_OFF(nxt1 + 1) = 0; /* just for consistency. */
4600 NEXT_OFF(nxt + 1) = 0; /* just for consistency. */
4603 while ( nxt1 && (OP(nxt1) != WHILEM)) {
4604 regnode *nnxt = regnext(nxt1);
4606 if (reg_off_by_arg[OP(nxt1)])
4607 ARG_SET(nxt1, nxt2 - nxt1);
4608 else if (nxt2 - nxt1 < U16_MAX)
4609 NEXT_OFF(nxt1) = nxt2 - nxt1;
4611 OP(nxt) = NOTHING; /* Cannot beautify */
4616 /* Optimize again: */
4617 study_chunk(pRExC_state, &nxt1, minlenp, &deltanext, nxt,
4618 NULL, stopparen, recursed_depth, NULL, 0,depth+1);
4623 else if ((OP(oscan) == CURLYX)
4624 && (flags & SCF_WHILEM_VISITED_POS)
4625 /* See the comment on a similar expression above.
4626 However, this time it's not a subexpression
4627 we care about, but the expression itself. */
4628 && (maxcount == REG_INFTY)
4629 && data && ++data->whilem_c < 16) {
4630 /* This stays as CURLYX, we can put the count/of pair. */
4631 /* Find WHILEM (as in regexec.c) */
4632 regnode *nxt = oscan + NEXT_OFF(oscan);
4634 if (OP(PREVOPER(nxt)) == NOTHING) /* LONGJMP */
4636 PREVOPER(nxt)->flags = (U8)(data->whilem_c
4637 | (RExC_whilem_seen << 4)); /* On WHILEM */
4639 if (data && fl & (SF_HAS_PAR|SF_IN_PAR))
4641 if (flags & SCF_DO_SUBSTR) {
4642 SV *last_str = NULL;
4643 STRLEN last_chrs = 0;
4644 int counted = mincount != 0;
4646 if (data->last_end > 0 && mincount != 0) { /* Ends with a
4648 SSize_t b = pos_before >= data->last_start_min
4649 ? pos_before : data->last_start_min;
4651 const char * const s = SvPV_const(data->last_found, l);
4652 SSize_t old = b - data->last_start_min;
4655 old = utf8_hop((U8*)s, old) - (U8*)s;
4657 /* Get the added string: */
4658 last_str = newSVpvn_utf8(s + old, l, UTF);
4659 last_chrs = UTF ? utf8_length((U8*)(s + old),
4660 (U8*)(s + old + l)) : l;
4661 if (deltanext == 0 && pos_before == b) {
4662 /* What was added is a constant string */
4665 SvGROW(last_str, (mincount * l) + 1);
4666 repeatcpy(SvPVX(last_str) + l,
4667 SvPVX_const(last_str), l,
4669 SvCUR_set(last_str, SvCUR(last_str) * mincount);
4670 /* Add additional parts. */
4671 SvCUR_set(data->last_found,
4672 SvCUR(data->last_found) - l);
4673 sv_catsv(data->last_found, last_str);
4675 SV * sv = data->last_found;
4677 SvUTF8(sv) && SvMAGICAL(sv) ?
4678 mg_find(sv, PERL_MAGIC_utf8) : NULL;
4679 if (mg && mg->mg_len >= 0)
4680 mg->mg_len += last_chrs * (mincount-1);
4682 last_chrs *= mincount;
4683 data->last_end += l * (mincount - 1);
4686 /* start offset must point into the last copy */
4687 data->last_start_min += minnext * (mincount - 1);
4688 data->last_start_max += is_inf ? SSize_t_MAX
4689 : (maxcount - 1) * (minnext + data->pos_delta);
4692 /* It is counted once already... */
4693 data->pos_min += minnext * (mincount - counted);
4695 PerlIO_printf(Perl_debug_log, "counted=%"UVdf" deltanext=%"UVdf
4696 " SSize_t_MAX=%"UVdf" minnext=%"UVdf
4697 " maxcount=%"UVdf" mincount=%"UVdf"\n",
4698 (UV)counted, (UV)deltanext, (UV)SSize_t_MAX, (UV)minnext, (UV)maxcount,
4700 if (deltanext != SSize_t_MAX)
4701 PerlIO_printf(Perl_debug_log, "LHS=%"UVdf" RHS=%"UVdf"\n",
4702 (UV)(-counted * deltanext + (minnext + deltanext) * maxcount
4703 - minnext * mincount), (UV)(SSize_t_MAX - data->pos_delta));
4705 if (deltanext == SSize_t_MAX
4706 || -counted * deltanext + (minnext + deltanext) * maxcount - minnext * mincount >= SSize_t_MAX - data->pos_delta)
4707 data->pos_delta = SSize_t_MAX;
4709 data->pos_delta += - counted * deltanext +
4710 (minnext + deltanext) * maxcount - minnext * mincount;
4711 if (mincount != maxcount) {
4712 /* Cannot extend fixed substrings found inside
4714 scan_commit(pRExC_state, data, minlenp, is_inf);
4715 if (mincount && last_str) {
4716 SV * const sv = data->last_found;
4717 MAGIC * const mg = SvUTF8(sv) && SvMAGICAL(sv) ?
4718 mg_find(sv, PERL_MAGIC_utf8) : NULL;
4722 sv_setsv(sv, last_str);
4723 data->last_end = data->pos_min;
4724 data->last_start_min = data->pos_min - last_chrs;
4725 data->last_start_max = is_inf
4727 : data->pos_min + data->pos_delta - last_chrs;
4729 data->longest = &(data->longest_float);
4731 SvREFCNT_dec(last_str);
4733 if (data && (fl & SF_HAS_EVAL))
4734 data->flags |= SF_HAS_EVAL;
4735 optimize_curly_tail:
4736 if (OP(oscan) != CURLYX) {
4737 while (PL_regkind[OP(next = regnext(oscan))] == NOTHING
4739 NEXT_OFF(oscan) += NEXT_OFF(next);
4745 Perl_croak(aTHX_ "panic: unexpected varying REx opcode %d",
4750 if (flags & SCF_DO_SUBSTR) {
4751 /* Cannot expect anything... */
4752 scan_commit(pRExC_state, data, minlenp, is_inf);
4753 data->longest = &(data->longest_float);
4755 is_inf = is_inf_internal = 1;
4756 if (flags & SCF_DO_STCLASS_OR) {
4757 if (OP(scan) == CLUMP) {
4758 /* Actually is any start char, but very few code points
4759 * aren't start characters */
4760 ssc_match_all_cp(data->start_class);
4763 ssc_anything(data->start_class);
4766 flags &= ~SCF_DO_STCLASS;
4770 else if (OP(scan) == LNBREAK) {
4771 if (flags & SCF_DO_STCLASS) {
4772 if (flags & SCF_DO_STCLASS_AND) {
4773 ssc_intersection(data->start_class,
4774 PL_XPosix_ptrs[_CC_VERTSPACE], FALSE);
4775 ssc_clear_locale(data->start_class);
4776 ANYOF_FLAGS(data->start_class) &= ~ANYOF_EMPTY_STRING;
4778 else if (flags & SCF_DO_STCLASS_OR) {
4779 ssc_union(data->start_class,
4780 PL_XPosix_ptrs[_CC_VERTSPACE],
4782 ssc_and(pRExC_state, data->start_class, (regnode_charclass *) and_withp);
4784 /* See commit msg for
4785 * 749e076fceedeb708a624933726e7989f2302f6a */
4786 ANYOF_FLAGS(data->start_class) &= ~ANYOF_EMPTY_STRING;
4788 flags &= ~SCF_DO_STCLASS;
4791 delta++; /* Because of the 2 char string cr-lf */
4792 if (flags & SCF_DO_SUBSTR) {
4793 /* Cannot expect anything... */
4794 scan_commit(pRExC_state, data, minlenp, is_inf);
4796 data->pos_delta += 1;
4797 data->longest = &(data->longest_float);
4800 else if (REGNODE_SIMPLE(OP(scan))) {
4802 if (flags & SCF_DO_SUBSTR) {
4803 scan_commit(pRExC_state, data, minlenp, is_inf);
4807 if (flags & SCF_DO_STCLASS) {
4809 SV* my_invlist = sv_2mortal(_new_invlist(0));
4812 /* See commit msg 749e076fceedeb708a624933726e7989f2302f6a */
4813 ANYOF_FLAGS(data->start_class) &= ~ANYOF_EMPTY_STRING;
4815 /* Some of the logic below assumes that switching
4816 locale on will only add false positives. */
4821 Perl_croak(aTHX_ "panic: unexpected simple REx opcode %d",
4826 if (flags & SCF_DO_STCLASS_OR) /* Allow everything */
4827 ssc_match_all_cp(data->start_class);
4832 SV* REG_ANY_invlist = _new_invlist(2);
4833 REG_ANY_invlist = add_cp_to_invlist(REG_ANY_invlist,
4835 if (flags & SCF_DO_STCLASS_OR) {
4836 ssc_union(data->start_class,
4838 TRUE /* TRUE => invert, hence all but \n
4842 else if (flags & SCF_DO_STCLASS_AND) {
4843 ssc_intersection(data->start_class,
4845 TRUE /* TRUE => invert */
4847 ssc_clear_locale(data->start_class);
4849 SvREFCNT_dec_NN(REG_ANY_invlist);
4854 if (flags & SCF_DO_STCLASS_AND)
4855 ssc_and(pRExC_state, data->start_class,
4856 (regnode_charclass *) scan);
4858 ssc_or(pRExC_state, data->start_class,
4859 (regnode_charclass *) scan);
4867 namedclass = classnum_to_namedclass(FLAGS(scan)) + invert;
4868 if (flags & SCF_DO_STCLASS_AND) {
4869 bool was_there = cBOOL(
4870 ANYOF_POSIXL_TEST(data->start_class,
4872 ANYOF_POSIXL_ZERO(data->start_class);
4873 if (was_there) { /* Do an AND */
4874 ANYOF_POSIXL_SET(data->start_class, namedclass);
4876 /* No individual code points can now match */
4877 data->start_class->invlist
4878 = sv_2mortal(_new_invlist(0));
4881 int complement = namedclass + ((invert) ? -1 : 1);
4883 assert(flags & SCF_DO_STCLASS_OR);
4885 /* If the complement of this class was already there,
4886 * the result is that they match all code points,
4887 * (\d + \D == everything). Remove the classes from
4888 * future consideration. Locale is not relevant in
4890 if (ANYOF_POSIXL_TEST(data->start_class, complement)) {
4891 ssc_match_all_cp(data->start_class);
4892 ANYOF_POSIXL_CLEAR(data->start_class, namedclass);
4893 ANYOF_POSIXL_CLEAR(data->start_class, complement);
4895 else { /* The usual case; just add this class to the
4897 ANYOF_POSIXL_SET(data->start_class, namedclass);
4902 case NPOSIXA: /* For these, we always know the exact set of
4907 if (FLAGS(scan) == _CC_ASCII) {
4908 my_invlist = PL_XPosix_ptrs[_CC_ASCII];
4911 _invlist_intersection(PL_XPosix_ptrs[FLAGS(scan)],
4912 PL_XPosix_ptrs[_CC_ASCII],
4923 my_invlist = invlist_clone(PL_XPosix_ptrs[FLAGS(scan)]);
4925 /* NPOSIXD matches all upper Latin1 code points unless the
4926 * target string being matched is UTF-8, which is
4927 * unknowable until match time. Since we are going to
4928 * invert, we want to get rid of all of them so that the
4929 * inversion will match all */
4930 if (OP(scan) == NPOSIXD) {
4931 _invlist_subtract(my_invlist, PL_UpperLatin1,
4937 if (flags & SCF_DO_STCLASS_AND) {
4938 ssc_intersection(data->start_class, my_invlist, invert);
4939 ssc_clear_locale(data->start_class);
4942 assert(flags & SCF_DO_STCLASS_OR);
4943 ssc_union(data->start_class, my_invlist, invert);
4946 if (flags & SCF_DO_STCLASS_OR)
4947 ssc_and(pRExC_state, data->start_class, (regnode_charclass *) and_withp);
4948 flags &= ~SCF_DO_STCLASS;
4951 else if (PL_regkind[OP(scan)] == EOL && flags & SCF_DO_SUBSTR) {
4952 data->flags |= (OP(scan) == MEOL
4955 scan_commit(pRExC_state, data, minlenp, is_inf);
4958 else if ( PL_regkind[OP(scan)] == BRANCHJ
4959 /* Lookbehind, or need to calculate parens/evals/stclass: */
4960 && (scan->flags || data || (flags & SCF_DO_STCLASS))
4961 && (OP(scan) == IFMATCH || OP(scan) == UNLESSM)) {
4962 if ( OP(scan) == UNLESSM &&
4964 OP(NEXTOPER(NEXTOPER(scan))) == NOTHING &&
4965 OP(regnext(NEXTOPER(NEXTOPER(scan)))) == SUCCEED
4968 regnode *upto= regnext(scan);
4970 SV * const mysv_val=sv_newmortal();
4971 DEBUG_STUDYDATA("OPFAIL",data,depth);
4973 /*DEBUG_PARSE_MSG("opfail");*/
4974 regprop(RExC_rx, mysv_val, upto, NULL);
4975 PerlIO_printf(Perl_debug_log,
4976 "~ replace with OPFAIL pointed at %s (%"IVdf") offset %"IVdf"\n",
4977 SvPV_nolen_const(mysv_val),
4978 (IV)REG_NODE_NUM(upto),
4983 NEXT_OFF(scan) = upto - scan;
4984 for (opt= scan + 1; opt < upto ; opt++)
4985 OP(opt) = OPTIMIZED;
4989 if ( !PERL_ENABLE_POSITIVE_ASSERTION_STUDY
4990 || OP(scan) == UNLESSM )
4992 /* Negative Lookahead/lookbehind
4993 In this case we can't do fixed string optimisation.
4996 SSize_t deltanext, minnext, fake = 0;
5001 data_fake.flags = 0;
5003 data_fake.whilem_c = data->whilem_c;
5004 data_fake.last_closep = data->last_closep;
5007 data_fake.last_closep = &fake;
5008 data_fake.pos_delta = delta;
5009 if ( flags & SCF_DO_STCLASS && !scan->flags
5010 && OP(scan) == IFMATCH ) { /* Lookahead */
5011 ssc_init(pRExC_state, &intrnl);
5012 data_fake.start_class = &intrnl;
5013 f |= SCF_DO_STCLASS_AND;
5015 if (flags & SCF_WHILEM_VISITED_POS)
5016 f |= SCF_WHILEM_VISITED_POS;
5017 next = regnext(scan);
5018 nscan = NEXTOPER(NEXTOPER(scan));
5019 minnext = study_chunk(pRExC_state, &nscan, minlenp, &deltanext,
5020 last, &data_fake, stopparen,
5021 recursed_depth, NULL, f, depth+1);
5024 FAIL("Variable length lookbehind not implemented");
5026 else if (minnext > (I32)U8_MAX) {
5027 FAIL2("Lookbehind longer than %"UVuf" not implemented",
5030 scan->flags = (U8)minnext;
5033 if (data_fake.flags & (SF_HAS_PAR|SF_IN_PAR))
5035 if (data_fake.flags & SF_HAS_EVAL)
5036 data->flags |= SF_HAS_EVAL;
5037 data->whilem_c = data_fake.whilem_c;
5039 if (f & SCF_DO_STCLASS_AND) {
5040 if (flags & SCF_DO_STCLASS_OR) {
5041 /* OR before, AND after: ideally we would recurse with
5042 * data_fake to get the AND applied by study of the
5043 * remainder of the pattern, and then derecurse;
5044 * *** HACK *** for now just treat as "no information".
5045 * See [perl #56690].
5047 ssc_init(pRExC_state, data->start_class);
5049 /* AND before and after: combine and continue. These
5050 * assertions are zero-length, so can match an EMPTY
5052 ssc_and(pRExC_state, data->start_class, (regnode_charclass *) &intrnl);
5053 ANYOF_FLAGS(data->start_class) |= ANYOF_EMPTY_STRING;
5057 #if PERL_ENABLE_POSITIVE_ASSERTION_STUDY
5059 /* Positive Lookahead/lookbehind
5060 In this case we can do fixed string optimisation,
5061 but we must be careful about it. Note in the case of
5062 lookbehind the positions will be offset by the minimum
5063 length of the pattern, something we won't know about
5064 until after the recurse.
5066 SSize_t deltanext, fake = 0;
5070 /* We use SAVEFREEPV so that when the full compile
5071 is finished perl will clean up the allocated
5072 minlens when it's all done. This way we don't
5073 have to worry about freeing them when we know
5074 they wont be used, which would be a pain.
5077 Newx( minnextp, 1, SSize_t );
5078 SAVEFREEPV(minnextp);
5081 StructCopy(data, &data_fake, scan_data_t);
5082 if ((flags & SCF_DO_SUBSTR) && data->last_found) {
5085 scan_commit(pRExC_state, &data_fake, minlenp, is_inf);
5086 data_fake.last_found=newSVsv(data->last_found);
5090 data_fake.last_closep = &fake;
5091 data_fake.flags = 0;
5092 data_fake.pos_delta = delta;
5094 data_fake.flags |= SF_IS_INF;
5095 if ( flags & SCF_DO_STCLASS && !scan->flags
5096 && OP(scan) == IFMATCH ) { /* Lookahead */
5097 ssc_init(pRExC_state, &intrnl);
5098 data_fake.start_class = &intrnl;
5099 f |= SCF_DO_STCLASS_AND;
5101 if (flags & SCF_WHILEM_VISITED_POS)
5102 f |= SCF_WHILEM_VISITED_POS;
5103 next = regnext(scan);
5104 nscan = NEXTOPER(NEXTOPER(scan));
5106 *minnextp = study_chunk(pRExC_state, &nscan, minnextp,
5107 &deltanext, last, &data_fake,
5108 stopparen, recursed_depth, NULL,
5112 FAIL("Variable length lookbehind not implemented");
5114 else if (*minnextp > (I32)U8_MAX) {
5115 FAIL2("Lookbehind longer than %"UVuf" not implemented",
5118 scan->flags = (U8)*minnextp;
5123 if (f & SCF_DO_STCLASS_AND) {
5124 ssc_and(pRExC_state, data->start_class, (regnode_charclass *) &intrnl);
5125 ANYOF_FLAGS(data->start_class) |= ANYOF_EMPTY_STRING;
5128 if (data_fake.flags & (SF_HAS_PAR|SF_IN_PAR))
5130 if (data_fake.flags & SF_HAS_EVAL)
5131 data->flags |= SF_HAS_EVAL;
5132 data->whilem_c = data_fake.whilem_c;
5133 if ((flags & SCF_DO_SUBSTR) && data_fake.last_found) {
5134 if (RExC_rx->minlen<*minnextp)
5135 RExC_rx->minlen=*minnextp;
5136 scan_commit(pRExC_state, &data_fake, minnextp, is_inf);
5137 SvREFCNT_dec_NN(data_fake.last_found);
5139 if ( data_fake.minlen_fixed != minlenp )
5141 data->offset_fixed= data_fake.offset_fixed;
5142 data->minlen_fixed= data_fake.minlen_fixed;
5143 data->lookbehind_fixed+= scan->flags;
5145 if ( data_fake.minlen_float != minlenp )
5147 data->minlen_float= data_fake.minlen_float;
5148 data->offset_float_min=data_fake.offset_float_min;
5149 data->offset_float_max=data_fake.offset_float_max;
5150 data->lookbehind_float+= scan->flags;
5157 else if (OP(scan) == OPEN) {
5158 if (stopparen != (I32)ARG(scan))
5161 else if (OP(scan) == CLOSE) {
5162 if (stopparen == (I32)ARG(scan)) {
5165 if ((I32)ARG(scan) == is_par) {
5166 next = regnext(scan);
5168 if ( next && (OP(next) != WHILEM) && next < last)
5169 is_par = 0; /* Disable optimization */
5172 *(data->last_closep) = ARG(scan);
5174 else if (OP(scan) == EVAL) {
5176 data->flags |= SF_HAS_EVAL;
5178 else if ( PL_regkind[OP(scan)] == ENDLIKE ) {
5179 if (flags & SCF_DO_SUBSTR) {
5180 scan_commit(pRExC_state, data, minlenp, is_inf);
5181 flags &= ~SCF_DO_SUBSTR;
5183 if (data && OP(scan)==ACCEPT) {
5184 data->flags |= SCF_SEEN_ACCEPT;
5189 else if (OP(scan) == LOGICAL && scan->flags == 2) /* Embedded follows */
5191 if (flags & SCF_DO_SUBSTR) {
5192 scan_commit(pRExC_state, data, minlenp, is_inf);
5193 data->longest = &(data->longest_float);
5195 is_inf = is_inf_internal = 1;
5196 if (flags & SCF_DO_STCLASS_OR) /* Allow everything */
5197 ssc_anything(data->start_class);
5198 flags &= ~SCF_DO_STCLASS;
5200 else if (OP(scan) == GPOS) {
5201 if (!(RExC_rx->intflags & PREGf_GPOS_FLOAT) &&
5202 !(delta || is_inf || (data && data->pos_delta)))
5204 if (!(RExC_rx->intflags & PREGf_ANCH) && (flags & SCF_DO_SUBSTR))
5205 RExC_rx->intflags |= PREGf_ANCH_GPOS;
5206 if (RExC_rx->gofs < (STRLEN)min)
5207 RExC_rx->gofs = min;
5209 RExC_rx->intflags |= PREGf_GPOS_FLOAT;
5213 #ifdef TRIE_STUDY_OPT
5214 #ifdef FULL_TRIE_STUDY
5215 else if (PL_regkind[OP(scan)] == TRIE) {
5216 /* NOTE - There is similar code to this block above for handling
5217 BRANCH nodes on the initial study. If you change stuff here
5219 regnode *trie_node= scan;
5220 regnode *tail= regnext(scan);
5221 reg_trie_data *trie = (reg_trie_data*)RExC_rxi->data->data[ ARG(scan) ];
5222 SSize_t max1 = 0, min1 = SSize_t_MAX;
5225 if (flags & SCF_DO_SUBSTR) { /* XXXX Add !SUSPEND? */
5226 /* Cannot merge strings after this. */
5227 scan_commit(pRExC_state, data, minlenp, is_inf);
5229 if (flags & SCF_DO_STCLASS)
5230 ssc_init_zero(pRExC_state, &accum);
5236 const regnode *nextbranch= NULL;
5239 for ( word=1 ; word <= trie->wordcount ; word++)
5241 SSize_t deltanext=0, minnext=0, f = 0, fake;
5242 regnode_ssc this_class;
5244 data_fake.flags = 0;
5246 data_fake.whilem_c = data->whilem_c;
5247 data_fake.last_closep = data->last_closep;
5250 data_fake.last_closep = &fake;
5251 data_fake.pos_delta = delta;
5252 if (flags & SCF_DO_STCLASS) {
5253 ssc_init(pRExC_state, &this_class);
5254 data_fake.start_class = &this_class;
5255 f = SCF_DO_STCLASS_AND;
5257 if (flags & SCF_WHILEM_VISITED_POS)
5258 f |= SCF_WHILEM_VISITED_POS;
5260 if (trie->jump[word]) {
5262 nextbranch = trie_node + trie->jump[0];
5263 scan= trie_node + trie->jump[word];
5264 /* We go from the jump point to the branch that follows
5265 it. Note this means we need the vestigal unused
5266 branches even though they arent otherwise used. */
5267 minnext = study_chunk(pRExC_state, &scan, minlenp,
5268 &deltanext, (regnode *)nextbranch, &data_fake,
5269 stopparen, recursed_depth, NULL, f,depth+1);
5271 if (nextbranch && PL_regkind[OP(nextbranch)]==BRANCH)
5272 nextbranch= regnext((regnode*)nextbranch);
5274 if (min1 > (SSize_t)(minnext + trie->minlen))
5275 min1 = minnext + trie->minlen;
5276 if (deltanext == SSize_t_MAX) {
5277 is_inf = is_inf_internal = 1;
5279 } else if (max1 < (SSize_t)(minnext + deltanext + trie->maxlen))
5280 max1 = minnext + deltanext + trie->maxlen;
5282 if (data_fake.flags & (SF_HAS_PAR|SF_IN_PAR))
5284 if (data_fake.flags & SCF_SEEN_ACCEPT) {
5285 if ( stopmin > min + min1)
5286 stopmin = min + min1;
5287 flags &= ~SCF_DO_SUBSTR;
5289 data->flags |= SCF_SEEN_ACCEPT;
5292 if (data_fake.flags & SF_HAS_EVAL)
5293 data->flags |= SF_HAS_EVAL;
5294 data->whilem_c = data_fake.whilem_c;
5296 if (flags & SCF_DO_STCLASS)
5297 ssc_or(pRExC_state, &accum, (regnode_charclass *) &this_class);
5300 if (flags & SCF_DO_SUBSTR) {
5301 data->pos_min += min1;
5302 data->pos_delta += max1 - min1;
5303 if (max1 != min1 || is_inf)
5304 data->longest = &(data->longest_float);
5307 delta += max1 - min1;
5308 if (flags & SCF_DO_STCLASS_OR) {
5309 ssc_or(pRExC_state, data->start_class, (regnode_charclass *) &accum);
5311 ssc_and(pRExC_state, data->start_class, (regnode_charclass *) and_withp);
5312 flags &= ~SCF_DO_STCLASS;
5315 else if (flags & SCF_DO_STCLASS_AND) {
5317 ssc_and(pRExC_state, data->start_class, (regnode_charclass *) &accum);
5318 flags &= ~SCF_DO_STCLASS;
5321 /* Switch to OR mode: cache the old value of
5322 * data->start_class */
5324 StructCopy(data->start_class, and_withp, regnode_ssc);
5325 flags &= ~SCF_DO_STCLASS_AND;
5326 StructCopy(&accum, data->start_class, regnode_ssc);
5327 flags |= SCF_DO_STCLASS_OR;
5334 else if (PL_regkind[OP(scan)] == TRIE) {
5335 reg_trie_data *trie = (reg_trie_data*)RExC_rxi->data->data[ ARG(scan) ];
5338 min += trie->minlen;
5339 delta += (trie->maxlen - trie->minlen);
5340 flags &= ~SCF_DO_STCLASS; /* xxx */
5341 if (flags & SCF_DO_SUBSTR) {
5342 /* Cannot expect anything... */
5343 scan_commit(pRExC_state, data, minlenp, is_inf);
5344 data->pos_min += trie->minlen;
5345 data->pos_delta += (trie->maxlen - trie->minlen);
5346 if (trie->maxlen != trie->minlen)
5347 data->longest = &(data->longest_float);
5349 if (trie->jump) /* no more substrings -- for now /grr*/
5350 flags &= ~SCF_DO_SUBSTR;
5352 #endif /* old or new */
5353 #endif /* TRIE_STUDY_OPT */
5355 /* Else: zero-length, ignore. */
5356 scan = regnext(scan);
5358 /* If we are exiting a recursion we can unset its recursed bit
5359 * and allow ourselves to enter it again - no danger of an
5360 * infinite loop there.
5361 if (stopparen > -1 && recursed) {
5362 DEBUG_STUDYDATA("unset:", data,depth);
5363 PAREN_UNSET( recursed, stopparen);
5367 DEBUG_STUDYDATA("frame-end:",data,depth);
5368 DEBUG_PEEP("fend", scan, depth);
5369 /* restore previous context */
5372 stopparen = frame->stop;
5373 recursed_depth = frame->prev_recursed_depth;
5376 frame = frame->prev;
5377 goto fake_study_recurse;
5382 DEBUG_STUDYDATA("pre-fin:",data,depth);
5385 *deltap = is_inf_internal ? SSize_t_MAX : delta;
5387 if (flags & SCF_DO_SUBSTR && is_inf)
5388 data->pos_delta = SSize_t_MAX - data->pos_min;
5389 if (is_par > (I32)U8_MAX)
5391 if (is_par && pars==1 && data) {
5392 data->flags |= SF_IN_PAR;
5393 data->flags &= ~SF_HAS_PAR;
5395 else if (pars && data) {
5396 data->flags |= SF_HAS_PAR;
5397 data->flags &= ~SF_IN_PAR;
5399 if (flags & SCF_DO_STCLASS_OR)
5400 ssc_and(pRExC_state, data->start_class, (regnode_charclass *) and_withp);
5401 if (flags & SCF_TRIE_RESTUDY)
5402 data->flags |= SCF_TRIE_RESTUDY;
5404 DEBUG_STUDYDATA("post-fin:",data,depth);
5407 SSize_t final_minlen= min < stopmin ? min : stopmin;
5409 if (!(RExC_seen & REG_UNBOUNDED_QUANTIFIER_SEEN) && (RExC_maxlen < final_minlen + delta)) {
5410 RExC_maxlen = final_minlen + delta;
5412 return final_minlen;
5418 S_add_data(RExC_state_t* const pRExC_state, const char* const s, const U32 n)
5420 U32 count = RExC_rxi->data ? RExC_rxi->data->count : 0;
5422 PERL_ARGS_ASSERT_ADD_DATA;
5424 Renewc(RExC_rxi->data,
5425 sizeof(*RExC_rxi->data) + sizeof(void*) * (count + n - 1),
5426 char, struct reg_data);
5428 Renew(RExC_rxi->data->what, count + n, U8);
5430 Newx(RExC_rxi->data->what, n, U8);
5431 RExC_rxi->data->count = count + n;
5432 Copy(s, RExC_rxi->data->what + count, n, U8);
5436 /*XXX: todo make this not included in a non debugging perl */
5437 #ifndef PERL_IN_XSUB_RE
5439 Perl_reginitcolors(pTHX)
5442 const char * const s = PerlEnv_getenv("PERL_RE_COLORS");
5444 char *t = savepv(s);
5448 t = strchr(t, '\t');
5454 PL_colors[i] = t = (char *)"";
5459 PL_colors[i++] = (char *)"";
5466 #ifdef TRIE_STUDY_OPT
5467 #define CHECK_RESTUDY_GOTO_butfirst(dOsomething) \
5470 (data.flags & SCF_TRIE_RESTUDY) \
5478 #define CHECK_RESTUDY_GOTO_butfirst
5482 * pregcomp - compile a regular expression into internal code
5484 * Decides which engine's compiler to call based on the hint currently in
5488 #ifndef PERL_IN_XSUB_RE
5490 /* return the currently in-scope regex engine (or the default if none) */
5492 regexp_engine const *
5493 Perl_current_re_engine(pTHX)
5497 if (IN_PERL_COMPILETIME) {
5498 HV * const table = GvHV(PL_hintgv);
5501 if (!table || !(PL_hints & HINT_LOCALIZE_HH))
5502 return &PL_core_reg_engine;
5503 ptr = hv_fetchs(table, "regcomp", FALSE);
5504 if ( !(ptr && SvIOK(*ptr) && SvIV(*ptr)))
5505 return &PL_core_reg_engine;
5506 return INT2PTR(regexp_engine*,SvIV(*ptr));
5510 if (!PL_curcop->cop_hints_hash)
5511 return &PL_core_reg_engine;
5512 ptr = cop_hints_fetch_pvs(PL_curcop, "regcomp", 0);
5513 if ( !(ptr && SvIOK(ptr) && SvIV(ptr)))
5514 return &PL_core_reg_engine;
5515 return INT2PTR(regexp_engine*,SvIV(ptr));
5521 Perl_pregcomp(pTHX_ SV * const pattern, const U32 flags)
5524 regexp_engine const *eng = current_re_engine();
5525 GET_RE_DEBUG_FLAGS_DECL;
5527 PERL_ARGS_ASSERT_PREGCOMP;
5529 /* Dispatch a request to compile a regexp to correct regexp engine. */
5531 PerlIO_printf(Perl_debug_log, "Using engine %"UVxf"\n",
5534 return CALLREGCOMP_ENG(eng, pattern, flags);
5538 /* public(ish) entry point for the perl core's own regex compiling code.
5539 * It's actually a wrapper for Perl_re_op_compile that only takes an SV
5540 * pattern rather than a list of OPs, and uses the internal engine rather
5541 * than the current one */
5544 Perl_re_compile(pTHX_ SV * const pattern, U32 rx_flags)
5546 SV *pat = pattern; /* defeat constness! */
5547 PERL_ARGS_ASSERT_RE_COMPILE;
5548 return Perl_re_op_compile(aTHX_ &pat, 1, NULL,
5549 #ifdef PERL_IN_XSUB_RE
5552 &PL_core_reg_engine,
5554 NULL, NULL, rx_flags, 0);
5558 /* upgrade pattern pat_p of length plen_p to UTF8, and if there are code
5559 * blocks, recalculate the indices. Update pat_p and plen_p in-place to
5560 * point to the realloced string and length.
5562 * This is essentially a copy of Perl_bytes_to_utf8() with the code index
5566 S_pat_upgrade_to_utf8(pTHX_ RExC_state_t * const pRExC_state,
5567 char **pat_p, STRLEN *plen_p, int num_code_blocks)
5569 U8 *const src = (U8*)*pat_p;
5572 STRLEN s = 0, d = 0;
5574 GET_RE_DEBUG_FLAGS_DECL;
5576 DEBUG_PARSE_r(PerlIO_printf(Perl_debug_log,
5577 "UTF8 mismatch! Converting to utf8 for resizing and compile\n"));
5579 Newx(dst, *plen_p * 2 + 1, U8);
5581 while (s < *plen_p) {
5582 if (NATIVE_BYTE_IS_INVARIANT(src[s]))
5585 dst[d++] = UTF8_EIGHT_BIT_HI(src[s]);
5586 dst[d] = UTF8_EIGHT_BIT_LO(src[s]);
5588 if (n < num_code_blocks) {
5589 if (!do_end && pRExC_state->code_blocks[n].start == s) {
5590 pRExC_state->code_blocks[n].start = d;
5591 assert(dst[d] == '(');
5594 else if (do_end && pRExC_state->code_blocks[n].end == s) {
5595 pRExC_state->code_blocks[n].end = d;
5596 assert(dst[d] == ')');
5606 *pat_p = (char*) dst;
5608 RExC_orig_utf8 = RExC_utf8 = 1;
5613 /* S_concat_pat(): concatenate a list of args to the pattern string pat,
5614 * while recording any code block indices, and handling overloading,
5615 * nested qr// objects etc. If pat is null, it will allocate a new
5616 * string, or just return the first arg, if there's only one.
5618 * Returns the malloced/updated pat.
5619 * patternp and pat_count is the array of SVs to be concatted;
5620 * oplist is the optional list of ops that generated the SVs;
5621 * recompile_p is a pointer to a boolean that will be set if
5622 * the regex will need to be recompiled.
5623 * delim, if non-null is an SV that will be inserted between each element
5627 S_concat_pat(pTHX_ RExC_state_t * const pRExC_state,
5628 SV *pat, SV ** const patternp, int pat_count,
5629 OP *oplist, bool *recompile_p, SV *delim)
5633 bool use_delim = FALSE;
5634 bool alloced = FALSE;
5636 /* if we know we have at least two args, create an empty string,
5637 * then concatenate args to that. For no args, return an empty string */
5638 if (!pat && pat_count != 1) {
5639 pat = newSVpvn("", 0);
5644 for (svp = patternp; svp < patternp + pat_count; svp++) {
5647 STRLEN orig_patlen = 0;
5649 SV *msv = use_delim ? delim : *svp;
5650 if (!msv) msv = &PL_sv_undef;
5652 /* if we've got a delimiter, we go round the loop twice for each
5653 * svp slot (except the last), using the delimiter the second
5662 if (SvTYPE(msv) == SVt_PVAV) {
5663 /* we've encountered an interpolated array within
5664 * the pattern, e.g. /...@a..../. Expand the list of elements,
5665 * then recursively append elements.
5666 * The code in this block is based on S_pushav() */
5668 AV *const av = (AV*)msv;
5669 const SSize_t maxarg = AvFILL(av) + 1;
5673 assert(oplist->op_type == OP_PADAV
5674 || oplist->op_type == OP_RV2AV);
5675 oplist = oplist->op_sibling;;
5678 if (SvRMAGICAL(av)) {
5681 Newx(array, maxarg, SV*);
5683 for (i=0; i < maxarg; i++) {
5684 SV ** const svp = av_fetch(av, i, FALSE);
5685 array[i] = svp ? *svp : &PL_sv_undef;
5689 array = AvARRAY(av);
5691 pat = S_concat_pat(aTHX_ pRExC_state, pat,
5692 array, maxarg, NULL, recompile_p,
5694 GvSV((gv_fetchpvs("\"", GV_ADDMULTI, SVt_PV))));
5700 /* we make the assumption here that each op in the list of
5701 * op_siblings maps to one SV pushed onto the stack,
5702 * except for code blocks, with have both an OP_NULL and
5704 * This allows us to match up the list of SVs against the
5705 * list of OPs to find the next code block.
5707 * Note that PUSHMARK PADSV PADSV ..
5709 * PADRANGE PADSV PADSV ..
5710 * so the alignment still works. */
5713 if (oplist->op_type == OP_NULL
5714 && (oplist->op_flags & OPf_SPECIAL))
5716 assert(n < pRExC_state->num_code_blocks);
5717 pRExC_state->code_blocks[n].start = pat ? SvCUR(pat) : 0;
5718 pRExC_state->code_blocks[n].block = oplist;
5719 pRExC_state->code_blocks[n].src_regex = NULL;
5722 oplist = oplist->op_sibling; /* skip CONST */
5725 oplist = oplist->op_sibling;;
5728 /* apply magic and QR overloading to arg */
5731 if (SvROK(msv) && SvAMAGIC(msv)) {
5732 SV *sv = AMG_CALLunary(msv, regexp_amg);
5736 if (SvTYPE(sv) != SVt_REGEXP)
5737 Perl_croak(aTHX_ "Overloaded qr did not return a REGEXP");
5742 /* try concatenation overload ... */
5743 if (pat && (SvAMAGIC(pat) || SvAMAGIC(msv)) &&
5744 (sv = amagic_call(pat, msv, concat_amg, AMGf_assign)))
5747 /* overloading involved: all bets are off over literal
5748 * code. Pretend we haven't seen it */
5749 pRExC_state->num_code_blocks -= n;
5753 /* ... or failing that, try "" overload */
5754 while (SvAMAGIC(msv)
5755 && (sv = AMG_CALLunary(msv, string_amg))
5759 && SvRV(msv) == SvRV(sv))
5764 if (SvROK(msv) && SvTYPE(SvRV(msv)) == SVt_REGEXP)
5768 /* this is a partially unrolled
5769 * sv_catsv_nomg(pat, msv);
5770 * that allows us to adjust code block indices if
5773 char *dst = SvPV_force_nomg(pat, dlen);
5775 if (SvUTF8(msv) && !SvUTF8(pat)) {
5776 S_pat_upgrade_to_utf8(aTHX_ pRExC_state, &dst, &dlen, n);
5777 sv_setpvn(pat, dst, dlen);
5780 sv_catsv_nomg(pat, msv);
5787 pRExC_state->code_blocks[n-1].end = SvCUR(pat)-1;
5790 /* extract any code blocks within any embedded qr//'s */
5791 if (rx && SvTYPE(rx) == SVt_REGEXP
5792 && RX_ENGINE((REGEXP*)rx)->op_comp)
5795 RXi_GET_DECL(ReANY((REGEXP *)rx), ri);
5796 if (ri->num_code_blocks) {
5798 /* the presence of an embedded qr// with code means
5799 * we should always recompile: the text of the
5800 * qr// may not have changed, but it may be a
5801 * different closure than last time */
5803 Renew(pRExC_state->code_blocks,
5804 pRExC_state->num_code_blocks + ri->num_code_blocks,
5805 struct reg_code_block);
5806 pRExC_state->num_code_blocks += ri->num_code_blocks;
5808 for (i=0; i < ri->num_code_blocks; i++) {
5809 struct reg_code_block *src, *dst;
5810 STRLEN offset = orig_patlen
5811 + ReANY((REGEXP *)rx)->pre_prefix;
5812 assert(n < pRExC_state->num_code_blocks);
5813 src = &ri->code_blocks[i];
5814 dst = &pRExC_state->code_blocks[n];
5815 dst->start = src->start + offset;
5816 dst->end = src->end + offset;
5817 dst->block = src->block;
5818 dst->src_regex = (REGEXP*) SvREFCNT_inc( (SV*)
5827 /* avoid calling magic multiple times on a single element e.g. =~ $qr */
5836 /* see if there are any run-time code blocks in the pattern.
5837 * False positives are allowed */
5840 S_has_runtime_code(pTHX_ RExC_state_t * const pRExC_state,
5841 char *pat, STRLEN plen)
5846 for (s = 0; s < plen; s++) {
5847 if (n < pRExC_state->num_code_blocks
5848 && s == pRExC_state->code_blocks[n].start)
5850 s = pRExC_state->code_blocks[n].end;
5854 /* TODO ideally should handle [..], (#..), /#.../x to reduce false
5856 if (pat[s] == '(' && s+2 <= plen && pat[s+1] == '?' &&
5858 || (s + 2 <= plen && pat[s+2] == '?' && pat[s+3] == '{'))
5865 /* Handle run-time code blocks. We will already have compiled any direct
5866 * or indirect literal code blocks. Now, take the pattern 'pat' and make a
5867 * copy of it, but with any literal code blocks blanked out and
5868 * appropriate chars escaped; then feed it into
5870 * eval "qr'modified_pattern'"
5874 * a\bc(?{"this was literal"})def'ghi\\jkl(?{"this is runtime"})mno
5878 * qr'a\\bc_______________________def\'ghi\\\\jkl(?{"this is runtime"})mno'
5880 * After eval_sv()-ing that, grab any new code blocks from the returned qr
5881 * and merge them with any code blocks of the original regexp.
5883 * If the pat is non-UTF8, while the evalled qr is UTF8, don't merge;
5884 * instead, just save the qr and return FALSE; this tells our caller that
5885 * the original pattern needs upgrading to utf8.
5889 S_compile_runtime_code(pTHX_ RExC_state_t * const pRExC_state,
5890 char *pat, STRLEN plen)
5894 GET_RE_DEBUG_FLAGS_DECL;
5896 if (pRExC_state->runtime_code_qr) {
5897 /* this is the second time we've been called; this should
5898 * only happen if the main pattern got upgraded to utf8
5899 * during compilation; re-use the qr we compiled first time
5900 * round (which should be utf8 too)
5902 qr = pRExC_state->runtime_code_qr;
5903 pRExC_state->runtime_code_qr = NULL;
5904 assert(RExC_utf8 && SvUTF8(qr));
5910 int newlen = plen + 6; /* allow for "qr''x\0" extra chars */
5914 /* determine how many extra chars we need for ' and \ escaping */
5915 for (s = 0; s < plen; s++) {
5916 if (pat[s] == '\'' || pat[s] == '\\')
5920 Newx(newpat, newlen, char);
5922 *p++ = 'q'; *p++ = 'r'; *p++ = '\'';
5924 for (s = 0; s < plen; s++) {
5925 if (n < pRExC_state->num_code_blocks
5926 && s == pRExC_state->code_blocks[n].start)
5928 /* blank out literal code block */
5929 assert(pat[s] == '(');
5930 while (s <= pRExC_state->code_blocks[n].end) {
5938 if (pat[s] == '\'' || pat[s] == '\\')
5943 if (pRExC_state->pm_flags & RXf_PMf_EXTENDED)
5947 PerlIO_printf(Perl_debug_log,
5948 "%sre-parsing pattern for runtime code:%s %s\n",
5949 PL_colors[4],PL_colors[5],newpat);
5952 sv = newSVpvn_flags(newpat, p-newpat-1, RExC_utf8 ? SVf_UTF8 : 0);
5958 PUSHSTACKi(PERLSI_REQUIRE);
5959 /* G_RE_REPARSING causes the toker to collapse \\ into \ when
5960 * parsing qr''; normally only q'' does this. It also alters
5962 eval_sv(sv, G_SCALAR|G_RE_REPARSING);
5963 SvREFCNT_dec_NN(sv);
5968 SV * const errsv = ERRSV;
5969 if (SvTRUE_NN(errsv))
5971 Safefree(pRExC_state->code_blocks);
5972 /* use croak_sv ? */
5973 Perl_croak_nocontext("%"SVf, SVfARG(errsv));
5976 assert(SvROK(qr_ref));
5978 assert(SvTYPE(qr) == SVt_REGEXP && RX_ENGINE((REGEXP*)qr)->op_comp);
5979 /* the leaving below frees the tmp qr_ref.
5980 * Give qr a life of its own */
5988 if (!RExC_utf8 && SvUTF8(qr)) {
5989 /* first time through; the pattern got upgraded; save the
5990 * qr for the next time through */
5991 assert(!pRExC_state->runtime_code_qr);
5992 pRExC_state->runtime_code_qr = qr;
5997 /* extract any code blocks within the returned qr// */
6000 /* merge the main (r1) and run-time (r2) code blocks into one */
6002 RXi_GET_DECL(ReANY((REGEXP *)qr), r2);
6003 struct reg_code_block *new_block, *dst;
6004 RExC_state_t * const r1 = pRExC_state; /* convenient alias */
6007 if (!r2->num_code_blocks) /* we guessed wrong */
6009 SvREFCNT_dec_NN(qr);
6014 r1->num_code_blocks + r2->num_code_blocks,
6015 struct reg_code_block);
6018 while ( i1 < r1->num_code_blocks
6019 || i2 < r2->num_code_blocks)
6021 struct reg_code_block *src;
6024 if (i1 == r1->num_code_blocks) {
6025 src = &r2->code_blocks[i2++];
6028 else if (i2 == r2->num_code_blocks)
6029 src = &r1->code_blocks[i1++];
6030 else if ( r1->code_blocks[i1].start
6031 < r2->code_blocks[i2].start)
6033 src = &r1->code_blocks[i1++];
6034 assert(src->end < r2->code_blocks[i2].start);
6037 assert( r1->code_blocks[i1].start
6038 > r2->code_blocks[i2].start);
6039 src = &r2->code_blocks[i2++];
6041 assert(src->end < r1->code_blocks[i1].start);
6044 assert(pat[src->start] == '(');
6045 assert(pat[src->end] == ')');
6046 dst->start = src->start;
6047 dst->end = src->end;
6048 dst->block = src->block;
6049 dst->src_regex = is_qr ? (REGEXP*) SvREFCNT_inc( (SV*) qr)
6053 r1->num_code_blocks += r2->num_code_blocks;
6054 Safefree(r1->code_blocks);
6055 r1->code_blocks = new_block;
6058 SvREFCNT_dec_NN(qr);
6064 S_setup_longest(pTHX_ RExC_state_t *pRExC_state, SV* sv_longest,
6065 SV** rx_utf8, SV** rx_substr, SSize_t* rx_end_shift,
6066 SSize_t lookbehind, SSize_t offset, SSize_t *minlen,
6067 STRLEN longest_length, bool eol, bool meol)
6069 /* This is the common code for setting up the floating and fixed length
6070 * string data extracted from Perl_re_op_compile() below. Returns a boolean
6071 * as to whether succeeded or not */
6076 if (! (longest_length
6077 || (eol /* Can't have SEOL and MULTI */
6078 && (! meol || (RExC_flags & RXf_PMf_MULTILINE)))
6080 /* See comments for join_exact for why REG_UNFOLDED_MULTI_SEEN */
6081 || (RExC_seen & REG_UNFOLDED_MULTI_SEEN))
6086 /* copy the information about the longest from the reg_scan_data
6087 over to the program. */
6088 if (SvUTF8(sv_longest)) {
6089 *rx_utf8 = sv_longest;
6092 *rx_substr = sv_longest;
6095 /* end_shift is how many chars that must be matched that
6096 follow this item. We calculate it ahead of time as once the
6097 lookbehind offset is added in we lose the ability to correctly
6099 ml = minlen ? *(minlen) : (SSize_t)longest_length;
6100 *rx_end_shift = ml - offset
6101 - longest_length + (SvTAIL(sv_longest) != 0)
6104 t = (eol/* Can't have SEOL and MULTI */
6105 && (! meol || (RExC_flags & RXf_PMf_MULTILINE)));
6106 fbm_compile(sv_longest, t ? FBMcf_TAIL : 0);
6112 * Perl_re_op_compile - the perl internal RE engine's function to compile a
6113 * regular expression into internal code.
