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 HAS_NONLATIN1_SIMPLE_FOLD_CLOSURE(i) \
97 _HAS_NONLATIN1_SIMPLE_FOLD_CLOSURE_ONLY_FOR_USE_BY_REGCOMP_DOT_C_AND_REGEXEC_DOT_C(i)
98 #define IS_NON_FINAL_FOLD(c) _IS_NON_FINAL_FOLD_ONLY_FOR_USE_BY_REGCOMP_DOT_C(c)
99 #define IS_IN_SOME_FOLD_L1(c) _IS_IN_SOME_FOLD_ONLY_FOR_USE_BY_REGCOMP_DOT_C(c)
102 #define STATIC static
106 struct RExC_state_t {
107 U32 flags; /* RXf_* are we folding, multilining? */
108 U32 pm_flags; /* PMf_* stuff from the calling PMOP */
109 char *precomp; /* uncompiled string. */
110 REGEXP *rx_sv; /* The SV that is the regexp. */
111 regexp *rx; /* perl core regexp structure */
112 regexp_internal *rxi; /* internal data for regexp object
114 char *start; /* Start of input for compile */
115 char *end; /* End of input for compile */
116 char *parse; /* Input-scan pointer. */
117 SSize_t whilem_seen; /* number of WHILEM in this expr */
118 regnode *emit_start; /* Start of emitted-code area */
119 regnode *emit_bound; /* First regnode outside of the
121 regnode *emit; /* Code-emit pointer; if = &emit_dummy,
122 implies compiling, so don't emit */
123 regnode_ssc emit_dummy; /* placeholder for emit to point to;
124 large enough for the largest
125 non-EXACTish node, so can use it as
127 I32 naughty; /* How bad is this pattern? */
128 I32 sawback; /* Did we see \1, ...? */
130 SSize_t size; /* Code size. */
131 I32 npar; /* Capture buffer count, (OPEN) plus
132 one. ("par" 0 is the whole
134 I32 nestroot; /* root parens we are in - used by
138 regnode **open_parens; /* pointers to open parens */
139 regnode **close_parens; /* pointers to close parens */
140 regnode *opend; /* END node in program */
141 I32 utf8; /* whether the pattern is utf8 or not */
142 I32 orig_utf8; /* whether the pattern was originally in utf8 */
143 /* XXX use this for future optimisation of case
144 * where pattern must be upgraded to utf8. */
145 I32 uni_semantics; /* If a d charset modifier should use unicode
146 rules, even if the pattern is not in
148 HV *paren_names; /* Paren names */
150 regnode **recurse; /* Recurse regops */
151 I32 recurse_count; /* Number of recurse regops */
152 U8 *study_chunk_recursed; /* bitmap of which parens we have moved
154 U32 study_chunk_recursed_bytes; /* bytes in bitmap */
158 I32 override_recoding;
159 I32 in_multi_char_class;
160 struct reg_code_block *code_blocks; /* positions of literal (?{})
162 int num_code_blocks; /* size of code_blocks[] */
163 int code_index