6114 * The pattern may be passed either as:
6115 * a list of SVs (patternp plus pat_count)
6116 * a list of OPs (expr)
6117 * If both are passed, the SV list is used, but the OP list indicates
6118 * which SVs are actually pre-compiled code blocks
6120 * The SVs in the list have magic and qr overloading applied to them (and
6121 * the list may be modified in-place with replacement SVs in the latter
6124 * If the pattern hasn't changed from old_re, then old_re will be
6127 * eng is the current engine. If that engine has an op_comp method, then
6128 * handle directly (i.e. we assume that op_comp was us); otherwise, just
6129 * do the initial concatenation of arguments and pass on to the external
6132 * If is_bare_re is not null, set it to a boolean indicating whether the
6133 * arg list reduced (after overloading) to a single bare regex which has
6134 * been returned (i.e. /$qr/).
6136 * orig_rx_flags contains RXf_* flags. See perlreapi.pod for more details.
6138 * pm_flags contains the PMf_* flags, typically based on those from the
6139 * pm_flags field of the related PMOP. Currently we're only interested in
6140 * PMf_HAS_CV, PMf_IS_QR, PMf_USE_RE_EVAL.
6142 * We can't allocate space until we know how big the compiled form will be,
6143 * but we can't compile it (and thus know how big it is) until we've got a
6144 * place to put the code. So we cheat: we compile it twice, once with code
6145 * generation turned off and size counting turned on, and once "for real".
6146 * This also means that we don't allocate space until we are sure that the
6147 * thing really will compile successfully, and we never have to move the
6148 * code and thus invalidate pointers into it. (Note that it has to be in
6149 * one piece because free() must be able to free it all.) [NB: not true in perl]
6151 * Beware that the optimization-preparation code in here knows about some
6152 * of the structure of the compiled regexp. [I'll say.]
6156 Perl_re_op_compile(pTHX_ SV ** const patternp, int pat_count,
6157 OP *expr, const regexp_engine* eng, REGEXP *old_re,
6158 bool *is_bare_re, U32 orig_rx_flags, U32 pm_flags)
6163 regexp_internal *ri;
6171 SV *code_blocksv = NULL;
6172 SV** new_patternp = patternp;
6174 /* these are all flags - maybe they should be turned
6175 * into a single int with different bit masks */
6176 I32 sawlookahead = 0;
6181 regex_charset initial_charset = get_regex_charset(orig_rx_flags);
6183 bool runtime_code = 0;
6185 RExC_state_t RExC_state;
6186 RExC_state_t * const pRExC_state = &RExC_state;
6187 #ifdef TRIE_STUDY_OPT
6189 RExC_state_t copyRExC_state;
6191 GET_RE_DEBUG_FLAGS_DECL;
6193 PERL_ARGS_ASSERT_RE_OP_COMPILE;
6195 DEBUG_r(if (!PL_colorset) reginitcolors());
6197 #ifndef PERL_IN_XSUB_RE
6198 /* Initialize these here instead of as-needed, as is quick and avoids
6199 * having to test them each time otherwise */
6200 if (! PL_AboveLatin1) {
6201 PL_AboveLatin1 = _new_invlist_C_array(AboveLatin1_invlist);
6202 PL_Latin1 = _new_invlist_C_array(Latin1_invlist);
6203 PL_UpperLatin1 = _new_invlist_C_array(UpperLatin1_invlist);
6204 PL_utf8_foldable = _new_invlist_C_array(_Perl_Any_Folds_invlist);
6205 PL_HasMultiCharFold =
6206 _new_invlist_C_array(_Perl_Folds_To_Multi_Char_invlist);
6210 pRExC_state->code_blocks = NULL;
6211 pRExC_state->num_code_blocks = 0;
6214 *is_bare_re = FALSE;
6216 if (expr && (expr->op_type == OP_LIST ||
6217 (expr->op_type == OP_NULL && expr->op_targ == OP_LIST))) {
6218 /* allocate code_blocks if needed */
6222 for (o = cLISTOPx(expr)->op_first; o; o = o->op_sibling)
6223 if (o->op_type == OP_NULL && (o->op_flags & OPf_SPECIAL))
6224 ncode++; /* count of DO blocks */
6226 pRExC_state->num_code_blocks = ncode;
6227 Newx(pRExC_state->code_blocks, ncode, struct reg_code_block);
6232 /* compile-time pattern with just OP_CONSTs and DO blocks */
6237 /* find how many CONSTs there are */
6240 if (expr->op_type == OP_CONST)
6243 for (o = cLISTOPx(expr)->op_first; o; o = o->op_sibling) {
6244 if (o->op_type == OP_CONST)
6248 /* fake up an SV array */
6250 assert(!new_patternp);
6251 Newx(new_patternp, n, SV*);
6252 SAVEFREEPV(new_patternp);
6256 if (expr->op_type == OP_CONST)
6257 new_patternp[n] = cSVOPx_sv(expr);
6259 for (o = cLISTOPx(expr)->op_first; o; o = o->op_sibling) {
6260 if (o->op_type == OP_CONST)
6261 new_patternp[n++] = cSVOPo_sv;
6266 DEBUG_PARSE_r(PerlIO_printf(Perl_debug_log,
6267 "Assembling pattern from %d elements%s\n", pat_count,
6268 orig_rx_flags & RXf_SPLIT ? " for split" : ""));
6270 /* set expr to the first arg op */
6272 if (pRExC_state->num_code_blocks
6273 && expr->op_type != OP_CONST)
6275 expr = cLISTOPx(expr)->op_first;
6276 assert( expr->op_type == OP_PUSHMARK
6277 || (expr->op_type == OP_NULL && expr->op_targ == OP_PUSHMARK)
6278 || expr->op_type == OP_PADRANGE);
6279 expr = expr->op_sibling;
6282 pat = S_concat_pat(aTHX_ pRExC_state, NULL, new_patternp, pat_count,
6283 expr, &recompile, NULL);
6285 /* handle bare (possibly after overloading) regex: foo =~ $re */
6290 if (SvTYPE(re) == SVt_REGEXP) {
6294 Safefree(pRExC_state->code_blocks);
6295 DEBUG_PARSE_r(PerlIO_printf(Perl_debug_log,
6296 "Precompiled pattern%s\n",
6297 orig_rx_flags & RXf_SPLIT ? " for split" : ""));
6303 exp = SvPV_nomg(pat, plen);
6305 if (!eng->op_comp) {
6306 if ((SvUTF8(pat) && IN_BYTES)
6307 || SvGMAGICAL(pat) || SvAMAGIC(pat))
6309 /* make a temporary copy; either to convert to bytes,
6310 * or to avoid repeating get-magic / overloaded stringify */
6311 pat = newSVpvn_flags(exp, plen, SVs_TEMP |
6312 (IN_BYTES ? 0 : SvUTF8(pat)));
6314 Safefree(pRExC_state->code_blocks);
6315 return CALLREGCOMP_ENG(eng, pat, orig_rx_flags);
6318 /* ignore the utf8ness if the pattern is 0 length */
6319 RExC_utf8 = RExC_orig_utf8 = (plen == 0 || IN_BYTES) ? 0 : SvUTF8(pat);
6320 RExC_uni_semantics = 0;
6321 RExC_contains_locale = 0;
6322 RExC_contains_i = 0;
6323 pRExC_state->runtime_code_qr = NULL;
6326 SV *dsv= sv_newmortal();
6327 RE_PV_QUOTED_DECL(s, RExC_utf8, dsv, exp, plen, 60);
6328 PerlIO_printf(Perl_debug_log, "%sCompiling REx%s %s\n",
6329 PL_colors[4],PL_colors[5],s);
6333 /* we jump here if we upgrade the pattern to utf8 and have to
6336 if ((pm_flags & PMf_USE_RE_EVAL)
6337 /* this second condition covers the non-regex literal case,
6338 * i.e. $foo =~ '(?{})'. */
6339 || (IN_PERL_COMPILETIME && (PL_hints & HINT_RE_EVAL))
6341 runtime_code = S_has_runtime_code(aTHX_ pRExC_state, exp, plen);
6343 /* return old regex if pattern hasn't changed */
6344 /* XXX: note in the below we have to check the flags as well as the
6347 * Things get a touch tricky as we have to compare the utf8 flag
6348 * independently from the compile flags. */
6352 && !!RX_UTF8(old_re) == !!RExC_utf8
6353 && ( RX_COMPFLAGS(old_re) == ( orig_rx_flags & RXf_PMf_FLAGCOPYMASK ) )
6354 && RX_PRECOMP(old_re)
6355 && RX_PRELEN(old_re) == plen
6356 && memEQ(RX_PRECOMP(old_re), exp, plen)
6357 && !runtime_code /* with runtime code, always recompile */ )
6359 Safefree(pRExC_state->code_blocks);
6363 rx_flags = orig_rx_flags;
6365 if (rx_flags & PMf_FOLD) {
6366 RExC_contains_i = 1;
6368 if (RExC_utf8 && initial_charset == REGEX_DEPENDS_CHARSET) {
6370 /* Set to use unicode semantics if the pattern is in utf8 and has the
6371 * 'depends' charset specified, as it means unicode when utf8 */
6372 set_regex_charset(&rx_flags, REGEX_UNICODE_CHARSET);
6376 RExC_flags = rx_flags;
6377 RExC_pm_flags = pm_flags;
6380 if (TAINTING_get && TAINT_get)
6381 Perl_croak(aTHX_ "Eval-group in insecure regular expression");
6383 if (!S_compile_runtime_code(aTHX_ pRExC_state, exp, plen)) {
6384 /* whoops, we have a non-utf8 pattern, whilst run-time code
6385 * got compiled as utf8. Try again with a utf8 pattern */
6386 S_pat_upgrade_to_utf8(aTHX_ pRExC_state, &exp, &plen,
6387 pRExC_state->num_code_blocks);
6388 goto redo_first_pass;
6391 assert(!pRExC_state->runtime_code_qr);
6397 RExC_in_lookbehind = 0;
6398 RExC_seen_zerolen = *exp == '^' ? -1 : 0;
6400 RExC_override_recoding = 0;
6401 RExC_in_multi_char_class = 0;
6403 /* First pass: determine size, legality. */
6406 RExC_end = exp + plen;
6411 RExC_emit = (regnode *) &RExC_emit_dummy;
6412 RExC_whilem_seen = 0;
6413 RExC_open_parens = NULL;
6414 RExC_close_parens = NULL;
6416 RExC_paren_names = NULL;
6418 RExC_paren_name_list = NULL;
6420 RExC_recurse = NULL;
6421 RExC_study_chunk_recursed = NULL;
6422 RExC_study_chunk_recursed_bytes= 0;
6423 RExC_recurse_count = 0;
6424 pRExC_state->code_index = 0;
6426 #if 0 /* REGC() is (currently) a NOP at the first pass.
6427 * Clever compilers notice this and complain. --jhi */
6428 REGC((U8)REG_MAGIC, (char*)RExC_emit);
6431 PerlIO_printf(Perl_debug_log, "Starting first pass (sizing)\n");
6433 RExC_lastparse=NULL;
6435 /* reg may croak on us, not giving us a chance to free
6436 pRExC_state->code_blocks. We cannot SAVEFREEPV it now, as we may
6437 need it to survive as long as the regexp (qr/(?{})/).
6438 We must check that code_blocksv is not already set, because we may
6439 have jumped back to restart the sizing pass. */
6440 if (pRExC_state->code_blocks && !code_blocksv) {
6441 code_blocksv = newSV_type(SVt_PV);
6442 SAVEFREESV(code_blocksv);
6443 SvPV_set(code_blocksv, (char *)pRExC_state->code_blocks);
6444 SvLEN_set(code_blocksv, 1); /*sufficient to make sv_clear free it*/
6446 if (reg(pRExC_state, 0, &flags,1) == NULL) {
6447 /* It's possible to write a regexp in ascii that represents Unicode
6448 codepoints outside of the byte range, such as via \x{100}. If we
6449 detect such a sequence we have to convert the entire pattern to utf8
6450 and then recompile, as our sizing calculation will have been based
6451 on 1 byte == 1 character, but we will need to use utf8 to encode
6452 at least some part of the pattern, and therefore must convert the whole
6455 if (flags & RESTART_UTF8) {
6456 S_pat_upgrade_to_utf8(aTHX_ pRExC_state, &exp, &plen,
6457 pRExC_state->num_code_blocks);
6458 goto redo_first_pass;
6460 Perl_croak(aTHX_ "panic: reg returned NULL to re_op_compile for sizing pass, flags=%#"UVxf"", (UV) flags);
6463 SvLEN_set(code_blocksv,0); /* no you can't have it, sv_clear */
6466 PerlIO_printf(Perl_debug_log,
6467 "Required size %"IVdf" nodes\n"
6468 "Starting second pass (creation)\n",
6471 RExC_lastparse=NULL;
6474 /* The first pass could have found things that force Unicode semantics */
6475 if ((RExC_utf8 || RExC_uni_semantics)
6476 && get_regex_charset(rx_flags) == REGEX_DEPENDS_CHARSET)
6478 set_regex_charset(&rx_flags, REGEX_UNICODE_CHARSET);
6481 /* Small enough for pointer-storage convention?
6482 If extralen==0, this means that we will not need long jumps. */
6483 if (RExC_size >= 0x10000L && RExC_extralen)
6484 RExC_size += RExC_extralen;
6487 if (RExC_whilem_seen > 15)
6488 RExC_whilem_seen = 15;
6490 /* Allocate space and zero-initialize. Note, the two step process
6491 of zeroing when in debug mode, thus anything assigned has to
6492 happen after that */
6493 rx = (REGEXP*) newSV_type(SVt_REGEXP);
6495 Newxc(ri, sizeof(regexp_internal) + (unsigned)RExC_size * sizeof(regnode),
6496 char, regexp_internal);
6497 if ( r == NULL || ri == NULL )
6498 FAIL("Regexp out of space");
6500 /* avoid reading uninitialized memory in DEBUGGING code in study_chunk() */
6501 Zero(ri, sizeof(regexp_internal) + (unsigned)RExC_size * sizeof(regnode),
6504 /* bulk initialize base fields with 0. */
6505 Zero(ri, sizeof(regexp_internal), char);
6508 /* non-zero initialization begins here */
6511 r->extflags = rx_flags;
6512 RXp_COMPFLAGS(r) = orig_rx_flags & RXf_PMf_FLAGCOPYMASK;
6514 if (pm_flags & PMf_IS_QR) {
6515 ri->code_blocks = pRExC_state->code_blocks;
6516 ri->num_code_blocks = pRExC_state->num_code_blocks;
6521 for (n = 0; n < pRExC_state->num_code_blocks; n++)
6522 if (pRExC_state->code_blocks[n].src_regex)
6523 SAVEFREESV(pRExC_state->code_blocks[n].src_regex);
6524 SAVEFREEPV(pRExC_state->code_blocks);
6528 bool has_p = ((r->extflags & RXf_PMf_KEEPCOPY) == RXf_PMf_KEEPCOPY);
6529 bool has_charset = (get_regex_charset(r->extflags)
6530 != REGEX_DEPENDS_CHARSET);
6532 /* The caret is output if there are any defaults: if not all the STD
6533 * flags are set, or if no character set specifier is needed */
6535 (((r->extflags & RXf_PMf_STD_PMMOD) != RXf_PMf_STD_PMMOD)
6537 bool has_runon = ((RExC_seen & REG_RUN_ON_COMMENT_SEEN)
6538 == REG_RUN_ON_COMMENT_SEEN);
6539 U16 reganch = (U16)((r->extflags & RXf_PMf_STD_PMMOD)
6540 >> RXf_PMf_STD_PMMOD_SHIFT);
6541 const char *fptr = STD_PAT_MODS; /*"msix"*/
6543 /* Allocate for the worst case, which is all the std flags are turned
6544 * on. If more precision is desired, we could do a population count of
6545 * the flags set. This could be done with a small lookup table, or by
6546 * shifting, masking and adding, or even, when available, assembly
6547 * language for a machine-language population count.
6548 * We never output a minus, as all those are defaults, so are
6549 * covered by the caret */
6550 const STRLEN wraplen = plen + has_p + has_runon
6551 + has_default /* If needs a caret */
6553 /* If needs a character set specifier */
6554 + ((has_charset) ? MAX_CHARSET_NAME_LENGTH : 0)
6555 + (sizeof(STD_PAT_MODS) - 1)
6556 + (sizeof("(?:)") - 1);
6558 Newx(p, wraplen + 1, char); /* +1 for the ending NUL */
6559 r->xpv_len_u.xpvlenu_pv = p;
6561 SvFLAGS(rx) |= SVf_UTF8;
6564 /* If a default, cover it using the caret */
6566 *p++= DEFAULT_PAT_MOD;
6570 const char* const name = get_regex_charset_name(r->extflags, &len);
6571 Copy(name, p, len, char);
6575 *p++ = KEEPCOPY_PAT_MOD; /*'p'*/
6578 while((ch = *fptr++)) {
6586 Copy(RExC_precomp, p, plen, char);
6587 assert ((RX_WRAPPED(rx) - p) < 16);
6588 r->pre_prefix = p - RX_WRAPPED(rx);
6594 SvCUR_set(rx, p - RX_WRAPPED(rx));
6598 r->nparens = RExC_npar - 1; /* set early to validate backrefs */
6600 /* setup various meta data about recursion, this all requires
6601 * RExC_npar to be correctly set, and a bit later on we clear it */
6602 if (RExC_seen & REG_RECURSE_SEEN) {
6603 Newxz(RExC_open_parens, RExC_npar,regnode *);
6604 SAVEFREEPV(RExC_open_parens);
6605 Newxz(RExC_close_parens,RExC_npar,regnode *);
6606 SAVEFREEPV(RExC_close_parens);
6608 if (RExC_seen & (REG_RECURSE_SEEN | REG_GOSTART_SEEN)) {
6609 /* Note, RExC_npar is 1 + the number of parens in a pattern.
6610 * So its 1 if there are no parens. */
6611 RExC_study_chunk_recursed_bytes= (RExC_npar >> 3) +
6612 ((RExC_npar & 0x07) != 0);
6613 Newx(RExC_study_chunk_recursed,
6614 RExC_study_chunk_recursed_bytes * RExC_npar, U8);
6615 SAVEFREEPV(RExC_study_chunk_recursed);
6618 /* Useful during FAIL. */
6619 #ifdef RE_TRACK_PATTERN_OFFSETS
6620 Newxz(ri->u.offsets, 2*RExC_size+1, U32); /* MJD 20001228 */
6621 DEBUG_OFFSETS_r(PerlIO_printf(Perl_debug_log,
6622 "%s %"UVuf" bytes for offset annotations.\n",
6623 ri->u.offsets ? "Got" : "Couldn't get",
6624 (UV)((2*RExC_size+1) * sizeof(U32))));
6626 SetProgLen(ri,RExC_size);
6631 /* Second pass: emit code. */
6632 RExC_flags = rx_flags; /* don't let top level (?i) bleed */
6633 RExC_pm_flags = pm_flags;
6635 RExC_end = exp + plen;
6638 RExC_emit_start = ri->program;
6639 RExC_emit = ri->program;
6640 RExC_emit_bound = ri->program + RExC_size + 1;
6641 pRExC_state->code_index = 0;
6643 REGC((U8)REG_MAGIC, (char*) RExC_emit++);
6644 if (reg(pRExC_state, 0, &flags,1) == NULL) {
6646 Perl_croak(aTHX_ "panic: reg returned NULL to re_op_compile for generation pass, flags=%#"UVxf"", (UV) flags);
6648 /* XXXX To minimize changes to RE engine we always allocate
6649 3-units-long substrs field. */
6650 Newx(r->substrs, 1, struct reg_substr_data);
6651 if (RExC_recurse_count) {
6652 Newxz(RExC_recurse,RExC_recurse_count,regnode *);
6653 SAVEFREEPV(RExC_recurse);
6657 r->minlen = minlen = sawlookahead = sawplus = sawopen = sawminmod = 0;
6658 Zero(r->substrs, 1, struct reg_substr_data);
6659 if (RExC_study_chunk_recursed)
6660 Zero(RExC_study_chunk_recursed,
6661 RExC_study_chunk_recursed_bytes * RExC_npar, U8);
6663 #ifdef TRIE_STUDY_OPT
6665 StructCopy(&zero_scan_data, &data, scan_data_t);
6666 copyRExC_state = RExC_state;
6669 DEBUG_OPTIMISE_r(PerlIO_printf(Perl_debug_log,"Restudying\n"));
6671 RExC_state = copyRExC_state;
6672 if (seen & REG_TOP_LEVEL_BRANCHES_SEEN)
6673 RExC_seen |= REG_TOP_LEVEL_BRANCHES_SEEN;
6675 RExC_seen &= ~REG_TOP_LEVEL_BRANCHES_SEEN;
6676 StructCopy(&zero_scan_data, &data, scan_data_t);
6679 StructCopy(&zero_scan_data, &data, scan_data_t);
6682 /* Dig out information for optimizations. */
6683 r->extflags = RExC_flags; /* was pm_op */
6684 /*dmq: removed as part of de-PMOP: pm->op_pmflags = RExC_flags; */
6687 SvUTF8_on(rx); /* Unicode in it? */
6688 ri->regstclass = NULL;
6689 if (RExC_naughty >= 10) /* Probably an expensive pattern. */
6690 r->intflags |= PREGf_NAUGHTY;
6691 scan = ri->program + 1; /* First BRANCH. */
6693 /* testing for BRANCH here tells us whether there is "must appear"
6694 data in the pattern. If there is then we can use it for optimisations */
6695 if (!(RExC_seen & REG_TOP_LEVEL_BRANCHES_SEEN)) { /* Only one top-level choice.
6698 STRLEN longest_float_length, longest_fixed_length;
6699 regnode_ssc ch_class; /* pointed to by data */
6701 SSize_t last_close = 0; /* pointed to by data */
6702 regnode *first= scan;
6703 regnode *first_next= regnext(first);
6705 * Skip introductions and multiplicators >= 1
6706 * so that we can extract the 'meat' of the pattern that must
6707 * match in the large if() sequence following.
6708 * NOTE that EXACT is NOT covered here, as it is normally
6709 * picked up by the optimiser separately.
6711 * This is unfortunate as the optimiser isnt handling lookahead
6712 * properly currently.
6715 while ((OP(first) == OPEN && (sawopen = 1)) ||
6716 /* An OR of *one* alternative - should not happen now. */
6717 (OP(first) == BRANCH && OP(first_next) != BRANCH) ||
6718 /* for now we can't handle lookbehind IFMATCH*/
6719 (OP(first) == IFMATCH && !first->flags && (sawlookahead = 1)) ||
6720 (OP(first) == PLUS) ||
6721 (OP(first) == MINMOD) ||
6722 /* An {n,m} with n>0 */
6723 (PL_regkind[OP(first)] == CURLY && ARG1(first) > 0) ||
6724 (OP(first) == NOTHING && PL_regkind[OP(first_next)] != END ))
6727 * the only op that could be a regnode is PLUS, all the rest
6728 * will be regnode_1 or regnode_2.
6730 * (yves doesn't think this is true)
6732 if (OP(first) == PLUS)
6735 if (OP(first) == MINMOD)
6737 first += regarglen[OP(first)];
6739 first = NEXTOPER(first);
6740 first_next= regnext(first);
6743 /* Starting-point info. */
6745 DEBUG_PEEP("first:",first,0);
6746 /* Ignore EXACT as we deal with it later. */
6747 if (PL_regkind[OP(first)] == EXACT) {
6748 if (OP(first) == EXACT)
6749 NOOP; /* Empty, get anchored substr later. */
6751 ri->regstclass = first;
6754 else if (PL_regkind[OP(first)] == TRIE &&
6755 ((reg_trie_data *)ri->data->data[ ARG(first) ])->minlen>0)
6758 /* this can happen only on restudy */
6759 if ( OP(first) == TRIE ) {
6760 struct regnode_1 *trieop = (struct regnode_1 *)
6761 PerlMemShared_calloc(1, sizeof(struct regnode_1));
6762 StructCopy(first,trieop,struct regnode_1);
6763 trie_op=(regnode *)trieop;
6765 struct regnode_charclass *trieop = (struct regnode_charclass *)
6766 PerlMemShared_calloc(1, sizeof(struct regnode_charclass));
6767 StructCopy(first,trieop,struct regnode_charclass);
6768 trie_op=(regnode *)trieop;
6771 make_trie_failtable(pRExC_state, (regnode *)first, trie_op, 0);
6772 ri->regstclass = trie_op;
6775 else if (REGNODE_SIMPLE(OP(first)))
6776 ri->regstclass = first;
6777 else if (PL_regkind[OP(first)] == BOUND ||
6778 PL_regkind[OP(first)] == NBOUND)
6779 ri->regstclass = first;
6780 else if (PL_regkind[OP(first)] == BOL) {
6781 r->intflags |= (OP(first) == MBOL
6783 : (OP(first) == SBOL
6786 first = NEXTOPER(first);
6789 else if (OP(first) == GPOS) {
6790 r->intflags |= PREGf_ANCH_GPOS;
6791 first = NEXTOPER(first);
6794 else if ((!sawopen || !RExC_sawback) &&
6795 (OP(first) == STAR &&
6796 PL_regkind[OP(NEXTOPER(first))] == REG_ANY) &&
6797 !(r->intflags & PREGf_ANCH) && !pRExC_state->num_code_blocks)
6799 /* turn .* into ^.* with an implied $*=1 */
6801 (OP(NEXTOPER(first)) == REG_ANY)
6804 r->intflags |= (type | PREGf_IMPLICIT);
6805 first = NEXTOPER(first);
6808 if (sawplus && !sawminmod && !sawlookahead
6809 && (!sawopen || !RExC_sawback)
6810 && !pRExC_state->num_code_blocks) /* May examine pos and $& */
6811 /* x+ must match at the 1st pos of run of x's */
6812 r->intflags |= PREGf_SKIP;
6814 /* Scan is after the zeroth branch, first is atomic matcher. */
6815 #ifdef TRIE_STUDY_OPT
6818 PerlIO_printf(Perl_debug_log, "first at %"IVdf"\n",
6819 (IV)(first - scan + 1))
6823 PerlIO_printf(Perl_debug_log, "first at %"IVdf"\n",
6824 (IV)(first - scan + 1))
6830 * If there's something expensive in the r.e., find the
6831 * longest literal string that must appear and make it the
6832 * regmust. Resolve ties in favor of later strings, since
6833 * the regstart check works with the beginning of the r.e.
6834 * and avoiding duplication strengthens checking. Not a
6835 * strong reason, but sufficient in the absence of others.
6836 * [Now we resolve ties in favor of the earlier string if
6837 * it happens that c_offset_min has been invalidated, since the
6838 * earlier string may buy us something the later one won't.]
6841 data.longest_fixed = newSVpvs("");
6842 data.longest_float = newSVpvs("");
6843 data.last_found = newSVpvs("");
6844 data.longest = &(data.longest_fixed);
6845 ENTER_with_name("study_chunk");
6846 SAVEFREESV(data.longest_fixed);
6847 SAVEFREESV(data.longest_float);
6848 SAVEFREESV(data.last_found);
6850 if (!ri->regstclass) {
6851 ssc_init(pRExC_state, &ch_class);
6852 data.start_class = &ch_class;
6853 stclass_flag = SCF_DO_STCLASS_AND;
6854 } else /* XXXX Check for BOUND? */
6856 data.last_closep = &last_close;
6859 minlen = study_chunk(pRExC_state, &first, &minlen, &fake,
6860 scan + RExC_size, /* Up to end */
6862 SCF_DO_SUBSTR | SCF_WHILEM_VISITED_POS | stclass_flag
6863 | (restudied ? SCF_TRIE_DOING_RESTUDY : 0),
6867 CHECK_RESTUDY_GOTO_butfirst(LEAVE_with_name("study_chunk"));
6870 if ( RExC_npar == 1 && data.longest == &(data.longest_fixed)
6871 && data.last_start_min == 0 && data.last_end > 0
6872 && !RExC_seen_zerolen
6873 && !(RExC_seen & REG_VERBARG_SEEN)
6874 && !(RExC_seen & REG_GPOS_SEEN)
6876 r->extflags |= RXf_CHECK_ALL;
6878 scan_commit(pRExC_state, &data,&minlen,0);
6880 longest_float_length = CHR_SVLEN(data.longest_float);
6882 if (! ((SvCUR(data.longest_fixed) /* ok to leave SvCUR */
6883 && data.offset_fixed == data.offset_float_min
6884 && SvCUR(data.longest_fixed) == SvCUR(data.longest_float)))
6885 && S_setup_longest (aTHX_ pRExC_state,
6889 &(r->float_end_shift),
6890 data.lookbehind_float,
6891 data.offset_float_min,
6893 longest_float_length,
6894 cBOOL(data.flags & SF_FL_BEFORE_EOL),
6895 cBOOL(data.flags & SF_FL_BEFORE_MEOL)))
6897 r->float_min_offset = data.offset_float_min - data.lookbehind_float;
6898 r->float_max_offset = data.offset_float_max;
6899 if (data.offset_float_max < SSize_t_MAX) /* Don't offset infinity */
6900 r->float_max_offset -= data.lookbehind_float;
6901 SvREFCNT_inc_simple_void_NN(data.longest_float);
6904 r->float_substr = r->float_utf8 = NULL;
6905 longest_float_length = 0;
6908 longest_fixed_length = CHR_SVLEN(data.longest_fixed);
6910 if (S_setup_longest (aTHX_ pRExC_state,
6912 &(r->anchored_utf8),
6913 &(r->anchored_substr),
6914 &(r->anchored_end_shift),
6915 data.lookbehind_fixed,
6918 longest_fixed_length,
6919 cBOOL(data.flags & SF_FIX_BEFORE_EOL),
6920 cBOOL(data.flags & SF_FIX_BEFORE_MEOL)))
6922 r->anchored_offset = data.offset_fixed - data.lookbehind_fixed;
6923 SvREFCNT_inc_simple_void_NN(data.longest_fixed);
6926 r->anchored_substr = r->anchored_utf8 = NULL;
6927 longest_fixed_length = 0;
6929 LEAVE_with_name("study_chunk");
6932 && (OP(ri->regstclass) == REG_ANY || OP(ri->regstclass) == SANY))
6933 ri->regstclass = NULL;
6935 if ((!(r->anchored_substr || r->anchored_utf8) || r->anchored_offset)
6937 && ! (ANYOF_FLAGS(data.start_class) & ANYOF_EMPTY_STRING)
6938 && !ssc_is_anything(data.start_class))
6940 const U32 n = add_data(pRExC_state, STR_WITH_LEN("f"));
6942 ssc_finalize(pRExC_state, data.start_class);
6944 Newx(RExC_rxi->data->data[n], 1, regnode_ssc);
6945 StructCopy(data.start_class,
6946 (regnode_ssc*)RExC_rxi->data->data[n],
6948 ri->regstclass = (regnode*)RExC_rxi->data->data[n];
6949 r->intflags &= ~PREGf_SKIP; /* Used in find_byclass(). */
6950 DEBUG_COMPILE_r({ SV *sv = sv_newmortal();
6951 regprop(r, sv, (regnode*)data.start_class, NULL);
6952 PerlIO_printf(Perl_debug_log,
6953 "synthetic stclass \"%s\".\n",
6954 SvPVX_const(sv));});
6955 data.start_class = NULL;
6958 /* A temporary algorithm prefers floated substr to fixed one to dig
6960 if (longest_fixed_length > longest_float_length) {
6961 r->substrs->check_ix = 0;
6962 r->check_end_shift = r->anchored_end_shift;
6963 r->check_substr = r->anchored_substr;
6964 r->check_utf8 = r->anchored_utf8;
6965 r->check_offset_min = r->check_offset_max = r->anchored_offset;
6966 if (r->intflags & (PREGf_ANCH_SBOL|PREGf_ANCH_GPOS))
6967 r->intflags |= PREGf_NOSCAN;
6970 r->substrs->check_ix = 1;
6971 r->check_end_shift = r->float_end_shift;
6972 r->check_substr = r->float_substr;
6973 r->check_utf8 = r->float_utf8;
6974 r->check_offset_min = r->float_min_offset;
6975 r->check_offset_max = r->float_max_offset;
6977 if ((r->check_substr || r->check_utf8) ) {
6978 r->extflags |= RXf_USE_INTUIT;
6979 if (SvTAIL(r->check_substr ? r->check_substr : r->check_utf8))
6980 r->extflags |= RXf_INTUIT_TAIL;
6982 r->substrs->data[0].max_offset = r->substrs->data[0].min_offset;
6984 /* XXX Unneeded? dmq (shouldn't as this is handled elsewhere)
6985 if ( (STRLEN)minlen < longest_float_length )
6986 minlen= longest_float_length;
6987 if ( (STRLEN)minlen < longest_fixed_length )
6988 minlen= longest_fixed_length;
6992 /* Several toplevels. Best we can is to set minlen. */
6994 regnode_ssc ch_class;
6995 SSize_t last_close = 0;
6997 DEBUG_PARSE_r(PerlIO_printf(Perl_debug_log, "\nMulti Top Level\n"));
6999 scan = ri->program + 1;
7000 ssc_init(pRExC_state, &ch_class);
7001 data.start_class = &ch_class;
7002 data.last_closep = &last_close;
7005 minlen = study_chunk(pRExC_state,
7006 &scan, &minlen, &fake, scan + RExC_size, &data, -1, 0, NULL,
7007 SCF_DO_STCLASS_AND|SCF_WHILEM_VISITED_POS|(restudied
7008 ? SCF_TRIE_DOING_RESTUDY
7012 CHECK_RESTUDY_GOTO_butfirst(NOOP);
7014 r->check_substr = r->check_utf8 = r->anchored_substr = r->anchored_utf8
7015 = r->float_substr = r->float_utf8 = NULL;
7017 if (! (ANYOF_FLAGS(data.start_class) & ANYOF_EMPTY_STRING)
7018 && ! ssc_is_anything(data.start_class))
7020 const U32 n = add_data(pRExC_state, STR_WITH_LEN("f"));
7022 ssc_finalize(pRExC_state, data.start_class);
7024 Newx(RExC_rxi->data->data[n], 1, regnode_ssc);
7025 StructCopy(data.start_class,
7026 (regnode_ssc*)RExC_rxi->data->data[n],
7028 ri->regstclass = (regnode*)RExC_rxi->data->data[n];
7029 r->intflags &= ~PREGf_SKIP; /* Used in find_byclass(). */
7030 DEBUG_COMPILE_r({ SV* sv = sv_newmortal();
7031 regprop(r, sv, (regnode*)data.start_class, NULL);
7032 PerlIO_printf(Perl_debug_log,
7033 "synthetic stclass \"%s\".\n",
7034 SvPVX_const(sv));});
7035 data.start_class = NULL;
7039 if (RExC_seen & REG_UNBOUNDED_QUANTIFIER_SEEN) {
7040 r->extflags |= RXf_UNBOUNDED_QUANTIFIER_SEEN;
7041 r->maxlen = REG_INFTY;
7044 r->maxlen = RExC_maxlen;
7047 /* Guard against an embedded (?=) or (?<=) with a longer minlen than
7048 the "real" pattern. */
7050 PerlIO_printf(Perl_debug_log,"minlen: %"IVdf" r->minlen:%"IVdf" maxlen:%ld\n",
7051 (IV)minlen, (IV)r->minlen, RExC_maxlen);
7053 r->minlenret = minlen;
7054 if (r->minlen < minlen)
7057 if (RExC_seen & REG_GPOS_SEEN)
7058 r->intflags |= PREGf_GPOS_SEEN;
7059 if (RExC_seen & REG_LOOKBEHIND_SEEN)
7060 r->extflags |= RXf_NO_INPLACE_SUBST; /* inplace might break the
7062 if (pRExC_state->num_code_blocks)
7063 r->extflags |= RXf_EVAL_SEEN;
7064 if (RExC_seen & REG_CANY_SEEN)
7065 r->intflags |= PREGf_CANY_SEEN;
7066 if (RExC_seen & REG_VERBARG_SEEN)
7068 r->intflags |= PREGf_VERBARG_SEEN;
7069 r->extflags |= RXf_NO_INPLACE_SUBST; /* don't understand this! Yves */
7071 if (RExC_seen & REG_CUTGROUP_SEEN)
7072 r->intflags |= PREGf_CUTGROUP_SEEN;
7073 if (pm_flags & PMf_USE_RE_EVAL)
7074 r->intflags |= PREGf_USE_RE_EVAL;
7075 if (RExC_paren_names)
7076 RXp_PAREN_NAMES(r) = MUTABLE_HV(SvREFCNT_inc(RExC_paren_names));
7078 RXp_PAREN_NAMES(r) = NULL;
7080 /* If we have seen an anchor in our pattern then we set the extflag RXf_IS_ANCHORED
7081 * so it can be used in pp.c */
7082 if (r->intflags & PREGf_ANCH)
7083 r->extflags |= RXf_IS_ANCHORED;
7087 /* this is used to identify "special" patterns that might result
7088 * in Perl NOT calling the regex engine and instead doing the match "itself",
7089 * particularly special cases in split//. By having the regex compiler
7090 * do this pattern matching at a regop level (instead of by inspecting the pattern)
7091 * we avoid weird issues with equivalent patterns resulting in different behavior,
7092 * AND we allow non Perl engines to get the same optimizations by the setting the
7093 * flags appropriately - Yves */
7094 regnode *first = ri->program + 1;
7096 regnode *next = NEXTOPER(first);
7099 if (PL_regkind[fop] == NOTHING && nop == END)
7100 r->extflags |= RXf_NULL;
7101 else if (PL_regkind[fop] == BOL && nop == END)
7102 r->extflags |= RXf_START_ONLY;
7103 else if (fop == PLUS
7104 && PL_regkind[nop] == POSIXD && FLAGS(next) == _CC_SPACE
7105 && OP(regnext(first)) == END)
7106 r->extflags |= RXf_WHITE;
7107 else if ( r->extflags & RXf_SPLIT
7109 && STR_LEN(first) == 1
7110 && *(STRING(first)) == ' '
7111 && OP(regnext(first)) == END )
7112 r->extflags |= (RXf_SKIPWHITE|RXf_WHITE);
7116 if (RExC_contains_locale) {
7117 RXp_EXTFLAGS(r) |= RXf_TAINTED;
7121 if (RExC_paren_names) {
7122 ri->name_list_idx = add_data( pRExC_state, STR_WITH_LEN("a"));
7123 ri->data->data[ri->name_list_idx]
7124 = (void*)SvREFCNT_inc(RExC_paren_name_list);
7127 ri->name_list_idx = 0;
7129 if (RExC_recurse_count) {
7130 for ( ; RExC_recurse_count ; RExC_recurse_count-- ) {
7131 const regnode *scan = RExC_recurse[RExC_recurse_count-1];
7132 ARG2L_SET( scan, RExC_open_parens[ARG(scan)-1] - scan );
7135 Newxz(r->offs, RExC_npar, regexp_paren_pair);
7136 /* assume we don't need to swap parens around before we match */
7140 PerlIO_printf(Perl_debug_log,"Final program:\n");
7143 #ifdef RE_TRACK_PATTERN_OFFSETS
7144 DEBUG_OFFSETS_r(if (ri->u.offsets) {
7145 const STRLEN len = ri->u.offsets[0];
7147 GET_RE_DEBUG_FLAGS_DECL;
7148 PerlIO_printf(Perl_debug_log,
7149 "Offsets: [%"UVuf"]\n\t", (UV)ri->u.offsets[0]);
7150 for (i = 1; i <= len; i++) {
7151 if (ri->u.offsets[i*2-1] || ri->u.offsets[i*2])
7152 PerlIO_printf(Perl_debug_log, "%"UVuf":%"UVuf"[%"UVuf"] ",
7153 (UV)i, (UV)ri->u.offsets[i*2-1], (UV)ri->u.offsets[i*2]);
7155 PerlIO_printf(Perl_debug_log, "\n");
7160 /* under ithreads the ?pat? PMf_USED flag on the pmop is simulated
7161 * by setting the regexp SV to readonly-only instead. If the
7162 * pattern's been recompiled, the USEDness should remain. */
7163 if (old_re && SvREADONLY(old_re))
7171 Perl_reg_named_buff(pTHX_ REGEXP * const rx, SV * const key, SV * const value,
7174 PERL_ARGS_ASSERT_REG_NAMED_BUFF;
7176 PERL_UNUSED_ARG(value);
7178 if (flags & RXapif_FETCH) {
7179 return reg_named_buff_fetch(rx, key, flags);
7180 } else if (flags & (RXapif_STORE | RXapif_DELETE | RXapif_CLEAR)) {
7181 Perl_croak_no_modify();
7183 } else if (flags & RXapif_EXISTS) {
7184 return reg_named_buff_exists(rx, key, flags)
7187 } else if (flags & RXapif_REGNAMES) {
7188 return reg_named_buff_all(rx, flags);
7189 } else if (flags & (RXapif_SCALAR | RXapif_REGNAMES_COUNT)) {
7190 return reg_named_buff_scalar(rx, flags);
7192 Perl_croak(aTHX_ "panic: Unknown flags %d in named_buff", (int)flags);
7198 Perl_reg_named_buff_iter(pTHX_ REGEXP * const rx, const SV * const lastkey,
7201 PERL_ARGS_ASSERT_REG_NAMED_BUFF_ITER;
7202 PERL_UNUSED_ARG(lastkey);
7204 if (flags & RXapif_FIRSTKEY)
7205 return reg_named_buff_firstkey(rx, flags);
7206 else if (flags & RXapif_NEXTKEY)
7207 return reg_named_buff_nextkey(rx, flags);
7209 Perl_croak(aTHX_ "panic: Unknown flags %d in named_buff_iter",
7216 Perl_reg_named_buff_fetch(pTHX_ REGEXP * const r, SV * const namesv,
7219 AV *retarray = NULL;
7221 struct regexp *const rx = ReANY(r);
7223 PERL_ARGS_ASSERT_REG_NAMED_BUFF_FETCH;
7225 if (flags & RXapif_ALL)
7228 if (rx && RXp_PAREN_NAMES(rx)) {
7229 HE *he_str = hv_fetch_ent( RXp_PAREN_NAMES(rx), namesv, 0, 0 );
7232 SV* sv_dat=HeVAL(he_str);
7233 I32 *nums=(I32*)SvPVX(sv_dat);
7234 for ( i=0; i<SvIVX(sv_dat); i++ ) {
7235 if ((I32)(rx->nparens) >= nums[i]
7236 && rx->offs[nums[i]].start != -1
7237 && rx->offs[nums[i]].end != -1)
7240 CALLREG_NUMBUF_FETCH(r,nums[i],ret);
7245 ret = newSVsv(&PL_sv_undef);
7248 av_push(retarray, ret);
7251 return newRV_noinc(MUTABLE_SV(retarray));
7258 Perl_reg_named_buff_exists(pTHX_ REGEXP * const r, SV * const key,
7261 struct regexp *const rx = ReANY(r);
7263 PERL_ARGS_ASSERT_REG_NAMED_BUFF_EXISTS;
7265 if (rx && RXp_PAREN_NAMES(rx)) {
7266 if (flags & RXapif_ALL) {
7267 return hv_exists_ent(RXp_PAREN_NAMES(rx), key, 0);
7269 SV *sv = CALLREG_NAMED_BUFF_FETCH(r, key, flags);
7271 SvREFCNT_dec_NN(sv);
7283 Perl_reg_named_buff_firstkey(pTHX_ REGEXP * const r, const U32 flags)
7285 struct regexp *const rx = ReANY(r);
7287 PERL_ARGS_ASSERT_REG_NAMED_BUFF_FIRSTKEY;
7289 if ( rx && RXp_PAREN_NAMES(rx) ) {
7290 (void)hv_iterinit(RXp_PAREN_NAMES(rx));
7292 return CALLREG_NAMED_BUFF_NEXTKEY(r, NULL, flags & ~RXapif_FIRSTKEY);
7299 Perl_reg_named_buff_nextkey(pTHX_ REGEXP * const r, const U32 flags)
7301 struct regexp *const rx = ReANY(r);
7302 GET_RE_DEBUG_FLAGS_DECL;
7304 PERL_ARGS_ASSERT_REG_NAMED_BUFF_NEXTKEY;
7306 if (rx && RXp_PAREN_NAMES(rx)) {
7307 HV *hv = RXp_PAREN_NAMES(rx);
7309 while ( (temphe = hv_iternext_flags(hv,0)) ) {
7312 SV* sv_dat = HeVAL(temphe);
7313 I32 *nums = (I32*)SvPVX(sv_dat);
7314 for ( i = 0; i < SvIVX(sv_dat); i++ ) {
7315 if ((I32)(rx->lastparen) >= nums[i] &&
7316 rx->offs[nums[i]].start != -1 &&
7317 rx->offs[nums[i]].end != -1)
7323 if (parno || flags & RXapif_ALL) {
7324 return newSVhek(HeKEY_hek(temphe));
7332 Perl_reg_named_buff_scalar(pTHX_ REGEXP * const r, const U32 flags)
7337 struct regexp *const rx = ReANY(r);
7339 PERL_ARGS_ASSERT_REG_NAMED_BUFF_SCALAR;
7341 if (rx && RXp_PAREN_NAMES(rx)) {
7342 if (flags & (RXapif_ALL | RXapif_REGNAMES_COUNT)) {
7343 return newSViv(HvTOTALKEYS(RXp_PAREN_NAMES(rx)));
7344 } else if (flags & RXapif_ONE) {
7345 ret = CALLREG_NAMED_BUFF_ALL(r, (flags | RXapif_REGNAMES));
7346 av = MUTABLE_AV(SvRV(ret));
7347 length = av_tindex(av);
7348 SvREFCNT_dec_NN(ret);
7349 return newSViv(length + 1);
7351 Perl_croak(aTHX_ "panic: Unknown flags %d in named_buff_scalar",
7356 return &PL_sv_undef;
7360 Perl_reg_named_buff_all(pTHX_ REGEXP * const r, const U32 flags)
7362 struct regexp *const rx = ReANY(r);
7365 PERL_ARGS_ASSERT_REG_NAMED_BUFF_ALL;
7367 if (rx && RXp_PAREN_NAMES(rx)) {
7368 HV *hv= RXp_PAREN_NAMES(rx);
7370 (void)hv_iterinit(hv);
7371 while ( (temphe = hv_iternext_flags(hv,0)) ) {
7374 SV* sv_dat = HeVAL(temphe);
7375 I32 *nums = (I32*)SvPVX(sv_dat);
7376 for ( i = 0; i < SvIVX(sv_dat); i++ ) {
7377 if ((I32)(rx->lastparen) >= nums[i] &&
7378 rx->offs[nums[i]].start != -1 &&
7379 rx->offs[nums[i]].end != -1)
7385 if (parno || flags & RXapif_ALL) {
7386 av_push(av, newSVhek(HeKEY_hek(temphe)));
7391 return newRV_noinc(MUTABLE_SV(av));
7395 Perl_reg_numbered_buff_fetch(pTHX_ REGEXP * const r, const I32 paren,
7398 struct regexp *const rx = ReANY(r);
7404 PERL_ARGS_ASSERT_REG_NUMBERED_BUFF_FETCH;
7406 if ( n == RX_BUFF_IDX_CARET_PREMATCH
7407 || n == RX_BUFF_IDX_CARET_FULLMATCH
7408 || n == RX_BUFF_IDX_CARET_POSTMATCH
7411 bool keepcopy = cBOOL(rx->extflags & RXf_PMf_KEEPCOPY);
7413 /* on something like
7416 * the KEEPCOPY is set on the PMOP rather than the regex */
7417 if (PL_curpm && r == PM_GETRE(PL_curpm))
7418 keepcopy = cBOOL(PL_curpm->op_pmflags & PMf_KEEPCOPY);
7427 if (n == RX_BUFF_IDX_CARET_FULLMATCH)
7428 /* no need to distinguish between them any more */
7429 n = RX_BUFF_IDX_FULLMATCH;
7431 if ((n == RX_BUFF_IDX_PREMATCH || n == RX_BUFF_IDX_CARET_PREMATCH)
7432 && rx->offs[0].start != -1)
7434 /* $`, ${^PREMATCH} */
7435 i = rx->offs[0].start;
7439 if ((n == RX_BUFF_IDX_POSTMATCH || n == RX_BUFF_IDX_CARET_POSTMATCH)
7440 && rx->offs[0].end != -1)
7442 /* $', ${^POSTMATCH} */
7443 s = rx->subbeg - rx->suboffset + rx->offs[0].end;
7444 i = rx->sublen + rx->suboffset - rx->offs[0].end;
7447 if ( 0 <= n && n <= (I32)rx->nparens &&
7448 (s1 = rx->offs[n].start) != -1 &&
7449 (t1 = rx->offs[n].end) != -1)
7451 /* $&, ${^MATCH}, $1 ... */
7453 s = rx->subbeg + s1 - rx->suboffset;
7458 assert(s >= rx->subbeg);
7459 assert((STRLEN)rx->sublen >= (STRLEN)((s - rx->subbeg) + i) );
7461 #ifdef NO_TAINT_SUPPORT
7462 sv_setpvn(sv, s, i);
7464 const int oldtainted = TAINT_get;
7466 sv_setpvn(sv, s, i);
7467 TAINT_set(oldtainted);
7469 if ( (rx->intflags & PREGf_CANY_SEEN)
7470 ? (RXp_MATCH_UTF8(rx)
7471 && (!i || is_utf8_string((U8*)s, i)))
7472 : (RXp_MATCH_UTF8(rx)) )
7479 if (RXp_MATCH_TAINTED(rx)) {
7480 if (SvTYPE(sv) >= SVt_PVMG) {
7481 MAGIC* const mg = SvMAGIC(sv);
7484 SvMAGIC_set(sv, mg->mg_moremagic);
7486 if ((mgt = SvMAGIC(sv))) {
7487 mg->mg_moremagic = mgt;
7488 SvMAGIC_set(sv, mg);
7499 sv_setsv(sv,&PL_sv_undef);
7505 Perl_reg_numbered_buff_store(pTHX_ REGEXP * const rx, const I32 paren,
7506 SV const * const value)
7508 PERL_ARGS_ASSERT_REG_NUMBERED_BUFF_STORE;
7510 PERL_UNUSED_ARG(rx);
7511 PERL_UNUSED_ARG(paren);
7512 PERL_UNUSED_ARG(value);
7515 Perl_croak_no_modify();
7519 Perl_reg_numbered_buff_length(pTHX_ REGEXP * const r, const SV * const sv,
7522 struct regexp *const rx = ReANY(r);
7526 PERL_ARGS_ASSERT_REG_NUMBERED_BUFF_LENGTH;
7528 if ( paren == RX_BUFF_IDX_CARET_PREMATCH
7529 || paren == RX_BUFF_IDX_CARET_FULLMATCH
7530 || paren == RX_BUFF_IDX_CARET_POSTMATCH
7533 bool keepcopy = cBOOL(rx->extflags & RXf_PMf_KEEPCOPY);
7535 /* on something like
7538 * the KEEPCOPY is set on the PMOP rather than the regex */
7539 if (PL_curpm && r == PM_GETRE(PL_curpm))
7540 keepcopy = cBOOL(PL_curpm->op_pmflags & PMf_KEEPCOPY);
7546 /* Some of this code was originally in C<Perl_magic_len> in F<mg.c> */
7548 case RX_BUFF_IDX_CARET_PREMATCH: /* ${^PREMATCH} */
7549 case RX_BUFF_IDX_PREMATCH: /* $` */
7550 if (rx->offs[0].start != -1) {
7551 i = rx->offs[0].start;
7560 case RX_BUFF_IDX_CARET_POSTMATCH: /* ${^POSTMATCH} */
7561 case RX_BUFF_IDX_POSTMATCH: /* $' */
7562 if (rx->offs[0].end != -1) {
7563 i = rx->sublen - rx->offs[0].end;
7565 s1 = rx->offs[0].end;
7572 default: /* $& / ${^MATCH}, $1, $2, ... */
7573 if (paren <= (I32)rx->nparens &&
7574 (s1 = rx->offs[paren].start) != -1 &&
7575 (t1 = rx->offs[paren].end) != -1)
7581 if (ckWARN(WARN_UNINITIALIZED))
7582 report_uninit((const SV *)sv);
7587 if (i > 0 && RXp_MATCH_UTF8(rx)) {
7588 const char * const s = rx->subbeg - rx->suboffset + s1;
7593 if (is_utf8_string_loclen((U8*)s, i, &ep, &el))
7600 Perl_reg_qr_package(pTHX_ REGEXP * const rx)
7602 PERL_ARGS_ASSERT_REG_QR_PACKAGE;
7603 PERL_UNUSED_ARG(rx);
7607 return newSVpvs("Regexp");
7610 /* Scans the name of a named buffer from the pattern.
7611 * If flags is REG_RSN_RETURN_NULL returns null.
7612 * If flags is REG_RSN_RETURN_NAME returns an SV* containing the name
7613 * If flags is REG_RSN_RETURN_DATA returns the data SV* corresponding
7614 * to the parsed name as looked up in the RExC_paren_names hash.
7615 * If there is an error throws a vFAIL().. type exception.
7618 #define REG_RSN_RETURN_NULL 0
7619 #define REG_RSN_RETURN_NAME 1
7620 #define REG_RSN_RETURN_DATA 2
7623 S_reg_scan_name(pTHX_ RExC_state_t *pRExC_state, U32 flags)
7625 char *name_start = RExC_parse;
7627 PERL_ARGS_ASSERT_REG_SCAN_NAME;
7629 assert (RExC_parse <= RExC_end);
7630 if (RExC_parse == RExC_end) NOOP;
7631 else if (isIDFIRST_lazy_if(RExC_parse, UTF)) {
7632 /* skip IDFIRST by using do...while */
7635 RExC_parse += UTF8SKIP(RExC_parse);
7636 } while (isWORDCHAR_utf8((U8*)RExC_parse));
7640 } while (isWORDCHAR(*RExC_parse));
7642 RExC_parse++; /* so the <- from the vFAIL is after the offending
7644 vFAIL("Group name must start with a non-digit word character");
7648 = newSVpvn_flags(name_start, (int)(RExC_parse - name_start),
7649 SVs_TEMP | (UTF ? SVf_UTF8 : 0));
7650 if ( flags == REG_RSN_RETURN_NAME)
7652 else if (flags==REG_RSN_RETURN_DATA) {
7655 if ( ! sv_name ) /* should not happen*/
7656 Perl_croak(aTHX_ "panic: no svname in reg_scan_name");
7657 if (RExC_paren_names)
7658 he_str = hv_fetch_ent( RExC_paren_names, sv_name, 0, 0 );
7660 sv_dat = HeVAL(he_str);
7662 vFAIL("Reference to nonexistent named group");
7666 Perl_croak(aTHX_ "panic: bad flag %lx in reg_scan_name",
7667 (unsigned long) flags);
7669 assert(0); /* NOT REACHED */
7674 #define DEBUG_PARSE_MSG(funcname) DEBUG_PARSE_r({ \
7675 int rem=(int)(RExC_end - RExC_parse); \
7684 if (RExC_lastparse!=RExC_parse) \
7685 PerlIO_printf(Perl_debug_log," >%.*s%-*s", \
7688 iscut ? "..." : "<" \
7691 PerlIO_printf(Perl_debug_log,"%16s",""); \
7694 num = RExC_size + 1; \
7696 num=REG_NODE_NUM(RExC_emit); \
7697 if (RExC_lastnum!=num) \
7698 PerlIO_printf(Perl_debug_log,"|%4d",num); \
7700 PerlIO_printf(Perl_debug_log,"|%4s",""); \
7701 PerlIO_printf(Perl_debug_log,"|%*s%-4s", \
7702 (int)((depth*2)), "", \
7706 RExC_lastparse=RExC_parse; \
7711 #define DEBUG_PARSE(funcname) DEBUG_PARSE_r({ \
7712 DEBUG_PARSE_MSG((funcname)); \
7713 PerlIO_printf(Perl_debug_log,"%4s","\n"); \
7715 #define DEBUG_PARSE_FMT(funcname,fmt,args) DEBUG_PARSE_r({ \
7716 DEBUG_PARSE_MSG((funcname)); \
7717 PerlIO_printf(Perl_debug_log,fmt "\n",args); \
7720 /* This section of code defines the inversion list object and its methods. The
7721 * interfaces are highly subject to change, so as much as possible is static to
7722 * this file. An inversion list is here implemented as a malloc'd C UV array
7723 * as an SVt_INVLIST scalar.
7725 * An inversion list for Unicode is an array of code points, sorted by ordinal
7726 * number. The zeroth element is the first code point in the list. The 1th
7727 * element is the first element beyond that not in the list. In other words,
7728 * the first range is
7729 * invlist[0]..(invlist[1]-1)
7730 * The other ranges follow. Thus every element whose index is divisible by two
7731 * marks the beginning of a range that is in the list, and every element not
7732 * divisible by two marks the beginning of a range not in the list. A single
7733 * element inversion list that contains the single code point N generally
7734 * consists of two elements
7737 * (The exception is when N is the highest representable value on the
7738 * machine, in which case the list containing just it would be a single
7739 * element, itself. By extension, if the last range in the list extends to
7740 * infinity, then the first element of that range will be in the inversion list
7741 * at a position that is divisible by two, and is the final element in the
7743 * Taking the complement (inverting) an inversion list is quite simple, if the
7744 * first element is 0, remove it; otherwise add a 0 element at the beginning.
7745 * This implementation reserves an element at the beginning of each inversion
7746 * list to always contain 0; there is an additional flag in the header which
7747 * indicates if the list begins at the 0, or is offset to begin at the next
7750 * More about inversion lists can be found in "Unicode Demystified"
7751 * Chapter 13 by Richard Gillam, published by Addison-Wesley.
7752 * More will be coming when functionality is added later.
7754 * The inversion list data structure is currently implemented as an SV pointing
7755 * to an array of UVs that the SV thinks are bytes. This allows us to have an
7756 * array of UV whose memory management is automatically handled by the existing
7757 * facilities for SV's.
7759 * Some of the methods should always be private to the implementation, and some
7760 * should eventually be made public */
7762 /* The header definitions are in F<inline_invlist.c> */
7764 PERL_STATIC_INLINE UV*
7765 S__invlist_array_init(pTHX_ SV* const invlist, const bool will_have_0)
7767 /* Returns a pointer to the first element in the inversion list's array.
7768 * This is called upon initialization of an inversion list. Where the
7769 * array begins depends on whether the list has the code point U+0000 in it
7770 * or not. The other parameter tells it whether the code that follows this
7771 * call is about to put a 0 in the inversion list or not. The first
7772 * element is either the element reserved for 0, if TRUE, or the element
7773 * after it, if FALSE */
7775 bool* offset = get_invlist_offset_addr(invlist);
7776 UV* zero_addr = (UV *) SvPVX(invlist);
7778 PERL_ARGS_ASSERT__INVLIST_ARRAY_INIT;
7781 assert(! _invlist_len(invlist));
7785 /* 1^1 = 0; 1^0 = 1 */
7786 *offset = 1 ^ will_have_0;
7787 return zero_addr + *offset;
7790 PERL_STATIC_INLINE UV*
7791 S_invlist_array(pTHX_ SV* const invlist)
7793 /* Returns the pointer to the inversion list's array. Every time the
7794 * length changes, this needs to be called in case malloc or realloc moved
7797 PERL_ARGS_ASSERT_INVLIST_ARRAY;
7799 /* Must not be empty. If these fail, you probably didn't check for <len>
7800 * being non-zero before trying to get the array */
7801 assert(_invlist_len(invlist));
7803 /* The very first element always contains zero, The array begins either
7804 * there, or if the inversion list is offset, at the element after it.
7805 * The offset header field determines which; it contains 0 or 1 to indicate
7806 * how much additionally to add */
7807 assert(0 == *(SvPVX(invlist)));
7808 return ((UV *) SvPVX(invlist) + *get_invlist_offset_addr(invlist));
7811 PERL_STATIC_INLINE void
7812 S_invlist_set_len(pTHX_ SV* const invlist, const UV len, const bool offset)
7814 /* Sets the current number of elements stored in the inversion list.
7815 * Updates SvCUR correspondingly */
7817 PERL_ARGS_ASSERT_INVLIST_SET_LEN;
7819 assert(SvTYPE(invlist) == SVt_INVLIST);
7824 : TO_INTERNAL_SIZE(len + offset));
7825 assert(SvLEN(invlist) == 0 || SvCUR(invlist) <= SvLEN(invlist));
7828 PERL_STATIC_INLINE IV*
7829 S_get_invlist_previous_index_addr(pTHX_ SV* invlist)
7831 /* Return the address of the IV that is reserved to hold the cached index
7834 PERL_ARGS_ASSERT_GET_INVLIST_PREVIOUS_INDEX_ADDR;
7836 assert(SvTYPE(invlist) == SVt_INVLIST);
7838 return &(((XINVLIST*) SvANY(invlist))->prev_index);
7841 PERL_STATIC_INLINE IV
7842 S_invlist_previous_index(pTHX_ SV* const invlist)
7844 /* Returns cached index of previous search */
7846 PERL_ARGS_ASSERT_INVLIST_PREVIOUS_INDEX;
7848 return *get_invlist_previous_index_addr(invlist);
7851 PERL_STATIC_INLINE void
7852 S_invlist_set_previous_index(pTHX_ SV* const invlist, const IV index)
7854 /* Caches <index> for later retrieval */
7856 PERL_ARGS_ASSERT_INVLIST_SET_PREVIOUS_INDEX;
7858 assert(index == 0 || index < (int) _invlist_len(invlist));
7860 *get_invlist_previous_index_addr(invlist) = index;
7863 PERL_STATIC_INLINE UV
7864 S_invlist_max(pTHX_ SV* const invlist)
7866 /* Returns the maximum number of elements storable in the inversion list's
7867 * array, without having to realloc() */
7869 PERL_ARGS_ASSERT_INVLIST_MAX;
7871 assert(SvTYPE(invlist) == SVt_INVLIST);
7873 /* Assumes worst case, in which the 0 element is not counted in the
7874 * inversion list, so subtracts 1 for that */
7875 return SvLEN(invlist) == 0 /* This happens under _new_invlist_C_array */
7876 ? FROM_INTERNAL_SIZE(SvCUR(invlist)) - 1
7877 : FROM_INTERNAL_SIZE(SvLEN(invlist)) - 1;
7880 #ifndef PERL_IN_XSUB_RE
7882 Perl__new_invlist(pTHX_ IV initial_size)
7885 /* Return a pointer to a newly constructed inversion list, with enough
7886 * space to store 'initial_size' elements. If that number is negative, a
7887 * system default is used instead */
7891 if (initial_size < 0) {
7895 /* Allocate the initial space */
7896 new_list = newSV_type(SVt_INVLIST);
7898 /* First 1 is in case the zero element isn't in the list; second 1 is for
7900 SvGROW(new_list, TO_INTERNAL_SIZE(initial_size + 1) + 1);
7901 invlist_set_len(new_list, 0, 0);
7903 /* Force iterinit() to be used to get iteration to work */
7904 *get_invlist_iter_addr(new_list) = (STRLEN) UV_MAX;
7906 *get_invlist_previous_index_addr(new_list) = 0;
7912 Perl__new_invlist_C_array(pTHX_ const UV* const list)
7914 /* Return a pointer to a newly constructed inversion list, initialized to
7915 * point to <list>, which has to be in the exact correct inversion list
7916 * form, including internal fields. Thus this is a dangerous routine that
7917 * should not be used in the wrong hands. The passed in 'list' contains
7918 * several header fields at the beginning that are not part of the
7919 * inversion list body proper */
7921 const STRLEN length = (STRLEN) list[0];
7922 const UV version_id = list[1];
7923 const bool offset = cBOOL(list[2]);
7924 #define HEADER_LENGTH 3
7925 /* If any of the above changes in any way, you must change HEADER_LENGTH
7926 * (if appropriate) and regenerate INVLIST_VERSION_ID by running
7927 * perl -E 'say int(rand 2**31-1)'
7929 #define INVLIST_VERSION_ID 148565664 /* This is a combination of a version and
7930 data structure type, so that one being
7931 passed in can be validated to be an
7932 inversion list of the correct vintage.
7935 SV* invlist = newSV_type(SVt_INVLIST);
7937 PERL_ARGS_ASSERT__NEW_INVLIST_C_ARRAY;
7939 if (version_id != INVLIST_VERSION_ID) {
7940 Perl_croak(aTHX_ "panic: Incorrect version for previously generated inversion list");
7943 /* The generated array passed in includes header elements that aren't part
7944 * of the list proper, so start it just after them */
7945 SvPV_set(invlist, (char *) (list + HEADER_LENGTH));
7947 SvLEN_set(invlist, 0); /* Means we own the contents, and the system
7948 shouldn't touch it */
7950 *(get_invlist_offset_addr(invlist)) = offset;
7952 /* The 'length' passed to us is the physical number of elements in the
7953 * inversion list. But if there is an offset the logical number is one
7955 invlist_set_len(invlist, length - offset, offset);
7957 invlist_set_previous_index(invlist, 0);
7959 /* Initialize the iteration pointer. */
7960 invlist_iterfinish(invlist);
7962 SvREADONLY_on(invlist);
7966 #endif /* ifndef PERL_IN_XSUB_RE */
7969 S_invlist_extend(pTHX_ SV* const invlist, const UV new_max)
7971 /* Grow the maximum size of an inversion list */
7973 PERL_ARGS_ASSERT_INVLIST_EXTEND;
7975 assert(SvTYPE(invlist) == SVt_INVLIST);
7977 /* Add one to account for the zero element at the beginning which may not
7978 * be counted by the calling parameters */
7979 SvGROW((SV *)invlist, TO_INTERNAL_SIZE(new_max + 1));
7982 PERL_STATIC_INLINE void
7983 S_invlist_trim(pTHX_ SV* const invlist)
7985 PERL_ARGS_ASSERT_INVLIST_TRIM;
7987 assert(SvTYPE(invlist) == SVt_INVLIST);
7989 /* Change the length of the inversion list to how many entries it currently
7991 SvPV_shrink_to_cur((SV *) invlist);
7995 S__append_range_to_invlist(pTHX_ SV* const invlist,
7996 const UV start, const UV end)
7998 /* Subject to change or removal. Append the range from 'start' to 'end' at
7999 * the end of the inversion list. The range must be above any existing
8003 UV max = invlist_max(invlist);
8004 UV len = _invlist_len(invlist);
8007 PERL_ARGS_ASSERT__APPEND_RANGE_TO_INVLIST;
8009 if (len == 0) { /* Empty lists must be initialized */
8010 offset = start != 0;
8011 array = _invlist_array_init(invlist, ! offset);
8014 /* Here, the existing list is non-empty. The current max entry in the
8015 * list is generally the first value not in the set, except when the
8016 * set extends to the end of permissible values, in which case it is
8017 * the first entry in that final set, and so this call is an attempt to
8018 * append out-of-order */
8020 UV final_element = len - 1;
8021 array = invlist_array(invlist);
8022 if (array[final_element] > start
8023 || ELEMENT_RANGE_MATCHES_INVLIST(final_element))
8025 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",
8026 array[final_element], start,
8027 ELEMENT_RANGE_MATCHES_INVLIST(final_element) ? 't' : 'f');
8030 /* Here, it is a legal append. If the new range begins with the first
8031 * value not in the set, it is extending the set, so the new first
8032 * value not in the set is one greater than the newly extended range.
8034 offset = *get_invlist_offset_addr(invlist);
8035 if (array[final_element] == start) {
8036 if (end != UV_MAX) {
8037 array[final_element] = end + 1;
8040 /* But if the end is the maximum representable on the machine,
8041 * just let the range that this would extend to have no end */
8042 invlist_set_len(invlist, len - 1, offset);
8048 /* Here the new range doesn't extend any existing set. Add it */
8050 len += 2; /* Includes an element each for the start and end of range */
8052 /* If wll overflow the existing space, extend, which may cause the array to
8055 invlist_extend(invlist, len);
8057 /* Have to set len here to avoid assert failure in invlist_array() */
8058 invlist_set_len(invlist, len, offset);
8060 array = invlist_array(invlist);
8063 invlist_set_len(invlist, len, offset);
8066 /* The next item on the list starts the range, the one after that is
8067 * one past the new range. */
8068 array[len - 2] = start;
8069 if (end != UV_MAX) {
8070 array[len - 1] = end + 1;
8073 /* But if the end is the maximum representable on the machine, just let
8074 * the range have no end */
8075 invlist_set_len(invlist, len - 1, offset);
8079 #ifndef PERL_IN_XSUB_RE
8082 Perl__invlist_search(pTHX_ SV* const invlist, const UV cp)
8084 /* Searches the inversion list for the entry that contains the input code
8085 * point <cp>. If <cp> is not in the list, -1 is returned. Otherwise, the
8086 * return value is the index into the list's array of the range that
8091 IV high = _invlist_len(invlist);
8092 const IV highest_element = high - 1;
8095 PERL_ARGS_ASSERT__INVLIST_SEARCH;
8097 /* If list is empty, return failure. */
8102 /* (We can't get the array unless we know the list is non-empty) */
8103 array = invlist_array(invlist);
8105 mid = invlist_previous_index(invlist);
8106 assert(mid >=0 && mid <= highest_element);
8108 /* <mid> contains the cache of the result of the previous call to this
8109 * function (0 the first time). See if this call is for the same result,
8110 * or if it is for mid-1. This is under the theory that calls to this
8111 * function will often be for related code points that are near each other.
8112 * And benchmarks show that caching gives better results. We also test
8113 * here if the code point is within the bounds of the list. These tests
8114 * replace others that would have had to be made anyway to make sure that
8115 * the array bounds were not exceeded, and these give us extra information
8116 * at the same time */
8117 if (cp >= array[mid]) {
8118 if (cp >= array[highest_element]) {
8119 return highest_element;
8122 /* Here, array[mid] <= cp < array[highest_element]. This means that
8123 * the final element is not the answer, so can exclude it; it also
8124 * means that <mid> is not the final element, so can refer to 'mid + 1'
8126 if (cp < array[mid + 1]) {
8132 else { /* cp < aray[mid] */
8133 if (cp < array[0]) { /* Fail if outside the array */
8137 if (cp >= array[mid - 1]) {
8142 /* Binary search. What we are looking for is <i> such that
8143 * array[i] <= cp < array[i+1]
8144 * The loop below converges on the i+1. Note that there may not be an
8145 * (i+1)th element in the array, and things work nonetheless */
8146 while (low < high) {
8147 mid = (low + high) / 2;
8148 assert(mid <= highest_element);
8149 if (array[mid] <= cp) { /* cp >= array[mid] */
8152 /* We could do this extra test to exit the loop early.
8153 if (cp < array[low]) {
8158 else { /* cp < array[mid] */
8165 invlist_set_previous_index(invlist, high);
8170 Perl__invlist_populate_swatch(pTHX_ SV* const invlist,
8171 const UV start, const UV end, U8* swatch)
8173 /* populates a swatch of a swash the same way swatch_get() does in utf8.c,
8174 * but is used when the swash has an inversion list. This makes this much
8175 * faster, as it uses a binary search instead of a linear one. This is
8176 * intimately tied to that function, and perhaps should be in utf8.c,
8177 * except it is intimately tied to inversion lists as well. It assumes
8178 * that <swatch> is all 0's on input */
8181 const IV len = _invlist_len(invlist);
8185 PERL_ARGS_ASSERT__INVLIST_POPULATE_SWATCH;
8187 if (len == 0) { /* Empty inversion list */
8191 array = invlist_array(invlist);
8193 /* Find which element it is */
8194 i = _invlist_search(invlist, start);
8196 /* We populate from <start> to <end> */
8197 while (current < end) {
8200 /* The inversion list gives the results for every possible code point
8201 * after the first one in the list. Only those ranges whose index is
8202 * even are ones that the inversion list matches. For the odd ones,
8203 * and if the initial code point is not in the list, we have to skip
8204 * forward to the next element */
8205 if (i == -1 || ! ELEMENT_RANGE_MATCHES_INVLIST(i)) {
8207 if (i >= len) { /* Finished if beyond the end of the array */
8211 if (current >= end) { /* Finished if beyond the end of what we
8213 if (LIKELY(end < UV_MAX)) {
8217 /* We get here when the upper bound is the maximum
8218 * representable on the machine, and we are looking for just
8219 * that code point. Have to special case it */
8221 goto join_end_of_list;
8224 assert(current >= start);
8226 /* The current range ends one below the next one, except don't go past
8229 upper = (i < len && array[i] < end) ? array[i] : end;
8231 /* Here we are in a range that matches. Populate a bit in the 3-bit U8
8232 * for each code point in it */
8233 for (; current < upper; current++) {
8234 const STRLEN offset = (STRLEN)(current - start);
8235 swatch[offset >> 3] |= 1 << (offset & 7);
8240 /* Quit if at the end of the list */
8243 /* But first, have to deal with the highest possible code point on
8244 * the platform. The previous code assumes that <end> is one
8245 * beyond where we want to populate, but that is impossible at the
8246 * platform's infinity, so have to handle it specially */
8247 if (UNLIKELY(end == UV_MAX && ELEMENT_RANGE_MATCHES_INVLIST(len-1)))
8249 const STRLEN offset = (STRLEN)(end - start);
8250 swatch[offset >> 3] |= 1 << (offset & 7);
8255 /* Advance to the next range, which will be for code points not in the
8264 Perl__invlist_union_maybe_complement_2nd(pTHX_ SV* const a, SV* const b,
8265 const bool complement_b, SV** output)
8267 /* Take the union of two inversion lists and point <output> to it. *output
8268 * SHOULD BE DEFINED upon input, and if it points to one of the two lists,
8269 * the reference count to that list will be decremented if not already a
8270 * temporary (mortal); otherwise *output will be made correspondingly
8271 * mortal. The first list, <a>, may be NULL, in which case a copy of the
8272 * second list is returned. If <complement_b> is TRUE, the union is taken
8273 * of the complement (inversion) of <b> instead of b itself.
8275 * The basis for this comes from "Unicode Demystified" Chapter 13 by
8276 * Richard Gillam, published by Addison-Wesley, and explained at some
8277 * length there. The preface says to incorporate its examples into your
8278 * code at your own risk.
8280 * The algorithm is like a merge sort.
8282 * XXX A potential performance improvement is to keep track as we go along
8283 * if only one of the inputs contributes to the result, meaning the other
8284 * is a subset of that one. In that case, we can skip the final copy and
8285 * return the larger of the input lists, but then outside code might need
8286 * to keep track of whether to free the input list or not */
8288 const UV* array_a; /* a's array */
8290 UV len_a; /* length of a's array */
8293 SV* u; /* the resulting union */
8297 UV i_a = 0; /* current index into a's array */
8301 /* running count, as explained in the algorithm source book; items are
8302 * stopped accumulating and are output when the count changes to/from 0.
8303 * The count is incremented when we start a range that's in the set, and
8304 * decremented when we start a range that's not in the set. So its range
8305 * is 0 to 2. Only when the count is zero is something not in the set.
8309 PERL_ARGS_ASSERT__INVLIST_UNION_MAYBE_COMPLEMENT_2ND;
8312 /* If either one is empty, the union is the other one */
8313 if (a == NULL || ((len_a = _invlist_len(a)) == 0)) {
8314 bool make_temp = FALSE; /* Should we mortalize the result? */
8318 if (! (make_temp = cBOOL(SvTEMP(a)))) {
8324 *output = invlist_clone(b);
8326 _invlist_invert(*output);
8328 } /* else *output already = b; */
8331 sv_2mortal(*output);
8335 else if ((len_b = _invlist_len(b)) == 0) {
8336 bool make_temp = FALSE;
8338 if (! (make_temp = cBOOL(SvTEMP(b)))) {
8343 /* The complement of an empty list is a list that has everything in it,
8344 * so the union with <a> includes everything too */
8347 if (! (make_temp = cBOOL(SvTEMP(a)))) {
8351 *output = _new_invlist(1);
8352 _append_range_to_invlist(*output, 0, UV_MAX);
8354 else if (*output != a) {
8355 *output = invlist_clone(a);
8357 /* else *output already = a; */
8360 sv_2mortal(*output);
8365 /* Here both lists exist and are non-empty */
8366 array_a = invlist_array(a);
8367 array_b = invlist_array(b);
8369 /* If are to take the union of 'a' with the complement of b, set it
8370 * up so are looking at b's complement. */
8373 /* To complement, we invert: if the first element is 0, remove it. To
8374 * do this, we just pretend the array starts one later */
8375 if (array_b[0] == 0) {
8381 /* But if the first element is not zero, we pretend the list starts
8382 * at the 0 that is always stored immediately before the array. */
8388 /* Size the union for the worst case: that the sets are completely
8390 u = _new_invlist(len_a + len_b);
8392 /* Will contain U+0000 if either component does */
8393 array_u = _invlist_array_init(u, (len_a > 0 && array_a[0] == 0)
8394 || (len_b > 0 && array_b[0] == 0));
8396 /* Go through each list item by item, stopping when exhausted one of
8398 while (i_a < len_a && i_b < len_b) {
8399 UV cp; /* The element to potentially add to the union's array */
8400 bool cp_in_set; /* is it in the the input list's set or not */
8402 /* We need to take one or the other of the two inputs for the union.
8403 * Since we are merging two sorted lists, we take the smaller of the
8404 * next items. In case of a tie, we take the one that is in its set
8405 * first. If we took one not in the set first, it would decrement the
8406 * count, possibly to 0 which would cause it to be output as ending the
8407 * range, and the next time through we would take the same number, and
8408 * output it again as beginning the next range. By doing it the
8409 * opposite way, there is no possibility that the count will be
8410 * momentarily decremented to 0, and thus the two adjoining ranges will
8411 * be seamlessly merged. (In a tie and both are in the set or both not
8412 * in the set, it doesn't matter which we take first.) */
8413 if (array_a[i_a] < array_b[i_b]
8414 || (array_a[i_a] == array_b[i_b]
8415 && ELEMENT_RANGE_MATCHES_INVLIST(i_a)))
8417 cp_in_set = ELEMENT_RANGE_MATCHES_INVLIST(i_a);
8421 cp_in_set = ELEMENT_RANGE_MATCHES_INVLIST(i_b);
8422 cp = array_b[i_b++];
8425 /* Here, have chosen which of the two inputs to look at. Only output
8426 * if the running count changes to/from 0, which marks the
8427 * beginning/end of a range in that's in the set */
8430 array_u[i_u++] = cp;
8437 array_u[i_u++] = cp;
8442 /* Here, we are finished going through at least one of the lists, which
8443 * means there is something remaining in at most one. We check if the list
8444 * that hasn't been exhausted is positioned such that we are in the middle
8445 * of a range in its set or not. (i_a and i_b point to the element beyond
8446 * the one we care about.) If in the set, we decrement 'count'; if 0, there
8447 * is potentially more to output.
8448 * There are four cases:
8449 * 1) Both weren't in their sets, count is 0, and remains 0. What's left
8450 * in the union is entirely from the non-exhausted set.
8451 * 2) Both were in their sets, count is 2. Nothing further should
8452 * be output, as everything that remains will be in the exhausted
8453 * list's set, hence in the union; decrementing to 1 but not 0 insures
8455 * 3) the exhausted was in its set, non-exhausted isn't, count is 1.
8456 * Nothing further should be output because the union includes
8457 * everything from the exhausted set. Not decrementing ensures that.
8458 * 4) the exhausted wasn't in its set, non-exhausted is, count is 1;
8459 * decrementing to 0 insures that we look at the remainder of the
8460 * non-exhausted set */
8461 if ((i_a != len_a && PREV_RANGE_MATCHES_INVLIST(i_a))
8462 || (i_b != len_b && PREV_RANGE_MATCHES_INVLIST(i_b)))
8467 /* The final length is what we've output so far, plus what else is about to
8468 * be output. (If 'count' is non-zero, then the input list we exhausted
8469 * has everything remaining up to the machine's limit in its set, and hence
8470 * in the union, so there will be no further output. */
8473 /* At most one of the subexpressions will be non-zero */
8474 len_u += (len_a - i_a) + (len_b - i_b);
8477 /* Set result to final length, which can change the pointer to array_u, so
8479 if (len_u != _invlist_len(u)) {
8480 invlist_set_len(u, len_u, *get_invlist_offset_addr(u));
8482 array_u = invlist_array(u);
8485 /* When 'count' is 0, the list that was exhausted (if one was shorter than
8486 * the other) ended with everything above it not in its set. That means
8487 * that the remaining part of the union is precisely the same as the
8488 * non-exhausted list, so can just copy it unchanged. (If both list were
8489 * exhausted at the same time, then the operations below will be both 0.)
8492 IV copy_count; /* At most one will have a non-zero copy count */
8493 if ((copy_count = len_a - i_a) > 0) {
8494 Copy(array_a + i_a, array_u + i_u, copy_count, UV);
8496 else if ((copy_count = len_b - i_b) > 0) {
8497 Copy(array_b + i_b, array_u + i_u, copy_count, UV);
8501 /* We may be removing a reference to one of the inputs. If so, the output
8502 * is made mortal if the input was. (Mortal SVs shouldn't have their ref
8503 * count decremented) */
8504 if (a == *output || b == *output) {
8505 assert(! invlist_is_iterating(*output));
8506 if ((SvTEMP(*output))) {
8510 SvREFCNT_dec_NN(*output);
8520 Perl__invlist_intersection_maybe_complement_2nd(pTHX_ SV* const a, SV* const b,
8521 const bool complement_b, SV** i)
8523 /* Take the intersection of two inversion lists and point <i> to it. *i
8524 * SHOULD BE DEFINED upon input, and if it points to one of the two lists,
8525 * the reference count to that list will be decremented if not already a
8526 * temporary (mortal); otherwise *i will be made correspondingly mortal.
8527 * The first list, <a>, may be NULL, in which case an empty list is
8528 * returned. If <complement_b> is TRUE, the result will be the
8529 * intersection of <a> and the complement (or inversion) of <b> instead of
8532 * The basis for this comes from "Unicode Demystified" Chapter 13 by
8533 * Richard Gillam, published by Addison-Wesley, and explained at some
8534 * length there. The preface says to incorporate its examples into your
8535 * code at your own risk. In fact, it had bugs
8537 * The algorithm is like a merge sort, and is essentially the same as the
8541 const UV* array_a; /* a's array */
8543 UV len_a; /* length of a's array */
8546 SV* r; /* the resulting intersection */
8550 UV i_a = 0; /* current index into a's array */
8554 /* running count, as explained in the algorithm source book; items are
8555 * stopped accumulating and are output when the count changes to/from 2.
8556 * The count is incremented when we start a range that's in the set, and
8557 * decremented when we start a range that's not in the set. So its range
8558 * is 0 to 2. Only when the count is 2 is something in the intersection.
8562 PERL_ARGS_ASSERT__INVLIST_INTERSECTION_MAYBE_COMPLEMENT_2ND;
8565 /* Special case if either one is empty */
8566 len_a = (a == NULL) ? 0 : _invlist_len(a);
8567 if ((len_a == 0) || ((len_b = _invlist_len(b)) == 0)) {
8568 bool make_temp = FALSE;
8570 if (len_a != 0 && complement_b) {
8572 /* Here, 'a' is not empty, therefore from the above 'if', 'b' must
8573 * be empty. Here, also we are using 'b's complement, which hence
8574 * must be every possible code point. Thus the intersection is
8578 if (! (make_temp = cBOOL(SvTEMP(b)))) {
8583 *i = invlist_clone(a);
8585 /* else *i is already 'a' */
8593 /* Here, 'a' or 'b' is empty and not using the complement of 'b'. The
8594 * intersection must be empty */
8596 if (! (make_temp = cBOOL(SvTEMP(a)))) {
8601 if (! (make_temp = cBOOL(SvTEMP(b)))) {
8605 *i = _new_invlist(0);
8613 /* Here both lists exist and are non-empty */
8614 array_a = invlist_array(a);
8615 array_b = invlist_array(b);
8617 /* If are to take the intersection of 'a' with the complement of b, set it
8618 * up so are looking at b's complement. */
8621 /* To complement, we invert: if the first element is 0, remove it. To
8622 * do this, we just pretend the array starts one later */
8623 if (array_b[0] == 0) {
8629 /* But if the first element is not zero, we pretend the list starts
8630 * at the 0 that is always stored immediately before the array. */
8636 /* Size the intersection for the worst case: that the intersection ends up
8637 * fragmenting everything to be completely disjoint */
8638 r= _new_invlist(len_a + len_b);
8640 /* Will contain U+0000 iff both components do */
8641 array_r = _invlist_array_init(r, len_a > 0 && array_a[0] == 0
8642 && len_b > 0 && array_b[0] == 0);
8644 /* Go through each list item by item, stopping when exhausted one of
8646 while (i_a < len_a && i_b < len_b) {
8647 UV cp; /* The element to potentially add to the intersection's
8649 bool cp_in_set; /* Is it in the input list's set or not */
8651 /* We need to take one or the other of the two inputs for the
8652 * intersection. Since we are merging two sorted lists, we take the
8653 * smaller of the next items. In case of a tie, we take the one that
8654 * is not in its set first (a difference from the union algorithm). If
8655 * we took one in the set first, it would increment the count, possibly
8656 * to 2 which would cause it to be output as starting a range in the
8657 * intersection, and the next time through we would take that same
8658 * number, and output it again as ending the set. By doing it the
8659 * opposite of this, there is no possibility that the count will be
8660 * momentarily incremented to 2. (In a tie and both are in the set or
8661 * both not in the set, it doesn't matter which we take first.) */
8662 if (array_a[i_a] < array_b[i_b]
8663 || (array_a[i_a] == array_b[i_b]
8664 && ! ELEMENT_RANGE_MATCHES_INVLIST(i_a)))
8666 cp_in_set = ELEMENT_RANGE_MATCHES_INVLIST(i_a);
8670 cp_in_set = ELEMENT_RANGE_MATCHES_INVLIST(i_b);
8674 /* Here, have chosen which of the two inputs to look at. Only output
8675 * if the running count changes to/from 2, which marks the
8676 * beginning/end of a range that's in the intersection */
8680 array_r[i_r++] = cp;
8685 array_r[i_r++] = cp;
8691 /* Here, we are finished going through at least one of the lists, which
8692 * means there is something remaining in at most one. We check if the list
8693 * that has been exhausted is positioned such that we are in the middle
8694 * of a range in its set or not. (i_a and i_b point to elements 1 beyond
8695 * the ones we care about.) There are four cases:
8696 * 1) Both weren't in their sets, count is 0, and remains 0. There's
8697 * nothing left in the intersection.
8698 * 2) Both were in their sets, count is 2 and perhaps is incremented to
8699 * above 2. What should be output is exactly that which is in the
8700 * non-exhausted set, as everything it has is also in the intersection
8701 * set, and everything it doesn't have can't be in the intersection
8702 * 3) The exhausted was in its set, non-exhausted isn't, count is 1, and
8703 * gets incremented to 2. Like the previous case, the intersection is
8704 * everything that remains in the non-exhausted set.
8705 * 4) the exhausted wasn't in its set, non-exhausted is, count is 1, and
8706 * remains 1. And the intersection has nothing more. */
8707 if ((i_a == len_a && PREV_RANGE_MATCHES_INVLIST(i_a))
8708 || (i_b == len_b && PREV_RANGE_MATCHES_INVLIST(i_b)))
8713 /* The final length is what we've output so far plus what else is in the
8714 * intersection. At most one of the subexpressions below will be non-zero
8718 len_r += (len_a - i_a) + (len_b - i_b);
8721 /* Set result to final length, which can change the pointer to array_r, so
8723 if (len_r != _invlist_len(r)) {
8724 invlist_set_len(r, len_r, *get_invlist_offset_addr(r));
8726 array_r = invlist_array(r);
8729 /* Finish outputting any remaining */
8730 if (count >= 2) { /* At most one will have a non-zero copy count */
8732 if ((copy_count = len_a - i_a) > 0) {
8733 Copy(array_a + i_a, array_r + i_r, copy_count, UV);
8735 else if ((copy_count = len_b - i_b) > 0) {
8736 Copy(array_b + i_b, array_r + i_r, copy_count, UV);
8740 /* We may be removing a reference to one of the inputs. If so, the output
8741 * is made mortal if the input was. (Mortal SVs shouldn't have their ref
8742 * count decremented) */
8743 if (a == *i || b == *i) {
8744 assert(! invlist_is_iterating(*i));
8749 SvREFCNT_dec_NN(*i);
8759 Perl__add_range_to_invlist(pTHX_ SV* invlist, const UV start, const UV end)
8761 /* Add the range from 'start' to 'end' inclusive to the inversion list's
8762 * set. A pointer to the inversion list is returned. This may actually be
8763 * a new list, in which case the passed in one has been destroyed. The
8764 * passed in inversion list can be NULL, in which case a new one is created
8765 * with just the one range in it */
8770 if (invlist == NULL) {
8771 invlist = _new_invlist(2);
8775 len = _invlist_len(invlist);
8778 /* If comes after the final entry actually in the list, can just append it
8781 || (! ELEMENT_RANGE_MATCHES_INVLIST(len - 1)
8782 && start >= invlist_array(invlist)[len - 1]))
8784 _append_range_to_invlist(invlist, start, end);
8788 /* Here, can't just append things, create and return a new inversion list
8789 * which is the union of this range and the existing inversion list */
8790 range_invlist = _new_invlist(2);
8791 _append_range_to_invlist(range_invlist, start, end);
8793 _invlist_union(invlist, range_invlist, &invlist);
8795 /* The temporary can be freed */
8796 SvREFCNT_dec_NN(range_invlist);
8802 Perl__setup_canned_invlist(pTHX_ const STRLEN size, const UV element0,
8803 UV** other_elements_ptr)
8805 /* Create and return an inversion list whose contents are to be populated
8806 * by the caller. The caller gives the number of elements (in 'size') and
8807 * the very first element ('element0'). This function will set
8808 * '*other_elements_ptr' to an array of UVs, where the remaining elements
8811 * Obviously there is some trust involved that the caller will properly
8812 * fill in the other elements of the array.
8814 * (The first element needs to be passed in, as the underlying code does
8815 * things differently depending on whether it is zero or non-zero) */
8817 SV* invlist = _new_invlist(size);
8820 PERL_ARGS_ASSERT__SETUP_CANNED_INVLIST;
8822 _append_range_to_invlist(invlist, element0, element0);
8823 offset = *get_invlist_offset_addr(invlist);
8825 invlist_set_len(invlist, size, offset);
8826 *other_elements_ptr = invlist_array(invlist) + 1;
8832 PERL_STATIC_INLINE SV*
8833 S_add_cp_to_invlist(pTHX_ SV* invlist, const UV cp) {
8834 return _add_range_to_invlist(invlist, cp, cp);
8837 #ifndef PERL_IN_XSUB_RE
8839 Perl__invlist_invert(pTHX_ SV* const invlist)
8841 /* Complement the input inversion list. This adds a 0 if the list didn't
8842 * have a zero; removes it otherwise. As described above, the data
8843 * structure is set up so that this is very efficient */
8845 PERL_ARGS_ASSERT__INVLIST_INVERT;
8847 assert(! invlist_is_iterating(invlist));
8849 /* The inverse of matching nothing is matching everything */
8850 if (_invlist_len(invlist) == 0) {
8851 _append_range_to_invlist(invlist, 0, UV_MAX);
8855 *get_invlist_offset_addr(invlist) = ! *get_invlist_offset_addr(invlist);
8860 PERL_STATIC_INLINE SV*
8861 S_invlist_clone(pTHX_ SV* const invlist)
8864 /* Return a new inversion list that is a copy of the input one, which is
8865 * unchanged. The new list will not be mortal even if the old one was. */
8867 /* Need to allocate extra space to accommodate Perl's addition of a
8868 * trailing NUL to SvPV's, since it thinks they are always strings */
8869 SV* new_invlist = _new_invlist(_invlist_len(invlist) + 1);
8870 STRLEN physical_length = SvCUR(invlist);
8871 bool offset = *(get_invlist_offset_addr(invlist));
8873 PERL_ARGS_ASSERT_INVLIST_CLONE;
8875 *(get_invlist_offset_addr(new_invlist)) = offset;
8876 invlist_set_len(new_invlist, _invlist_len(invlist), offset);
8877 Copy(SvPVX(invlist), SvPVX(new_invlist), physical_length, char);
8882 PERL_STATIC_INLINE STRLEN*
8883 S_get_invlist_iter_addr(pTHX_ SV* invlist)
8885 /* Return the address of the UV that contains the current iteration
8888 PERL_ARGS_ASSERT_GET_INVLIST_ITER_ADDR;
8890 assert(SvTYPE(invlist) == SVt_INVLIST);
8892 return &(((XINVLIST*) SvANY(invlist))->iterator);
8895 PERL_STATIC_INLINE void
8896 S_invlist_iterinit(pTHX_ SV* invlist) /* Initialize iterator for invlist */
8898 PERL_ARGS_ASSERT_INVLIST_ITERINIT;
8900 *get_invlist_iter_addr(invlist) = 0;
8903 PERL_STATIC_INLINE void
8904 S_invlist_iterfinish(pTHX_ SV* invlist)
8906 /* Terminate iterator for invlist. This is to catch development errors.
8907 * Any iteration that is interrupted before completed should call this
8908 * function. Functions that add code points anywhere else but to the end
8909 * of an inversion list assert that they are not in the middle of an
8910 * iteration. If they were, the addition would make the iteration
8911 * problematical: if the iteration hadn't reached the place where things
8912 * were being added, it would be ok */
8914 PERL_ARGS_ASSERT_INVLIST_ITERFINISH;
8916 *get_invlist_iter_addr(invlist) = (STRLEN) UV_MAX;
8920 S_invlist_iternext(pTHX_ SV* invlist, UV* start, UV* end)
8922 /* An C<invlist_iterinit> call on <invlist> must be used to set this up.
8923 * This call sets in <*start> and <*end>, the next range in <invlist>.
8924 * Returns <TRUE> if successful and the next call will return the next
8925 * range; <FALSE> if was already at the end of the list. If the latter,
8926 * <*start> and <*end> are unchanged, and the next call to this function
8927 * will start over at the beginning of the list */
8929 STRLEN* pos = get_invlist_iter_addr(invlist);
8930 UV len = _invlist_len(invlist);
8933 PERL_ARGS_ASSERT_INVLIST_ITERNEXT;
8936 *pos = (STRLEN) UV_MAX; /* Force iterinit() to be required next time */
8940 array = invlist_array(invlist);
8942 *start = array[(*pos)++];
8948 *end = array[(*pos)++] - 1;
8954 PERL_STATIC_INLINE bool
8955 S_invlist_is_iterating(pTHX_ SV* const invlist)
8957 PERL_ARGS_ASSERT_INVLIST_IS_ITERATING;
8959 return *(get_invlist_iter_addr(invlist)) < (STRLEN) UV_MAX;
8962 PERL_STATIC_INLINE UV
8963 S_invlist_highest(pTHX_ SV* const invlist)
8965 /* Returns the highest code point that matches an inversion list. This API
8966 * has an ambiguity, as it returns 0 under either the highest is actually
8967 * 0, or if the list is empty. If this distinction matters to you, check
8968 * for emptiness before calling this function */
8970 UV len = _invlist_len(invlist);
8973 PERL_ARGS_ASSERT_INVLIST_HIGHEST;
8979 array = invlist_array(invlist);
8981 /* The last element in the array in the inversion list always starts a
8982 * range that goes to infinity. That range may be for code points that are
8983 * matched in the inversion list, or it may be for ones that aren't
8984 * matched. In the latter case, the highest code point in the set is one
8985 * less than the beginning of this range; otherwise it is the final element
8986 * of this range: infinity */
8987 return (ELEMENT_RANGE_MATCHES_INVLIST(len - 1))
8989 : array[len - 1] - 1;
8992 #ifndef PERL_IN_XSUB_RE
8994 Perl__invlist_contents(pTHX_ SV* const invlist)
8996 /* Get the contents of an inversion list into a string SV so that they can
8997 * be printed out. It uses the format traditionally done for debug tracing
9001 SV* output = newSVpvs("\n");
9003 PERL_ARGS_ASSERT__INVLIST_CONTENTS;
9005 assert(! invlist_is_iterating(invlist));
9007 invlist_iterinit(invlist);
9008 while (invlist_iternext(invlist, &start, &end)) {
9009 if (end == UV_MAX) {
9010 Perl_sv_catpvf(aTHX_ output, "%04"UVXf"\tINFINITY\n", start);
9012 else if (end != start) {
9013 Perl_sv_catpvf(aTHX_ output, "%04"UVXf"\t%04"UVXf"\n",
9017 Perl_sv_catpvf(aTHX_ output, "%04"UVXf"\n", start);
9025 #ifndef PERL_IN_XSUB_RE
9027 Perl__invlist_dump(pTHX_ PerlIO *file, I32 level,
9028 const char * const indent, SV* const invlist)
9030 /* Designed to be called only by do_sv_dump(). Dumps out the ranges of the
9031 * inversion list 'invlist' to 'file' at 'level' Each line is prefixed by
9032 * the string 'indent'. The output looks like this:
9033 [0] 0x000A .. 0x000D
9035 [4] 0x2028 .. 0x2029
9036 [6] 0x3104 .. INFINITY
9037 * This means that the first range of code points matched by the list are
9038 * 0xA through 0xD; the second range contains only the single code point
9039 * 0x85, etc. An inversion list is an array of UVs. Two array elements
9040 * are used to define each range (except if the final range extends to
9041 * infinity, only a single element is needed). The array index of the
9042 * first element for the corresponding range is given in brackets. */
9047 PERL_ARGS_ASSERT__INVLIST_DUMP;
9049 if (invlist_is_iterating(invlist)) {
9050 Perl_dump_indent(aTHX_ level, file,
9051 "%sCan't dump inversion list because is in middle of iterating\n",
9056 invlist_iterinit(invlist);
9057 while (invlist_iternext(invlist, &start, &end)) {
9058 if (end == UV_MAX) {
9059 Perl_dump_indent(aTHX_ level, file,
9060 "%s[%"UVuf"] 0x%04"UVXf" .. INFINITY\n",
9061 indent, (UV)count, start);
9063 else if (end != start) {
9064 Perl_dump_indent(aTHX_ level, file,
9065 "%s[%"UVuf"] 0x%04"UVXf" .. 0x%04"UVXf"\n",
9066 indent, (UV)count, start, end);
9069 Perl_dump_indent(aTHX_ level, file, "%s[%"UVuf"] 0x%04"UVXf"\n",
9070 indent, (UV)count, start);
9077 #ifdef PERL_ARGS_ASSERT__INVLISTEQ
9079 S__invlistEQ(pTHX_ SV* const a, SV* const b, const bool complement_b)
9081 /* Return a boolean as to if the two passed in inversion lists are
9082 * identical. The final argument, if TRUE, says to take the complement of
9083 * the second inversion list before doing the comparison */
9085 const UV* array_a = invlist_array(a);
9086 const UV* array_b = invlist_array(b);
9087 UV len_a = _invlist_len(a);
9088 UV len_b = _invlist_len(b);
9090 UV i = 0; /* current index into the arrays */
9091 bool retval = TRUE; /* Assume are identical until proven otherwise */
9093 PERL_ARGS_ASSERT__INVLISTEQ;
9095 /* If are to compare 'a' with the complement of b, set it
9096 * up so are looking at b's complement. */
9099 /* The complement of nothing is everything, so <a> would have to have
9100 * just one element, starting at zero (ending at infinity) */
9102 return (len_a == 1 && array_a[0] == 0);
9104 else if (array_b[0] == 0) {
9106 /* Otherwise, to complement, we invert. Here, the first element is
9107 * 0, just remove it. To do this, we just pretend the array starts
9115 /* But if the first element is not zero, we pretend the list starts
9116 * at the 0 that is always stored immediately before the array. */
9122 /* Make sure that the lengths are the same, as well as the final element
9123 * before looping through the remainder. (Thus we test the length, final,
9124 * and first elements right off the bat) */
9125 if (len_a != len_b || array_a[len_a-1] != array_b[len_a-1]) {
9128 else for (i = 0; i < len_a - 1; i++) {
9129 if (array_a[i] != array_b[i]) {
9139 #undef HEADER_LENGTH
9140 #undef TO_INTERNAL_SIZE
9141 #undef FROM_INTERNAL_SIZE
9142 #undef INVLIST_VERSION_ID
9144 /* End of inversion list object */
9147 S_parse_lparen_question_flags(pTHX_ RExC_state_t *pRExC_state)
9149 /* This parses the flags that are in either the '(?foo)' or '(?foo:bar)'
9150 * constructs, and updates RExC_flags with them. On input, RExC_parse
9151 * should point to the first flag; it is updated on output to point to the
9152 * final ')' or ':'. There needs to be at least one flag, or this will
9155 /* for (?g), (?gc), and (?o) warnings; warning
9156 about (?c) will warn about (?g) -- japhy */
9158 #define WASTED_O 0x01
9159 #define WASTED_G 0x02
9160 #define WASTED_C 0x04
9161 #define WASTED_GC (WASTED_G|WASTED_C)
9162 I32 wastedflags = 0x00;
9163 U32 posflags = 0, negflags = 0;
9164 U32 *flagsp = &posflags;
9165 char has_charset_modifier = '\0';
9167 bool has_use_defaults = FALSE;
9168 const char* const seqstart = RExC_parse - 1; /* Point to the '?' */
9170 PERL_ARGS_ASSERT_PARSE_LPAREN_QUESTION_FLAGS;
9172 /* '^' as an initial flag sets certain defaults */
9173 if (UCHARAT(RExC_parse) == '^') {
9175 has_use_defaults = TRUE;
9176 STD_PMMOD_FLAGS_CLEAR(&RExC_flags);
9177 set_regex_charset(&RExC_flags, (RExC_utf8 || RExC_uni_semantics)
9178 ? REGEX_UNICODE_CHARSET
9179 : REGEX_DEPENDS_CHARSET);
9182 cs = get_regex_charset(RExC_flags);
9183 if (cs == REGEX_DEPENDS_CHARSET
9184 && (RExC_utf8 || RExC_uni_semantics))
9186 cs = REGEX_UNICODE_CHARSET;
9189 while (*RExC_parse) {
9190 /* && strchr("iogcmsx", *RExC_parse) */
9191 /* (?g), (?gc) and (?o) are useless here
9192 and must be globally applied -- japhy */
9193 switch (*RExC_parse) {
9195 /* Code for the imsx flags */
9196 CASE_STD_PMMOD_FLAGS_PARSE_SET(flagsp);
9198 case LOCALE_PAT_MOD:
9199 if (has_charset_modifier) {
9200 goto excess_modifier;
9202 else if (flagsp == &negflags) {
9205 cs = REGEX_LOCALE_CHARSET;
9206 has_charset_modifier = LOCALE_PAT_MOD;
9208 case UNICODE_PAT_MOD:
9209 if (has_charset_modifier) {
9210 goto excess_modifier;
9212 else if (flagsp == &negflags) {
9215 cs = REGEX_UNICODE_CHARSET;
9216 has_charset_modifier = UNICODE_PAT_MOD;
9218 case ASCII_RESTRICT_PAT_MOD:
9219 if (flagsp == &negflags) {
9222 if (has_charset_modifier) {
9223 if (cs != REGEX_ASCII_RESTRICTED_CHARSET) {
9224 goto excess_modifier;
9226 /* Doubled modifier implies more restricted */
9227 cs = REGEX_ASCII_MORE_RESTRICTED_CHARSET;
9230 cs = REGEX_ASCII_RESTRICTED_CHARSET;
9232 has_charset_modifier = ASCII_RESTRICT_PAT_MOD;
9234 case DEPENDS_PAT_MOD:
9235 if (has_use_defaults) {
9236 goto fail_modifiers;
9238 else if (flagsp == &negflags) {
9241 else if (has_charset_modifier) {
9242 goto excess_modifier;
9245 /* The dual charset means unicode semantics if the
9246 * pattern (or target, not known until runtime) are
9247 * utf8, or something in the pattern indicates unicode
9249 cs = (RExC_utf8 || RExC_uni_semantics)
9250 ? REGEX_UNICODE_CHARSET
9251 : REGEX_DEPENDS_CHARSET;
9252 has_charset_modifier = DEPENDS_PAT_MOD;
9256 if (has_charset_modifier == ASCII_RESTRICT_PAT_MOD) {
9257 vFAIL2("Regexp modifier \"%c\" may appear a maximum of twice", ASCII_RESTRICT_PAT_MOD);
9259 else if (has_charset_modifier == *(RExC_parse - 1)) {
9260 vFAIL2("Regexp modifier \"%c\" may not appear twice",
9264 vFAIL3("Regexp modifiers \"%c\" and \"%c\" are mutually exclusive", has_charset_modifier, *(RExC_parse - 1));
9269 vFAIL2("Regexp modifier \"%c\" may not appear after the \"-\"",
9272 case ONCE_PAT_MOD: /* 'o' */
9273 case GLOBAL_PAT_MOD: /* 'g' */
9274 if (SIZE_ONLY && ckWARN(WARN_REGEXP)) {
9275 const I32 wflagbit = *RExC_parse == 'o'
9278 if (! (wastedflags & wflagbit) ) {
9279 wastedflags |= wflagbit;
9280 /* diag_listed_as: Useless (?-%s) - don't use /%s modifier in regex; marked by <-- HERE in m/%s/ */
9283 "Useless (%s%c) - %suse /%c modifier",
9284 flagsp == &negflags ? "?-" : "?",
9286 flagsp == &negflags ? "don't " : "",
9293 case CONTINUE_PAT_MOD: /* 'c' */
9294 if (SIZE_ONLY && ckWARN(WARN_REGEXP)) {
9295 if (! (wastedflags & WASTED_C) ) {
9296 wastedflags |= WASTED_GC;
9297 /* diag_listed_as: Useless (?-%s) - don't use /%s modifier in regex; marked by <-- HERE in m/%s/ */
9300 "Useless (%sc) - %suse /gc modifier",
9301 flagsp == &negflags ? "?-" : "?",
9302 flagsp == &negflags ? "don't " : ""
9307 case KEEPCOPY_PAT_MOD: /* 'p' */
9308 if (flagsp == &negflags) {
9310 ckWARNreg(RExC_parse + 1,"Useless use of (?-p)");
9312 *flagsp |= RXf_PMf_KEEPCOPY;
9316 /* A flag is a default iff it is following a minus, so
9317 * if there is a minus, it means will be trying to
9318 * re-specify a default which is an error */
9319 if (has_use_defaults || flagsp == &negflags) {
9320 goto fail_modifiers;
9323 wastedflags = 0; /* reset so (?g-c) warns twice */
9327 RExC_flags |= posflags;
9328 RExC_flags &= ~negflags;
9329 set_regex_charset(&RExC_flags, cs);
9330 if (RExC_flags & RXf_PMf_FOLD) {
9331 RExC_contains_i = 1;
9337 RExC_parse += UTF ? UTF8SKIP(RExC_parse) : 1;
9338 /* diag_listed_as: Sequence (?%s...) not recognized in regex; marked by <-- HERE in m/%s/ */
9339 vFAIL2utf8f("Sequence (%"UTF8f"...) not recognized",
9340 UTF8fARG(UTF, RExC_parse-seqstart, seqstart));
9349 - reg - regular expression, i.e. main body or parenthesized thing
9351 * Caller must absorb opening parenthesis.
9353 * Combining parenthesis handling with the base level of regular expression
9354 * is a trifle forced, but the need to tie the tails of the branches to what
9355 * follows makes it hard to avoid.
9357 #define REGTAIL(x,y,z) regtail((x),(y),(z),depth+1)
9359 #define REGTAIL_STUDY(x,y,z) regtail_study((x),(y),(z),depth+1)
9361 #define REGTAIL_STUDY(x,y,z) regtail((x),(y),(z),depth+1)
9364 /* Returns NULL, setting *flagp to TRYAGAIN at the end of (?) that only sets
9365 flags. Returns NULL, setting *flagp to RESTART_UTF8 if the sizing scan
9366 needs to be restarted.
9367 Otherwise would only return NULL if regbranch() returns NULL, which
9370 S_reg(pTHX_ RExC_state_t *pRExC_state, I32 paren, I32 *flagp,U32 depth)
9371 /* paren: Parenthesized? 0=top; 1,2=inside '(': changed to letter.
9372 * 2 is like 1, but indicates that nextchar() has been called to advance
9373 * RExC_parse beyond the '('. Things like '(?' are indivisible tokens, and
9374 * this flag alerts us to the need to check for that */
9377 regnode *ret; /* Will be the head of the group. */
9380 regnode *ender = NULL;
9383 U32 oregflags = RExC_flags;
9384 bool have_branch = 0;
9386 I32 freeze_paren = 0;
9387 I32 after_freeze = 0;
9389 char * parse_start = RExC_parse; /* MJD */
9390 char * const oregcomp_parse = RExC_parse;
9392 GET_RE_DEBUG_FLAGS_DECL;
9394 PERL_ARGS_ASSERT_REG;
9395 DEBUG_PARSE("reg ");
9397 *flagp = 0; /* Tentatively. */
9400 /* Make an OPEN node, if parenthesized. */
9403 /* Under /x, space and comments can be gobbled up between the '(' and
9404 * here (if paren ==2). The forms '(*VERB' and '(?...' disallow such
9405 * intervening space, as the sequence is a token, and a token should be
9407 bool has_intervening_patws = paren == 2 && *(RExC_parse - 1) != '(';
9409 if ( *RExC_parse == '*') { /* (*VERB:ARG) */
9410 char *start_verb = RExC_parse;
9411 STRLEN verb_len = 0;
9412 char *start_arg = NULL;
9413 unsigned char op = 0;
9415 int internal_argval = 0; /* internal_argval is only useful if
9418 if (has_intervening_patws && SIZE_ONLY) {
9419 ckWARNregdep(RExC_parse + 1, "In '(*VERB...)', splitting the initial '(*' is deprecated");
9421 while ( *RExC_parse && *RExC_parse != ')' ) {
9422 if ( *RExC_parse == ':' ) {
9423 start_arg = RExC_parse + 1;
9429 verb_len = RExC_parse - start_verb;
9432 while ( *RExC_parse && *RExC_parse != ')' )
9434 if ( *RExC_parse != ')' )
9435 vFAIL("Unterminated verb pattern argument");
9436 if ( RExC_parse == start_arg )
9439 if ( *RExC_parse != ')' )
9440 vFAIL("Unterminated verb pattern");
9443 switch ( *start_verb ) {
9444 case 'A': /* (*ACCEPT) */
9445 if ( memEQs(start_verb,verb_len,"ACCEPT") ) {
9447 internal_argval = RExC_nestroot;
9450 case 'C': /* (*COMMIT) */
9451 if ( memEQs(start_verb,verb_len,"COMMIT") )
9454 case 'F': /* (*FAIL) */
9455 if ( verb_len==1 || memEQs(start_verb,verb_len,"FAIL") ) {
9460 case ':': /* (*:NAME) */
9461 case 'M': /* (*MARK:NAME) */
9462 if ( verb_len==0 || memEQs(start_verb,verb_len,"MARK") ) {
9467 case 'P': /* (*PRUNE) */
9468 if ( memEQs(start_verb,verb_len,"PRUNE") )
9471 case 'S': /* (*SKIP) */
9472 if ( memEQs(start_verb,verb_len,"SKIP") )
9475 case 'T': /* (*THEN) */
9476 /* [19:06] <TimToady> :: is then */
9477 if ( memEQs(start_verb,verb_len,"THEN") ) {
9479 RExC_seen |= REG_CUTGROUP_SEEN;
9484 RExC_parse += UTF ? UTF8SKIP(RExC_parse) : 1;
9486 "Unknown verb pattern '%"UTF8f"'",
9487 UTF8fARG(UTF, verb_len, start_verb));
9490 if ( start_arg && internal_argval ) {
9491 vFAIL3("Verb pattern '%.*s' may not have an argument",
9492 verb_len, start_verb);
9493 } else if ( argok < 0 && !start_arg ) {
9494 vFAIL3("Verb pattern '%.*s' has a mandatory argument",
9495 verb_len, start_verb);
9497 ret = reganode(pRExC_state, op, internal_argval);
9498 if ( ! internal_argval && ! SIZE_ONLY ) {
9500 SV *sv = newSVpvn( start_arg,
9501 RExC_parse - start_arg);
9502 ARG(ret) = add_data( pRExC_state,
9504 RExC_rxi->data->data[ARG(ret)]=(void*)sv;
9511 if (!internal_argval)
9512 RExC_seen |= REG_VERBARG_SEEN;
9513 } else if ( start_arg ) {
9514 vFAIL3("Verb pattern '%.*s' may not have an argument",
9515 verb_len, start_verb);
9517 ret = reg_node(pRExC_state, op);
9519 nextchar(pRExC_state);
9522 else if (*RExC_parse == '?') { /* (?...) */
9523 bool is_logical = 0;
9524 const char * const seqstart = RExC_parse;
9525 if (has_intervening_patws && SIZE_ONLY) {
9526 ckWARNregdep(RExC_parse + 1, "In '(?...)', splitting the initial '(?' is deprecated");
9530 paren = *RExC_parse++;
9531 ret = NULL; /* For look-ahead/behind. */
9534 case 'P': /* (?P...) variants for those used to PCRE/Python */
9535 paren = *RExC_parse++;
9536 if ( paren == '<') /* (?P<...>) named capture */
9538 else if (paren == '>') { /* (?P>name) named recursion */
9539 goto named_recursion;
9541 else if (paren == '=') { /* (?P=...) named backref */
9542 /* this pretty much dupes the code for \k<NAME> in
9543 * regatom(), if you change this make sure you change that
9545 char* name_start = RExC_parse;
9547 SV *sv_dat = reg_scan_name(pRExC_state,
9548 SIZE_ONLY ? REG_RSN_RETURN_NULL : REG_RSN_RETURN_DATA);
9549 if (RExC_parse == name_start || *RExC_parse != ')')
9550 /* diag_listed_as: Sequence ?P=... not terminated in regex; marked by <-- HERE in m/%s/ */
9551 vFAIL2("Sequence %.3s... not terminated",parse_start);
9554 num = add_data( pRExC_state, STR_WITH_LEN("S"));
9555 RExC_rxi->data->data[num]=(void*)sv_dat;
9556 SvREFCNT_inc_simple_void(sv_dat);
9559 ret = reganode(pRExC_state,
9562 : (ASCII_FOLD_RESTRICTED)
9564 : (AT_LEAST_UNI_SEMANTICS)
9572 Set_Node_Offset(ret, parse_start+1);
9573 Set_Node_Cur_Length(ret, parse_start);
9575 nextchar(pRExC_state);
9579 /* diag_listed_as: Sequence (?%s...) not recognized in regex; marked by <-- HERE in m/%s/ */
9580 vFAIL3("Sequence (%.*s...) not recognized",
9581 RExC_parse-seqstart, seqstart);
9583 case '<': /* (?<...) */
9584 if (*RExC_parse == '!')
9586 else if (*RExC_parse != '=')
9592 case '\'': /* (?'...') */
9593 name_start= RExC_parse;
9594 svname = reg_scan_name(pRExC_state,
9595 SIZE_ONLY /* reverse test from the others */
9596 ? REG_RSN_RETURN_NAME
9597 : REG_RSN_RETURN_NULL);
9598 if (RExC_parse == name_start || *RExC_parse != paren)
9599 vFAIL2("Sequence (?%c... not terminated",
9600 paren=='>' ? '<' : paren);
9604 if (!svname) /* shouldn't happen */
9606 "panic: reg_scan_name returned NULL");
9607 if (!RExC_paren_names) {
9608 RExC_paren_names= newHV();
9609 sv_2mortal(MUTABLE_SV(RExC_paren_names));
9611 RExC_paren_name_list= newAV();
9612 sv_2mortal(MUTABLE_SV(RExC_paren_name_list));
9615 he_str = hv_fetch_ent( RExC_paren_names, svname, 1, 0 );
9617 sv_dat = HeVAL(he_str);
9619 /* croak baby croak */
9621 "panic: paren_name hash element allocation failed");
9622 } else if ( SvPOK(sv_dat) ) {
9623 /* (?|...) can mean we have dupes so scan to check
9624 its already been stored. Maybe a flag indicating
9625 we are inside such a construct would be useful,
9626 but the arrays are likely to be quite small, so
9627 for now we punt -- dmq */
9628 IV count = SvIV(sv_dat);
9629 I32 *pv = (I32*)SvPVX(sv_dat);
9631 for ( i = 0 ; i < count ; i++ ) {
9632 if ( pv[i] == RExC_npar ) {
9638 pv = (I32*)SvGROW(sv_dat,
9639 SvCUR(sv_dat) + sizeof(I32)+1);
9640 SvCUR_set(sv_dat, SvCUR(sv_dat) + sizeof(I32));
9641 pv[count] = RExC_npar;
9642 SvIV_set(sv_dat, SvIVX(sv_dat) + 1);
9645 (void)SvUPGRADE(sv_dat,SVt_PVNV);
9646 sv_setpvn(sv_dat, (char *)&(RExC_npar),
9649 SvIV_set(sv_dat, 1);
9652 /* Yes this does cause a memory leak in debugging Perls
9654 if (!av_store(RExC_paren_name_list,
9655 RExC_npar, SvREFCNT_inc(svname)))
9656 SvREFCNT_dec_NN(svname);
9659 /*sv_dump(sv_dat);*/
9661 nextchar(pRExC_state);
9663 goto capturing_parens;
9665 RExC_seen |= REG_LOOKBEHIND_SEEN;
9666 RExC_in_lookbehind++;
9668 case '=': /* (?=...) */
9669 RExC_seen_zerolen++;
9671 case '!': /* (?!...) */
9672 RExC_seen_zerolen++;
9673 if (*RExC_parse == ')') {
9674 ret=reg_node(pRExC_state, OPFAIL);
9675 nextchar(pRExC_state);
9679 case '|': /* (?|...) */
9680 /* branch reset, behave like a (?:...) except that
9681 buffers in alternations share the same numbers */
9683 after_freeze = freeze_paren = RExC_npar;
9685 case ':': /* (?:...) */
9686 case '>': /* (?>...) */
9688 case '$': /* (?$...) */
9689 case '@': /* (?@...) */
9690 vFAIL2("Sequence (?%c...) not implemented", (int)paren);
9692 case '#': /* (?#...) */
9693 /* XXX As soon as we disallow separating the '?' and '*' (by
9694 * spaces or (?#...) comment), it is believed that this case
9695 * will be unreachable and can be removed. See
9697 while (*RExC_parse && *RExC_parse != ')')
9699 if (*RExC_parse != ')')
9700 FAIL("Sequence (?#... not terminated");
9701 nextchar(pRExC_state);
9704 case '0' : /* (?0) */
9705 case 'R' : /* (?R) */
9706 if (*RExC_parse != ')')
9707 FAIL("Sequence (?R) not terminated");
9708 ret = reg_node(pRExC_state, GOSTART);
9709 RExC_seen |= REG_GOSTART_SEEN;
9710 *flagp |= POSTPONED;
9711 nextchar(pRExC_state);
9714 { /* named and numeric backreferences */
9716 case '&': /* (?&NAME) */
9717 parse_start = RExC_parse - 1;
9720 SV *sv_dat = reg_scan_name(pRExC_state,
9721 SIZE_ONLY ? REG_RSN_RETURN_NULL : REG_RSN_RETURN_DATA);
9722 num = sv_dat ? *((I32 *)SvPVX(sv_dat)) : 0;
9724 if (RExC_parse == RExC_end || *RExC_parse != ')')
9725 vFAIL("Sequence (?&... not terminated");
9726 goto gen_recurse_regop;
9727 assert(0); /* NOT REACHED */
9729 if (!(RExC_parse[0] >= '1' && RExC_parse[0] <= '9')) {
9731 vFAIL("Illegal pattern");
9733 goto parse_recursion;
9735 case '-': /* (?-1) */
9736 if (!(RExC_parse[0] >= '1' && RExC_parse[0] <= '9')) {
9737 RExC_parse--; /* rewind to let it be handled later */
9741 case '1': case '2': case '3': case '4': /* (?1) */
9742 case '5': case '6': case '7': case '8': case '9':
9745 num = atoi(RExC_parse);
9746 parse_start = RExC_parse - 1; /* MJD */
9747 if (*RExC_parse == '-')
9749 while (isDIGIT(*RExC_parse))
9751 if (*RExC_parse!=')')
9752 vFAIL("Expecting close bracket");
9755 if ( paren == '-' ) {
9757 Diagram of capture buffer numbering.
9758 Top line is the normal capture buffer numbers
9759 Bottom line is the negative indexing as from
9763 /(a(x)y)(a(b(c(?-2)d)e)f)(g(h))/
9767 num = RExC_npar + num;
9770 vFAIL("Reference to nonexistent group");
9772 } else if ( paren == '+' ) {
9773 num = RExC_npar + num - 1;
9776 ret = reganode(pRExC_state, GOSUB, num);
9778 if (num > (I32)RExC_rx->nparens) {
9780 vFAIL("Reference to nonexistent group");
9782 ARG2L_SET( ret, RExC_recurse_count++);
9784 DEBUG_OPTIMISE_MORE_r(PerlIO_printf(Perl_debug_log,
9785 "Recurse #%"UVuf" to %"IVdf"\n",
9786 (UV)ARG(ret), (IV)ARG2L(ret)));
9790 RExC_seen |= REG_RECURSE_SEEN;
9791 Set_Node_Length(ret, 1 + regarglen[OP(ret)]); /* MJD */
9792 Set_Node_Offset(ret, parse_start); /* MJD */
9794 *flagp |= POSTPONED;
9795 nextchar(pRExC_state);
9797 } /* named and numeric backreferences */
9798 assert(0); /* NOT REACHED */
9800 case '?': /* (??...) */
9802 if (*RExC_parse != '{') {
9804 /* diag_listed_as: Sequence (?%s...) not recognized in regex; marked by <-- HERE in m/%s/ */
9806 "Sequence (%"UTF8f"...) not recognized",
9807 UTF8fARG(UTF, RExC_parse-seqstart, seqstart));
9810 *flagp |= POSTPONED;
9811 paren = *RExC_parse++;
9813 case '{': /* (?{...}) */
9816 struct reg_code_block *cb;
9818 RExC_seen_zerolen++;
9820 if ( !pRExC_state->num_code_blocks
9821 || pRExC_state->code_index >= pRExC_state->num_code_blocks
9822 || pRExC_state->code_blocks[pRExC_state->code_index].start
9823 != (STRLEN)((RExC_parse -3 - (is_logical ? 1 : 0))
9826 if (RExC_pm_flags & PMf_USE_RE_EVAL)
9827 FAIL("panic: Sequence (?{...}): no code block found\n");
9828 FAIL("Eval-group not allowed at runtime, use re 'eval'");
9830 /* this is a pre-compiled code block (?{...}) */
9831 cb = &pRExC_state->code_blocks[pRExC_state->code_index];
9832 RExC_parse = RExC_start + cb->end;
9835 if (cb->src_regex) {
9836 n = add_data(pRExC_state, STR_WITH_LEN("rl"));
9837 RExC_rxi->data->data[n] =
9838 (void*)SvREFCNT_inc((SV*)cb->src_regex);
9839 RExC_rxi->data->data[n+1] = (void*)o;
9842 n = add_data(pRExC_state,
9843 (RExC_pm_flags & PMf_HAS_CV) ? "L" : "l", 1);
9844 RExC_rxi->data->data[n] = (void*)o;
9847 pRExC_state->code_index++;
9848 nextchar(pRExC_state);
9852 ret = reg_node(pRExC_state, LOGICAL);
9853 eval = reganode(pRExC_state, EVAL, n);
9856 /* for later propagation into (??{}) return value */
9857 eval->flags = (U8) (RExC_flags & RXf_PMf_COMPILETIME);
9859 REGTAIL(pRExC_state, ret, eval);
9860 /* deal with the length of this later - MJD */
9863 ret = reganode(pRExC_state, EVAL, n);
9864 Set_Node_Length(ret, RExC_parse - parse_start + 1);
9865 Set_Node_Offset(ret, parse_start);
9868 case '(': /* (?(?{...})...) and (?(?=...)...) */
9871 if (RExC_parse[0] == '?') { /* (?(?...)) */
9872 if (RExC_parse[1] == '=' || RExC_parse[1] == '!'
9873 || RExC_parse[1] == '<'
9874 || RExC_parse[1] == '{') { /* Lookahead or eval. */
9878 ret = reg_node(pRExC_state, LOGICAL);
9882 tail = reg(pRExC_state, 1, &flag, depth+1);
9883 if (flag & RESTART_UTF8) {
9884 *flagp = RESTART_UTF8;
9887 REGTAIL(pRExC_state, ret, tail);
9891 else if ( RExC_parse[0] == '<' /* (?(<NAME>)...) */
9892 || RExC_parse[0] == '\'' ) /* (?('NAME')...) */
9894 char ch = RExC_parse[0] == '<' ? '>' : '\'';
9895 char *name_start= RExC_parse++;
9897 SV *sv_dat=reg_scan_name(pRExC_state,
9898 SIZE_ONLY ? REG_RSN_RETURN_NULL : REG_RSN_RETURN_DATA);
9899 if (RExC_parse == name_start || *RExC_parse != ch)
9900 vFAIL2("Sequence (?(%c... not terminated",
9901 (ch == '>' ? '<' : ch));
9904 num = add_data( pRExC_state, STR_WITH_LEN("S"));
9905 RExC_rxi->data->data[num]=(void*)sv_dat;
9906 SvREFCNT_inc_simple_void(sv_dat);
9908 ret = reganode(pRExC_state,NGROUPP,num);
9909 goto insert_if_check_paren;
9911 else if (RExC_parse[0] == 'D' &&
9912 RExC_parse[1] == 'E' &&
9913 RExC_parse[2] == 'F' &&
9914 RExC_parse[3] == 'I' &&
9915 RExC_parse[4] == 'N' &&
9916 RExC_parse[5] == 'E')
9918 ret = reganode(pRExC_state,DEFINEP,0);
9921 goto insert_if_check_paren;
9923 else if (RExC_parse[0] == 'R') {
9926 if (RExC_parse[0] >= '1' && RExC_parse[0] <= '9' ) {
9927 parno = atoi(RExC_parse++);
9928 while (isDIGIT(*RExC_parse))
9930 } else if (RExC_parse[0] == '&') {
9933 sv_dat = reg_scan_name(pRExC_state,
9935 ? REG_RSN_RETURN_NULL
9936 : REG_RSN_RETURN_DATA);
9937 parno = sv_dat ? *((I32 *)SvPVX(sv_dat)) : 0;
9939 ret = reganode(pRExC_state,INSUBP,parno);
9940 goto insert_if_check_paren;
9942 else if (RExC_parse[0] >= '1' && RExC_parse[0] <= '9' ) {
9946 parno = atoi(RExC_parse++);
9948 while (isDIGIT(*RExC_parse))
9950 ret = reganode(pRExC_state, GROUPP, parno);
9952 insert_if_check_paren:
9953 if (*(tmp = nextchar(pRExC_state)) != ')') {
9954 /* nextchar also skips comments, so undo its work
9955 * and skip over the the next character.
9958 RExC_parse += UTF ? UTF8SKIP(RExC_parse) : 1;
9959 vFAIL("Switch condition not recognized");
9962 REGTAIL(pRExC_state, ret, reganode(pRExC_state, IFTHEN, 0));
9963 br = regbranch(pRExC_state, &flags, 1,depth+1);
9965 if (flags & RESTART_UTF8) {
9966 *flagp = RESTART_UTF8;
9969 FAIL2("panic: regbranch returned NULL, flags=%#"UVxf"",
9972 REGTAIL(pRExC_state, br, reganode(pRExC_state,
9974 c = *nextchar(pRExC_state);
9979 vFAIL("(?(DEFINE)....) does not allow branches");
9981 /* Fake one for optimizer. */
9982 lastbr = reganode(pRExC_state, IFTHEN, 0);
9984 if (!regbranch(pRExC_state, &flags, 1,depth+1)) {
9985 if (flags & RESTART_UTF8) {
9986 *flagp = RESTART_UTF8;
9989 FAIL2("panic: regbranch returned NULL, flags=%#"UVxf"",
9992 REGTAIL(pRExC_state, ret, lastbr);
9995 c = *nextchar(pRExC_state);
10000 vFAIL("Switch (?(condition)... contains too many branches");
10001 ender = reg_node(pRExC_state, TAIL);
10002 REGTAIL(pRExC_state, br, ender);
10004 REGTAIL(pRExC_state, lastbr, ender);
10005 REGTAIL(pRExC_state, NEXTOPER(NEXTOPER(lastbr)), ender);
10008 REGTAIL(pRExC_state, ret, ender);
10009 RExC_size++; /* XXX WHY do we need this?!!
10010 For large programs it seems to be required
10011 but I can't figure out why. -- dmq*/
10015 RExC_parse += UTF ? UTF8SKIP(RExC_parse) : 1;
10016 vFAIL("Unknown switch condition (?(...))");
10019 case '[': /* (?[ ... ]) */
10020 return handle_regex_sets(pRExC_state, NULL, flagp, depth,
10023 RExC_parse--; /* for vFAIL to print correctly */
10024 vFAIL("Sequence (? incomplete");
10026 default: /* e.g., (?i) */
10029 parse_lparen_question_flags(pRExC_state);
10030 if (UCHARAT(RExC_parse) != ':') {
10031 nextchar(pRExC_state);
10036 nextchar(pRExC_state);
10046 ret = reganode(pRExC_state, OPEN, parno);
10048 if (!RExC_nestroot)
10049 RExC_nestroot = parno;
10050 if (RExC_seen & REG_RECURSE_SEEN
10051 && !RExC_open_parens[parno-1])
10053 DEBUG_OPTIMISE_MORE_r(PerlIO_printf(Perl_debug_log,
10054 "Setting open paren #%"IVdf" to %d\n",
10055 (IV)parno, REG_NODE_NUM(ret)));
10056 RExC_open_parens[parno-1]= ret;
10059 Set_Node_Length(ret, 1); /* MJD */
10060 Set_Node_Offset(ret, RExC_parse); /* MJD */
10068 /* Pick up the branches, linking them together. */
10069 parse_start = RExC_parse; /* MJD */
10070 br = regbranch(pRExC_state, &flags, 1,depth+1);
10072 /* branch_len = (paren != 0); */
10075 if (flags & RESTART_UTF8) {
10076 *flagp = RESTART_UTF8;
10079 FAIL2("panic: regbranch returned NULL, flags=%#"UVxf"", (UV) flags);
10081 if (*RExC_parse == '|') {
10082 if (!SIZE_ONLY && RExC_extralen) {
10083 reginsert(pRExC_state, BRANCHJ, br, depth+1);
10086 reginsert(pRExC_state, BRANCH, br, depth+1);
10087 Set_Node_Length(br, paren != 0);
10088 Set_Node_Offset_To_R(br-RExC_emit_start, parse_start-RExC_start);
10092 RExC_extralen += 1; /* For BRANCHJ-BRANCH. */
10094 else if (paren == ':') {
10095 *flagp |= flags&SIMPLE;
10097 if (is_open) { /* Starts with OPEN. */
10098 REGTAIL(pRExC_state, ret, br); /* OPEN -> first. */
10100 else if (paren != '?') /* Not Conditional */
10102 *flagp |= flags & (SPSTART | HASWIDTH | POSTPONED);
10104 while (*RExC_parse == '|') {
10105 if (!SIZE_ONLY && RExC_extralen) {
10106 ender = reganode(pRExC_state, LONGJMP,0);
10108 /* Append to the previous. */
10109 REGTAIL(pRExC_state, NEXTOPER(NEXTOPER(lastbr)), ender);
10112 RExC_extralen += 2; /* Account for LONGJMP. */
10113 nextchar(pRExC_state);
10114 if (freeze_paren) {
10115 if (RExC_npar > after_freeze)
10116 after_freeze = RExC_npar;
10117 RExC_npar = freeze_paren;
10119 br = regbranch(pRExC_state, &flags, 0, depth+1);
10122 if (flags & RESTART_UTF8) {
10123 *flagp = RESTART_UTF8;
10126 FAIL2("panic: regbranch returned NULL, flags=%#"UVxf"", (UV) flags);
10128 REGTAIL(pRExC_state, lastbr, br); /* BRANCH -> BRANCH. */
10130 *flagp |= flags & (SPSTART | HASWIDTH | POSTPONED);
10133 if (have_branch || paren != ':') {
10134 /* Make a closing node, and hook it on the end. */
10137 ender = reg_node(pRExC_state, TAIL);
10140 ender = reganode(pRExC_state, CLOSE, parno);
10141 if (!SIZE_ONLY && RExC_seen & REG_RECURSE_SEEN) {
10142 DEBUG_OPTIMISE_MORE_r(PerlIO_printf(Perl_debug_log,
10143 "Setting close paren #%"IVdf" to %d\n",
10144 (IV)parno, REG_NODE_NUM(ender)));
10145 RExC_close_parens[parno-1]= ender;
10146 if (RExC_nestroot == parno)
10149 Set_Node_Offset(ender,RExC_parse+1); /* MJD */
10150 Set_Node_Length(ender,1); /* MJD */
10156 *flagp &= ~HASWIDTH;
10159 ender = reg_node(pRExC_state, SUCCEED);
10162 ender = reg_node(pRExC_state, END);
10164 assert(!RExC_opend); /* there can only be one! */
10165 RExC_opend = ender;
10169 DEBUG_PARSE_r(if (!SIZE_ONLY) {
10170 SV * const mysv_val1=sv_newmortal();
10171 SV * const mysv_val2=sv_newmortal();
10172 DEBUG_PARSE_MSG("lsbr");
10173 regprop(RExC_rx, mysv_val1, lastbr, NULL);
10174 regprop(RExC_rx, mysv_val2, ender, NULL);
10175 PerlIO_printf(Perl_debug_log, "~ tying lastbr %s (%"IVdf") to ender %s (%"IVdf") offset %"IVdf"\n",
10176 SvPV_nolen_const(mysv_val1),
10177 (IV)REG_NODE_NUM(lastbr),
10178 SvPV_nolen_const(mysv_val2),
10179 (IV)REG_NODE_NUM(ender),
10180 (IV)(ender - lastbr)
10183 REGTAIL(pRExC_state, lastbr, ender);
10185 if (have_branch && !SIZE_ONLY) {
10186 char is_nothing= 1;
10188 RExC_seen |= REG_TOP_LEVEL_BRANCHES_SEEN;
10190 /* Hook the tails of the branches to the closing node. */
10191 for (br = ret; br; br = regnext(br)) {
10192 const U8 op = PL_regkind[OP(br)];
10193 if (op == BRANCH) {
10194 REGTAIL_STUDY(pRExC_state, NEXTOPER(br), ender);
10195 if ( OP(NEXTOPER(br)) != NOTHING
10196 || regnext(NEXTOPER(br)) != ender)
10199 else if (op == BRANCHJ) {
10200 REGTAIL_STUDY(pRExC_state, NEXTOPER(NEXTOPER(br)), ender);
10201 /* for now we always disable this optimisation * /
10202 if ( OP(NEXTOPER(NEXTOPER(br))) != NOTHING
10203 || regnext(NEXTOPER(NEXTOPER(br))) != ender)
10209 br= PL_regkind[OP(ret)] != BRANCH ? regnext(ret) : ret;
10210 DEBUG_PARSE_r(if (!SIZE_ONLY) {
10211 SV * const mysv_val1=sv_newmortal();
10212 SV * const mysv_val2=sv_newmortal();
10213 DEBUG_PARSE_MSG("NADA");
10214 regprop(RExC_rx, mysv_val1, ret, NULL);
10215 regprop(RExC_rx, mysv_val2, ender, NULL);
10216 PerlIO_printf(Perl_debug_log, "~ converting ret %s (%"IVdf") to ender %s (%"IVdf") offset %"IVdf"\n",
10217 SvPV_nolen_const(mysv_val1),
10218 (IV)REG_NODE_NUM(ret),
10219 SvPV_nolen_const(mysv_val2),
10220 (IV)REG_NODE_NUM(ender),
10225 if (OP(ender) == TAIL) {
10230 for ( opt= br + 1; opt < ender ; opt++ )
10231 OP(opt)= OPTIMIZED;
10232 NEXT_OFF(br)= ender - br;
10240 static const char parens[] = "=!<,>";
10242 if (paren && (p = strchr(parens, paren))) {
10243 U8 node = ((p - parens) % 2) ? UNLESSM : IFMATCH;
10244 int flag = (p - parens) > 1;
10247 node = SUSPEND, flag = 0;
10248 reginsert(pRExC_state, node,ret, depth+1);
10249 Set_Node_Cur_Length(ret, parse_start);
10250 Set_Node_Offset(ret, parse_start + 1);
10252 REGTAIL_STUDY(pRExC_state, ret, reg_node(pRExC_state, TAIL));
10256 /* Check for proper termination. */
10258 /* restore original flags, but keep (?p) */
10259 RExC_flags = oregflags | (RExC_flags & RXf_PMf_KEEPCOPY);
10260 if (RExC_parse >= RExC_end || *nextchar(pRExC_state) != ')') {
10261 RExC_parse = oregcomp_parse;
10262 vFAIL("Unmatched (");
10265 else if (!paren && RExC_parse < RExC_end) {
10266 if (*RExC_parse == ')') {
10268 vFAIL("Unmatched )");
10271 FAIL("Junk on end of regexp"); /* "Can't happen". */
10272 assert(0); /* NOTREACHED */
10275 if (RExC_in_lookbehind) {
10276 RExC_in_lookbehind--;
10278 if (after_freeze > RExC_npar)
10279 RExC_npar = after_freeze;
10284 - regbranch - one alternative of an | operator
10286 * Implements the concatenation operator.
10288 * Returns NULL, setting *flagp to RESTART_UTF8 if the sizing scan needs to be
10292 S_regbranch(pTHX_ RExC_state_t *pRExC_state, I32 *flagp, I32 first, U32 depth)
10296 regnode *chain = NULL;
10298 I32 flags = 0, c = 0;
10299 GET_RE_DEBUG_FLAGS_DECL;
10301 PERL_ARGS_ASSERT_REGBRANCH;
10303 DEBUG_PARSE("brnc");
10308 if (!SIZE_ONLY && RExC_extralen)
10309 ret = reganode(pRExC_state, BRANCHJ,0);
10311 ret = reg_node(pRExC_state, BRANCH);
10312 Set_Node_Length(ret, 1);
10316 if (!first && SIZE_ONLY)
10317 RExC_extralen += 1; /* BRANCHJ */
10319 *flagp = WORST; /* Tentatively. */
10322 nextchar(pRExC_state);
10323 while (RExC_parse < RExC_end && *RExC_parse != '|' && *RExC_parse != ')') {
10324 flags &= ~TRYAGAIN;
10325 latest = regpiece(pRExC_state, &flags,depth+1);
10326 if (latest == NULL) {
10327 if (flags & TRYAGAIN)
10329 if (flags & RESTART_UTF8) {
10330 *flagp = RESTART_UTF8;
10333 FAIL2("panic: regpiece returned NULL, flags=%#"UVxf"", (UV) flags);
10335 else if (ret == NULL)
10337 *flagp |= flags&(HASWIDTH|POSTPONED);
10338 if (chain == NULL) /* First piece. */
10339 *flagp |= flags&SPSTART;
10342 REGTAIL(pRExC_state, chain, latest);
10347 if (chain == NULL) { /* Loop ran zero times. */
10348 chain = reg_node(pRExC_state, NOTHING);
10353 *flagp |= flags&SIMPLE;
10360 - regpiece - something followed by possible [*+?]
10362 * Note that the branching code sequences used for ? and the general cases
10363 * of * and + are somewhat optimized: they use the same NOTHING node as
10364 * both the endmarker for their branch list and the body of the last branch.
10365 * It might seem that this node could be dispensed with entirely, but the
10366 * endmarker role is not redundant.
10368 * Returns NULL, setting *flagp to TRYAGAIN if regatom() returns NULL with
10370 * Returns NULL, setting *flagp to RESTART_UTF8 if the sizing scan needs to be
10374 S_regpiece(pTHX_ RExC_state_t *pRExC_state, I32 *flagp, U32 depth)
10381 const char * const origparse = RExC_parse;
10383 I32 max = REG_INFTY;
10384 #ifdef RE_TRACK_PATTERN_OFFSETS
10387 const char *maxpos = NULL;
10389 /* Save the original in case we change the emitted regop to a FAIL. */
10390 regnode * const orig_emit = RExC_emit;
10392 GET_RE_DEBUG_FLAGS_DECL;
10394 PERL_ARGS_ASSERT_REGPIECE;
10396 DEBUG_PARSE("piec");
10398 ret = regatom(pRExC_state, &flags,depth+1);
10400 if (flags & (TRYAGAIN|RESTART_UTF8))
10401 *flagp |= flags & (TRYAGAIN|RESTART_UTF8);
10403 FAIL2("panic: regatom returned NULL, flags=%#"UVxf"", (UV) flags);
10409 if (op == '{' && regcurly(RExC_parse, FALSE)) {
10411 #ifdef RE_TRACK_PATTERN_OFFSETS
10412 parse_start = RExC_parse; /* MJD */
10414 next = RExC_parse + 1;
10415 while (isDIGIT(*next) || *next == ',') {
10416 if (*next == ',') {
10424 if (*next == '}') { /* got one */
10428 min = atoi(RExC_parse);
10429 if (*maxpos == ',')
10432 maxpos = RExC_parse;
10433 max = atoi(maxpos);
10434 if (!max && *maxpos != '0')
10435 max = REG_INFTY; /* meaning "infinity" */
10436 else if (max >= REG_INFTY)
10437 vFAIL2("Quantifier in {,} bigger than %d", REG_INFTY - 1);
10439 nextchar(pRExC_state);
10440 if (max < min) { /* If can't match, warn and optimize to fail
10443 ckWARNreg(RExC_parse, "Quantifier {n,m} with n > m can't match");
10445 /* We can't back off the size because we have to reserve
10446 * enough space for all the things we are about to throw
10447 * away, but we can shrink it by the ammount we are about
10448 * to re-use here */
10449 RExC_size = PREVOPER(RExC_size) - regarglen[(U8)OPFAIL];
10452 RExC_emit = orig_emit;
10454 ret = reg_node(pRExC_state, OPFAIL);
10457 else if (min == max
10458 && RExC_parse < RExC_end
10459 && (*RExC_parse == '?' || *RExC_parse == '+'))
10462 ckWARN2reg(RExC_parse + 1,
10463 "Useless use of greediness modifier '%c'",
10466 /* Absorb the modifier, so later code doesn't see nor use
10468 nextchar(pRExC_state);
10472 if ((flags&SIMPLE)) {
10473 RExC_naughty += 2 + RExC_naughty / 2;
10474 reginsert(pRExC_state, CURLY, ret, depth+1);
10475 Set_Node_Offset(ret, parse_start+1); /* MJD */
10476 Set_Node_Cur_Length(ret, parse_start);
10479 regnode * const w = reg_node(pRExC_state, WHILEM);
10482 REGTAIL(pRExC_state, ret, w);
10483 if (!SIZE_ONLY && RExC_extralen) {
10484 reginsert(pRExC_state, LONGJMP,ret, depth+1);
10485 reginsert(pRExC_state, NOTHING,ret, depth+1);
10486 NEXT_OFF(ret) = 3; /* Go over LONGJMP. */
10488 reginsert(pRExC_state, CURLYX,ret, depth+1);
10490 Set_Node_Offset(ret, parse_start+1);
10491 Set_Node_Length(ret,
10492 op == '{' ? (RExC_parse - parse_start) : 1);
10494 if (!SIZE_ONLY && RExC_extralen)
10495 NEXT_OFF(ret) = 3; /* Go over NOTHING to LONGJMP. */
10496 REGTAIL(pRExC_state, ret, reg_node(pRExC_state, NOTHING));
10498 RExC_whilem_seen++, RExC_extralen += 3;
10499 RExC_naughty += 4 + RExC_naughty; /* compound interest */
10506 *flagp |= HASWIDTH;
10508 ARG1_SET(ret, (U16)min);
10509 ARG2_SET(ret, (U16)max);
10511 if (max == REG_INFTY)
10512 RExC_seen |= REG_UNBOUNDED_QUANTIFIER_SEEN;
10518 if (!ISMULT1(op)) {
10523 #if 0 /* Now runtime fix should be reliable. */
10525 /* if this is reinstated, don't forget to put this back into perldiag:
10527 =item Regexp *+ operand could be empty at {#} in regex m/%s/
10529 (F) The part of the regexp subject to either the * or + quantifier
10530 could match an empty string. The {#} shows in the regular
10531 expression about where the problem was discovered.
10535 if (!(flags&HASWIDTH) && op != '?')
10536 vFAIL("Regexp *+ operand could be empty");
10539 #ifdef RE_TRACK_PATTERN_OFFSETS
10540 parse_start = RExC_parse;
10542 nextchar(pRExC_state);
10544 *flagp = (op != '+') ? (WORST|SPSTART|HASWIDTH) : (WORST|HASWIDTH);
10546 if (op == '*' && (flags&SIMPLE)) {
10547 reginsert(pRExC_state, STAR, ret, depth+1);
10550 RExC_seen |= REG_UNBOUNDED_QUANTIFIER_SEEN;
10552 else if (op == '*') {
10556 else if (op == '+' && (flags&SIMPLE)) {
10557 reginsert(pRExC_state, PLUS, ret, depth+1);
10560 RExC_seen |= REG_UNBOUNDED_QUANTIFIER_SEEN;
10562 else if (op == '+') {
10566 else if (op == '?') {
10571 if (!SIZE_ONLY && !(flags&(HASWIDTH|POSTPONED)) && max > REG_INFTY/3) {
10572 SAVEFREESV(RExC_rx_sv); /* in case of fatal warnings */
10573 ckWARN2reg(RExC_parse,
10574 "%"UTF8f" matches null string many times",
10575 UTF8fARG(UTF, (RExC_parse >= origparse
10576 ? RExC_parse - origparse
10579 (void)ReREFCNT_inc(RExC_rx_sv);
10582 if (RExC_parse < RExC_end && *RExC_parse == '?') {
10583 nextchar(pRExC_state);
10584 reginsert(pRExC_state, MINMOD, ret, depth+1);
10585 REGTAIL(pRExC_state, ret, ret + NODE_STEP_REGNODE);
10588 if (RExC_parse < RExC_end && *RExC_parse == '+') {
10590 nextchar(pRExC_state);
10591 ender = reg_node(pRExC_state, SUCCEED);
10592 REGTAIL(pRExC_state, ret, ender);
10593 reginsert(pRExC_state, SUSPEND, ret, depth+1);
10595 ender = reg_node(pRExC_state, TAIL);
10596 REGTAIL(pRExC_state, ret, ender);
10599 if (RExC_parse < RExC_end && ISMULT2(RExC_parse)) {
10601 vFAIL("Nested quantifiers");
10608 S_grok_bslash_N(pTHX_ RExC_state_t *pRExC_state, regnode** node_p,
10609 UV *valuep, I32 *flagp, U32 depth, bool in_char_class,
10610 const bool strict /* Apply stricter parsing rules? */
10614 /* This is expected to be called by a parser routine that has recognized '\N'
10615 and needs to handle the rest. RExC_parse is expected to point at the first
10616 char following the N at the time of the call. On successful return,
10617 RExC_parse has been updated to point to just after the sequence identified
10618 by this routine, and <*flagp> has been updated.
10620 The \N may be inside (indicated by the boolean <in_char_class>) or outside a
10623 \N may begin either a named sequence, or if outside a character class, mean
10624 to match a non-newline. For non single-quoted regexes, the tokenizer has
10625 attempted to decide which, and in the case of a named sequence, converted it
10626 into one of the forms: \N{} (if the sequence is null), or \N{U+c1.c2...},
10627 where c1... are the characters in the sequence. For single-quoted regexes,
10628 the tokenizer passes the \N sequence through unchanged; this code will not
10629 attempt to determine this nor expand those, instead raising a syntax error.
10630 The net effect is that if the beginning of the passed-in pattern isn't '{U+'
10631 or there is no '}', it signals that this \N occurrence means to match a
10634 Only the \N{U+...} form should occur in a character class, for the same
10635 reason that '.' inside a character class means to just match a period: it
10636 just doesn't make sense.
10638 The function raises an error (via vFAIL), and doesn't return for various
10639 syntax errors. Otherwise it returns TRUE and sets <node_p> or <valuep> on
10640 success; it returns FALSE otherwise. Returns FALSE, setting *flagp to
10641 RESTART_UTF8 if the sizing scan needs to be restarted. Such a restart is
10642 only possible if node_p is non-NULL.
10645 If <valuep> is non-null, it means the caller can accept an input sequence
10646 consisting of a just a single code point; <*valuep> is set to that value
10647 if the input is such.
10649 If <node_p> is non-null it signifies that the caller can accept any other
10650 legal sequence (i.e., one that isn't just a single code point). <*node_p>
10652 1) \N means not-a-NL: points to a newly created REG_ANY node;
10653 2) \N{}: points to a new NOTHING node;
10654 3) otherwise: points to a new EXACT node containing the resolved
10656 Note that FALSE is returned for single code point sequences if <valuep> is
10660 char * endbrace; /* '}' following the name */
10662 char *endchar; /* Points to '.' or '}' ending cur char in the input
10664 bool has_multiple_chars; /* true if the input stream contains a sequence of
10665 more than one character */
10667 GET_RE_DEBUG_FLAGS_DECL;
10669 PERL_ARGS_ASSERT_GROK_BSLASH_N;
10671 GET_RE_DEBUG_FLAGS;
10673 assert(cBOOL(node_p) ^ cBOOL(valuep)); /* Exactly one should be set */
10675 /* The [^\n] meaning of \N ignores spaces and comments under the /x
10676 * modifier. The other meaning does not, so use a temporary until we find
10677 * out which we are being called with */
10678 p = (RExC_flags & RXf_PMf_EXTENDED)
10679 ? regwhite( pRExC_state, RExC_parse )
10682 /* Disambiguate between \N meaning a named character versus \N meaning
10683 * [^\n]. The former is assumed when it can't be the latter. */
10684 if (*p != '{' || regcurly(p, FALSE)) {
10687 /* no bare \N allowed in a charclass */
10688 if (in_char_class) {
10689 vFAIL("\\N in a character class must be a named character: \\N{...}");
10693 RExC_parse--; /* Need to back off so nextchar() doesn't skip the
10695 nextchar(pRExC_state);
10696 *node_p = reg_node(pRExC_state, REG_ANY);
10697 *flagp |= HASWIDTH|SIMPLE;
10699 Set_Node_Length(*node_p, 1); /* MJD */
10703 /* Here, we have decided it should be a named character or sequence */
10705 /* The test above made sure that the next real character is a '{', but
10706 * under the /x modifier, it could be separated by space (or a comment and
10707 * \n) and this is not allowed (for consistency with \x{...} and the
10708 * tokenizer handling of \N{NAME}). */
10709 if (*RExC_parse != '{') {
10710 vFAIL("Missing braces on \\N{}");
10713 RExC_parse++; /* Skip past the '{' */
10715 if (! (endbrace = strchr(RExC_parse, '}')) /* no trailing brace */
10716 || ! (endbrace == RExC_parse /* nothing between the {} */
10717 || (endbrace - RExC_parse >= 2 /* U+ (bad hex is checked below
10719 && strnEQ(RExC_parse, "U+", 2)))) /* for a better error msg)
10722 if (endbrace) RExC_parse = endbrace; /* position msg's '<--HERE' */
10723 vFAIL("\\N{NAME} must be resolved by the lexer");
10726 if (endbrace == RExC_parse) { /* empty: \N{} */
10729 *node_p = reg_node(pRExC_state,NOTHING);
10731 else if (in_char_class) {
10732 if (SIZE_ONLY && in_char_class) {
10734 RExC_parse++; /* Position after the "}" */
10735 vFAIL("Zero length \\N{}");
10738 ckWARNreg(RExC_parse,
10739 "Ignoring zero length \\N{} in character class");
10747 nextchar(pRExC_state);
10751 RExC_uni_semantics = 1; /* Unicode named chars imply Unicode semantics */
10752 RExC_parse += 2; /* Skip past the 'U+' */
10754 endchar = RExC_parse + strcspn(RExC_parse, ".}");
10756 /* Code points are separated by dots. If none, there is only one code
10757 * point, and is terminated by the brace */
10758 has_multiple_chars = (endchar < endbrace);
10760 if (valuep && (! has_multiple_chars || in_char_class)) {
10761 /* We only pay attention to the first char of
10762 multichar strings being returned in char classes. I kinda wonder
10763 if this makes sense as it does change the behaviour
10764 from earlier versions, OTOH that behaviour was broken
10765 as well. XXX Solution is to recharacterize as
10766 [rest-of-class]|multi1|multi2... */
10768 STRLEN length_of_hex = (STRLEN)(endchar - RExC_parse);
10769 I32 grok_hex_flags = PERL_SCAN_ALLOW_UNDERSCORES
10770 | PERL_SCAN_DISALLOW_PREFIX
10771 | (SIZE_ONLY ? PERL_SCAN_SILENT_ILLDIGIT : 0);
10773 *valuep = grok_hex(RExC_parse, &length_of_hex, &grok_hex_flags, NULL);
10775 /* The tokenizer should have guaranteed validity, but it's possible to
10776 * bypass it by using single quoting, so check */
10777 if (length_of_hex == 0
10778 || length_of_hex != (STRLEN)(endchar - RExC_parse) )
10780 RExC_parse += length_of_hex; /* Includes all the valid */
10781 RExC_parse += (RExC_orig_utf8) /* point to after 1st invalid */
10782 ? UTF8SKIP(RExC_parse)
10784 /* Guard against malformed utf8 */
10785 if (RExC_parse >= endchar) {
10786 RExC_parse = endchar;
10788 vFAIL("Invalid hexadecimal number in \\N{U+...}");
10791 if (in_char_class && has_multiple_chars) {
10793 RExC_parse = endbrace;
10794 vFAIL("\\N{} in character class restricted to one character");
10797 ckWARNreg(endchar, "Using just the first character returned by \\N{} in character class");
10801 RExC_parse = endbrace + 1;
10803 else if (! node_p || ! has_multiple_chars) {
10805 /* Here, the input is legal, but not according to the caller's
10806 * options. We fail without advancing the parse, so that the
10807 * caller can try again */
10813 /* What is done here is to convert this to a sub-pattern of the form
10814 * (?:\x{char1}\x{char2}...)
10815 * and then call reg recursively. That way, it retains its atomicness,
10816 * while not having to worry about special handling that some code
10817 * points may have. toke.c has converted the original Unicode values
10818 * to native, so that we can just pass on the hex values unchanged. We
10819 * do have to set a flag to keep recoding from happening in the
10822 SV * substitute_parse = newSVpvn_flags("?:", 2, SVf_UTF8|SVs_TEMP);
10824 char *orig_end = RExC_end;
10827 while (RExC_parse < endbrace) {
10829 /* Convert to notation the rest of the code understands */
10830 sv_catpv(substitute_parse, "\\x{");
10831 sv_catpvn(substitute_parse, RExC_parse, endchar - RExC_parse);
10832 sv_catpv(substitute_parse, "}");
10834 /* Point to the beginning of the next character in the sequence. */
10835 RExC_parse = endchar + 1;
10836 endchar = RExC_parse + strcspn(RExC_parse, ".}");
10838 sv_catpv(substitute_parse, ")");
10840 RExC_parse = SvPV(substitute_parse, len);
10842 /* Don't allow empty number */
10844 vFAIL("Invalid hexadecimal number in \\N{U+...}");
10846 RExC_end = RExC_parse + len;
10848 /* The values are Unicode, and therefore not subject to recoding */
10849 RExC_override_recoding = 1;
10851 if (!(*node_p = reg(pRExC_state, 1, &flags, depth+1))) {
10852 if (flags & RESTART_UTF8) {
10853 *flagp = RESTART_UTF8;
10856 FAIL2("panic: reg returned NULL to grok_bslash_N, flags=%#"UVxf"",
10859 *flagp |= flags&(HASWIDTH|SPSTART|SIMPLE|POSTPONED);
10861 RExC_parse = endbrace;
10862 RExC_end = orig_end;
10863 RExC_override_recoding = 0;
10865 nextchar(pRExC_state);
10875 * It returns the code point in utf8 for the value in *encp.
10876 * value: a code value in the source encoding
10877 * encp: a pointer to an Encode object
10879 * If the result from Encode is not a single character,
10880 * it returns U+FFFD (Replacement character) and sets *encp to NULL.
10883 S_reg_recode(pTHX_ const char value, SV **encp)
10886 SV * const sv = newSVpvn_flags(&value, numlen, SVs_TEMP);
10887 const char * const s = *encp ? sv_recode_to_utf8(sv, *encp) : SvPVX(sv);
10888 const STRLEN newlen = SvCUR(sv);
10889 UV uv = UNICODE_REPLACEMENT;
10891 PERL_ARGS_ASSERT_REG_RECODE;
10895 ? utf8n_to_uvchr((U8*)s, newlen, &numlen, UTF8_ALLOW_DEFAULT)
10898 if (!newlen || numlen != newlen) {
10899 uv = UNICODE_REPLACEMENT;
10905 PERL_STATIC_INLINE U8
10906 S_compute_EXACTish(pTHX_ RExC_state_t *pRExC_state)
10910 PERL_ARGS_ASSERT_COMPUTE_EXACTISH;
10916 op = get_regex_charset(RExC_flags);
10917 if (op >= REGEX_ASCII_RESTRICTED_CHARSET) {
10918 op--; /* /a is same as /u, and map /aa's offset to what /a's would have
10919 been, so there is no hole */
10922 return op + EXACTF;
10925 PERL_STATIC_INLINE void
10926 S_alloc_maybe_populate_EXACT(pTHX_ RExC_state_t *pRExC_state,
10927 regnode *node, I32* flagp, STRLEN len, UV code_point,
10930 /* This knows the details about sizing an EXACTish node, setting flags for
10931 * it (by setting <*flagp>, and potentially populating it with a single
10934 * If <len> (the length in bytes) is non-zero, this function assumes that
10935 * the node has already been populated, and just does the sizing. In this
10936 * case <code_point> should be the final code point that has already been
10937 * placed into the node. This value will be ignored except that under some
10938 * circumstances <*flagp> is set based on it.
10940 * If <len> is zero, the function assumes that the node is to contain only
10941 * the single character given by <code_point> and calculates what <len>
10942 * should be. In pass 1, it sizes the node appropriately. In pass 2, it
10943 * additionally will populate the node's STRING with <code_point> or its
10946 * In both cases <*flagp> is appropriately set
10948 * It knows that under FOLD, the Latin Sharp S and UTF characters above
10949 * 255, must be folded (the former only when the rules indicate it can
10952 * When it does the populating, it looks at the flag 'downgradable'. If
10953 * true with a node that folds, it checks if the single code point
10954 * participates in a fold, and if not downgrades the node to an EXACT.
10955 * This helps the optimizer */
10957 bool len_passed_in = cBOOL(len != 0);
10958 U8 character[UTF8_MAXBYTES_CASE+1];
10960 PERL_ARGS_ASSERT_ALLOC_MAYBE_POPULATE_EXACT;
10962 /* Don't bother to check for downgrading in PASS1, as it doesn't make any
10963 * sizing difference, and is extra work that is thrown away */
10964 if (downgradable && ! PASS2) {
10965 downgradable = FALSE;
10968 if (! len_passed_in) {
10970 if (UNI_IS_INVARIANT(code_point)) {
10971 if (LOC || ! FOLD) { /* /l defers folding until runtime */
10972 *character = (U8) code_point;
10974 else { /* Here is /i and not /l (toFOLD() is defined on just
10975 ASCII, which isn't the same thing as INVARIANT on
10976 EBCDIC, but it works there, as the extra invariants
10977 fold to themselves) */
10978 *character = toFOLD((U8) code_point);
10980 /* We can downgrade to an EXACT node if this character
10981 * isn't a folding one. Note that this assumes that
10982 * nothing above Latin1 folds to some other invariant than
10983 * one of these alphabetics; otherwise we would also have
10985 * && (! HAS_NONLATIN1_FOLD_CLOSURE(code_point)
10986 * || ASCII_FOLD_RESTRICTED))
10988 if (downgradable && PL_fold[code_point] == code_point) {
10994 else if (FOLD && (! LOC
10995 || ! is_PROBLEMATIC_LOCALE_FOLD_cp(code_point)))
10996 { /* Folding, and ok to do so now */
10997 UV folded = _to_uni_fold_flags(
11001 FOLD_FLAGS_FULL | ((ASCII_FOLD_RESTRICTED)
11002 ? FOLD_FLAGS_NOMIX_ASCII
11005 && folded == code_point
11006 && ! _invlist_contains_cp(PL_utf8_foldable, code_point))
11011 else if (code_point <= MAX_UTF8_TWO_BYTE) {
11013 /* Not folding this cp, and can output it directly */
11014 *character = UTF8_TWO_BYTE_HI(code_point);
11015 *(character + 1) = UTF8_TWO_BYTE_LO(code_point);
11019 uvchr_to_utf8( character, code_point);
11020 len = UTF8SKIP(character);
11022 } /* Else pattern isn't UTF8. */
11024 *character = (U8) code_point;
11026 } /* Else is folded non-UTF8 */
11027 else if (LIKELY(code_point != LATIN_SMALL_LETTER_SHARP_S)) {
11029 /* We don't fold any non-UTF8 except possibly the Sharp s (see
11030 * comments at join_exact()); */
11031 *character = (U8) code_point;
11034 /* Can turn into an EXACT node if we know the fold at compile time,
11035 * and it folds to itself and doesn't particpate in other folds */
11038 && PL_fold_latin1[code_point] == code_point
11039 && (! HAS_NONLATIN1_FOLD_CLOSURE(code_point)
11040 || (isASCII(code_point) && ASCII_FOLD_RESTRICTED)))
11044 } /* else is Sharp s. May need to fold it */
11045 else if (AT_LEAST_UNI_SEMANTICS && ! ASCII_FOLD_RESTRICTED) {
11047 *(character + 1) = 's';
11051 *character = LATIN_SMALL_LETTER_SHARP_S;
11057 RExC_size += STR_SZ(len);
11060 RExC_emit += STR_SZ(len);
11061 STR_LEN(node) = len;
11062 if (! len_passed_in) {
11063 Copy((char *) character, STRING(node), len, char);
11067 *flagp |= HASWIDTH;
11069 /* A single character node is SIMPLE, except for the special-cased SHARP S
11071 if ((len == 1 || (UTF && len == UNISKIP(code_point)))
11072 && (code_point != LATIN_SMALL_LETTER_SHARP_S
11073 || ! FOLD || ! DEPENDS_SEMANTICS))
11078 /* The OP may not be well defined in PASS1 */
11079 if (PASS2 && OP(node) == EXACTFL) {
11080 RExC_contains_locale = 1;
11085 /* return atoi(p), unless it's too big to sensibly be a backref,
11086 * in which case return I32_MAX (rather than possibly 32-bit wrapping) */
11089 S_backref_value(char *p)
11093 for (;isDIGIT(*q); q++); /* calculate length of num */
11094 if (q - p == 0 || q - p > 9)
11101 - regatom - the lowest level
11103 Try to identify anything special at the start of the pattern. If there
11104 is, then handle it as required. This may involve generating a single regop,
11105 such as for an assertion; or it may involve recursing, such as to
11106 handle a () structure.
11108 If the string doesn't start with something special then we gobble up
11109 as much literal text as we can.
11111 Once we have been able to handle whatever type of thing started the
11112 sequence, we return.
11114 Note: we have to be careful with escapes, as they can be both literal
11115 and special, and in the case of \10 and friends, context determines which.
11117 A summary of the code structure is:
11119 switch (first_byte) {
11120 cases for each special:
11121 handle this special;
11124 switch (2nd byte) {
11125 cases for each unambiguous special:
11126 handle this special;
11128 cases for each ambigous special/literal:
11130 if (special) handle here
11132 default: // unambiguously literal:
11135 default: // is a literal char
11138 create EXACTish node for literal;
11139 while (more input and node isn't full) {
11140 switch (input_byte) {
11141 cases for each special;
11142 make sure parse pointer is set so that the next call to
11143 regatom will see this special first
11144 goto loopdone; // EXACTish node terminated by prev. char
11146 append char to EXACTISH node;
11148 get next input byte;
11152 return the generated node;
11154 Specifically there are two separate switches for handling
11155 escape sequences, with the one for handling literal escapes requiring
11156 a dummy entry for all of the special escapes that are actually handled
11159 Returns NULL, setting *flagp to TRYAGAIN if reg() returns NULL with
11161 Returns NULL, setting *flagp to RESTART_UTF8 if the sizing scan needs to be
11163 Otherwise does not return NULL.
11167 S_regatom(pTHX_ RExC_state_t *pRExC_state, I32 *flagp, U32 depth)
11170 regnode *ret = NULL;
11172 char *parse_start = RExC_parse;
11176 GET_RE_DEBUG_FLAGS_DECL;
11178 *flagp = WORST; /* Tentatively. */
11180 DEBUG_PARSE("atom");
11182 PERL_ARGS_ASSERT_REGATOM;
11185 switch ((U8)*RExC_parse) {
11187 RExC_seen_zerolen++;
11188 nextchar(pRExC_state);
11189 if (RExC_flags & RXf_PMf_MULTILINE)
11190 ret = reg_node(pRExC_state, MBOL);
11191 else if (RExC_flags & RXf_PMf_SINGLELINE)
11192 ret = reg_node(pRExC_state, SBOL);
11194 ret = reg_node(pRExC_state, BOL);
11195 Set_Node_Length(ret, 1); /* MJD */
11198 nextchar(pRExC_state);
11200 RExC_seen_zerolen++;
11201 if (RExC_flags & RXf_PMf_MULTILINE)
11202 ret = reg_node(pRExC_state, MEOL);
11203 else if (RExC_flags & RXf_PMf_SINGLELINE)
11204 ret = reg_node(pRExC_state, SEOL);
11206 ret = reg_node(pRExC_state, EOL);
11207 Set_Node_Length(ret, 1); /* MJD */
11210 nextchar(pRExC_state);
11211 if (RExC_flags & RXf_PMf_SINGLELINE)
11212 ret = reg_node(pRExC_state, SANY);
11214 ret = reg_node(pRExC_state, REG_ANY);
11215 *flagp |= HASWIDTH|SIMPLE;
11217 Set_Node_Length(ret, 1); /* MJD */
11221 char * const oregcomp_parse = ++RExC_parse;
11222 ret = regclass(pRExC_state, flagp,depth+1,
11223 FALSE, /* means parse the whole char class */
11224 TRUE, /* allow multi-char folds */
11225 FALSE, /* don't silence non-portable warnings. */
11227 if (*RExC_parse != ']') {
11228 RExC_parse = oregcomp_parse;
11229 vFAIL("Unmatched [");
11232 if (*flagp & RESTART_UTF8)
11234 FAIL2("panic: regclass returned NULL to regatom, flags=%#"UVxf"",
11237 nextchar(pRExC_state);
11238 Set_Node_Length(ret, RExC_parse - oregcomp_parse + 1); /* MJD */
11242 nextchar(pRExC_state);
11243 ret = reg(pRExC_state, 2, &flags,depth+1);
11245 if (flags & TRYAGAIN) {
11246 if (RExC_parse == RExC_end) {
11247 /* Make parent create an empty node if needed. */
11248 *flagp |= TRYAGAIN;
11253 if (flags & RESTART_UTF8) {
11254 *flagp = RESTART_UTF8;
11257 FAIL2("panic: reg returned NULL to regatom, flags=%#"UVxf"",
11260 *flagp |= flags&(HASWIDTH|SPSTART|SIMPLE|POSTPONED);
11264 if (flags & TRYAGAIN) {
11265 *flagp |= TRYAGAIN;
11268 vFAIL("Internal urp");
11269 /* Supposed to be caught earlier. */
11272 if (!regcurly(RExC_parse, FALSE)) {
11281 vFAIL("Quantifier follows nothing");
11286 This switch handles escape sequences that resolve to some kind
11287 of special regop and not to literal text. Escape sequnces that
11288 resolve to literal text are handled below in the switch marked
11291 Every entry in this switch *must* have a corresponding entry
11292 in the literal escape switch. However, the opposite is not
11293 required, as the default for this switch is to jump to the
11294 literal text handling code.
11296 switch ((U8)*++RExC_parse) {
11298 /* Special Escapes */
11300 RExC_seen_zerolen++;
11301 ret = reg_node(pRExC_state, SBOL);
11303 goto finish_meta_pat;
11305 ret = reg_node(pRExC_state, GPOS);
11306 RExC_seen |= REG_GPOS_SEEN;
11308 goto finish_meta_pat;
11310 RExC_seen_zerolen++;
11311 ret = reg_node(pRExC_state, KEEPS);
11313 /* XXX:dmq : disabling in-place substitution seems to
11314 * be necessary here to avoid cases of memory corruption, as
11315 * with: C<$_="x" x 80; s/x\K/y/> -- rgs
11317 RExC_seen |= REG_LOOKBEHIND_SEEN;
11318 goto finish_meta_pat;
11320 ret = reg_node(pRExC_state, SEOL);
11322 RExC_seen_zerolen++; /* Do not optimize RE away */
11323 goto finish_meta_pat;
11325 ret = reg_node(pRExC_state, EOS);
11327 RExC_seen_zerolen++; /* Do not optimize RE away */
11328 goto finish_meta_pat;
11330 ret = reg_node(pRExC_state, CANY);
11331 RExC_seen |= REG_CANY_SEEN;
11332 *flagp |= HASWIDTH|SIMPLE;
11333 goto finish_meta_pat;
11335 ret = reg_node(pRExC_state, CLUMP);
11336 *flagp |= HASWIDTH;
11337 goto finish_meta_pat;
11343 arg = ANYOF_WORDCHAR;
11347 RExC_seen_zerolen++;
11348 RExC_seen |= REG_LOOKBEHIND_SEEN;
11349 op = BOUND + get_regex_charset(RExC_flags);
11350 if (op > BOUNDA) { /* /aa is same as /a */
11353 else if (op == BOUNDL) {
11354 RExC_contains_locale = 1;
11356 ret = reg_node(pRExC_state, op);
11357 FLAGS(ret) = get_regex_charset(RExC_flags);
11359 if (! SIZE_ONLY && (U8) *(RExC_parse + 1) == '{') {
11360 /* diag_listed_as: Use "%s" instead of "%s" */
11361 vFAIL("Use \"\\b\\{\" instead of \"\\b{\"");
11363 goto finish_meta_pat;
11365 RExC_seen_zerolen++;
11366 RExC_seen |= REG_LOOKBEHIND_SEEN;
11367 op = NBOUND + get_regex_charset(RExC_flags);
11368 if (op > NBOUNDA) { /* /aa is same as /a */
11371 else if (op == NBOUNDL) {
11372 RExC_contains_locale = 1;
11374 ret = reg_node(pRExC_state, op);
11375 FLAGS(ret) = get_regex_charset(RExC_flags);
11377 if (! SIZE_ONLY && (U8) *(RExC_parse + 1) == '{') {
11378 /* diag_listed_as: Use "%s" instead of "%s" */
11379 vFAIL("Use \"\\B\\{\" instead of \"\\B{\"");
11381 goto finish_meta_pat;
11391 ret = reg_node(pRExC_state, LNBREAK);
11392 *flagp |= HASWIDTH|SIMPLE;
11393 goto finish_meta_pat;
11401 goto join_posix_op_known;
11407 arg = ANYOF_VERTWS;
11409 goto join_posix_op_known;
11419 op = POSIXD + get_regex_charset(RExC_flags);
11420 if (op > POSIXA) { /* /aa is same as /a */
11423 else if (op == POSIXL) {
11424 RExC_contains_locale = 1;
11427 join_posix_op_known:
11430 op += NPOSIXD - POSIXD;
11433 ret = reg_node(pRExC_state, op);
11435 FLAGS(ret) = namedclass_to_classnum(arg);
11438 *flagp |= HASWIDTH|SIMPLE;
11442 nextchar(pRExC_state);
11443 Set_Node_Length(ret, 2); /* MJD */
11449 char* parse_start = RExC_parse - 2;
11454 ret = regclass(pRExC_state, flagp,depth+1,
11455 TRUE, /* means just parse this element */
11456 FALSE, /* don't allow multi-char folds */
11457 FALSE, /* don't silence non-portable warnings.
11458 It would be a bug if these returned
11461 /* regclass() can only return RESTART_UTF8 if multi-char folds
11464 FAIL2("panic: regclass returned NULL to regatom, flags=%#"UVxf"",
11469 Set_Node_Offset(ret, parse_start + 2);
11470 Set_Node_Cur_Length(ret, parse_start);
11471 nextchar(pRExC_state);
11475 /* Handle \N and \N{NAME} with multiple code points here and not
11476 * below because it can be multicharacter. join_exact() will join
11477 * them up later on. Also this makes sure that things like
11478 * /\N{BLAH}+/ and \N{BLAH} being multi char Just Happen. dmq.
11479 * The options to the grok function call causes it to fail if the
11480 * sequence is just a single code point. We then go treat it as
11481 * just another character in the current EXACT node, and hence it
11482 * gets uniform treatment with all the other characters. The
11483 * special treatment for quantifiers is not needed for such single
11484 * character sequences */
11486 if (! grok_bslash_N(pRExC_state, &ret, NULL, flagp, depth, FALSE,
11487 FALSE /* not strict */ )) {
11488 if (*flagp & RESTART_UTF8)
11494 case 'k': /* Handle \k<NAME> and \k'NAME' */
11497 char ch= RExC_parse[1];
11498 if (ch != '<' && ch != '\'' && ch != '{') {
11500 /* diag_listed_as: Sequence \%s... not terminated in regex; marked by <-- HERE in m/%s/ */
11501 vFAIL2("Sequence %.2s... not terminated",parse_start);
11503 /* this pretty much dupes the code for (?P=...) in reg(), if
11504 you change this make sure you change that */
11505 char* name_start = (RExC_parse += 2);
11507 SV *sv_dat = reg_scan_name(pRExC_state,
11508 SIZE_ONLY ? REG_RSN_RETURN_NULL : REG_RSN_RETURN_DATA);
11509 ch= (ch == '<') ? '>' : (ch == '{') ? '}' : '\'';
11510 if (RExC_parse == name_start || *RExC_parse != ch)
11511 /* diag_listed_as: Sequence \%s... not terminated in regex; marked by <-- HERE in m/%s/ */
11512 vFAIL2("Sequence %.3s... not terminated",parse_start);
11515 num = add_data( pRExC_state, STR_WITH_LEN("S"));
11516 RExC_rxi->data->data[num]=(void*)sv_dat;
11517 SvREFCNT_inc_simple_void(sv_dat);
11521 ret = reganode(pRExC_state,
11524 : (ASCII_FOLD_RESTRICTED)
11526 : (AT_LEAST_UNI_SEMANTICS)
11532 *flagp |= HASWIDTH;
11534 /* override incorrect value set in reganode MJD */
11535 Set_Node_Offset(ret, parse_start+1);
11536 Set_Node_Cur_Length(ret, parse_start);
11537 nextchar(pRExC_state);
11543 case '1': case '2': case '3': case '4':
11544 case '5': case '6': case '7': case '8': case '9':
11549 if (*RExC_parse == 'g') {
11553 if (*RExC_parse == '{') {
11557 if (*RExC_parse == '-') {
11561 if (hasbrace && !isDIGIT(*RExC_parse)) {
11562 if (isrel) RExC_parse--;
11564 goto parse_named_seq;
11567 num = S_backref_value(RExC_parse);
11569 vFAIL("Reference to invalid group 0");
11570 else if (num == I32_MAX) {
11571 if (isDIGIT(*RExC_parse))
11572 vFAIL("Reference to nonexistent group");
11574 vFAIL("Unterminated \\g... pattern");
11578 num = RExC_npar - num;
11580 vFAIL("Reference to nonexistent or unclosed group");
11584 num = S_backref_value(RExC_parse);
11585 /* bare \NNN might be backref or octal - if it is larger than or equal
11586 * RExC_npar then it is assumed to be and octal escape.
11587 * Note RExC_npar is +1 from the actual number of parens*/
11588 if (num == I32_MAX || (num > 9 && num >= RExC_npar
11589 && *RExC_parse != '8' && *RExC_parse != '9'))
11591 /* Probably a character specified in octal, e.g. \35 */
11596 /* at this point RExC_parse definitely points to a backref
11599 #ifdef RE_TRACK_PATTERN_OFFSETS
11600 char * const parse_start = RExC_parse - 1; /* MJD */
11602 while (isDIGIT(*RExC_parse))
11605 if (*RExC_parse != '}')
11606 vFAIL("Unterminated \\g{...} pattern");
11610 if (num > (I32)RExC_rx->nparens)
11611 vFAIL("Reference to nonexistent group");
11614 ret = reganode(pRExC_state,
11617 : (ASCII_FOLD_RESTRICTED)
11619 : (AT_LEAST_UNI_SEMANTICS)
11625 *flagp |= HASWIDTH;
11627 /* override incorrect value set in reganode MJD */
11628 Set_Node_Offset(ret, parse_start+1);
11629 Set_Node_Cur_Length(ret, parse_start);
11631 nextchar(pRExC_state);
11636 if (RExC_parse >= RExC_end)
11637 FAIL("Trailing \\");
11640 /* Do not generate "unrecognized" warnings here, we fall
11641 back into the quick-grab loop below */
11648 if (RExC_flags & RXf_PMf_EXTENDED) {
11649 if ( reg_skipcomment( pRExC_state ) )
11656 parse_start = RExC_parse - 1;
11665 #define MAX_NODE_STRING_SIZE 127
11666 char foldbuf[MAX_NODE_STRING_SIZE+UTF8_MAXBYTES_CASE];
11668 U8 upper_parse = MAX_NODE_STRING_SIZE;
11669 U8 node_type = compute_EXACTish(pRExC_state);
11670 bool next_is_quantifier;
11671 char * oldp = NULL;
11673 /* We can convert EXACTF nodes to EXACTFU if they contain only
11674 * characters that match identically regardless of the target
11675 * string's UTF8ness. The reason to do this is that EXACTF is not
11676 * trie-able, EXACTFU is.
11678 * Similarly, we can convert EXACTFL nodes to EXACTFU if they
11679 * contain only above-Latin1 characters (hence must be in UTF8),
11680 * which don't participate in folds with Latin1-range characters,
11681 * as the latter's folds aren't known until runtime. (We don't
11682 * need to figure this out until pass 2) */
11683 bool maybe_exactfu = PASS2
11684 && (node_type == EXACTF || node_type == EXACTFL);
11686 /* If a folding node contains only code points that don't
11687 * participate in folds, it can be changed into an EXACT node,
11688 * which allows the optimizer more things to look for */
11691 ret = reg_node(pRExC_state, node_type);
11693 /* In pass1, folded, we use a temporary buffer instead of the
11694 * actual node, as the node doesn't exist yet */
11695 s = (SIZE_ONLY && FOLD) ? foldbuf : STRING(ret);
11701 /* We do the EXACTFish to EXACT node only if folding. (And we
11702 * don't need to figure this out until pass 2) */
11703 maybe_exact = FOLD && PASS2;
11705 /* XXX The node can hold up to 255 bytes, yet this only goes to
11706 * 127. I (khw) do not know why. Keeping it somewhat less than
11707 * 255 allows us to not have to worry about overflow due to
11708 * converting to utf8 and fold expansion, but that value is
11709 * 255-UTF8_MAXBYTES_CASE. join_exact() may join adjacent nodes
11710 * split up by this limit into a single one using the real max of
11711 * 255. Even at 127, this breaks under rare circumstances. If
11712 * folding, we do not want to split a node at a character that is a
11713 * non-final in a multi-char fold, as an input string could just
11714 * happen to want to match across the node boundary. The join
11715 * would solve that problem if the join actually happens. But a
11716 * series of more than two nodes in a row each of 127 would cause
11717 * the first join to succeed to get to 254, but then there wouldn't
11718 * be room for the next one, which could at be one of those split
11719 * multi-char folds. I don't know of any fool-proof solution. One
11720 * could back off to end with only a code point that isn't such a
11721 * non-final, but it is possible for there not to be any in the
11723 for (p = RExC_parse - 1;
11724 len < upper_parse && p < RExC_end;
11729 if (RExC_flags & RXf_PMf_EXTENDED)
11730 p = regwhite( pRExC_state, p );
11741 /* Literal Escapes Switch
11743 This switch is meant to handle escape sequences that
11744 resolve to a literal character.
11746 Every escape sequence that represents something
11747 else, like an assertion or a char class, is handled
11748 in the switch marked 'Special Escapes' above in this
11749 routine, but also has an entry here as anything that
11750 isn't explicitly mentioned here will be treated as
11751 an unescaped equivalent literal.
11754 switch ((U8)*++p) {
11755 /* These are all the special escapes. */
11756 case 'A': /* Start assertion */
11757 case 'b': case 'B': /* Word-boundary assertion*/
11758 case 'C': /* Single char !DANGEROUS! */
11759 case 'd': case 'D': /* digit class */
11760 case 'g': case 'G': /* generic-backref, pos assertion */
11761 case 'h': case 'H': /* HORIZWS */
11762 case 'k': case 'K': /* named backref, keep marker */
11763 case 'p': case 'P': /* Unicode property */
11764 case 'R': /* LNBREAK */
11765 case 's': case 'S': /* space class */
11766 case 'v': case 'V': /* VERTWS */
11767 case 'w': case 'W': /* word class */
11768 case 'X': /* eXtended Unicode "combining
11769 character sequence" */
11770 case 'z': case 'Z': /* End of line/string assertion */
11774 /* Anything after here is an escape that resolves to a
11775 literal. (Except digits, which may or may not)
11781 case 'N': /* Handle a single-code point named character. */
11782 /* The options cause it to fail if a multiple code
11783 * point sequence. Handle those in the switch() above
11785 RExC_parse = p + 1;
11786 if (! grok_bslash_N(pRExC_state, NULL, &ender,
11787 flagp, depth, FALSE,
11788 FALSE /* not strict */ ))
11790 if (*flagp & RESTART_UTF8)
11791 FAIL("panic: grok_bslash_N set RESTART_UTF8");
11792 RExC_parse = p = oldp;
11796 if (ender > 0xff) {
11813 ender = ASCII_TO_NATIVE('\033');
11823 const char* error_msg;
11825 bool valid = grok_bslash_o(&p,
11828 TRUE, /* out warnings */
11829 FALSE, /* not strict */
11830 TRUE, /* Output warnings
11835 RExC_parse = p; /* going to die anyway; point
11836 to exact spot of failure */
11840 if (PL_encoding && ender < 0x100) {
11841 goto recode_encoding;
11843 if (ender > 0xff) {
11850 UV result = UV_MAX; /* initialize to erroneous
11852 const char* error_msg;
11854 bool valid = grok_bslash_x(&p,
11857 TRUE, /* out warnings */
11858 FALSE, /* not strict */
11859 TRUE, /* Output warnings
11864 RExC_parse = p; /* going to die anyway; point
11865 to exact spot of failure */
11870 if (PL_encoding && ender < 0x100) {
11871 goto recode_encoding;
11873 if (ender > 0xff) {
11880 ender = grok_bslash_c(*p++, SIZE_ONLY);
11882 case '8': case '9': /* must be a backreference */
11885 case '1': case '2': case '3':case '4':
11886 case '5': case '6': case '7':
11887 /* When we parse backslash escapes there is ambiguity
11888 * between backreferences and octal escapes. Any escape
11889 * from \1 - \9 is a backreference, any multi-digit
11890 * escape which does not start with 0 and which when
11891 * evaluated as decimal could refer to an already
11892 * parsed capture buffer is a backslash. Anything else
11895 * Note this implies that \118 could be interpreted as
11896 * 118 OR as "\11" . "8" depending on whether there
11897 * were 118 capture buffers defined already in the
11900 /* NOTE, RExC_npar is 1 more than the actual number of
11901 * parens we have seen so far, hence the < RExC_npar below. */
11903 if ( !isDIGIT(p[1]) || S_backref_value(p) < RExC_npar)
11904 { /* Not to be treated as an octal constant, go
11911 I32 flags = PERL_SCAN_SILENT_ILLDIGIT;
11913 ender = grok_oct(p, &numlen, &flags, NULL);
11914 if (ender > 0xff) {
11918 if (SIZE_ONLY /* like \08, \178 */
11921 && isDIGIT(*p) && ckWARN(WARN_REGEXP))
11923 reg_warn_non_literal_string(
11925 form_short_octal_warning(p, numlen));
11928 if (PL_encoding && ender < 0x100)
11929 goto recode_encoding;
11932 if (! RExC_override_recoding) {
11933 SV* enc = PL_encoding;
11934 ender = reg_recode((const char)(U8)ender, &enc);
11935 if (!enc && SIZE_ONLY)
11936 ckWARNreg(p, "Invalid escape in the specified encoding");
11942 FAIL("Trailing \\");
11945 if (!SIZE_ONLY&& isALPHANUMERIC(*p)) {
11946 /* Include any { following the alpha to emphasize
11947 * that it could be part of an escape at some point
11949 int len = (isALPHA(*p) && *(p + 1) == '{') ? 2 : 1;
11950 ckWARN3reg(p + len, "Unrecognized escape \\%.*s passed through", len, p);
11952 goto normal_default;
11953 } /* End of switch on '\' */
11955 default: /* A literal character */
11958 && RExC_flags & RXf_PMf_EXTENDED
11959 && ckWARN_d(WARN_DEPRECATED)
11960 && is_PATWS_non_low_safe(p, RExC_end, UTF))
11962 vWARN_dep(p + ((UTF) ? UTF8SKIP(p) : 1),
11963 "Escape literal pattern white space under /x");
11967 if (UTF8_IS_START(*p) && UTF) {
11969 ender = utf8n_to_uvchr((U8*)p, RExC_end - p,
11970 &numlen, UTF8_ALLOW_DEFAULT);
11976 } /* End of switch on the literal */
11978 /* Here, have looked at the literal character and <ender>
11979 * contains its ordinal, <p> points to the character after it
11982 if ( RExC_flags & RXf_PMf_EXTENDED)
11983 p = regwhite( pRExC_state, p );
11985 /* If the next thing is a quantifier, it applies to this
11986 * character only, which means that this character has to be in
11987 * its own node and can't just be appended to the string in an
11988 * existing node, so if there are already other characters in
11989 * the node, close the node with just them, and set up to do
11990 * this character again next time through, when it will be the
11991 * only thing in its new node */
11992 if ((next_is_quantifier = (p < RExC_end && ISMULT2(p))) && len)
11998 if (! FOLD /* The simple case, just append the literal */
11999 || (LOC /* Also don't fold for tricky chars under /l */
12000 && is_PROBLEMATIC_LOCALE_FOLD_cp(ender)))
12003 const STRLEN unilen = reguni(pRExC_state, ender, s);
12009 /* The loop increments <len> each time, as all but this
12010 * path (and one other) through it add a single byte to
12011 * the EXACTish node. But this one has changed len to
12012 * be the correct final value, so subtract one to
12013 * cancel out the increment that follows */
12017 REGC((char)ender, s++);
12020 /* Can get here if folding only if is one of the /l
12021 * characters whose fold depends on the locale. The
12022 * occurrence of any of these indicate that we can't
12023 * simplify things */
12025 maybe_exact = FALSE;
12026 maybe_exactfu = FALSE;
12031 /* See comments for join_exact() as to why we fold this
12032 * non-UTF at compile time */
12033 || (node_type == EXACTFU
12034 && ender == LATIN_SMALL_LETTER_SHARP_S)))
12036 /* Here, are folding and are not UTF-8 encoded; therefore
12037 * the character must be in the range 0-255, and is not /l
12038 * (Not /l because we already handled these under /l in
12039 * is_PROBLEMATIC_LOCALE_FOLD_cp */
12040 if (IS_IN_SOME_FOLD_L1(ender)) {
12041 maybe_exact = FALSE;
12043 /* See if the character's fold differs between /d and
12044 * /u. This includes the multi-char fold SHARP S to
12047 && (PL_fold[ender] != PL_fold_latin1[ender]
12048 || ender == LATIN_SMALL_LETTER_SHARP_S
12050 && isARG2_lower_or_UPPER_ARG1('s', ender)
12051 && isARG2_lower_or_UPPER_ARG1('s',
12054 maybe_exactfu = FALSE;
12058 /* Even when folding, we store just the input character, as
12059 * we have an array that finds its fold quickly */
12060 *(s++) = (char) ender;
12062 else { /* FOLD and UTF */
12063 /* Unlike the non-fold case, we do actually have to
12064 * calculate the results here in pass 1. This is for two
12065 * reasons, the folded length may be longer than the
12066 * unfolded, and we have to calculate how many EXACTish
12067 * nodes it will take; and we may run out of room in a node
12068 * in the middle of a potential multi-char fold, and have
12069 * to back off accordingly. (Hence we can't use REGC for
12070 * the simple case just below.) */
12073 if (isASCII(ender)) {
12074 folded = toFOLD(ender);
12075 *(s)++ = (U8) folded;
12080 folded = _to_uni_fold_flags(
12084 FOLD_FLAGS_FULL | ((ASCII_FOLD_RESTRICTED)
12085 ? FOLD_FLAGS_NOMIX_ASCII
12089 /* The loop increments <len> each time, as all but this
12090 * path (and one other) through it add a single byte to
12091 * the EXACTish node. But this one has changed len to
12092 * be the correct final value, so subtract one to
12093 * cancel out the increment that follows */
12094 len += foldlen - 1;
12096 /* If this node only contains non-folding code points so
12097 * far, see if this new one is also non-folding */
12099 if (folded != ender) {
12100 maybe_exact = FALSE;
12103 /* Here the fold is the original; we have to check
12104 * further to see if anything folds to it */
12105 if (_invlist_contains_cp(PL_utf8_foldable,
12108 maybe_exact = FALSE;
12115 if (next_is_quantifier) {
12117 /* Here, the next input is a quantifier, and to get here,
12118 * the current character is the only one in the node.
12119 * Also, here <len> doesn't include the final byte for this
12125 } /* End of loop through literal characters */
12127 /* Here we have either exhausted the input or ran out of room in
12128 * the node. (If we encountered a character that can't be in the
12129 * node, transfer is made directly to <loopdone>, and so we
12130 * wouldn't have fallen off the end of the loop.) In the latter
12131 * case, we artificially have to split the node into two, because
12132 * we just don't have enough space to hold everything. This
12133 * creates a problem if the final character participates in a
12134 * multi-character fold in the non-final position, as a match that
12135 * should have occurred won't, due to the way nodes are matched,
12136 * and our artificial boundary. So back off until we find a non-
12137 * problematic character -- one that isn't at the beginning or
12138 * middle of such a fold. (Either it doesn't participate in any
12139 * folds, or appears only in the final position of all the folds it
12140 * does participate in.) A better solution with far fewer false
12141 * positives, and that would fill the nodes more completely, would
12142 * be to actually have available all the multi-character folds to
12143 * test against, and to back-off only far enough to be sure that
12144 * this node isn't ending with a partial one. <upper_parse> is set
12145 * further below (if we need to reparse the node) to include just
12146 * up through that final non-problematic character that this code
12147 * identifies, so when it is set to less than the full node, we can
12148 * skip the rest of this */
12149 if (FOLD && p < RExC_end && upper_parse == MAX_NODE_STRING_SIZE) {
12151 const STRLEN full_len = len;
12153 assert(len >= MAX_NODE_STRING_SIZE);
12155 /* Here, <s> points to the final byte of the final character.
12156 * Look backwards through the string until find a non-
12157 * problematic character */
12161 /* This has no multi-char folds to non-UTF characters */
12162 if (ASCII_FOLD_RESTRICTED) {
12166 while (--s >= s0 && IS_NON_FINAL_FOLD(*s)) { }
12170 if (! PL_NonL1NonFinalFold) {
12171 PL_NonL1NonFinalFold = _new_invlist_C_array(
12172 NonL1_Perl_Non_Final_Folds_invlist);
12175 /* Point to the first byte of the final character */
12176 s = (char *) utf8_hop((U8 *) s, -1);
12178 while (s >= s0) { /* Search backwards until find
12179 non-problematic char */
12180 if (UTF8_IS_INVARIANT(*s)) {
12182 /* There are no ascii characters that participate
12183 * in multi-char folds under /aa. In EBCDIC, the
12184 * non-ascii invariants are all control characters,
12185 * so don't ever participate in any folds. */
12186 if (ASCII_FOLD_RESTRICTED
12187 || ! IS_NON_FINAL_FOLD(*s))
12192 else if (UTF8_IS_DOWNGRADEABLE_START(*s)) {
12193 if (! IS_NON_FINAL_FOLD(TWO_BYTE_UTF8_TO_NATIVE(
12199 else if (! _invlist_contains_cp(
12200 PL_NonL1NonFinalFold,
12201 valid_utf8_to_uvchr((U8 *) s, NULL)))
12206 /* Here, the current character is problematic in that
12207 * it does occur in the non-final position of some
12208 * fold, so try the character before it, but have to
12209 * special case the very first byte in the string, so
12210 * we don't read outside the string */
12211 s = (s == s0) ? s -1 : (char *) utf8_hop((U8 *) s, -1);
12212 } /* End of loop backwards through the string */
12214 /* If there were only problematic characters in the string,
12215 * <s> will point to before s0, in which case the length
12216 * should be 0, otherwise include the length of the
12217 * non-problematic character just found */
12218 len = (s < s0) ? 0 : s - s0 + UTF8SKIP(s);
12221 /* Here, have found the final character, if any, that is
12222 * non-problematic as far as ending the node without splitting
12223 * it across a potential multi-char fold. <len> contains the
12224 * number of bytes in the node up-to and including that
12225 * character, or is 0 if there is no such character, meaning
12226 * the whole node contains only problematic characters. In
12227 * this case, give up and just take the node as-is. We can't
12232 /* If the node ends in an 's' we make sure it stays EXACTF,
12233 * as if it turns into an EXACTFU, it could later get
12234 * joined with another 's' that would then wrongly match
12236 if (maybe_exactfu && isARG2_lower_or_UPPER_ARG1('s', ender))
12238 maybe_exactfu = FALSE;
12242 /* Here, the node does contain some characters that aren't
12243 * problematic. If one such is the final character in the
12244 * node, we are done */
12245 if (len == full_len) {
12248 else if (len + ((UTF) ? UTF8SKIP(s) : 1) == full_len) {
12250 /* If the final character is problematic, but the
12251 * penultimate is not, back-off that last character to
12252 * later start a new node with it */
12257 /* Here, the final non-problematic character is earlier
12258 * in the input than the penultimate character. What we do
12259 * is reparse from the beginning, going up only as far as
12260 * this final ok one, thus guaranteeing that the node ends
12261 * in an acceptable character. The reason we reparse is
12262 * that we know how far in the character is, but we don't
12263 * know how to correlate its position with the input parse.
12264 * An alternate implementation would be to build that
12265 * correlation as we go along during the original parse,
12266 * but that would entail extra work for every node, whereas
12267 * this code gets executed only when the string is too
12268 * large for the node, and the final two characters are
12269 * problematic, an infrequent occurrence. Yet another
12270 * possible strategy would be to save the tail of the
12271 * string, and the next time regatom is called, initialize
12272 * with that. The problem with this is that unless you
12273 * back off one more character, you won't be guaranteed
12274 * regatom will get called again, unless regbranch,
12275 * regpiece ... are also changed. If you do back off that
12276 * extra character, so that there is input guaranteed to
12277 * force calling regatom, you can't handle the case where
12278 * just the first character in the node is acceptable. I
12279 * (khw) decided to try this method which doesn't have that
12280 * pitfall; if performance issues are found, we can do a
12281 * combination of the current approach plus that one */
12287 } /* End of verifying node ends with an appropriate char */
12289 loopdone: /* Jumped to when encounters something that shouldn't be in
12292 /* I (khw) don't know if you can get here with zero length, but the
12293 * old code handled this situation by creating a zero-length EXACT
12294 * node. Might as well be NOTHING instead */
12300 /* If 'maybe_exact' is still set here, means there are no
12301 * code points in the node that participate in folds;
12302 * similarly for 'maybe_exactfu' and code points that match
12303 * differently depending on UTF8ness of the target string
12304 * (for /u), or depending on locale for /l */
12308 else if (maybe_exactfu) {
12312 alloc_maybe_populate_EXACT(pRExC_state, ret, flagp, len, ender,
12313 FALSE /* Don't look to see if could
12314 be turned into an EXACT
12315 node, as we have already
12320 RExC_parse = p - 1;
12321 Set_Node_Cur_Length(ret, parse_start);
12322 nextchar(pRExC_state);
12324 /* len is STRLEN which is unsigned, need to copy to signed */
12327 vFAIL("Internal disaster");
12330 } /* End of label 'defchar:' */
12332 } /* End of giant switch on input character */
12338 S_regwhite( RExC_state_t *pRExC_state, char *p )
12340 const char *e = RExC_end;
12342 PERL_ARGS_ASSERT_REGWHITE;
12347 else if (*p == '#') {
12350 if (*p++ == '\n') {
12356 RExC_seen |= REG_RUN_ON_COMMENT_SEEN;
12365 S_regpatws( RExC_state_t *pRExC_state, char *p , const bool recognize_comment )
12367 /* Returns the next non-pattern-white space, non-comment character (the
12368 * latter only if 'recognize_comment is true) in the string p, which is
12369 * ended by RExC_end. If there is no line break ending a comment,
12370 * RExC_seen has added the REG_RUN_ON_COMMENT_SEEN flag; */
12371 const char *e = RExC_end;
12373 PERL_ARGS_ASSERT_REGPATWS;
12377 if ((len = is_PATWS_safe(p, e, UTF))) {
12380 else if (recognize_comment && *p == '#') {
12384 if (is_LNBREAK_safe(p, e, UTF)) {
12390 RExC_seen |= REG_RUN_ON_COMMENT_SEEN;
12399 S_populate_ANYOF_from_invlist(pTHX_ regnode *node, SV** invlist_ptr)
12401 /* Uses the inversion list '*invlist_ptr' to populate the ANYOF 'node'. It
12402 * sets up the bitmap and any flags, removing those code points from the
12403 * inversion list, setting it to NULL should it become completely empty */
12405 PERL_ARGS_ASSERT_POPULATE_ANYOF_FROM_INVLIST;
12406 assert(PL_regkind[OP(node)] == ANYOF);
12408 ANYOF_BITMAP_ZERO(node);
12409 if (*invlist_ptr) {
12411 /* This gets set if we actually need to modify things */
12412 bool change_invlist = FALSE;
12416 /* Start looking through *invlist_ptr */
12417 invlist_iterinit(*invlist_ptr);
12418 while (invlist_iternext(*invlist_ptr, &start, &end)) {
12422 if (end == UV_MAX && start <= 256) {
12423 ANYOF_FLAGS(node) |= ANYOF_ABOVE_LATIN1_ALL;
12425 else if (end >= 256) {
12426 ANYOF_FLAGS(node) |= ANYOF_UTF8;
12429 /* Quit if are above what we should change */
12434 change_invlist = TRUE;
12436 /* Set all the bits in the range, up to the max that we are doing */
12437 high = (end < 255) ? end : 255;
12438 for (i = start; i <= (int) high; i++) {
12439 if (! ANYOF_BITMAP_TEST(node, i)) {
12440 ANYOF_BITMAP_SET(node, i);
12444 invlist_iterfinish(*invlist_ptr);
12446 /* Done with loop; remove any code points that are in the bitmap from
12447 * *invlist_ptr; similarly for code points above latin1 if we have a
12448 * flag to match all of them anyways */
12449 if (change_invlist) {
12450 _invlist_subtract(*invlist_ptr, PL_Latin1, invlist_ptr);
12452 if (ANYOF_FLAGS(node) & ANYOF_ABOVE_LATIN1_ALL) {
12453 _invlist_intersection(*invlist_ptr, PL_Latin1, invlist_ptr);
12456 /* If have completely emptied it, remove it completely */
12457 if (_invlist_len(*invlist_ptr) == 0) {
12458 SvREFCNT_dec_NN(*invlist_ptr);
12459 *invlist_ptr = NULL;
12464 /* Parse POSIX character classes: [[:foo:]], [[=foo=]], [[.foo.]].
12465 Character classes ([:foo:]) can also be negated ([:^foo:]).
12466 Returns a named class id (ANYOF_XXX) if successful, -1 otherwise.
12467 Equivalence classes ([=foo=]) and composites ([.foo.]) are parsed,
12468 but trigger failures because they are currently unimplemented. */
12470 #define POSIXCC_DONE(c) ((c) == ':')
12471 #define POSIXCC_NOTYET(c) ((c) == '=' || (c) == '.')
12472 #define POSIXCC(c) (POSIXCC_DONE(c) || POSIXCC_NOTYET(c))
12474 PERL_STATIC_INLINE I32
12475 S_regpposixcc(pTHX_ RExC_state_t *pRExC_state, I32 value, const bool strict)
12478 I32 namedclass = OOB_NAMEDCLASS;
12480 PERL_ARGS_ASSERT_REGPPOSIXCC;
12482 if (value == '[' && RExC_parse + 1 < RExC_end &&
12483 /* I smell either [: or [= or [. -- POSIX has been here, right? */
12484 POSIXCC(UCHARAT(RExC_parse)))
12486 const char c = UCHARAT(RExC_parse);
12487 char* const s = RExC_parse++;
12489 while (RExC_parse < RExC_end && UCHARAT(RExC_parse) != c)
12491 if (RExC_parse == RExC_end) {
12494 /* Try to give a better location for the error (than the end of
12495 * the string) by looking for the matching ']' */
12497 while (RExC_parse < RExC_end && UCHARAT(RExC_parse) != ']') {
12500 vFAIL2("Unmatched '%c' in POSIX class", c);
12502 /* Grandfather lone [:, [=, [. */
12506 const char* const t = RExC_parse++; /* skip over the c */
12509 if (UCHARAT(RExC_parse) == ']') {
12510 const char *posixcc = s + 1;
12511 RExC_parse++; /* skip over the ending ] */
12514 const I32 complement = *posixcc == '^' ? *posixcc++ : 0;
12515 const I32 skip = t - posixcc;
12517 /* Initially switch on the length of the name. */
12520 if (memEQ(posixcc, "word", 4)) /* this is not POSIX,
12521 this is the Perl \w
12523 namedclass = ANYOF_WORDCHAR;
12526 /* Names all of length 5. */
12527 /* alnum alpha ascii blank cntrl digit graph lower
12528 print punct space upper */
12529 /* Offset 4 gives the best switch position. */
12530 switch (posixcc[4]) {
12532 if (memEQ(posixcc, "alph", 4)) /* alpha */
12533 namedclass = ANYOF_ALPHA;
12536 if (memEQ(posixcc, "spac", 4)) /* space */
12537 namedclass = ANYOF_PSXSPC;
12540 if (memEQ(posixcc, "grap", 4)) /* graph */
12541 namedclass = ANYOF_GRAPH;
12544 if (memEQ(posixcc, "asci", 4)) /* ascii */
12545 namedclass = ANYOF_ASCII;
12548 if (memEQ(posixcc, "blan", 4)) /* blank */
12549 namedclass = ANYOF_BLANK;
12552 if (memEQ(posixcc, "cntr", 4)) /* cntrl */
12553 namedclass = ANYOF_CNTRL;
12556 if (memEQ(posixcc, "alnu", 4)) /* alnum */
12557 namedclass = ANYOF_ALPHANUMERIC;
12560 if (memEQ(posixcc, "lowe", 4)) /* lower */
12561 namedclass = (FOLD) ? ANYOF_CASED : ANYOF_LOWER;
12562 else if (memEQ(posixcc, "uppe", 4)) /* upper */
12563 namedclass = (FOLD) ? ANYOF_CASED : ANYOF_UPPER;
12566 if (memEQ(posixcc, "digi", 4)) /* digit */
12567 namedclass = ANYOF_DIGIT;
12568 else if (memEQ(posixcc, "prin", 4)) /* print */
12569 namedclass = ANYOF_PRINT;
12570 else if (memEQ(posixcc, "punc", 4)) /* punct */
12571 namedclass = ANYOF_PUNCT;
12576 if (memEQ(posixcc, "xdigit", 6))
12577 namedclass = ANYOF_XDIGIT;
12581 if (namedclass == OOB_NAMEDCLASS)
12583 "POSIX class [:%"UTF8f":] unknown",
12584 UTF8fARG(UTF, t - s - 1, s + 1));
12586 /* The #defines are structured so each complement is +1 to
12587 * the normal one */
12591 assert (posixcc[skip] == ':');
12592 assert (posixcc[skip+1] == ']');
12593 } else if (!SIZE_ONLY) {
12594 /* [[=foo=]] and [[.foo.]] are still future. */
12596 /* adjust RExC_parse so the warning shows after
12597 the class closes */
12598 while (UCHARAT(RExC_parse) && UCHARAT(RExC_parse) != ']')
12600 vFAIL3("POSIX syntax [%c %c] is reserved for future extensions", c, c);
12603 /* Maternal grandfather:
12604 * "[:" ending in ":" but not in ":]" */
12606 vFAIL("Unmatched '[' in POSIX class");
12609 /* Grandfather lone [:, [=, [. */
12619 S_could_it_be_a_POSIX_class(pTHX_ RExC_state_t *pRExC_state)
12621 /* This applies some heuristics at the current parse position (which should
12622 * be at a '[') to see if what follows might be intended to be a [:posix:]
12623 * class. It returns true if it really is a posix class, of course, but it
12624 * also can return true if it thinks that what was intended was a posix
12625 * class that didn't quite make it.
12627 * It will return true for
12629 * [:alphanumerics] (as long as the ] isn't followed immediately by a
12630 * ')' indicating the end of the (?[
12631 * [:any garbage including %^&$ punctuation:]
12633 * This is designed to be called only from S_handle_regex_sets; it could be
12634 * easily adapted to be called from the spot at the beginning of regclass()
12635 * that checks to see in a normal bracketed class if the surrounding []
12636 * have been omitted ([:word:] instead of [[:word:]]). But doing so would
12637 * change long-standing behavior, so I (khw) didn't do that */
12638 char* p = RExC_parse + 1;
12639 char first_char = *p;
12641 PERL_ARGS_ASSERT_COULD_IT_BE_A_POSIX_CLASS;
12643 assert(*(p - 1) == '[');
12645 if (! POSIXCC(first_char)) {
12650 while (p < RExC_end && isWORDCHAR(*p)) p++;
12652 if (p >= RExC_end) {
12656 if (p - RExC_parse > 2 /* Got at least 1 word character */
12657 && (*p == first_char
12658 || (*p == ']' && p + 1 < RExC_end && *(p + 1) != ')')))
12663 p = (char *) memchr(RExC_parse, ']', RExC_end - RExC_parse);
12666 && p - RExC_parse > 2 /* [:] evaluates to colon;
12667 [::] is a bad posix class. */
12668 && first_char == *(p - 1));
12672 S_handle_regex_sets(pTHX_ RExC_state_t *pRExC_state, SV** return_invlist,
12673 I32 *flagp, U32 depth,
12674 char * const oregcomp_parse)
12676 /* Handle the (?[...]) construct to do set operations */
12679 UV start, end; /* End points of code point ranges */
12681 char *save_end, *save_parse;
12686 const bool save_fold = FOLD;
12688 GET_RE_DEBUG_FLAGS_DECL;
12690 PERL_ARGS_ASSERT_HANDLE_REGEX_SETS;
12693 vFAIL("(?[...]) not valid in locale");
12695 RExC_uni_semantics = 1;
12697 /* This will return only an ANYOF regnode, or (unlikely) something smaller
12698 * (such as EXACT). Thus we can skip most everything if just sizing. We
12699 * call regclass to handle '[]' so as to not have to reinvent its parsing
12700 * rules here (throwing away the size it computes each time). And, we exit
12701 * upon an unescaped ']' that isn't one ending a regclass. To do both
12702 * these things, we need to realize that something preceded by a backslash
12703 * is escaped, so we have to keep track of backslashes */
12705 UV depth = 0; /* how many nested (?[...]) constructs */
12707 Perl_ck_warner_d(aTHX_
12708 packWARN(WARN_EXPERIMENTAL__REGEX_SETS),
12709 "The regex_sets feature is experimental" REPORT_LOCATION,
12710 UTF8fARG(UTF, (RExC_parse - RExC_precomp), RExC_precomp),
12712 RExC_end - RExC_start - (RExC_parse - RExC_precomp),
12713 RExC_precomp + (RExC_parse - RExC_precomp)));
12715 while (RExC_parse < RExC_end) {
12716 SV* current = NULL;
12717 RExC_parse = regpatws(pRExC_state, RExC_parse,
12718 TRUE); /* means recognize comments */
12719 switch (*RExC_parse) {
12721 if (RExC_parse[1] == '[') depth++, RExC_parse++;
12726 /* Skip the next byte (which could cause us to end up in
12727 * the middle of a UTF-8 character, but since none of those
12728 * are confusable with anything we currently handle in this
12729 * switch (invariants all), it's safe. We'll just hit the
12730 * default: case next time and keep on incrementing until
12731 * we find one of the invariants we do handle. */
12736 /* If this looks like it is a [:posix:] class, leave the
12737 * parse pointer at the '[' to fool regclass() into
12738 * thinking it is part of a '[[:posix:]]'. That function
12739 * will use strict checking to force a syntax error if it
12740 * doesn't work out to a legitimate class */
12741 bool is_posix_class
12742 = could_it_be_a_POSIX_class(pRExC_state);
12743 if (! is_posix_class) {
12747 /* regclass() can only return RESTART_UTF8 if multi-char
12748 folds are allowed. */
12749 if (!regclass(pRExC_state, flagp,depth+1,
12750 is_posix_class, /* parse the whole char
12751 class only if not a
12753 FALSE, /* don't allow multi-char folds */
12754 TRUE, /* silence non-portable warnings. */
12756 FAIL2("panic: regclass returned NULL to handle_sets, flags=%#"UVxf"",
12759 /* function call leaves parse pointing to the ']', except
12760 * if we faked it */
12761 if (is_posix_class) {
12765 SvREFCNT_dec(current); /* In case it returned something */
12770 if (depth--) break;
12772 if (RExC_parse < RExC_end
12773 && *RExC_parse == ')')
12775 node = reganode(pRExC_state, ANYOF, 0);
12776 RExC_size += ANYOF_SKIP;
12777 nextchar(pRExC_state);
12778 Set_Node_Length(node,
12779 RExC_parse - oregcomp_parse + 1); /* MJD */
12788 FAIL("Syntax error in (?[...])");
12791 /* Pass 2 only after this. Everything in this construct is a
12792 * metacharacter. Operands begin with either a '\' (for an escape
12793 * sequence), or a '[' for a bracketed character class. Any other
12794 * character should be an operator, or parenthesis for grouping. Both
12795 * types of operands are handled by calling regclass() to parse them. It
12796 * is called with a parameter to indicate to return the computed inversion
12797 * list. The parsing here is implemented via a stack. Each entry on the
12798 * stack is a single character representing one of the operators, or the
12799 * '('; or else a pointer to an operand inversion list. */
12801 #define IS_OPERAND(a) (! SvIOK(a))
12803 /* The stack starts empty. It is a syntax error if the first thing parsed
12804 * is a binary operator; everything else is pushed on the stack. When an
12805 * operand is parsed, the top of the stack is examined. If it is a binary
12806 * operator, the item before it should be an operand, and both are replaced
12807 * by the result of doing that operation on the new operand and the one on
12808 * the stack. Thus a sequence of binary operands is reduced to a single
12809 * one before the next one is parsed.
12811 * A unary operator may immediately follow a binary in the input, for
12814 * When an operand is parsed and the top of the stack is a unary operator,
12815 * the operation is performed, and then the stack is rechecked to see if
12816 * this new operand is part of a binary operation; if so, it is handled as
12819 * A '(' is simply pushed on the stack; it is valid only if the stack is
12820 * empty, or the top element of the stack is an operator or another '('
12821 * (for which the parenthesized expression will become an operand). By the
12822 * time the corresponding ')' is parsed everything in between should have
12823 * been parsed and evaluated to a single operand (or else is a syntax
12824 * error), and is handled as a regular operand */
12826 sv_2mortal((SV *)(stack = newAV()));
12828 while (RExC_parse < RExC_end) {
12829 I32 top_index = av_tindex(stack);
12831 SV* current = NULL;
12833 /* Skip white space */
12834 RExC_parse = regpatws(pRExC_state, RExC_parse,
12835 TRUE); /* means recognize comments */
12836 if (RExC_parse >= RExC_end) {
12837 Perl_croak(aTHX_ "panic: Read past end of '(?[ ])'");
12839 if ((curchar = UCHARAT(RExC_parse)) == ']') {
12846 if (av_tindex(stack) >= 0 /* This makes sure that we can
12847 safely subtract 1 from
12848 RExC_parse in the next clause.
12849 If we have something on the
12850 stack, we have parsed something
12852 && UCHARAT(RExC_parse - 1) == '('
12853 && RExC_parse < RExC_end)
12855 /* If is a '(?', could be an embedded '(?flags:(?[...])'.
12856 * This happens when we have some thing like
12858 * my $thai_or_lao = qr/(?[ \p{Thai} + \p{Lao} ])/;
12860 * qr/(?[ \p{Digit} & $thai_or_lao ])/;
12862 * Here we would be handling the interpolated
12863 * '$thai_or_lao'. We handle this by a recursive call to
12864 * ourselves which returns the inversion list the
12865 * interpolated expression evaluates to. We use the flags
12866 * from the interpolated pattern. */
12867 U32 save_flags = RExC_flags;
12868 const char * const save_parse = ++RExC_parse;
12870 parse_lparen_question_flags(pRExC_state);
12872 if (RExC_parse == save_parse /* Makes sure there was at
12873 least one flag (or this
12874 embedding wasn't compiled)
12876 || RExC_parse >= RExC_end - 4
12877 || UCHARAT(RExC_parse) != ':'
12878 || UCHARAT(++RExC_parse) != '('
12879 || UCHARAT(++RExC_parse) != '?'
12880 || UCHARAT(++RExC_parse) != '[')
12883 /* In combination with the above, this moves the
12884 * pointer to the point just after the first erroneous
12885 * character (or if there are no flags, to where they
12886 * should have been) */
12887 if (RExC_parse >= RExC_end - 4) {
12888 RExC_parse = RExC_end;
12890 else if (RExC_parse != save_parse) {
12891 RExC_parse += (UTF) ? UTF8SKIP(RExC_parse) : 1;
12893 vFAIL("Expecting '(?flags:(?[...'");
12896 (void) handle_regex_sets(pRExC_state, ¤t, flagp,
12897 depth+1, oregcomp_parse);
12899 /* Here, 'current' contains the embedded expression's
12900 * inversion list, and RExC_parse points to the trailing
12901 * ']'; the next character should be the ')' which will be
12902 * paired with the '(' that has been put on the stack, so
12903 * the whole embedded expression reduces to '(operand)' */
12906 RExC_flags = save_flags;
12907 goto handle_operand;
12912 RExC_parse += (UTF) ? UTF8SKIP(RExC_parse) : 1;
12913 vFAIL("Unexpected character");
12916 /* regclass() can only return RESTART_UTF8 if multi-char
12917 folds are allowed. */
12918 if (!regclass(pRExC_state, flagp,depth+1,
12919 TRUE, /* means parse just the next thing */
12920 FALSE, /* don't allow multi-char folds */
12921 FALSE, /* don't silence non-portable warnings. */
12923 FAIL2("panic: regclass returned NULL to handle_sets, flags=%#"UVxf"",
12925 /* regclass() will return with parsing just the \ sequence,
12926 * leaving the parse pointer at the next thing to parse */
12928 goto handle_operand;
12930 case '[': /* Is a bracketed character class */
12932 bool is_posix_class = could_it_be_a_POSIX_class(pRExC_state);
12934 if (! is_posix_class) {
12938 /* regclass() can only return RESTART_UTF8 if multi-char
12939 folds are allowed. */
12940 if(!regclass(pRExC_state, flagp,depth+1,
12941 is_posix_class, /* parse the whole char class
12942 only if not a posix class */
12943 FALSE, /* don't allow multi-char folds */
12944 FALSE, /* don't silence non-portable warnings. */
12946 FAIL2("panic: regclass returned NULL to handle_sets, flags=%#"UVxf"",
12948 /* function call leaves parse pointing to the ']', except if we
12950 if (is_posix_class) {
12954 goto handle_operand;
12963 || ( ! (top_ptr = av_fetch(stack, top_index, FALSE)))
12964 || ! IS_OPERAND(*top_ptr))
12967 vFAIL2("Unexpected binary operator '%c' with no preceding operand", curchar);
12969 av_push(stack, newSVuv(curchar));
12973 av_push(stack, newSVuv(curchar));
12977 if (top_index >= 0) {
12978 top_ptr = av_fetch(stack, top_index, FALSE);
12980 if (IS_OPERAND(*top_ptr)) {
12982 vFAIL("Unexpected '(' with no preceding operator");
12985 av_push(stack, newSVuv(curchar));
12992 || ! (current = av_pop(stack))
12993 || ! IS_OPERAND(current)
12994 || ! (lparen = av_pop(stack))
12995 || IS_OPERAND(lparen)
12996 || SvUV(lparen) != '(')
12998 SvREFCNT_dec(current);
13000 vFAIL("Unexpected ')'");
13003 SvREFCNT_dec_NN(lparen);
13010 /* Here, we have an operand to process, in 'current' */
13012 if (top_index < 0) { /* Just push if stack is empty */
13013 av_push(stack, current);
13016 SV* top = av_pop(stack);
13018 char current_operator;
13020 if (IS_OPERAND(top)) {
13021 SvREFCNT_dec_NN(top);
13022 SvREFCNT_dec_NN(current);
13023 vFAIL("Operand with no preceding operator");
13025 current_operator = (char) SvUV(top);
13026 switch (current_operator) {
13027 case '(': /* Push the '(' back on followed by the new
13029 av_push(stack, top);
13030 av_push(stack, current);
13031 SvREFCNT_inc(top); /* Counters the '_dec' done
13032 just after the 'break', so
13033 it doesn't get wrongly freed
13038 _invlist_invert(current);
13040 /* Unlike binary operators, the top of the stack,
13041 * now that this unary one has been popped off, may
13042 * legally be an operator, and we now have operand
13045 SvREFCNT_dec_NN(top);
13046 goto handle_operand;
13049 prev = av_pop(stack);
13050 _invlist_intersection(prev,
13053 av_push(stack, current);
13058 prev = av_pop(stack);
13059 _invlist_union(prev, current, ¤t);
13060 av_push(stack, current);
13064 prev = av_pop(stack);;
13065 _invlist_subtract(prev, current, ¤t);
13066 av_push(stack, current);
13069 case '^': /* The union minus the intersection */
13075 prev = av_pop(stack);
13076 _invlist_union(prev, current, &u);
13077 _invlist_intersection(prev, current, &i);
13078 /* _invlist_subtract will overwrite current
13079 without freeing what it already contains */
13081 _invlist_subtract(u, i, ¤t);
13082 av_push(stack, current);
13083 SvREFCNT_dec_NN(i);
13084 SvREFCNT_dec_NN(u);
13085 SvREFCNT_dec_NN(element);
13090 Perl_croak(aTHX_ "panic: Unexpected item on '(?[ ])' stack");
13092 SvREFCNT_dec_NN(top);
13093 SvREFCNT_dec(prev);
13097 RExC_parse += (UTF) ? UTF8SKIP(RExC_parse) : 1;
13100 if (av_tindex(stack) < 0 /* Was empty */
13101 || ((final = av_pop(stack)) == NULL)
13102 || ! IS_OPERAND(final)
13103 || av_tindex(stack) >= 0) /* More left on stack */
13105 vFAIL("Incomplete expression within '(?[ ])'");
13108 /* Here, 'final' is the resultant inversion list from evaluating the
13109 * expression. Return it if so requested */
13110 if (return_invlist) {
13111 *return_invlist = final;
13115 /* Otherwise generate a resultant node, based on 'final'. regclass() is
13116 * expecting a string of ranges and individual code points */
13117 invlist_iterinit(final);
13118 result_string = newSVpvs("");
13119 while (invlist_iternext(final, &start, &end)) {
13120 if (start == end) {
13121 Perl_sv_catpvf(aTHX_ result_string, "\\x{%"UVXf"}", start);
13124 Perl_sv_catpvf(aTHX_ result_string, "\\x{%"UVXf"}-\\x{%"UVXf"}",
13129 save_parse = RExC_parse;
13130 RExC_parse = SvPV(result_string, len);
13131 save_end = RExC_end;
13132 RExC_end = RExC_parse + len;
13134 /* We turn off folding around the call, as the class we have constructed
13135 * already has all folding taken into consideration, and we don't want
13136 * regclass() to add to that */
13137 RExC_flags &= ~RXf_PMf_FOLD;
13138 /* regclass() can only return RESTART_UTF8 if multi-char folds are allowed.
13140 node = regclass(pRExC_state, flagp,depth+1,
13141 FALSE, /* means parse the whole char class */
13142 FALSE, /* don't allow multi-char folds */
13143 TRUE, /* silence non-portable warnings. The above may very
13144 well have generated non-portable code points, but
13145 they're valid on this machine */
13148 FAIL2("panic: regclass returned NULL to handle_sets, flags=%#"UVxf,
13151 RExC_flags |= RXf_PMf_FOLD;
13153 RExC_parse = save_parse + 1;
13154 RExC_end = save_end;
13155 SvREFCNT_dec_NN(final);
13156 SvREFCNT_dec_NN(result_string);
13158 nextchar(pRExC_state);
13159 Set_Node_Length(node, RExC_parse - oregcomp_parse + 1); /* MJD */
13164 /* The names of properties whose definitions are not known at compile time are
13165 * stored in this SV, after a constant heading. So if the length has been
13166 * changed since initialization, then there is a run-time definition. */
13167 #define HAS_NONLOCALE_RUNTIME_PROPERTY_DEFINITION \
13168 (SvCUR(listsv) != initial_listsv_len)
13171 S_regclass(pTHX_ RExC_state_t *pRExC_state, I32 *flagp, U32 depth,
13172 const bool stop_at_1, /* Just parse the next thing, don't
13173 look for a full character class */
13174 bool allow_multi_folds,
13175 const bool silence_non_portable, /* Don't output warnings
13178 SV** ret_invlist) /* Return an inversion list, not a node */
13180 /* parse a bracketed class specification. Most of these will produce an
13181 * ANYOF node; but something like [a] will produce an EXACT node; [aA], an
13182 * EXACTFish node; [[:ascii:]], a POSIXA node; etc. It is more complex
13183 * under /i with multi-character folds: it will be rewritten following the
13184 * paradigm of this example, where the <multi-fold>s are characters which
13185 * fold to multiple character sequences:
13186 * /[abc\x{multi-fold1}def\x{multi-fold2}ghi]/i
13187 * gets effectively rewritten as:
13188 * /(?:\x{multi-fold1}|\x{multi-fold2}|[abcdefghi]/i
13189 * reg() gets called (recursively) on the rewritten version, and this
13190 * function will return what it constructs. (Actually the <multi-fold>s
13191 * aren't physically removed from the [abcdefghi], it's just that they are
13192 * ignored in the recursion by means of a flag:
13193 * <RExC_in_multi_char_class>.)
13195 * ANYOF nodes contain a bit map for the first 256 characters, with the
13196 * corresponding bit set if that character is in the list. For characters
13197 * above 255, a range list or swash is used. There are extra bits for \w,
13198 * etc. in locale ANYOFs, as what these match is not determinable at
13201 * Returns NULL, setting *flagp to RESTART_UTF8 if the sizing scan needs
13202 * to be restarted. This can only happen if ret_invlist is non-NULL.
13206 UV prevvalue = OOB_UNICODE, save_prevvalue = OOB_UNICODE;
13208 UV value = OOB_UNICODE, save_value = OOB_UNICODE;
13211 IV namedclass = OOB_NAMEDCLASS;
13212 char *rangebegin = NULL;
13213 bool need_class = 0;
13215 STRLEN initial_listsv_len = 0; /* Kind of a kludge to see if it is more
13216 than just initialized. */
13217 SV* properties = NULL; /* Code points that match \p{} \P{} */
13218 SV* posixes = NULL; /* Code points that match classes like [:word:],
13219 extended beyond the Latin1 range. These have to
13220 be kept separate from other code points for much
13221 of this function because their handling is
13222 different under /i, and for most classes under
13224 SV* nposixes = NULL; /* Similarly for [:^word:]. These are kept
13225 separate for a while from the non-complemented
13226 versions because of complications with /d
13228 UV element_count = 0; /* Number of distinct elements in the class.
13229 Optimizations may be possible if this is tiny */
13230 AV * multi_char_matches = NULL; /* Code points that fold to more than one
13231 character; used under /i */
13233 char * stop_ptr = RExC_end; /* where to stop parsing */
13234 const bool skip_white = cBOOL(ret_invlist); /* ignore unescaped white
13236 const bool strict = cBOOL(ret_invlist); /* Apply strict parsing rules? */
13238 /* Unicode properties are stored in a swash; this holds the current one
13239 * being parsed. If this swash is the only above-latin1 component of the
13240 * character class, an optimization is to pass it directly on to the
13241 * execution engine. Otherwise, it is set to NULL to indicate that there
13242 * are other things in the class that have to be dealt with at execution
13244 SV* swash = NULL; /* Code points that match \p{} \P{} */
13246 /* Set if a component of this character class is user-defined; just passed
13247 * on to the engine */
13248 bool has_user_defined_property = FALSE;
13250 /* inversion list of code points this node matches only when the target
13251 * string is in UTF-8. (Because is under /d) */
13252 SV* depends_list = NULL;
13254 /* Inversion list of code points this node matches regardless of things
13255 * like locale, folding, utf8ness of the target string */
13256 SV* cp_list = NULL;
13258 /* Like cp_list, but code points on this list need to be checked for things
13259 * that fold to/from them under /i */
13260 SV* cp_foldable_list = NULL;
13262 /* Like cp_list, but code points on this list are valid only when the
13263 * runtime locale is UTF-8 */
13264 SV* only_utf8_locale_list = NULL;
13267 /* In a range, counts how many 0-2 of the ends of it came from literals,
13268 * not escapes. Thus we can tell if 'A' was input vs \x{C1} */
13269 UV literal_endpoint = 0;
13271 bool invert = FALSE; /* Is this class to be complemented */
13273 bool warn_super = ALWAYS_WARN_SUPER;
13275 regnode * const orig_emit = RExC_emit; /* Save the original RExC_emit in
13276 case we need to change the emitted regop to an EXACT. */
13277 const char * orig_parse = RExC_parse;
13278 const SSize_t orig_size = RExC_size;
13279 bool posixl_matches_all = FALSE; /* Does /l class have both e.g. \W,\w ? */
13280 GET_RE_DEBUG_FLAGS_DECL;
13282 PERL_ARGS_ASSERT_REGCLASS;
13284 PERL_UNUSED_ARG(depth);
13287 DEBUG_PARSE("clas");
13289 /* Assume we are going to generate an ANYOF node. */
13290 ret = reganode(pRExC_state, ANYOF, 0);
13293 RExC_size += ANYOF_SKIP;
13294 listsv = &PL_sv_undef; /* For code scanners: listsv always non-NULL. */
13297 ANYOF_FLAGS(ret) = 0;
13299 RExC_emit += ANYOF_SKIP;
13300 listsv = newSVpvs_flags("# comment\n", SVs_TEMP);
13301 initial_listsv_len = SvCUR(listsv);
13302 SvTEMP_off(listsv); /* Grr, TEMPs and mortals are conflated. */
13306 RExC_parse = regpatws(pRExC_state, RExC_parse,
13307 FALSE /* means don't recognize comments */);
13310 if (UCHARAT(RExC_parse) == '^') { /* Complement of range. */
13313 allow_multi_folds = FALSE;
13316 RExC_parse = regpatws(pRExC_state, RExC_parse,
13317 FALSE /* means don't recognize comments */);
13321 /* Check that they didn't say [:posix:] instead of [[:posix:]] */
13322 if (!SIZE_ONLY && RExC_parse < RExC_end && POSIXCC(UCHARAT(RExC_parse))) {
13323 const char *s = RExC_parse;
13324 const char c = *s++;
13326 while (isWORDCHAR(*s))
13328 if (*s && c == *s && s[1] == ']') {
13329 SAVEFREESV(RExC_rx_sv);
13331 "POSIX syntax [%c %c] belongs inside character classes",
13333 (void)ReREFCNT_inc(RExC_rx_sv);
13337 /* If the caller wants us to just parse a single element, accomplish this
13338 * by faking the loop ending condition */
13339 if (stop_at_1 && RExC_end > RExC_parse) {
13340 stop_ptr = RExC_parse + 1;
13343 /* allow 1st char to be ']' (allowing it to be '-' is dealt with later) */
13344 if (UCHARAT(RExC_parse) == ']')
13345 goto charclassloop;
13349 if (RExC_parse >= stop_ptr) {
13354 RExC_parse = regpatws(pRExC_state, RExC_parse,
13355 FALSE /* means don't recognize comments */);
13358 if (UCHARAT(RExC_parse) == ']') {
13364 namedclass = OOB_NAMEDCLASS; /* initialize as illegal */
13365 save_value = value;
13366 save_prevvalue = prevvalue;
13369 rangebegin = RExC_parse;
13373 value = utf8n_to_uvchr((U8*)RExC_parse,
13374 RExC_end - RExC_parse,
13375 &numlen, UTF8_ALLOW_DEFAULT);
13376 RExC_parse += numlen;
13379 value = UCHARAT(RExC_parse++);
13382 && RExC_parse < RExC_end
13383 && POSIXCC(UCHARAT(RExC_parse)))
13385 namedclass = regpposixcc(pRExC_state, value, strict);
13387 else if (value == '\\') {
13389 value = utf8n_to_uvchr((U8*)RExC_parse,
13390 RExC_end - RExC_parse,
13391 &numlen, UTF8_ALLOW_DEFAULT);
13392 RExC_parse += numlen;
13395 value = UCHARAT(RExC_parse++);
13397 /* Some compilers cannot handle switching on 64-bit integer
13398 * values, therefore value cannot be an UV. Yes, this will
13399 * be a problem later if we want switch on Unicode.
13400 * A similar issue a little bit later when switching on
13401 * namedclass. --jhi */
13403 /* If the \ is escaping white space when white space is being
13404 * skipped, it means that that white space is wanted literally, and
13405 * is already in 'value'. Otherwise, need to translate the escape
13406 * into what it signifies. */
13407 if (! skip_white || ! is_PATWS_cp(value)) switch ((I32)value) {
13409 case 'w': namedclass = ANYOF_WORDCHAR; break;
13410 case 'W': namedclass = ANYOF_NWORDCHAR; break;
13411 case 's': namedclass = ANYOF_SPACE; break;
13412 case 'S': namedclass = ANYOF_NSPACE; break;
13413 case 'd': namedclass = ANYOF_DIGIT; break;
13414 case 'D': namedclass = ANYOF_NDIGIT; break;
13415 case 'v': namedclass = ANYOF_VERTWS; break;
13416 case 'V': namedclass = ANYOF_NVERTWS; break;
13417 case 'h': namedclass = ANYOF_HORIZWS; break;
13418 case 'H': namedclass = ANYOF_NHORIZWS; break;
13419 case 'N': /* Handle \N{NAME} in class */
13421 /* We only pay attention to the first char of
13422 multichar strings being returned. I kinda wonder
13423 if this makes sense as it does change the behaviour
13424 from earlier versions, OTOH that behaviour was broken
13426 if (! grok_bslash_N(pRExC_state, NULL, &value, flagp, depth,
13427 TRUE, /* => charclass */
13430 if (*flagp & RESTART_UTF8)
13431 FAIL("panic: grok_bslash_N set RESTART_UTF8");
13441 /* We will handle any undefined properties ourselves */
13442 U8 swash_init_flags = _CORE_SWASH_INIT_RETURN_IF_UNDEF
13443 /* And we actually would prefer to get
13444 * the straight inversion list of the
13445 * swash, since we will be accessing it
13446 * anyway, to save a little time */
13447 |_CORE_SWASH_INIT_ACCEPT_INVLIST;
13449 if (RExC_parse >= RExC_end)
13450 vFAIL2("Empty \\%c{}", (U8)value);
13451 if (*RExC_parse == '{') {
13452 const U8 c = (U8)value;
13453 e = strchr(RExC_parse++, '}');
13455 vFAIL2("Missing right brace on \\%c{}", c);
13456 while (isSPACE(UCHARAT(RExC_parse)))
13458 if (e == RExC_parse)
13459 vFAIL2("Empty \\%c{}", c);
13460 n = e - RExC_parse;
13461 while (isSPACE(UCHARAT(RExC_parse + n - 1)))
13473 if (UCHARAT(RExC_parse) == '^') {
13476 /* toggle. (The rhs xor gets the single bit that
13477 * differs between P and p; the other xor inverts just
13479 value ^= 'P' ^ 'p';
13481 while (isSPACE(UCHARAT(RExC_parse))) {
13486 /* Try to get the definition of the property into
13487 * <invlist>. If /i is in effect, the effective property
13488 * will have its name be <__NAME_i>. The design is
13489 * discussed in commit
13490 * 2f833f5208e26b208886e51e09e2c072b5eabb46 */
13491 formatted = Perl_form(aTHX_
13493 (FOLD) ? "__" : "",
13498 name = savepvn(formatted, strlen(formatted));
13500 /* Look up the property name, and get its swash and
13501 * inversion list, if the property is found */
13503 SvREFCNT_dec_NN(swash);
13505 swash = _core_swash_init("utf8", name, &PL_sv_undef,
13508 NULL, /* No inversion list */
13511 if (! swash || ! (invlist = _get_swash_invlist(swash))) {
13513 SvREFCNT_dec_NN(swash);
13517 /* Here didn't find it. It could be a user-defined
13518 * property that will be available at run-time. If we
13519 * accept only compile-time properties, is an error;
13520 * otherwise add it to the list for run-time look up */
13522 RExC_parse = e + 1;
13524 "Property '%"UTF8f"' is unknown",
13525 UTF8fARG(UTF, n, name));
13527 Perl_sv_catpvf(aTHX_ listsv, "%cutf8::%"UTF8f"\n",
13528 (value == 'p' ? '+' : '!'),
13529 UTF8fARG(UTF, n, name));
13530 has_user_defined_property = TRUE;
13532 /* We don't know yet, so have to assume that the
13533 * property could match something in the Latin1 range,
13534 * hence something that isn't utf8. Note that this
13535 * would cause things in <depends_list> to match
13536 * inappropriately, except that any \p{}, including
13537 * this one forces Unicode semantics, which means there
13538 * is no <depends_list> */
13539 ANYOF_FLAGS(ret) |= ANYOF_NONBITMAP_NON_UTF8;
13543 /* Here, did get the swash and its inversion list. If
13544 * the swash is from a user-defined property, then this
13545 * whole character class should be regarded as such */
13546 if (swash_init_flags
13547 & _CORE_SWASH_INIT_USER_DEFINED_PROPERTY)
13549 has_user_defined_property = TRUE;
13552 /* We warn on matching an above-Unicode code point
13553 * if the match would return true, except don't
13554 * warn for \p{All}, which has exactly one element
13556 (_invlist_contains_cp(invlist, 0x110000)
13557 && (! (_invlist_len(invlist) == 1
13558 && *invlist_array(invlist) == 0)))
13564 /* Invert if asking for the complement */
13565 if (value == 'P') {
13566 _invlist_union_complement_2nd(properties,
13570 /* The swash can't be used as-is, because we've
13571 * inverted things; delay removing it to here after
13572 * have copied its invlist above */
13573 SvREFCNT_dec_NN(swash);
13577 _invlist_union(properties, invlist, &properties);
13582 RExC_parse = e + 1;
13583 namedclass = ANYOF_UNIPROP; /* no official name, but it's
13586 /* \p means they want Unicode semantics */
13587 RExC_uni_semantics = 1;
13590 case 'n': value = '\n'; break;
13591 case 'r': value = '\r'; break;
13592 case 't': value = '\t'; break;
13593 case 'f': value = '\f'; break;
13594 case 'b': value = '\b'; break;
13595 case 'e': value = ASCII_TO_NATIVE('\033');break;
13596 case 'a': value = '\a'; break;
13598 RExC_parse--; /* function expects to be pointed at the 'o' */
13600 const char* error_msg;
13601 bool valid = grok_bslash_o(&RExC_parse,
13604 SIZE_ONLY, /* warnings in pass
13607 silence_non_portable,
13613 if (PL_encoding && value < 0x100) {
13614 goto recode_encoding;
13618 RExC_parse--; /* function expects to be pointed at the 'x' */
13620 const char* error_msg;
13621 bool valid = grok_bslash_x(&RExC_parse,
13624 TRUE, /* Output warnings */
13626 silence_non_portable,
13632 if (PL_encoding && value < 0x100)
13633 goto recode_encoding;
13636 value = grok_bslash_c(*RExC_parse++, SIZE_ONLY);
13638 case '0': case '1': case '2': case '3': case '4':
13639 case '5': case '6': case '7':
13641 /* Take 1-3 octal digits */
13642 I32 flags = PERL_SCAN_SILENT_ILLDIGIT;
13643 numlen = (strict) ? 4 : 3;
13644 value = grok_oct(--RExC_parse, &numlen, &flags, NULL);
13645 RExC_parse += numlen;
13648 RExC_parse += (UTF) ? UTF8SKIP(RExC_parse) : 1;
13649 vFAIL("Need exactly 3 octal digits");
13651 else if (! SIZE_ONLY /* like \08, \178 */
13653 && RExC_parse < RExC_end
13654 && isDIGIT(*RExC_parse)
13655 && ckWARN(WARN_REGEXP))
13657 SAVEFREESV(RExC_rx_sv);
13658 reg_warn_non_literal_string(
13660 form_short_octal_warning(RExC_parse, numlen));
13661 (void)ReREFCNT_inc(RExC_rx_sv);
13664 if (PL_encoding && value < 0x100)
13665 goto recode_encoding;
13669 if (! RExC_override_recoding) {
13670 SV* enc = PL_encoding;
13671 value = reg_recode((const char)(U8)value, &enc);
13674 vFAIL("Invalid escape in the specified encoding");
13676 else if (SIZE_ONLY) {
13677 ckWARNreg(RExC_parse,
13678 "Invalid escape in the specified encoding");
13684 /* Allow \_ to not give an error */
13685 if (!SIZE_ONLY && isWORDCHAR(value) && value != '_') {
13687 vFAIL2("Unrecognized escape \\%c in character class",
13691 SAVEFREESV(RExC_rx_sv);
13692 ckWARN2reg(RExC_parse,
13693 "Unrecognized escape \\%c in character class passed through",
13695 (void)ReREFCNT_inc(RExC_rx_sv);
13699 } /* End of switch on char following backslash */
13700 } /* end of handling backslash escape sequences */
13703 literal_endpoint++;
13706 /* Here, we have the current token in 'value' */
13708 if (namedclass > OOB_NAMEDCLASS) { /* this is a named class \blah */
13711 /* a bad range like a-\d, a-[:digit:]. The '-' is taken as a
13712 * literal, as is the character that began the false range, i.e.
13713 * the 'a' in the examples */
13716 const int w = (RExC_parse >= rangebegin)
13717 ? RExC_parse - rangebegin
13721 "False [] range \"%"UTF8f"\"",
13722 UTF8fARG(UTF, w, rangebegin));
13725 SAVEFREESV(RExC_rx_sv); /* in case of fatal warnings */
13726 ckWARN2reg(RExC_parse,
13727 "False [] range \"%"UTF8f"\"",
13728 UTF8fARG(UTF, w, rangebegin));
13729 (void)ReREFCNT_inc(RExC_rx_sv);
13730 cp_list = add_cp_to_invlist(cp_list, '-');
13731 cp_foldable_list = add_cp_to_invlist(cp_foldable_list,
13736 range = 0; /* this was not a true range */
13737 element_count += 2; /* So counts for three values */
13740 classnum = namedclass_to_classnum(namedclass);
13742 if (LOC && namedclass < ANYOF_POSIXL_MAX
13743 #ifndef HAS_ISASCII
13744 && classnum != _CC_ASCII
13747 /* What the Posix classes (like \w, [:space:]) match in locale
13748 * isn't knowable under locale until actual match time. Room
13749 * must be reserved (one time per outer bracketed class) to
13750 * store such classes. The space will contain a bit for each
13751 * named class that is to be matched against. This isn't
13752 * needed for \p{} and pseudo-classes, as they are not affected
13753 * by locale, and hence are dealt with separately */
13754 if (! need_class) {
13757 RExC_size += ANYOF_POSIXL_SKIP - ANYOF_SKIP;
13760 RExC_emit += ANYOF_POSIXL_SKIP - ANYOF_SKIP;
13762 ANYOF_FLAGS(ret) |= ANYOF_POSIXL;
13763 ANYOF_POSIXL_ZERO(ret);
13766 /* See if it already matches the complement of this POSIX
13768 if ((ANYOF_FLAGS(ret) & ANYOF_POSIXL)
13769 && ANYOF_POSIXL_TEST(ret, namedclass + ((namedclass % 2)
13773 posixl_matches_all = TRUE;
13774 break; /* No need to continue. Since it matches both
13775 e.g., \w and \W, it matches everything, and the
13776 bracketed class can be optimized into qr/./s */
13779 /* Add this class to those that should be checked at runtime */
13780 ANYOF_POSIXL_SET(ret, namedclass);
13782 /* The above-Latin1 characters are not subject to locale rules.
13783 * Just add them, in the second pass, to the
13784 * unconditionally-matched list */
13786 SV* scratch_list = NULL;
13788 /* Get the list of the above-Latin1 code points this
13790 _invlist_intersection_maybe_complement_2nd(PL_AboveLatin1,
13791 PL_XPosix_ptrs[classnum],
13793 /* Odd numbers are complements, like
13794 * NDIGIT, NASCII, ... */
13795 namedclass % 2 != 0,
13797 /* Checking if 'cp_list' is NULL first saves an extra
13798 * clone. Its reference count will be decremented at the
13799 * next union, etc, or if this is the only instance, at the
13800 * end of the routine */
13802 cp_list = scratch_list;
13805 _invlist_union(cp_list, scratch_list, &cp_list);
13806 SvREFCNT_dec_NN(scratch_list);
13808 continue; /* Go get next character */
13811 else if (! SIZE_ONLY) {
13813 /* Here, not in pass1 (in that pass we skip calculating the
13814 * contents of this class), and is /l, or is a POSIX class for
13815 * which /l doesn't matter (or is a Unicode property, which is
13816 * skipped here). */
13817 if (namedclass >= ANYOF_POSIXL_MAX) { /* If a special class */
13818 if (namedclass != ANYOF_UNIPROP) { /* UNIPROP = \p and \P */
13820 /* Here, should be \h, \H, \v, or \V. None of /d, /i
13821 * nor /l make a difference in what these match,
13822 * therefore we just add what they match to cp_list. */
13823 if (classnum != _CC_VERTSPACE) {
13824 assert( namedclass == ANYOF_HORIZWS
13825 || namedclass == ANYOF_NHORIZWS);
13827 /* It turns out that \h is just a synonym for
13829 classnum = _CC_BLANK;
13832 _invlist_union_maybe_complement_2nd(
13834 PL_XPosix_ptrs[classnum],
13835 namedclass % 2 != 0, /* Complement if odd
13836 (NHORIZWS, NVERTWS)
13841 else { /* Garden variety class. If is NASCII, NDIGIT, ...
13842 complement and use nposixes */
13843 SV** posixes_ptr = namedclass % 2 == 0
13846 SV** source_ptr = &PL_XPosix_ptrs[classnum];
13847 _invlist_union_maybe_complement_2nd(
13850 namedclass % 2 != 0,
13853 continue; /* Go get next character */
13855 } /* end of namedclass \blah */
13857 /* Here, we have a single value. If 'range' is set, it is the ending
13858 * of a range--check its validity. Later, we will handle each
13859 * individual code point in the range. If 'range' isn't set, this
13860 * could be the beginning of a range, so check for that by looking
13861 * ahead to see if the next real character to be processed is the range
13862 * indicator--the minus sign */
13865 RExC_parse = regpatws(pRExC_state, RExC_parse,
13866 FALSE /* means don't recognize comments */);
13870 if (prevvalue > value) /* b-a */ {
13871 const int w = RExC_parse - rangebegin;
13873 "Invalid [] range \"%"UTF8f"\"",
13874 UTF8fARG(UTF, w, rangebegin));
13875 range = 0; /* not a valid range */
13879 prevvalue = value; /* save the beginning of the potential range */
13880 if (! stop_at_1 /* Can't be a range if parsing just one thing */
13881 && *RExC_parse == '-')
13883 char* next_char_ptr = RExC_parse + 1;
13884 if (skip_white) { /* Get the next real char after the '-' */
13885 next_char_ptr = regpatws(pRExC_state,
13887 FALSE); /* means don't recognize
13891 /* If the '-' is at the end of the class (just before the ']',
13892 * it is a literal minus; otherwise it is a range */
13893 if (next_char_ptr < RExC_end && *next_char_ptr != ']') {
13894 RExC_parse = next_char_ptr;
13896 /* a bad range like \w-, [:word:]- ? */
13897 if (namedclass > OOB_NAMEDCLASS) {
13898 if (strict || ckWARN(WARN_REGEXP)) {
13900 RExC_parse >= rangebegin ?
13901 RExC_parse - rangebegin : 0;
13903 vFAIL4("False [] range \"%*.*s\"",
13908 "False [] range \"%*.*s\"",
13913 cp_list = add_cp_to_invlist(cp_list, '-');
13917 range = 1; /* yeah, it's a range! */
13918 continue; /* but do it the next time */
13923 /* Here, <prevvalue> is the beginning of the range, if any; or <value>
13926 /* non-Latin1 code point implies unicode semantics. Must be set in
13927 * pass1 so is there for the whole of pass 2 */
13929 RExC_uni_semantics = 1;
13932 /* Ready to process either the single value, or the completed range.
13933 * For single-valued non-inverted ranges, we consider the possibility
13934 * of multi-char folds. (We made a conscious decision to not do this
13935 * for the other cases because it can often lead to non-intuitive
13936 * results. For example, you have the peculiar case that:
13937 * "s s" =~ /^[^\xDF]+$/i => Y
13938 * "ss" =~ /^[^\xDF]+$/i => N
13940 * See [perl #89750] */
13941 if (FOLD && allow_multi_folds && value == prevvalue) {
13942 if (value == LATIN_SMALL_LETTER_SHARP_S
13943 || (value > 255 && _invlist_contains_cp(PL_HasMultiCharFold,
13946 /* Here <value> is indeed a multi-char fold. Get what it is */
13948 U8 foldbuf[UTF8_MAXBYTES_CASE];
13951 UV folded = _to_uni_fold_flags(
13955 FOLD_FLAGS_FULL | (ASCII_FOLD_RESTRICTED
13956 ? FOLD_FLAGS_NOMIX_ASCII
13960 /* Here, <folded> should be the first character of the
13961 * multi-char fold of <value>, with <foldbuf> containing the
13962 * whole thing. But, if this fold is not allowed (because of
13963 * the flags), <fold> will be the same as <value>, and should
13964 * be processed like any other character, so skip the special
13966 if (folded != value) {
13968 /* Skip if we are recursed, currently parsing the class
13969 * again. Otherwise add this character to the list of
13970 * multi-char folds. */
13971 if (! RExC_in_multi_char_class) {
13972 AV** this_array_ptr;
13974 STRLEN cp_count = utf8_length(foldbuf,
13975 foldbuf + foldlen);
13976 SV* multi_fold = sv_2mortal(newSVpvn("", 0));
13978 Perl_sv_catpvf(aTHX_ multi_fold, "\\x{%"UVXf"}", value);
13981 if (! multi_char_matches) {
13982 multi_char_matches = newAV();
13985 /* <multi_char_matches> is actually an array of arrays.
13986 * There will be one or two top-level elements: [2],
13987 * and/or [3]. The [2] element is an array, each
13988 * element thereof is a character which folds to TWO
13989 * characters; [3] is for folds to THREE characters.
13990 * (Unicode guarantees a maximum of 3 characters in any
13991 * fold.) When we rewrite the character class below,
13992 * we will do so such that the longest folds are
13993 * written first, so that it prefers the longest
13994 * matching strings first. This is done even if it
13995 * turns out that any quantifier is non-greedy, out of
13996 * programmer laziness. Tom Christiansen has agreed
13997 * that this is ok. This makes the test for the
13998 * ligature 'ffi' come before the test for 'ff' */
13999 if (av_exists(multi_char_matches, cp_count)) {
14000 this_array_ptr = (AV**) av_fetch(multi_char_matches,
14002 this_array = *this_array_ptr;
14005 this_array = newAV();
14006 av_store(multi_char_matches, cp_count,
14009 av_push(this_array, multi_fold);
14012 /* This element should not be processed further in this
14015 value = save_value;
14016 prevvalue = save_prevvalue;
14022 /* Deal with this element of the class */
14025 cp_foldable_list = _add_range_to_invlist(cp_foldable_list,
14028 SV* this_range = _new_invlist(1);
14029 _append_range_to_invlist(this_range, prevvalue, value);
14031 /* In EBCDIC, the ranges 'A-Z' and 'a-z' are each not contiguous.
14032 * If this range was specified using something like 'i-j', we want
14033 * to include only the 'i' and the 'j', and not anything in
14034 * between, so exclude non-ASCII, non-alphabetics from it.
14035 * However, if the range was specified with something like
14036 * [\x89-\x91] or [\x89-j], all code points within it should be
14037 * included. literal_endpoint==2 means both ends of the range used
14038 * a literal character, not \x{foo} */
14039 if (literal_endpoint == 2
14040 && ((prevvalue >= 'a' && value <= 'z')
14041 || (prevvalue >= 'A' && value <= 'Z')))
14043 _invlist_intersection(this_range, PL_ASCII,
14046 /* Since this above only contains ascii, the intersection of it
14047 * with anything will still yield only ascii */
14048 _invlist_intersection(this_range, PL_XPosix_ptrs[_CC_ALPHA],
14051 _invlist_union(cp_foldable_list, this_range, &cp_foldable_list);
14052 literal_endpoint = 0;
14056 range = 0; /* this range (if it was one) is done now */
14057 } /* End of loop through all the text within the brackets */
14059 /* If anything in the class expands to more than one character, we have to
14060 * deal with them by building up a substitute parse string, and recursively
14061 * calling reg() on it, instead of proceeding */
14062 if (multi_char_matches) {
14063 SV * substitute_parse = newSVpvn_flags("?:", 2, SVs_TEMP);
14066 char *save_end = RExC_end;
14067 char *save_parse = RExC_parse;
14068 bool first_time = TRUE; /* First multi-char occurrence doesn't get
14073 #if 0 /* Have decided not to deal with multi-char folds in inverted classes,
14074 because too confusing */
14076 sv_catpv(substitute_parse, "(?:");
14080 /* Look at the longest folds first */
14081 for (cp_count = av_tindex(multi_char_matches); cp_count > 0; cp_count--) {
14083 if (av_exists(multi_char_matches, cp_count)) {
14084 AV** this_array_ptr;
14087 this_array_ptr = (AV**) av_fetch(multi_char_matches,
14089 while ((this_sequence = av_pop(*this_array_ptr)) !=
14092 if (! first_time) {
14093 sv_catpv(substitute_parse, "|");
14095 first_time = FALSE;
14097 sv_catpv(substitute_parse, SvPVX(this_sequence));
14102 /* If the character class contains anything else besides these
14103 * multi-character folds, have to include it in recursive parsing */
14104 if (element_count) {
14105 sv_catpv(substitute_parse, "|[");
14106 sv_catpvn(substitute_parse, orig_parse, RExC_parse - orig_parse);
14107 sv_catpv(substitute_parse, "]");
14110 sv_catpv(substitute_parse, ")");
14113 /* This is a way to get the parse to skip forward a whole named
14114 * sequence instead of matching the 2nd character when it fails the
14116 sv_catpv(substitute_parse, "(*THEN)(*SKIP)(*FAIL)|.)");
14120 RExC_parse = SvPV(substitute_parse, len);
14121 RExC_end = RExC_parse + len;
14122 RExC_in_multi_char_class = 1;
14123 RExC_emit = (regnode *)orig_emit;
14125 ret = reg(pRExC_state, 1, ®_flags, depth+1);
14127 *flagp |= reg_flags&(HASWIDTH|SIMPLE|SPSTART|POSTPONED|RESTART_UTF8);
14129 RExC_parse = save_parse;
14130 RExC_end = save_end;
14131 RExC_in_multi_char_class = 0;
14132 SvREFCNT_dec_NN(multi_char_matches);
14136 /* Here, we've gone through the entire class and dealt with multi-char
14137 * folds. We are now in a position that we can do some checks to see if we
14138 * can optimize this ANYOF node into a simpler one, even in Pass 1.
14139 * Currently we only do two checks:
14140 * 1) is in the unlikely event that the user has specified both, eg. \w and
14141 * \W under /l, then the class matches everything. (This optimization
14142 * is done only to make the optimizer code run later work.)
14143 * 2) if the character class contains only a single element (including a
14144 * single range), we see if there is an equivalent node for it.
14145 * Other checks are possible */
14146 if (! ret_invlist /* Can't optimize if returning the constructed
14148 && (UNLIKELY(posixl_matches_all) || element_count == 1))
14153 if (UNLIKELY(posixl_matches_all)) {
14156 else if (namedclass > OOB_NAMEDCLASS) { /* this is a named class, like
14157 \w or [:digit:] or \p{foo}
14160 /* All named classes are mapped into POSIXish nodes, with its FLAG
14161 * argument giving which class it is */
14162 switch ((I32)namedclass) {
14163 case ANYOF_UNIPROP:
14166 /* These don't depend on the charset modifiers. They always
14167 * match under /u rules */
14168 case ANYOF_NHORIZWS:
14169 case ANYOF_HORIZWS:
14170 namedclass = ANYOF_BLANK + namedclass - ANYOF_HORIZWS;
14173 case ANYOF_NVERTWS:
14178 /* The actual POSIXish node for all the rest depends on the
14179 * charset modifier. The ones in the first set depend only on
14180 * ASCII or, if available on this platform, locale */
14184 op = (LOC) ? POSIXL : POSIXA;
14195 /* under /a could be alpha */
14197 if (ASCII_RESTRICTED) {
14198 namedclass = ANYOF_ALPHA + (namedclass % 2);
14206 /* The rest have more possibilities depending on the charset.
14207 * We take advantage of the enum ordering of the charset
14208 * modifiers to get the exact node type, */
14210 op = POSIXD + get_regex_charset(RExC_flags);
14211 if (op > POSIXA) { /* /aa is same as /a */
14216 /* The odd numbered ones are the complements of the
14217 * next-lower even number one */
14218 if (namedclass % 2 == 1) {
14222 arg = namedclass_to_classnum(namedclass);
14226 else if (value == prevvalue) {
14228 /* Here, the class consists of just a single code point */
14231 if (! LOC && value == '\n') {
14232 op = REG_ANY; /* Optimize [^\n] */
14233 *flagp |= HASWIDTH|SIMPLE;
14237 else if (value < 256 || UTF) {
14239 /* Optimize a single value into an EXACTish node, but not if it
14240 * would require converting the pattern to UTF-8. */
14241 op = compute_EXACTish(pRExC_state);
14243 } /* Otherwise is a range */
14244 else if (! LOC) { /* locale could vary these */
14245 if (prevvalue == '0') {
14246 if (value == '9') {
14253 /* Here, we have changed <op> away from its initial value iff we found
14254 * an optimization */
14257 /* Throw away this ANYOF regnode, and emit the calculated one,
14258 * which should correspond to the beginning, not current, state of
14260 const char * cur_parse = RExC_parse;
14261 RExC_parse = (char *)orig_parse;
14265 /* To get locale nodes to not use the full ANYOF size would
14266 * require moving the code above that writes the portions
14267 * of it that aren't in other nodes to after this point.
14268 * e.g. ANYOF_POSIXL_SET */
14269 RExC_size = orig_size;
14273 RExC_emit = (regnode *)orig_emit;
14274 if (PL_regkind[op] == POSIXD) {
14275 if (op == POSIXL) {
14276 RExC_contains_locale = 1;
14279 op += NPOSIXD - POSIXD;
14284 ret = reg_node(pRExC_state, op);
14286 if (PL_regkind[op] == POSIXD || PL_regkind[op] == NPOSIXD) {
14290 *flagp |= HASWIDTH|SIMPLE;
14292 else if (PL_regkind[op] == EXACT) {
14293 alloc_maybe_populate_EXACT(pRExC_state, ret, flagp, 0, value,
14294 TRUE /* downgradable to EXACT */
14298 RExC_parse = (char *) cur_parse;
14300 SvREFCNT_dec(posixes);
14301 SvREFCNT_dec(nposixes);
14302 SvREFCNT_dec(cp_list);
14303 SvREFCNT_dec(cp_foldable_list);
14310 /****** !SIZE_ONLY (Pass 2) AFTER HERE *********/
14312 /* If folding, we calculate all characters that could fold to or from the
14313 * ones already on the list */
14314 if (cp_foldable_list) {
14316 UV start, end; /* End points of code point ranges */
14318 SV* fold_intersection = NULL;
14321 /* Our calculated list will be for Unicode rules. For locale
14322 * matching, we have to keep a separate list that is consulted at
14323 * runtime only when the locale indicates Unicode rules. For
14324 * non-locale, we just use to the general list */
14326 use_list = &only_utf8_locale_list;
14329 use_list = &cp_list;
14332 /* Only the characters in this class that participate in folds need
14333 * be checked. Get the intersection of this class and all the
14334 * possible characters that are foldable. This can quickly narrow
14335 * down a large class */
14336 _invlist_intersection(PL_utf8_foldable, cp_foldable_list,
14337 &fold_intersection);
14339 /* The folds for all the Latin1 characters are hard-coded into this
14340 * program, but we have to go out to disk to get the others. */
14341 if (invlist_highest(cp_foldable_list) >= 256) {
14343 /* This is a hash that for a particular fold gives all
14344 * characters that are involved in it */
14345 if (! PL_utf8_foldclosures) {
14347 /* If the folds haven't been read in, call a fold function
14349 if (! PL_utf8_tofold) {
14350 U8 dummy[UTF8_MAXBYTES_CASE+1];
14352 /* This string is just a short named one above \xff */
14353 to_utf8_fold((U8*) HYPHEN_UTF8, dummy, NULL);
14354 assert(PL_utf8_tofold); /* Verify that worked */
14356 PL_utf8_foldclosures
14357 = _swash_inversion_hash(PL_utf8_tofold);
14361 /* Now look at the foldable characters in this class individually */
14362 invlist_iterinit(fold_intersection);
14363 while (invlist_iternext(fold_intersection, &start, &end)) {
14366 /* Look at every character in the range */
14367 for (j = start; j <= end; j++) {
14368 U8 foldbuf[UTF8_MAXBYTES_CASE+1];
14374 /* We have the latin1 folding rules hard-coded here so
14375 * that an innocent-looking character class, like
14376 * /[ks]/i won't have to go out to disk to find the
14377 * possible matches. XXX It would be better to
14378 * generate these via regen, in case a new version of
14379 * the Unicode standard adds new mappings, though that
14380 * is not really likely, and may be caught by the
14381 * default: case of the switch below. */
14383 if (IS_IN_SOME_FOLD_L1(j)) {
14385 /* ASCII is always matched; non-ASCII is matched
14386 * only under Unicode rules (which could happen
14387 * under /l if the locale is a UTF-8 one */
14388 if (isASCII(j) || ! DEPENDS_SEMANTICS) {
14389 *use_list = add_cp_to_invlist(*use_list,
14390 PL_fold_latin1[j]);
14394 add_cp_to_invlist(depends_list,
14395 PL_fold_latin1[j]);
14399 if (HAS_NONLATIN1_FOLD_CLOSURE(j)
14400 && (! isASCII(j) || ! ASCII_FOLD_RESTRICTED))
14402 /* Certain Latin1 characters have matches outside
14403 * Latin1. To get here, <j> is one of those
14404 * characters. None of these matches is valid for
14405 * ASCII characters under /aa, which is why the 'if'
14406 * just above excludes those. These matches only
14407 * happen when the target string is utf8. The code
14408 * below adds the single fold closures for <j> to the
14409 * inversion list. */
14415 add_cp_to_invlist(*use_list, KELVIN_SIGN);
14419 *use_list = add_cp_to_invlist(*use_list,
14420 LATIN_SMALL_LETTER_LONG_S);
14423 *use_list = add_cp_to_invlist(*use_list,
14424 GREEK_CAPITAL_LETTER_MU);
14425 *use_list = add_cp_to_invlist(*use_list,
14426 GREEK_SMALL_LETTER_MU);
14428 case LATIN_CAPITAL_LETTER_A_WITH_RING_ABOVE:
14429 case LATIN_SMALL_LETTER_A_WITH_RING_ABOVE:
14431 add_cp_to_invlist(*use_list, ANGSTROM_SIGN);
14433 case LATIN_SMALL_LETTER_Y_WITH_DIAERESIS:
14434 *use_list = add_cp_to_invlist(*use_list,
14435 LATIN_CAPITAL_LETTER_Y_WITH_DIAERESIS);
14437 case LATIN_SMALL_LETTER_SHARP_S:
14438 *use_list = add_cp_to_invlist(*use_list,
14439 LATIN_CAPITAL_LETTER_SHARP_S);
14441 case 'F': case 'f':
14442 case 'I': case 'i':
14443 case 'L': case 'l':
14444 case 'T': case 't':
14445 case 'A': case 'a':
14446 case 'H': case 'h':
14447 case 'J': case 'j':
14448 case 'N': case 'n':
14449 case 'W': case 'w':
14450 case 'Y': case 'y':
14451 /* These all are targets of multi-character
14452 * folds from code points that require UTF8
14453 * to express, so they can't match unless
14454 * the target string is in UTF-8, so no
14455 * action here is necessary, as regexec.c
14456 * properly handles the general case for
14457 * UTF-8 matching and multi-char folds */
14460 /* Use deprecated warning to increase the
14461 * chances of this being output */
14462 ckWARN2reg_d(RExC_parse, "Perl folding rules are not up-to-date for 0x%"UVXf"; please use the perlbug utility to report;", j);
14469 /* Here is an above Latin1 character. We don't have the
14470 * rules hard-coded for it. First, get its fold. This is
14471 * the simple fold, as the multi-character folds have been
14472 * handled earlier and separated out */
14473 _to_uni_fold_flags(j, foldbuf, &foldlen,
14474 (ASCII_FOLD_RESTRICTED)
14475 ? FOLD_FLAGS_NOMIX_ASCII
14478 /* Single character fold of above Latin1. Add everything in
14479 * its fold closure to the list that this node should match.
14480 * The fold closures data structure is a hash with the keys
14481 * being the UTF-8 of every character that is folded to, like
14482 * 'k', and the values each an array of all code points that
14483 * fold to its key. e.g. [ 'k', 'K', KELVIN_SIGN ].
14484 * Multi-character folds are not included */
14485 if ((listp = hv_fetch(PL_utf8_foldclosures,
14486 (char *) foldbuf, foldlen, FALSE)))
14488 AV* list = (AV*) *listp;
14490 for (k = 0; k <= av_tindex(list); k++) {
14491 SV** c_p = av_fetch(list, k, FALSE);
14494 Perl_croak(aTHX_ "panic: invalid PL_utf8_foldclosures structure");
14498 /* /aa doesn't allow folds between ASCII and non- */
14499 if ((ASCII_FOLD_RESTRICTED
14500 && (isASCII(c) != isASCII(j))))
14505 /* Folds under /l which cross the 255/256 boundary
14506 * are added to a separate list. (These are valid
14507 * only when the locale is UTF-8.) */
14508 if (c < 256 && LOC) {
14509 *use_list = add_cp_to_invlist(*use_list, c);
14513 if (isASCII(c) || c > 255 || AT_LEAST_UNI_SEMANTICS)
14515 cp_list = add_cp_to_invlist(cp_list, c);
14518 /* Similarly folds involving non-ascii Latin1
14519 * characters under /d are added to their list */
14520 depends_list = add_cp_to_invlist(depends_list,
14527 SvREFCNT_dec_NN(fold_intersection);
14530 /* Now that we have finished adding all the folds, there is no reason
14531 * to keep the foldable list separate */
14532 _invlist_union(cp_list, cp_foldable_list, &cp_list);
14533 SvREFCNT_dec_NN(cp_foldable_list);
14536 /* And combine the result (if any) with any inversion list from posix
14537 * classes. The lists are kept separate up to now because we don't want to
14538 * fold the classes (folding of those is automatically handled by the swash
14539 * fetching code) */
14540 if (posixes || nposixes) {
14541 if (posixes && AT_LEAST_ASCII_RESTRICTED) {
14542 /* Under /a and /aa, nothing above ASCII matches these */
14543 _invlist_intersection(posixes,
14544 PL_XPosix_ptrs[_CC_ASCII],
14548 if (DEPENDS_SEMANTICS) {
14549 /* Under /d, everything in the upper half of the Latin1 range
14550 * matches these complements */
14551 ANYOF_FLAGS(ret) |= ANYOF_NON_UTF8_NON_ASCII_ALL;
14553 else if (AT_LEAST_ASCII_RESTRICTED) {
14554 /* Under /a and /aa, everything above ASCII matches these
14556 _invlist_union_complement_2nd(nposixes,
14557 PL_XPosix_ptrs[_CC_ASCII],
14561 _invlist_union(posixes, nposixes, &posixes);
14562 SvREFCNT_dec_NN(nposixes);
14565 posixes = nposixes;
14568 if (! DEPENDS_SEMANTICS) {
14570 _invlist_union(cp_list, posixes, &cp_list);
14571 SvREFCNT_dec_NN(posixes);
14578 /* Under /d, we put into a separate list the Latin1 things that
14579 * match only when the target string is utf8 */
14580 SV* nonascii_but_latin1_properties = NULL;
14581 _invlist_intersection(posixes, PL_UpperLatin1,
14582 &nonascii_but_latin1_properties);
14583 _invlist_subtract(posixes, nonascii_but_latin1_properties,
14586 _invlist_union(cp_list, posixes, &cp_list);
14587 SvREFCNT_dec_NN(posixes);
14593 if (depends_list) {
14594 _invlist_union(depends_list, nonascii_but_latin1_properties,
14596 SvREFCNT_dec_NN(nonascii_but_latin1_properties);
14599 depends_list = nonascii_but_latin1_properties;
14604 /* And combine the result (if any) with any inversion list from properties.
14605 * The lists are kept separate up to now so that we can distinguish the two
14606 * in regards to matching above-Unicode. A run-time warning is generated
14607 * if a Unicode property is matched against a non-Unicode code point. But,
14608 * we allow user-defined properties to match anything, without any warning,
14609 * and we also suppress the warning if there is a portion of the character
14610 * class that isn't a Unicode property, and which matches above Unicode, \W
14611 * or [\x{110000}] for example.
14612 * (Note that in this case, unlike the Posix one above, there is no
14613 * <depends_list>, because having a Unicode property forces Unicode
14618 /* If it matters to the final outcome, see if a non-property
14619 * component of the class matches above Unicode. If so, the
14620 * warning gets suppressed. This is true even if just a single
14621 * such code point is specified, as though not strictly correct if
14622 * another such code point is matched against, the fact that they
14623 * are using above-Unicode code points indicates they should know
14624 * the issues involved */
14626 warn_super = ! (invert
14627 ^ (invlist_highest(cp_list) > PERL_UNICODE_MAX));
14630 _invlist_union(properties, cp_list, &cp_list);
14631 SvREFCNT_dec_NN(properties);
14634 cp_list = properties;
14638 ANYOF_FLAGS(ret) |= ANYOF_WARN_SUPER;
14642 /* Here, we have calculated what code points should be in the character
14645 * Now we can see about various optimizations. Fold calculation (which we
14646 * did above) needs to take place before inversion. Otherwise /[^k]/i
14647 * would invert to include K, which under /i would match k, which it
14648 * shouldn't. Therefore we can't invert folded locale now, as it won't be
14649 * folded until runtime */
14651 /* If we didn't do folding, it's because some information isn't available
14652 * until runtime; set the run-time fold flag for these. (We don't have to
14653 * worry about properties folding, as that is taken care of by the swash
14654 * fetching). We know to set the flag if we have a non-NULL list for UTF-8
14655 * locales, or the class matches at least one 0-255 range code point */
14657 if (only_utf8_locale_list) {
14658 ANYOF_FLAGS(ret) |= ANYOF_LOC_FOLD;
14660 else if (cp_list) { /* Look to see if there a 0-255 code point is in
14663 invlist_iterinit(cp_list);
14664 if (invlist_iternext(cp_list, &start, &end) && start < 256) {
14665 ANYOF_FLAGS(ret) |= ANYOF_LOC_FOLD;
14667 invlist_iterfinish(cp_list);
14671 /* Optimize inverted simple patterns (e.g. [^a-z]) when everything is known
14672 * at compile time. Besides not inverting folded locale now, we can't
14673 * invert if there are things such as \w, which aren't known until runtime
14677 && ! (ANYOF_FLAGS(ret) & (ANYOF_LOCALE_FLAGS))
14679 && ! HAS_NONLOCALE_RUNTIME_PROPERTY_DEFINITION)
14681 _invlist_invert(cp_list);
14683 /* Any swash can't be used as-is, because we've inverted things */
14685 SvREFCNT_dec_NN(swash);
14689 /* Clear the invert flag since have just done it here */
14694 *ret_invlist = cp_list;
14695 SvREFCNT_dec(swash);
14697 /* Discard the generated node */
14699 RExC_size = orig_size;
14702 RExC_emit = orig_emit;
14707 /* Some character classes are equivalent to other nodes. Such nodes take
14708 * up less room and generally fewer operations to execute than ANYOF nodes.
14709 * Above, we checked for and optimized into some such equivalents for
14710 * certain common classes that are easy to test. Getting to this point in
14711 * the code means that the class didn't get optimized there. Since this
14712 * code is only executed in Pass 2, it is too late to save space--it has
14713 * been allocated in Pass 1, and currently isn't given back. But turning
14714 * things into an EXACTish node can allow the optimizer to join it to any
14715 * adjacent such nodes. And if the class is equivalent to things like /./,
14716 * expensive run-time swashes can be avoided. Now that we have more
14717 * complete information, we can find things necessarily missed by the
14718 * earlier code. I (khw) am not sure how much to look for here. It would
14719 * be easy, but perhaps too slow, to check any candidates against all the
14720 * node types they could possibly match using _invlistEQ(). */
14725 && ! (ANYOF_FLAGS(ret) & (ANYOF_LOCALE_FLAGS))
14726 && ! HAS_NONLOCALE_RUNTIME_PROPERTY_DEFINITION
14728 /* We don't optimize if we are supposed to make sure all non-Unicode
14729 * code points raise a warning, as only ANYOF nodes have this check.
14731 && ! ((ANYOF_FLAGS(ret) | ANYOF_WARN_SUPER) && ALWAYS_WARN_SUPER))
14734 U8 op = END; /* The optimzation node-type */
14735 const char * cur_parse= RExC_parse;
14737 invlist_iterinit(cp_list);
14738 if (! invlist_iternext(cp_list, &start, &end)) {
14740 /* Here, the list is empty. This happens, for example, when a
14741 * Unicode property is the only thing in the character class, and
14742 * it doesn't match anything. (perluniprops.pod notes such
14745 *flagp |= HASWIDTH|SIMPLE;
14747 else if (start == end) { /* The range is a single code point */
14748 if (! invlist_iternext(cp_list, &start, &end)
14750 /* Don't do this optimization if it would require changing
14751 * the pattern to UTF-8 */
14752 && (start < 256 || UTF))
14754 /* Here, the list contains a single code point. Can optimize
14755 * into an EXACTish node */
14764 /* A locale node under folding with one code point can be
14765 * an EXACTFL, as its fold won't be calculated until
14771 /* Here, we are generally folding, but there is only one
14772 * code point to match. If we have to, we use an EXACT
14773 * node, but it would be better for joining with adjacent
14774 * nodes in the optimization pass if we used the same
14775 * EXACTFish node that any such are likely to be. We can
14776 * do this iff the code point doesn't participate in any
14777 * folds. For example, an EXACTF of a colon is the same as
14778 * an EXACT one, since nothing folds to or from a colon. */
14780 if (IS_IN_SOME_FOLD_L1(value)) {
14785 if (_invlist_contains_cp(PL_utf8_foldable, value)) {
14790 /* If we haven't found the node type, above, it means we
14791 * can use the prevailing one */
14793 op = compute_EXACTish(pRExC_state);
14798 else if (start == 0) {
14799 if (end == UV_MAX) {
14801 *flagp |= HASWIDTH|SIMPLE;
14804 else if (end == '\n' - 1
14805 && invlist_iternext(cp_list, &start, &end)
14806 && start == '\n' + 1 && end == UV_MAX)
14809 *flagp |= HASWIDTH|SIMPLE;
14813 invlist_iterfinish(cp_list);
14816 RExC_parse = (char *)orig_parse;
14817 RExC_emit = (regnode *)orig_emit;
14819 ret = reg_node(pRExC_state, op);
14821 RExC_parse = (char *)cur_parse;
14823 if (PL_regkind[op] == EXACT) {
14824 alloc_maybe_populate_EXACT(pRExC_state, ret, flagp, 0, value,
14825 TRUE /* downgradable to EXACT */
14829 SvREFCNT_dec_NN(cp_list);
14834 /* Here, <cp_list> contains all the code points we can determine at
14835 * compile time that match under all conditions. Go through it, and
14836 * for things that belong in the bitmap, put them there, and delete from
14837 * <cp_list>. While we are at it, see if everything above 255 is in the
14838 * list, and if so, set a flag to speed up execution */
14840 populate_ANYOF_from_invlist(ret, &cp_list);
14843 ANYOF_FLAGS(ret) |= ANYOF_INVERT;
14846 /* Here, the bitmap has been populated with all the Latin1 code points that
14847 * always match. Can now add to the overall list those that match only
14848 * when the target string is UTF-8 (<depends_list>). */
14849 if (depends_list) {
14851 _invlist_union(cp_list, depends_list, &cp_list);
14852 SvREFCNT_dec_NN(depends_list);
14855 cp_list = depends_list;
14857 ANYOF_FLAGS(ret) |= ANYOF_UTF8;
14860 /* If there is a swash and more than one element, we can't use the swash in
14861 * the optimization below. */
14862 if (swash && element_count > 1) {
14863 SvREFCNT_dec_NN(swash);
14867 set_ANYOF_arg(pRExC_state, ret, cp_list,
14868 (HAS_NONLOCALE_RUNTIME_PROPERTY_DEFINITION)
14870 only_utf8_locale_list,
14871 swash, has_user_defined_property);
14873 *flagp |= HASWIDTH|SIMPLE;
14875 if (ANYOF_FLAGS(ret) & ANYOF_LOCALE_FLAGS) {
14876 RExC_contains_locale = 1;
14882 #undef HAS_NONLOCALE_RUNTIME_PROPERTY_DEFINITION
14885 S_set_ANYOF_arg(pTHX_ RExC_state_t* const pRExC_state,
14886 regnode* const node,
14888 SV* const runtime_defns,
14889 SV* const only_utf8_locale_list,
14891 const bool has_user_defined_property)
14893 /* Sets the arg field of an ANYOF-type node 'node', using information about
14894 * the node passed-in. If there is nothing outside the node's bitmap, the
14895 * arg is set to ANYOF_NONBITMAP_EMPTY. Otherwise, it sets the argument to
14896 * the count returned by add_data(), having allocated and stored an array,
14897 * av, that that count references, as follows:
14898 * av[0] stores the character class description in its textual form.
14899 * This is used later (regexec.c:Perl_regclass_swash()) to
14900 * initialize the appropriate swash, and is also useful for dumping
14901 * the regnode. This is set to &PL_sv_undef if the textual
14902 * description is not needed at run-time (as happens if the other
14903 * elements completely define the class)
14904 * av[1] if &PL_sv_undef, is a placeholder to later contain the swash
14905 * computed from av[0]. But if no further computation need be done,
14906 * the swash is stored here now (and av[0] is &PL_sv_undef).
14907 * av[2] stores the inversion list of code points that match only if the
14908 * current locale is UTF-8
14909 * av[3] stores the cp_list inversion list for use in addition or instead
14910 * of av[0]; used only if cp_list exists and av[1] is &PL_sv_undef.
14911 * (Otherwise everything needed is already in av[0] and av[1])
14912 * av[4] is set if any component of the class is from a user-defined
14913 * property; used only if av[3] exists */
14917 PERL_ARGS_ASSERT_SET_ANYOF_ARG;
14919 if (! cp_list && ! runtime_defns && ! only_utf8_locale_list) {
14920 assert(! (ANYOF_FLAGS(node)
14921 & (ANYOF_UTF8|ANYOF_NONBITMAP_NON_UTF8)));
14922 ARG_SET(node, ANYOF_NONBITMAP_EMPTY);
14925 AV * const av = newAV();
14928 assert(ANYOF_FLAGS(node)
14929 & (ANYOF_UTF8|ANYOF_NONBITMAP_NON_UTF8|ANYOF_LOC_FOLD));
14931 av_store(av, 0, (runtime_defns)
14932 ? SvREFCNT_inc(runtime_defns) : &PL_sv_undef);
14934 av_store(av, 1, swash);
14935 SvREFCNT_dec_NN(cp_list);
14938 av_store(av, 1, &PL_sv_undef);
14940 av_store(av, 3, cp_list);
14941 av_store(av, 4, newSVuv(has_user_defined_property));
14945 if (only_utf8_locale_list) {
14946 av_store(av, 2, only_utf8_locale_list);
14949 av_store(av, 2, &PL_sv_undef);
14952 rv = newRV_noinc(MUTABLE_SV(av));
14953 n = add_data(pRExC_state, STR_WITH_LEN("s"));
14954 RExC_rxi->data->data[n] = (void*)rv;
14960 /* reg_skipcomment()
14962 Absorbs an /x style # comments from the input stream.
14963 Returns true if there is more text remaining in the stream.
14964 Will set the REG_RUN_ON_COMMENT_SEEN flag if the comment
14965 terminates the pattern without including a newline.
14967 Note its the callers responsibility to ensure that we are
14968 actually in /x mode
14973 S_reg_skipcomment(pTHX_ RExC_state_t *pRExC_state)
14977 PERL_ARGS_ASSERT_REG_SKIPCOMMENT;
14979 while (RExC_parse < RExC_end)
14980 if (*RExC_parse++ == '\n') {
14985 /* we ran off the end of the pattern without ending
14986 the comment, so we have to add an \n when wrapping */
14987 RExC_seen |= REG_RUN_ON_COMMENT_SEEN;
14995 Advances the parse position, and optionally absorbs
14996 "whitespace" from the inputstream.
14998 Without /x "whitespace" means (?#...) style comments only,
14999 with /x this means (?#...) and # comments and whitespace proper.
15001 Returns the RExC_parse point from BEFORE the scan occurs.
15003 This is the /x friendly way of saying RExC_parse++.
15007 S_nextchar(pTHX_ RExC_state_t *pRExC_state)
15009 char* const retval = RExC_parse++;
15011 PERL_ARGS_ASSERT_NEXTCHAR;
15014 if (RExC_end - RExC_parse >= 3
15015 && *RExC_parse == '('
15016 && RExC_parse[1] == '?'
15017 && RExC_parse[2] == '#')
15019 while (*RExC_parse != ')') {
15020 if (RExC_parse == RExC_end)
15021 FAIL("Sequence (?#... not terminated");
15027 if (RExC_flags & RXf_PMf_EXTENDED) {
15028 if (isSPACE(*RExC_parse)) {
15032 else if (*RExC_parse == '#') {
15033 if ( reg_skipcomment( pRExC_state ) )
15042 - reg_node - emit a node
15044 STATIC regnode * /* Location. */
15045 S_reg_node(pTHX_ RExC_state_t *pRExC_state, U8 op)
15049 regnode * const ret = RExC_emit;
15050 GET_RE_DEBUG_FLAGS_DECL;
15052 PERL_ARGS_ASSERT_REG_NODE;
15055 SIZE_ALIGN(RExC_size);
15059 if (RExC_emit >= RExC_emit_bound)
15060 Perl_croak(aTHX_ "panic: reg_node overrun trying to emit %d, %p>=%p",
15061 op, RExC_emit, RExC_emit_bound);
15063 NODE_ALIGN_FILL(ret);
15065 FILL_ADVANCE_NODE(ptr, op);
15066 #ifdef RE_TRACK_PATTERN_OFFSETS
15067 if (RExC_offsets) { /* MJD */
15069 ("%s:%d: (op %s) %s %"UVuf" (len %"UVuf") (max %"UVuf").\n",
15070 "reg_node", __LINE__,
15072 (UV)(RExC_emit - RExC_emit_start) > RExC_offsets[0]
15073 ? "Overwriting end of array!\n" : "OK",
15074 (UV)(RExC_emit - RExC_emit_start),
15075 (UV)(RExC_parse - RExC_start),
15076 (UV)RExC_offsets[0]));
15077 Set_Node_Offset(RExC_emit, RExC_parse + (op == END));
15085 - reganode - emit a node with an argument
15087 STATIC regnode * /* Location. */
15088 S_reganode(pTHX_ RExC_state_t *pRExC_state, U8 op, U32 arg)
15092 regnode * const ret = RExC_emit;
15093 GET_RE_DEBUG_FLAGS_DECL;
15095 PERL_ARGS_ASSERT_REGANODE;
15098 SIZE_ALIGN(RExC_size);
15103 assert(2==regarglen[op]+1);
15105 Anything larger than this has to allocate the extra amount.
15106 If we changed this to be:
15108 RExC_size += (1 + regarglen[op]);
15110 then it wouldn't matter. Its not clear what side effect
15111 might come from that so its not done so far.
15116 if (RExC_emit >= RExC_emit_bound)
15117 Perl_croak(aTHX_ "panic: reg_node overrun trying to emit %d, %p>=%p",
15118 op, RExC_emit, RExC_emit_bound);
15120 NODE_ALIGN_FILL(ret);
15122 FILL_ADVANCE_NODE_ARG(ptr, op, arg);
15123 #ifdef RE_TRACK_PATTERN_OFFSETS
15124 if (RExC_offsets) { /* MJD */
15126 ("%s(%d): (op %s) %s %"UVuf" <- %"UVuf" (max %"UVuf").\n",
15130 (UV)(RExC_emit - RExC_emit_start) > RExC_offsets[0] ?
15131 "Overwriting end of array!\n" : "OK",
15132 (UV)(RExC_emit - RExC_emit_start),
15133 (UV)(RExC_parse - RExC_start),
15134 (UV)RExC_offsets[0]));
15135 Set_Cur_Node_Offset;
15143 - reguni - emit (if appropriate) a Unicode character
15145 PERL_STATIC_INLINE STRLEN
15146 S_reguni(pTHX_ const RExC_state_t *pRExC_state, UV uv, char* s)
15150 PERL_ARGS_ASSERT_REGUNI;
15152 return SIZE_ONLY ? UNISKIP(uv) : (uvchr_to_utf8((U8*)s, uv) - (U8*)s);
15156 - reginsert - insert an operator in front of already-emitted operand
15158 * Means relocating the operand.
15161 S_reginsert(pTHX_ RExC_state_t *pRExC_state, U8 op, regnode *opnd, U32 depth)
15167 const int offset = regarglen[(U8)op];
15168 const int size = NODE_STEP_REGNODE + offset;
15169 GET_RE_DEBUG_FLAGS_DECL;
15171 PERL_ARGS_ASSERT_REGINSERT;
15172 PERL_UNUSED_ARG(depth);
15173 /* (PL_regkind[(U8)op] == CURLY ? EXTRA_STEP_2ARGS : 0); */
15174 DEBUG_PARSE_FMT("inst"," - %s",PL_reg_name[op]);
15183 if (RExC_open_parens) {
15185 /*DEBUG_PARSE_FMT("inst"," - %"IVdf, (IV)RExC_npar);*/
15186 for ( paren=0 ; paren < RExC_npar ; paren++ ) {
15187 if ( RExC_open_parens[paren] >= opnd ) {
15188 /*DEBUG_PARSE_FMT("open"," - %d",size);*/
15189 RExC_open_parens[paren] += size;
15191 /*DEBUG_PARSE_FMT("open"," - %s","ok");*/
15193 if ( RExC_close_parens[paren] >= opnd ) {
15194 /*DEBUG_PARSE_FMT("close"," - %d",size);*/
15195 RExC_close_parens[paren] += size;
15197 /*DEBUG_PARSE_FMT("close"," - %s","ok");*/
15202 while (src > opnd) {
15203 StructCopy(--src, --dst, regnode);
15204 #ifdef RE_TRACK_PATTERN_OFFSETS
15205 if (RExC_offsets) { /* MJD 20010112 */
15207 ("%s(%d): (op %s) %s copy %"UVuf" -> %"UVuf" (max %"UVuf").\n",
15211 (UV)(dst - RExC_emit_start) > RExC_offsets[0]
15212 ? "Overwriting end of array!\n" : "OK",
15213 (UV)(src - RExC_emit_start),
15214 (UV)(dst - RExC_emit_start),
15215 (UV)RExC_offsets[0]));
15216 Set_Node_Offset_To_R(dst-RExC_emit_start, Node_Offset(src));
15217 Set_Node_Length_To_R(dst-RExC_emit_start, Node_Length(src));
15223 place = opnd; /* Op node, where operand used to be. */
15224 #ifdef RE_TRACK_PATTERN_OFFSETS
15225 if (RExC_offsets) { /* MJD */
15227 ("%s(%d): (op %s) %s %"UVuf" <- %"UVuf" (max %"UVuf").\n",
15231 (UV)(place - RExC_emit_start) > RExC_offsets[0]
15232 ? "Overwriting end of array!\n" : "OK",
15233 (UV)(place - RExC_emit_start),
15234 (UV)(RExC_parse - RExC_start),
15235 (UV)RExC_offsets[0]));
15236 Set_Node_Offset(place, RExC_parse);
15237 Set_Node_Length(place, 1);
15240 src = NEXTOPER(place);
15241 FILL_ADVANCE_NODE(place, op);
15242 Zero(src, offset, regnode);
15246 - regtail - set the next-pointer at the end of a node chain of p to val.
15247 - SEE ALSO: regtail_study
15249 /* TODO: All three parms should be const */
15251 S_regtail(pTHX_ RExC_state_t *pRExC_state, regnode *p,
15252 const regnode *val,U32 depth)
15256 GET_RE_DEBUG_FLAGS_DECL;
15258 PERL_ARGS_ASSERT_REGTAIL;
15260 PERL_UNUSED_ARG(depth);
15266 /* Find last node. */
15269 regnode * const temp = regnext(scan);
15271 SV * const mysv=sv_newmortal();
15272 DEBUG_PARSE_MSG((scan==p ? "tail" : ""));
15273 regprop(RExC_rx, mysv, scan, NULL);
15274 PerlIO_printf(Perl_debug_log, "~ %s (%d) %s %s\n",
15275 SvPV_nolen_const(mysv), REG_NODE_NUM(scan),
15276 (temp == NULL ? "->" : ""),
15277 (temp == NULL ? PL_reg_name[OP(val)] : "")
15285 if (reg_off_by_arg[OP(scan)]) {
15286 ARG_SET(scan, val - scan);
15289 NEXT_OFF(scan) = val - scan;
15295 - regtail_study - set the next-pointer at the end of a node chain of p to val.
15296 - Look for optimizable sequences at the same time.
15297 - currently only looks for EXACT chains.
15299 This is experimental code. The idea is to use this routine to perform
15300 in place optimizations on branches and groups as they are constructed,
15301 with the long term intention of removing optimization from study_chunk so
15302 that it is purely analytical.
15304 Currently only used when in DEBUG mode. The macro REGTAIL_STUDY() is used
15305 to control which is which.
15308 /* TODO: All four parms should be const */
15311 S_regtail_study(pTHX_ RExC_state_t *pRExC_state, regnode *p,
15312 const regnode *val,U32 depth)
15317 #ifdef EXPERIMENTAL_INPLACESCAN
15320 GET_RE_DEBUG_FLAGS_DECL;
15322 PERL_ARGS_ASSERT_REGTAIL_STUDY;
15328 /* Find last node. */
15332 regnode * const temp = regnext(scan);
15333 #ifdef EXPERIMENTAL_INPLACESCAN
15334 if (PL_regkind[OP(scan)] == EXACT) {
15335 bool unfolded_multi_char; /* Unexamined in this routine */
15336 if (join_exact(pRExC_state, scan, &min,
15337 &unfolded_multi_char, 1, val, depth+1))
15342 switch (OP(scan)) {
15345 case EXACTFA_NO_TRIE:
15350 if( exact == PSEUDO )
15352 else if ( exact != OP(scan) )
15361 SV * const mysv=sv_newmortal();
15362 DEBUG_PARSE_MSG((scan==p ? "tsdy" : ""));
15363 regprop(RExC_rx, mysv, scan, NULL);
15364 PerlIO_printf(Perl_debug_log, "~ %s (%d) -> %s\n",
15365 SvPV_nolen_const(mysv),
15366 REG_NODE_NUM(scan),
15367 PL_reg_name[exact]);
15374 SV * const mysv_val=sv_newmortal();
15375 DEBUG_PARSE_MSG("");
15376 regprop(RExC_rx, mysv_val, val, NULL);
15377 PerlIO_printf(Perl_debug_log,
15378 "~ attach to %s (%"IVdf") offset to %"IVdf"\n",
15379 SvPV_nolen_const(mysv_val),
15380 (IV)REG_NODE_NUM(val),
15384 if (reg_off_by_arg[OP(scan)]) {
15385 ARG_SET(scan, val - scan);
15388 NEXT_OFF(scan) = val - scan;
15396 - regdump - dump a regexp onto Perl_debug_log in vaguely comprehensible form
15401 S_regdump_intflags(pTHX_ const char *lead, const U32 flags)
15406 ASSUME(REG_INTFLAGS_NAME_SIZE <= sizeof(flags)*8);
15408 for (bit=0; bit<REG_INTFLAGS_NAME_SIZE; bit++) {
15409 if (flags & (1<<bit)) {
15410 if (!set++ && lead)
15411 PerlIO_printf(Perl_debug_log, "%s",lead);
15412 PerlIO_printf(Perl_debug_log, "%s ",PL_reg_intflags_name[bit]);
15417 PerlIO_printf(Perl_debug_log, "\n");
15419 PerlIO_printf(Perl_debug_log, "%s[none-set]\n",lead);
15424 S_regdump_extflags(pTHX_ const char *lead, const U32 flags)
15430 ASSUME(REG_EXTFLAGS_NAME_SIZE <= sizeof(flags)*8);
15432 for (bit=0; bit<REG_EXTFLAGS_NAME_SIZE; bit++) {
15433 if (flags & (1<<bit)) {
15434 if ((1<<bit) & RXf_PMf_CHARSET) { /* Output separately, below */
15437 if (!set++ && lead)
15438 PerlIO_printf(Perl_debug_log, "%s",lead);
15439 PerlIO_printf(Perl_debug_log, "%s ",PL_reg_extflags_name[bit]);
15442 if ((cs = get_regex_charset(flags)) != REGEX_DEPENDS_CHARSET) {
15443 if (!set++ && lead) {
15444 PerlIO_printf(Perl_debug_log, "%s",lead);
15447 case REGEX_UNICODE_CHARSET:
15448 PerlIO_printf(Perl_debug_log, "UNICODE");
15450 case REGEX_LOCALE_CHARSET:
15451 PerlIO_printf(Perl_debug_log, "LOCALE");
15453 case REGEX_ASCII_RESTRICTED_CHARSET:
15454 PerlIO_printf(Perl_debug_log, "ASCII-RESTRICTED");
15456 case REGEX_ASCII_MORE_RESTRICTED_CHARSET:
15457 PerlIO_printf(Perl_debug_log, "ASCII-MORE_RESTRICTED");
15460 PerlIO_printf(Perl_debug_log, "UNKNOWN CHARACTER SET");
15466 PerlIO_printf(Perl_debug_log, "\n");
15468 PerlIO_printf(Perl_debug_log, "%s[none-set]\n",lead);
15474 Perl_regdump(pTHX_ const regexp *r)
15478 SV * const sv = sv_newmortal();
15479 SV *dsv= sv_newmortal();
15480 RXi_GET_DECL(r,ri);
15481 GET_RE_DEBUG_FLAGS_DECL;
15483 PERL_ARGS_ASSERT_REGDUMP;
15485 (void)dumpuntil(r, ri->program, ri->program + 1, NULL, NULL, sv, 0, 0);
15487 /* Header fields of interest. */
15488 if (r->anchored_substr) {
15489 RE_PV_QUOTED_DECL(s, 0, dsv, SvPVX_const(r->anchored_substr),
15490 RE_SV_DUMPLEN(r->anchored_substr), 30);
15491 PerlIO_printf(Perl_debug_log,
15492 "anchored %s%s at %"IVdf" ",
15493 s, RE_SV_TAIL(r->anchored_substr),
15494 (IV)r->anchored_offset);
15495 } else if (r->anchored_utf8) {
15496 RE_PV_QUOTED_DECL(s, 1, dsv, SvPVX_const(r->anchored_utf8),
15497 RE_SV_DUMPLEN(r->anchored_utf8), 30);
15498 PerlIO_printf(Perl_debug_log,
15499 "anchored utf8 %s%s at %"IVdf" ",
15500 s, RE_SV_TAIL(r->anchored_utf8),
15501 (IV)r->anchored_offset);
15503 if (r->float_substr) {
15504 RE_PV_QUOTED_DECL(s, 0, dsv, SvPVX_const(r->float_substr),
15505 RE_SV_DUMPLEN(r->float_substr), 30);
15506 PerlIO_printf(Perl_debug_log,
15507 "floating %s%s at %"IVdf"..%"UVuf" ",
15508 s, RE_SV_TAIL(r->float_substr),
15509 (IV)r->float_min_offset, (UV)r->float_max_offset);
15510 } else if (r->float_utf8) {
15511 RE_PV_QUOTED_DECL(s, 1, dsv, SvPVX_const(r->float_utf8),
15512 RE_SV_DUMPLEN(r->float_utf8), 30);
15513 PerlIO_printf(Perl_debug_log,
15514 "floating utf8 %s%s at %"IVdf"..%"UVuf" ",
15515 s, RE_SV_TAIL(r->float_utf8),
15516 (IV)r->float_min_offset, (UV)r->float_max_offset);
15518 if (r->check_substr || r->check_utf8)
15519 PerlIO_printf(Perl_debug_log,
15521 (r->check_substr == r->float_substr
15522 && r->check_utf8 == r->float_utf8
15523 ? "(checking floating" : "(checking anchored"));
15524 if (r->intflags & PREGf_NOSCAN)
15525 PerlIO_printf(Perl_debug_log, " noscan");
15526 if (r->extflags & RXf_CHECK_ALL)
15527 PerlIO_printf(Perl_debug_log, " isall");
15528 if (r->check_substr || r->check_utf8)
15529 PerlIO_printf(Perl_debug_log, ") ");
15531 if (ri->regstclass) {
15532 regprop(r, sv, ri->regstclass, NULL);
15533 PerlIO_printf(Perl_debug_log, "stclass %s ", SvPVX_const(sv));
15535 if (r->intflags & PREGf_ANCH) {
15536 PerlIO_printf(Perl_debug_log, "anchored");
15537 if (r->intflags & PREGf_ANCH_BOL)
15538 PerlIO_printf(Perl_debug_log, "(BOL)");
15539 if (r->intflags & PREGf_ANCH_MBOL)
15540 PerlIO_printf(Perl_debug_log, "(MBOL)");
15541 if (r->intflags & PREGf_ANCH_SBOL)
15542 PerlIO_printf(Perl_debug_log, "(SBOL)");
15543 if (r->intflags & PREGf_ANCH_GPOS)
15544 PerlIO_printf(Perl_debug_log, "(GPOS)");
15545 PerlIO_putc(Perl_debug_log, ' ');
15547 if (r->intflags & PREGf_GPOS_SEEN)
15548 PerlIO_printf(Perl_debug_log, "GPOS:%"UVuf" ", (UV)r->gofs);
15549 if (r->intflags & PREGf_SKIP)
15550 PerlIO_printf(Perl_debug_log, "plus ");
15551 if (r->intflags & PREGf_IMPLICIT)
15552 PerlIO_printf(Perl_debug_log, "implicit ");
15553 PerlIO_printf(Perl_debug_log, "minlen %"IVdf" ", (IV)r->minlen);
15554 if (r->extflags & RXf_EVAL_SEEN)
15555 PerlIO_printf(Perl_debug_log, "with eval ");
15556 PerlIO_printf(Perl_debug_log, "\n");
15558 regdump_extflags("r->extflags: ",r->extflags);
15559 regdump_intflags("r->intflags: ",r->intflags);
15562 PERL_ARGS_ASSERT_REGDUMP;
15563 PERL_UNUSED_CONTEXT;
15564 PERL_UNUSED_ARG(r);
15565 #endif /* DEBUGGING */
15569 - regprop - printable representation of opcode, with run time support
15573 Perl_regprop(pTHX_ const regexp *prog, SV *sv, const regnode *o, const regmatch_info *reginfo)
15579 /* Should be synchronized with * ANYOF_ #xdefines in regcomp.h */
15580 static const char * const anyofs[] = {
15581 #if _CC_WORDCHAR != 0 || _CC_DIGIT != 1 || _CC_ALPHA != 2 || _CC_LOWER != 3 \
15582 || _CC_UPPER != 4 || _CC_PUNCT != 5 || _CC_PRINT != 6 \
15583 || _CC_ALPHANUMERIC != 7 || _CC_GRAPH != 8 || _CC_CASED != 9 \
15584 || _CC_SPACE != 10 || _CC_BLANK != 11 || _CC_XDIGIT != 12 \
15585 || _CC_PSXSPC != 13 || _CC_CNTRL != 14 || _CC_ASCII != 15 \
15586 || _CC_VERTSPACE != 16
15587 #error Need to adjust order of anyofs[]
15624 RXi_GET_DECL(prog,progi);
15625 GET_RE_DEBUG_FLAGS_DECL;
15627 PERL_ARGS_ASSERT_REGPROP;
15631 if (OP(o) > REGNODE_MAX) /* regnode.type is unsigned */
15632 /* It would be nice to FAIL() here, but this may be called from
15633 regexec.c, and it would be hard to supply pRExC_state. */
15634 Perl_croak(aTHX_ "Corrupted regexp opcode %d > %d",
15635 (int)OP(o), (int)REGNODE_MAX);
15636 sv_catpv(sv, PL_reg_name[OP(o)]); /* Take off const! */
15638 k = PL_regkind[OP(o)];
15641 sv_catpvs(sv, " ");
15642 /* Using is_utf8_string() (via PERL_PV_UNI_DETECT)
15643 * is a crude hack but it may be the best for now since
15644 * we have no flag "this EXACTish node was UTF-8"
15646 pv_pretty(sv, STRING(o), STR_LEN(o), 60, PL_colors[0], PL_colors[1],
15647 PERL_PV_ESCAPE_UNI_DETECT |
15648 PERL_PV_ESCAPE_NONASCII |
15649 PERL_PV_PRETTY_ELLIPSES |
15650 PERL_PV_PRETTY_LTGT |
15651 PERL_PV_PRETTY_NOCLEAR
15653 } else if (k == TRIE) {
15654 /* print the details of the trie in dumpuntil instead, as
15655 * progi->data isn't available here */
15656 const char op = OP(o);
15657 const U32 n = ARG(o);
15658 const reg_ac_data * const ac = IS_TRIE_AC(op) ?
15659 (reg_ac_data *)progi->data->data[n] :
15661 const reg_trie_data * const trie
15662 = (reg_trie_data*)progi->data->data[!IS_TRIE_AC(op) ? n : ac->trie];
15664 Perl_sv_catpvf(aTHX_ sv, "-%s",PL_reg_name[o->flags]);
15665 DEBUG_TRIE_COMPILE_r(
15666 Perl_sv_catpvf(aTHX_ sv,
15667 "<S:%"UVuf"/%"IVdf" W:%"UVuf" L:%"UVuf"/%"UVuf" C:%"UVuf"/%"UVuf">",
15668 (UV)trie->startstate,
15669 (IV)trie->statecount-1, /* -1 because of the unused 0 element */
15670 (UV)trie->wordcount,
15673 (UV)TRIE_CHARCOUNT(trie),
15674 (UV)trie->uniquecharcount
15677 if ( IS_ANYOF_TRIE(op) || trie->bitmap ) {
15678 sv_catpvs(sv, "[");
15679 (void) put_latin1_charclass_innards(sv, IS_ANYOF_TRIE(op)
15681 : TRIE_BITMAP(trie));
15682 sv_catpvs(sv, "]");
15685 } else if (k == CURLY) {
15686 if (OP(o) == CURLYM || OP(o) == CURLYN || OP(o) == CURLYX)
15687 Perl_sv_catpvf(aTHX_ sv, "[%d]", o->flags); /* Parenth number */
15688 Perl_sv_catpvf(aTHX_ sv, " {%d,%d}", ARG1(o), ARG2(o));
15690 else if (k == WHILEM && o->flags) /* Ordinal/of */
15691 Perl_sv_catpvf(aTHX_ sv, "[%d/%d]", o->flags & 0xf, o->flags>>4);
15692 else if (k == REF || k == OPEN || k == CLOSE
15693 || k == GROUPP || OP(o)==ACCEPT)
15695 Perl_sv_catpvf(aTHX_ sv, "%d", (int)ARG(o)); /* Parenth number */
15696 if ( RXp_PAREN_NAMES(prog) ) {
15697 if ( k != REF || (OP(o) < NREF)) {
15698 AV *list= MUTABLE_AV(progi->data->data[progi->name_list_idx]);
15699 SV **name= av_fetch(list, ARG(o), 0 );
15701 Perl_sv_catpvf(aTHX_ sv, " '%"SVf"'", SVfARG(*name));
15704 AV *list= MUTABLE_AV(progi->data->data[ progi->name_list_idx ]);
15705 SV *sv_dat= MUTABLE_SV(progi->data->data[ ARG( o ) ]);
15706 I32 *nums=(I32*)SvPVX(sv_dat);
15707 SV **name= av_fetch(list, nums[0], 0 );
15710 for ( n=0; n<SvIVX(sv_dat); n++ ) {
15711 Perl_sv_catpvf(aTHX_ sv, "%s%"IVdf,
15712 (n ? "," : ""), (IV)nums[n]);
15714 Perl_sv_catpvf(aTHX_ sv, " '%"SVf"'", SVfARG(*name));
15718 if ( k == REF && reginfo) {
15719 U32 n = ARG(o); /* which paren pair */
15720 I32 ln = prog->offs[n].start;
15721 if (prog->lastparen < n || ln == -1)
15722 Perl_sv_catpvf(aTHX_ sv, ": FAIL");
15723 else if (ln == prog->offs[n].end)
15724 Perl_sv_catpvf(aTHX_ sv, ": ACCEPT - EMPTY STRING");
15726 const char *s = reginfo->strbeg + ln;
15727 Perl_sv_catpvf(aTHX_ sv, ": ");
15728 Perl_pv_pretty( aTHX_ sv, s, prog->offs[n].end - prog->offs[n].start, 32, 0, 0,
15729 PERL_PV_ESCAPE_UNI_DETECT|PERL_PV_PRETTY_NOCLEAR|PERL_PV_PRETTY_ELLIPSES|PERL_PV_PRETTY_QUOTE );
15732 } else if (k == GOSUB)
15733 /* Paren and offset */
15734 Perl_sv_catpvf(aTHX_ sv, "%d[%+d]", (int)ARG(o),(int)ARG2L(o));
15735 else if (k == VERB) {
15737 Perl_sv_catpvf(aTHX_ sv, ":%"SVf,
15738 SVfARG((MUTABLE_SV(progi->data->data[ ARG( o ) ]))));
15739 } else if (k == LOGICAL)
15740 /* 2: embedded, otherwise 1 */
15741 Perl_sv_catpvf(aTHX_ sv, "[%d]", o->flags);
15742 else if (k == ANYOF) {
15743 const U8 flags = ANYOF_FLAGS(o);
15747 if (flags & ANYOF_LOCALE_FLAGS)
15748 sv_catpvs(sv, "{loc}");
15749 if (flags & ANYOF_LOC_FOLD)
15750 sv_catpvs(sv, "{i}");
15751 Perl_sv_catpvf(aTHX_ sv, "[%s", PL_colors[0]);
15752 if (flags & ANYOF_INVERT)
15753 sv_catpvs(sv, "^");
15755 /* output what the standard cp 0-255 bitmap matches */
15756 do_sep = put_latin1_charclass_innards(sv, ANYOF_BITMAP(o));
15758 /* output any special charclass tests (used entirely under use
15760 if (ANYOF_POSIXL_TEST_ANY_SET(o)) {
15762 for (i = 0; i < ANYOF_POSIXL_MAX; i++) {
15763 if (ANYOF_POSIXL_TEST(o,i)) {
15764 sv_catpv(sv, anyofs[i]);
15770 if ((flags & (ANYOF_ABOVE_LATIN1_ALL
15772 |ANYOF_NONBITMAP_NON_UTF8
15776 Perl_sv_catpvf(aTHX_ sv,"%s][%s",PL_colors[1],PL_colors[0]);
15777 if (flags & ANYOF_INVERT)
15778 /*make sure the invert info is in each */
15779 sv_catpvs(sv, "^");
15782 if (flags & ANYOF_NON_UTF8_NON_ASCII_ALL) {
15783 sv_catpvs(sv, "{non-utf8-latin1-all}");
15786 /* output information about the unicode matching */
15787 if (flags & ANYOF_ABOVE_LATIN1_ALL)
15788 sv_catpvs(sv, "{unicode_all}");
15789 else if (ARG(o) != ANYOF_NONBITMAP_EMPTY) {
15790 SV *lv; /* Set if there is something outside the bit map. */
15791 bool byte_output = FALSE; /* If something in the bitmap has
15793 SV *only_utf8_locale;
15795 /* Get the stuff that wasn't in the bitmap */
15796 (void) _get_regclass_nonbitmap_data(prog, o, FALSE,
15797 &lv, &only_utf8_locale);
15798 if (lv && lv != &PL_sv_undef) {
15799 char *s = savesvpv(lv);
15800 char * const origs = s;
15802 while (*s && *s != '\n')
15806 const char * const t = ++s;
15808 if (flags & ANYOF_NONBITMAP_NON_UTF8) {
15809 sv_catpvs(sv, "{outside bitmap}");
15812 sv_catpvs(sv, "{utf8}");
15816 sv_catpvs(sv, " ");
15822 /* Truncate very long output */
15823 if (s - origs > 256) {
15824 Perl_sv_catpvf(aTHX_ sv,
15826 (int) (s - origs - 1),
15832 else if (*s == '\t') {
15846 SvREFCNT_dec_NN(lv);
15849 if ((flags & ANYOF_LOC_FOLD)
15850 && only_utf8_locale
15851 && only_utf8_locale != &PL_sv_undef)
15854 int max_entries = 256;
15856 sv_catpvs(sv, "{utf8 locale}");
15857 invlist_iterinit(only_utf8_locale);
15858 while (invlist_iternext(only_utf8_locale,
15860 put_range(sv, start, end);
15862 if (max_entries < 0) {
15863 sv_catpvs(sv, "...");
15867 invlist_iterfinish(only_utf8_locale);
15872 Perl_sv_catpvf(aTHX_ sv, "%s]", PL_colors[1]);
15874 else if (k == POSIXD || k == NPOSIXD) {
15875 U8 index = FLAGS(o) * 2;
15876 if (index < C_ARRAY_LENGTH(anyofs)) {
15877 if (*anyofs[index] != '[') {
15880 sv_catpv(sv, anyofs[index]);
15881 if (*anyofs[index] != '[') {
15886 Perl_sv_catpvf(aTHX_ sv, "[illegal type=%d])", index);
15889 else if (k == BRANCHJ && (OP(o) == UNLESSM || OP(o) == IFMATCH))
15890 Perl_sv_catpvf(aTHX_ sv, "[%d]", -(o->flags));
15892 PERL_UNUSED_CONTEXT;
15893 PERL_UNUSED_ARG(sv);
15894 PERL_UNUSED_ARG(o);
15895 PERL_UNUSED_ARG(prog);
15896 PERL_UNUSED_ARG(reginfo);
15897 #endif /* DEBUGGING */
15903 Perl_re_intuit_string(pTHX_ REGEXP * const r)
15904 { /* Assume that RE_INTUIT is set */
15906 struct regexp *const prog = ReANY(r);
15907 GET_RE_DEBUG_FLAGS_DECL;
15909 PERL_ARGS_ASSERT_RE_INTUIT_STRING;
15910 PERL_UNUSED_CONTEXT;
15914 const char * const s = SvPV_nolen_const(prog->check_substr
15915 ? prog->check_substr : prog->check_utf8);
15917 if (!PL_colorset) reginitcolors();
15918 PerlIO_printf(Perl_debug_log,
15919 "%sUsing REx %ssubstr:%s \"%s%.60s%s%s\"\n",
15921 prog->check_substr ? "" : "utf8 ",
15922 PL_colors[5],PL_colors[0],
15925 (strlen(s) > 60 ? "..." : ""));
15928 return prog->check_substr ? prog->check_substr : prog->check_utf8;
15934 handles refcounting and freeing the perl core regexp structure. When
15935 it is necessary to actually free the structure the first thing it
15936 does is call the 'free' method of the regexp_engine associated to
15937 the regexp, allowing the handling of the void *pprivate; member
15938 first. (This routine is not overridable by extensions, which is why
15939 the extensions free is called first.)
15941 See regdupe and regdupe_internal if you change anything here.
15943 #ifndef PERL_IN_XSUB_RE
15945 Perl_pregfree(pTHX_ REGEXP *r)
15951 Perl_pregfree2(pTHX_ REGEXP *rx)
15954 struct regexp *const r = ReANY(rx);
15955 GET_RE_DEBUG_FLAGS_DECL;
15957 PERL_ARGS_ASSERT_PREGFREE2;
15959 if (r->mother_re) {
15960 ReREFCNT_dec(r->mother_re);
15962 CALLREGFREE_PVT(rx); /* free the private data */
15963 SvREFCNT_dec(RXp_PAREN_NAMES(r));
15964 Safefree(r->xpv_len_u.xpvlenu_pv);
15967 SvREFCNT_dec(r->anchored_substr);
15968 SvREFCNT_dec(r->anchored_utf8);
15969 SvREFCNT_dec(r->float_substr);
15970 SvREFCNT_dec(r->float_utf8);
15971 Safefree(r->substrs);
15973 RX_MATCH_COPY_FREE(rx);
15974 #ifdef PERL_ANY_COW
15975 SvREFCNT_dec(r->saved_copy);
15978 SvREFCNT_dec(r->qr_anoncv);
15979 rx->sv_u.svu_rx = 0;
15984 This is a hacky workaround to the structural issue of match results
15985 being stored in the regexp structure which is in turn stored in
15986 PL_curpm/PL_reg_curpm. The problem is that due to qr// the pattern
15987 could be PL_curpm in multiple contexts, and could require multiple
15988 result sets being associated with the pattern simultaneously, such
15989 as when doing a recursive match with (??{$qr})
15991 The solution is to make a lightweight copy of the regexp structure
15992 when a qr// is returned from the code executed by (??{$qr}) this
15993 lightweight copy doesn't actually own any of its data except for
15994 the starp/end and the actual regexp structure itself.
16000 Perl_reg_temp_copy (pTHX_ REGEXP *ret_x, REGEXP *rx)
16002 struct regexp *ret;
16003 struct regexp *const r = ReANY(rx);
16004 const bool islv = ret_x && SvTYPE(ret_x) == SVt_PVLV;
16006 PERL_ARGS_ASSERT_REG_TEMP_COPY;
16009 ret_x = (REGEXP*) newSV_type(SVt_REGEXP);
16011 SvOK_off((SV *)ret_x);
16013 /* For PVLVs, SvANY points to the xpvlv body while sv_u points
16014 to the regexp. (For SVt_REGEXPs, sv_upgrade has already
16015 made both spots point to the same regexp body.) */
16016 REGEXP *temp = (REGEXP *)newSV_type(SVt_REGEXP);
16017 assert(!SvPVX(ret_x));
16018 ret_x->sv_u.svu_rx = temp->sv_any;
16019 temp->sv_any = NULL;
16020 SvFLAGS(temp) = (SvFLAGS(temp) & ~SVTYPEMASK) | SVt_NULL;
16021 SvREFCNT_dec_NN(temp);
16022 /* SvCUR still resides in the xpvlv struct, so the regexp copy-
16023 ing below will not set it. */
16024 SvCUR_set(ret_x, SvCUR(rx));
16027 /* This ensures that SvTHINKFIRST(sv) is true, and hence that
16028 sv_force_normal(sv) is called. */
16030 ret = ReANY(ret_x);
16032 SvFLAGS(ret_x) |= SvUTF8(rx);
16033 /* We share the same string buffer as the original regexp, on which we
16034 hold a reference count, incremented when mother_re is set below.
16035 The string pointer is copied here, being part of the regexp struct.
16037 memcpy(&(ret->xpv_cur), &(r->xpv_cur),
16038 sizeof(regexp) - STRUCT_OFFSET(regexp, xpv_cur));
16040 const I32 npar = r->nparens+1;
16041 Newx(ret->offs, npar, regexp_paren_pair);
16042 Copy(r->offs, ret->offs, npar, regexp_paren_pair);
16045 Newx(ret->substrs, 1, struct reg_substr_data);
16046 StructCopy(r->substrs, ret->substrs, struct reg_substr_data);
16048 SvREFCNT_inc_void(ret->anchored_substr);
16049 SvREFCNT_inc_void(ret->anchored_utf8);
16050 SvREFCNT_inc_void(ret->float_substr);
16051 SvREFCNT_inc_void(ret->float_utf8);
16053 /* check_substr and check_utf8, if non-NULL, point to either their
16054 anchored or float namesakes, and don't hold a second reference. */
16056 RX_MATCH_COPIED_off(ret_x);
16057 #ifdef PERL_ANY_COW
16058 ret->saved_copy = NULL;
16060 ret->mother_re = ReREFCNT_inc(r->mother_re ? r->mother_re : rx);
16061 SvREFCNT_inc_void(ret->qr_anoncv);
16067 /* regfree_internal()
16069 Free the private data in a regexp. This is overloadable by
16070 extensions. Perl takes care of the regexp structure in pregfree(),
16071 this covers the *pprivate pointer which technically perl doesn't
16072 know about, however of course we have to handle the
16073 regexp_internal structure when no extension is in use.
16075 Note this is called before freeing anything in the regexp
16080 Perl_regfree_internal(pTHX_ REGEXP * const rx)
16083 struct regexp *const r = ReANY(rx);
16084 RXi_GET_DECL(r,ri);
16085 GET_RE_DEBUG_FLAGS_DECL;
16087 PERL_ARGS_ASSERT_REGFREE_INTERNAL;
16093 SV *dsv= sv_newmortal();
16094 RE_PV_QUOTED_DECL(s, RX_UTF8(rx),
16095 dsv, RX_PRECOMP(rx), RX_PRELEN(rx), 60);
16096 PerlIO_printf(Perl_debug_log,"%sFreeing REx:%s %s\n",
16097 PL_colors[4],PL_colors[5],s);
16100 #ifdef RE_TRACK_PATTERN_OFFSETS
16102 Safefree(ri->u.offsets); /* 20010421 MJD */
16104 if (ri->code_blocks) {
16106 for (n = 0; n < ri->num_code_blocks; n++)
16107 SvREFCNT_dec(ri->code_blocks[n].src_regex);
16108 Safefree(ri->code_blocks);
16112 int n = ri->data->count;
16115 /* If you add a ->what type here, update the comment in regcomp.h */
16116 switch (ri->data->what[n]) {
16122 SvREFCNT_dec(MUTABLE_SV(ri->data->data[n]));
16125 Safefree(ri->data->data[n]);
16131 { /* Aho Corasick add-on structure for a trie node.
16132 Used in stclass optimization only */
16134 reg_ac_data *aho=(reg_ac_data*)ri->data->data[n];
16136 refcount = --aho->refcount;
16139 PerlMemShared_free(aho->states);
16140 PerlMemShared_free(aho->fail);
16141 /* do this last!!!! */
16142 PerlMemShared_free(ri->data->data[n]);
16143 PerlMemShared_free(ri->regstclass);
16149 /* trie structure. */
16151 reg_trie_data *trie=(reg_trie_data*)ri->data->data[n];
16153 refcount = --trie->refcount;
16156 PerlMemShared_free(trie->charmap);
16157 PerlMemShared_free(trie->states);
16158 PerlMemShared_free(trie->trans);
16160 PerlMemShared_free(trie->bitmap);
16162 PerlMemShared_free(trie->jump);
16163 PerlMemShared_free(trie->wordinfo);
16164 /* do this last!!!! */
16165 PerlMemShared_free(ri->data->data[n]);
16170 Perl_croak(aTHX_ "panic: regfree data code '%c'",
16171 ri->data->what[n]);
16174 Safefree(ri->data->what);
16175 Safefree(ri->data);
16181 #define av_dup_inc(s,t) MUTABLE_AV(sv_dup_inc((const SV *)s,t))
16182 #define hv_dup_inc(s,t) MUTABLE_HV(sv_dup_inc((const SV *)s,t))
16183 #define SAVEPVN(p,n) ((p) ? savepvn(p,n) : NULL)
16186 re_dup - duplicate a regexp.
16188 This routine is expected to clone a given regexp structure. It is only
16189 compiled under USE_ITHREADS.
16191 After all of the core data stored in struct regexp is duplicated
16192 the regexp_engine.dupe method is used to copy any private data
16193 stored in the *pprivate pointer. This allows extensions to handle
16194 any duplication it needs to do.
16196 See pregfree() and regfree_internal() if you change anything here.
16198 #if defined(USE_ITHREADS)
16199 #ifndef PERL_IN_XSUB_RE
16201 Perl_re_dup_guts(pTHX_ const REGEXP *sstr, REGEXP *dstr, CLONE_PARAMS *param)
16205 const struct regexp *r = ReANY(sstr);
16206 struct regexp *ret = ReANY(dstr);
16208 PERL_ARGS_ASSERT_RE_DUP_GUTS;
16210 npar = r->nparens+1;
16211 Newx(ret->offs, npar, regexp_paren_pair);
16212 Copy(r->offs, ret->offs, npar, regexp_paren_pair);
16214 if (ret->substrs) {
16215 /* Do it this way to avoid reading from *r after the StructCopy().
16216 That way, if any of the sv_dup_inc()s dislodge *r from the L1
16217 cache, it doesn't matter. */
16218 const bool anchored = r->check_substr
16219 ? r->check_substr == r->anchored_substr
16220 : r->check_utf8 == r->anchored_utf8;
16221 Newx(ret->substrs, 1, struct reg_substr_data);
16222 StructCopy(r->substrs, ret->substrs, struct reg_substr_data);
16224 ret->anchored_substr = sv_dup_inc(ret->anchored_substr, param);
16225 ret->anchored_utf8 = sv_dup_inc(ret->anchored_utf8, param);
16226 ret->float_substr = sv_dup_inc(ret->float_substr, param);
16227 ret->float_utf8 = sv_dup_inc(ret->float_utf8, param);
16229 /* check_substr and check_utf8, if non-NULL, point to either their
16230 anchored or float namesakes, and don't hold a second reference. */
16232 if (ret->check_substr) {
16234 assert(r->check_utf8 == r->anchored_utf8);
16235 ret->check_substr = ret->anchored_substr;
16236 ret->check_utf8 = ret->anchored_utf8;
16238 assert(r->check_substr == r->float_substr);
16239 assert(r->check_utf8 == r->float_utf8);
16240 ret->check_substr = ret->float_substr;
16241 ret->check_utf8 = ret->float_utf8;
16243 } else if (ret->check_utf8) {
16245 ret->check_utf8 = ret->anchored_utf8;
16247 ret->check_utf8 = ret->float_utf8;
16252 RXp_PAREN_NAMES(ret) = hv_dup_inc(RXp_PAREN_NAMES(ret), param);
16253 ret->qr_anoncv = MUTABLE_CV(sv_dup_inc((const SV *)ret->qr_anoncv, param));
16256 RXi_SET(ret,CALLREGDUPE_PVT(dstr,param));
16258 if (RX_MATCH_COPIED(dstr))
16259 ret->subbeg = SAVEPVN(ret->subbeg, ret->sublen);
16261 ret->subbeg = NULL;
16262 #ifdef PERL_ANY_COW
16263 ret->saved_copy = NULL;
16266 /* Whether mother_re be set or no, we need to copy the string. We
16267 cannot refrain from copying it when the storage points directly to
16268 our mother regexp, because that's
16269 1: a buffer in a different thread
16270 2: something we no longer hold a reference on
16271 so we need to copy it locally. */
16272 RX_WRAPPED(dstr) = SAVEPVN(RX_WRAPPED(sstr), SvCUR(sstr)+1);
16273 ret->mother_re = NULL;
16275 #endif /* PERL_IN_XSUB_RE */
16280 This is the internal complement to regdupe() which is used to copy
16281 the structure pointed to by the *pprivate pointer in the regexp.
16282 This is the core version of the extension overridable cloning hook.
16283 The regexp structure being duplicated will be copied by perl prior
16284 to this and will be provided as the regexp *r argument, however
16285 with the /old/ structures pprivate pointer value. Thus this routine
16286 may override any copying normally done by perl.
16288 It returns a pointer to the new regexp_internal structure.
16292 Perl_regdupe_internal(pTHX_ REGEXP * const rx, CLONE_PARAMS *param)
16295 struct regexp *const r = ReANY(rx);
16296 regexp_internal *reti;
16298 RXi_GET_DECL(r,ri);
16300 PERL_ARGS_ASSERT_REGDUPE_INTERNAL;
16304 Newxc(reti, sizeof(regexp_internal) + len*sizeof(regnode),
16305 char, regexp_internal);
16306 Copy(ri->program, reti->program, len+1, regnode);
16308 reti->num_code_blocks = ri->num_code_blocks;
16309 if (ri->code_blocks) {
16311 Newxc(reti->code_blocks, ri->num_code_blocks, struct reg_code_block,
16312 struct reg_code_block);
16313 Copy(ri->code_blocks, reti->code_blocks, ri->num_code_blocks,
16314 struct reg_code_block);
16315 for (n = 0; n < ri->num_code_blocks; n++)
16316 reti->code_blocks[n].src_regex = (REGEXP*)
16317 sv_dup_inc((SV*)(ri->code_blocks[n].src_regex), param);
16320 reti->code_blocks = NULL;
16322 reti->regstclass = NULL;
16325 struct reg_data *d;
16326 const int count = ri->data->count;
16329 Newxc(d, sizeof(struct reg_data) + count*sizeof(void *),
16330 char, struct reg_data);
16331 Newx(d->what, count, U8);
16334 for (i = 0; i < count; i++) {
16335 d->what[i] = ri->data->what[i];
16336 switch (d->what[i]) {
16337 /* see also regcomp.h and regfree_internal() */
16338 case 'a': /* actually an AV, but the dup function is identical. */
16342 case 'u': /* actually an HV, but the dup function is identical. */
16343 d->data[i] = sv_dup_inc((const SV *)ri->data->data[i], param);
16346 /* This is cheating. */
16347 Newx(d->data[i], 1, regnode_ssc);
16348 StructCopy(ri->data->data[i], d->data[i], regnode_ssc);
16349 reti->regstclass = (regnode*)d->data[i];
16352 /* Trie stclasses are readonly and can thus be shared
16353 * without duplication. We free the stclass in pregfree
16354 * when the corresponding reg_ac_data struct is freed.
16356 reti->regstclass= ri->regstclass;
16360 ((reg_trie_data*)ri->data->data[i])->refcount++;
16365 d->data[i] = ri->data->data[i];
16368 Perl_croak(aTHX_ "panic: re_dup unknown data code '%c'",
16369 ri->data->what[i]);
16378 reti->name_list_idx = ri->name_list_idx;
16380 #ifdef RE_TRACK_PATTERN_OFFSETS
16381 if (ri->u.offsets) {
16382 Newx(reti->u.offsets, 2*len+1, U32);
16383 Copy(ri->u.offsets, reti->u.offsets, 2*len+1, U32);
16386 SetProgLen(reti,len);
16389 return (void*)reti;
16392 #endif /* USE_ITHREADS */
16394 #ifndef PERL_IN_XSUB_RE
16397 - regnext - dig the "next" pointer out of a node
16400 Perl_regnext(pTHX_ regnode *p)
16408 if (OP(p) > REGNODE_MAX) { /* regnode.type is unsigned */
16409 Perl_croak(aTHX_ "Corrupted regexp opcode %d > %d",
16410 (int)OP(p), (int)REGNODE_MAX);
16413 offset = (reg_off_by_arg[OP(p)] ? ARG(p) : NEXT_OFF(p));
16422 S_re_croak2(pTHX_ bool utf8, const char* pat1,const char* pat2,...)
16425 STRLEN l1 = strlen(pat1);
16426 STRLEN l2 = strlen(pat2);
16429 const char *message;
16431 PERL_ARGS_ASSERT_RE_CROAK2;
16437 Copy(pat1, buf, l1 , char);
16438 Copy(pat2, buf + l1, l2 , char);
16439 buf[l1 + l2] = '\n';
16440 buf[l1 + l2 + 1] = '\0';
16441 va_start(args, pat2);
16442 msv = vmess(buf, &args);
16444 message = SvPV_const(msv,l1);
16447 Copy(message, buf, l1 , char);
16448 /* l1-1 to avoid \n */
16449 Perl_croak(aTHX_ "%"UTF8f, UTF8fARG(utf8, l1-1, buf));
16452 /* XXX Here's a total kludge. But we need to re-enter for swash routines. */
16454 #ifndef PERL_IN_XSUB_RE
16456 Perl_save_re_context(pTHX)
16460 /* Save $1..$n (#18107: UTF-8 s/(\w+)/uc($1)/e); AMS 20021106. */
16462 const REGEXP * const rx = PM_GETRE(PL_curpm);
16465 for (i = 1; i <= RX_NPARENS(rx); i++) {
16466 char digits[TYPE_CHARS(long)];
16467 const STRLEN len = my_snprintf(digits, sizeof(digits),
16469 GV *const *const gvp
16470 = (GV**)hv_fetch(PL_defstash, digits, len, 0);
16473 GV * const gv = *gvp;
16474 if (SvTYPE(gv) == SVt_PVGV && GvSV(gv))
16486 S_put_byte(pTHX_ SV *sv, int c)
16488 PERL_ARGS_ASSERT_PUT_BYTE;
16492 case '\r': Perl_sv_catpvf(aTHX_ sv, "\\r"); break;
16493 case '\n': Perl_sv_catpvf(aTHX_ sv, "\\n"); break;
16494 case '\t': Perl_sv_catpvf(aTHX_ sv, "\\t"); break;
16495 case '\f': Perl_sv_catpvf(aTHX_ sv, "\\f"); break;
16496 case '\a': Perl_sv_catpvf(aTHX_ sv, "\\a"); break;
16499 Perl_sv_catpvf(aTHX_ sv, "\\x{%x}", c);
16504 const char string = c;
16505 if (c == '-' || c == ']' || c == '\\' || c == '^')
16506 sv_catpvs(sv, "\\");
16507 sv_catpvn(sv, &string, 1);
16512 S_put_range(pTHX_ SV *sv, UV start, UV end)
16515 /* Appends to 'sv' a displayable version of the range of code points from
16516 * 'start' to 'end' */
16518 assert(start <= end);
16520 PERL_ARGS_ASSERT_PUT_RANGE;
16522 if (end - start < 3) { /* Individual chars in short ranges */
16523 for (; start <= end; start++)
16524 put_byte(sv, start);
16526 else if ( end > 255
16527 || ! isALPHANUMERIC(start)
16528 || ! isALPHANUMERIC(end)
16529 || isDIGIT(start) != isDIGIT(end)
16530 || isUPPER(start) != isUPPER(end)
16531 || isLOWER(start) != isLOWER(end)
16533 /* This final test should get optimized out except on EBCDIC
16534 * platforms, where it causes ranges that cross discontinuities
16535 * like i/j to be shown as hex instead of the misleading,
16536 * e.g. H-K (since that range includes more than H, I, J, K).
16538 || (end - start) != NATIVE_TO_ASCII(end) - NATIVE_TO_ASCII(start))
16540 Perl_sv_catpvf(aTHX_ sv, "\\x{%02" UVXf "}-\\x{%02" UVXf "}",
16542 (end < 256) ? end : 255);
16544 else { /* Here, the ends of the range are both digits, or both uppercase,
16545 or both lowercase; and there's no discontinuity in the range
16546 (which could happen on EBCDIC platforms) */
16547 put_byte(sv, start);
16548 sv_catpvs(sv, "-");
16554 S_put_latin1_charclass_innards(pTHX_ SV *sv, char *bitmap)
16556 /* Appends to 'sv' a displayable version of the innards of the bracketed
16557 * character class whose bitmap is 'bitmap'; Returns 'TRUE' if it actually
16558 * output anything */
16561 bool has_output_anything = FALSE;
16563 PERL_ARGS_ASSERT_PUT_LATIN1_CHARCLASS_INNARDS;
16565 for (i = 0; i < 256; i++) {
16566 if (BITMAP_TEST((U8 *) bitmap,i)) {
16568 /* The character at index i should be output. Find the next
16569 * character that should NOT be output */
16571 for (j = i + 1; j < 256; j++) {
16572 if (! BITMAP_TEST((U8 *) bitmap, j)) {
16577 /* Everything between them is a single range that should be output
16579 put_range(sv, i, j - 1);
16580 has_output_anything = TRUE;
16585 return has_output_anything;
16588 #define CLEAR_OPTSTART \
16589 if (optstart) STMT_START { \
16590 DEBUG_OPTIMISE_r(PerlIO_printf(Perl_debug_log, \
16591 " (%"IVdf" nodes)\n", (IV)(node - optstart))); \
16595 #define DUMPUNTIL(b,e) \
16597 node=dumpuntil(r,start,(b),(e),last,sv,indent+1,depth+1);
16599 STATIC const regnode *
16600 S_dumpuntil(pTHX_ const regexp *r, const regnode *start, const regnode *node,
16601 const regnode *last, const regnode *plast,
16602 SV* sv, I32 indent, U32 depth)
16605 U8 op = PSEUDO; /* Arbitrary non-END op. */
16606 const regnode *next;
16607 const regnode *optstart= NULL;
16609 RXi_GET_DECL(r,ri);
16610 GET_RE_DEBUG_FLAGS_DECL;
16612 PERL_ARGS_ASSERT_DUMPUNTIL;
16614 #ifdef DEBUG_DUMPUNTIL
16615 PerlIO_printf(Perl_debug_log, "--- %d : %d - %d - %d\n",indent,node-start,
16616 last ? last-start : 0,plast ? plast-start : 0);
16619 if (plast && plast < last)
16622 while (PL_regkind[op] != END && (!last || node < last)) {
16623 /* While that wasn't END last time... */
16626 if (op == CLOSE || op == WHILEM)
16628 next = regnext((regnode *)node);
16631 if (OP(node) == OPTIMIZED) {
16632 if (!optstart && RE_DEBUG_FLAG(RE_DEBUG_COMPILE_OPTIMISE))
16639 regprop(r, sv, node, NULL);
16640 PerlIO_printf(Perl_debug_log, "%4"IVdf":%*s%s", (IV)(node - start),
16641 (int)(2*indent + 1), "", SvPVX_const(sv));
16643 if (OP(node) != OPTIMIZED) {
16644 if (next == NULL) /* Next ptr. */
16645 PerlIO_printf(Perl_debug_log, " (0)");
16646 else if (PL_regkind[(U8)op] == BRANCH
16647 && PL_regkind[OP(next)] != BRANCH )
16648 PerlIO_printf(Perl_debug_log, " (FAIL)");
16650 PerlIO_printf(Perl_debug_log, " (%"IVdf")", (IV)(next - start));
16651 (void)PerlIO_putc(Perl_debug_log, '\n');
16655 if (PL_regkind[(U8)op] == BRANCHJ) {
16658 const regnode *nnode = (OP(next) == LONGJMP
16659 ? regnext((regnode *)next)
16661 if (last && nnode > last)
16663 DUMPUNTIL(NEXTOPER(NEXTOPER(node)), nnode);
16666 else if (PL_regkind[(U8)op] == BRANCH) {
16668 DUMPUNTIL(NEXTOPER(node), next);
16670 else if ( PL_regkind[(U8)op] == TRIE ) {
16671 const regnode *this_trie = node;
16672 const char op = OP(node);
16673 const U32 n = ARG(node);
16674 const reg_ac_data * const ac = op>=AHOCORASICK ?
16675 (reg_ac_data *)ri->data->data[n] :
16677 const reg_trie_data * const trie =
16678 (reg_trie_data*)ri->data->data[op<AHOCORASICK ? n : ac->trie];
16680 AV *const trie_words
16681 = MUTABLE_AV(ri->data->data[n + TRIE_WORDS_OFFSET]);
16683 const regnode *nextbranch= NULL;
16686 for (word_idx= 0; word_idx < (I32)trie->wordcount; word_idx++) {
16687 SV ** const elem_ptr = av_fetch(trie_words,word_idx,0);
16689 PerlIO_printf(Perl_debug_log, "%*s%s ",
16690 (int)(2*(indent+3)), "",
16692 ? pv_pretty(sv, SvPV_nolen_const(*elem_ptr),
16693 SvCUR(*elem_ptr), 60,
16694 PL_colors[0], PL_colors[1],
16696 ? PERL_PV_ESCAPE_UNI
16698 | PERL_PV_PRETTY_ELLIPSES
16699 | PERL_PV_PRETTY_LTGT
16704 U16 dist= trie->jump[word_idx+1];
16705 PerlIO_printf(Perl_debug_log, "(%"UVuf")\n",
16706 (UV)((dist ? this_trie + dist : next) - start));
16709 nextbranch= this_trie + trie->jump[0];
16710 DUMPUNTIL(this_trie + dist, nextbranch);
16712 if (nextbranch && PL_regkind[OP(nextbranch)]==BRANCH)
16713 nextbranch= regnext((regnode *)nextbranch);
16715 PerlIO_printf(Perl_debug_log, "\n");
16718 if (last && next > last)
16723 else if ( op == CURLY ) { /* "next" might be very big: optimizer */
16724 DUMPUNTIL(NEXTOPER(node) + EXTRA_STEP_2ARGS,
16725 NEXTOPER(node) + EXTRA_STEP_2ARGS + 1);
16727 else if (PL_regkind[(U8)op] == CURLY && op != CURLYX) {
16729 DUMPUNTIL(NEXTOPER(node) + EXTRA_STEP_2ARGS, next);
16731 else if ( op == PLUS || op == STAR) {
16732 DUMPUNTIL(NEXTOPER(node), NEXTOPER(node) + 1);
16734 else if (PL_regkind[(U8)op] == ANYOF) {
16735 /* arglen 1 + class block */
16736 node += 1 + ((ANYOF_FLAGS(node) & ANYOF_POSIXL)
16737 ? ANYOF_POSIXL_SKIP
16739 node = NEXTOPER(node);
16741 else if (PL_regkind[(U8)op] == EXACT) {
16742 /* Literal string, where present. */
16743 node += NODE_SZ_STR(node) - 1;
16744 node = NEXTOPER(node);
16747 node = NEXTOPER(node);
16748 node += regarglen[(U8)op];
16750 if (op == CURLYX || op == OPEN)
16754 #ifdef DEBUG_DUMPUNTIL
16755 PerlIO_printf(Perl_debug_log, "--- %d\n", (int)indent);
16760 #endif /* DEBUGGING */
16764 * c-indentation-style: bsd
16765 * c-basic-offset: 4
16766 * indent-tabs-mode: nil
16769 * ex: set ts=8 sts=4 sw=4 et: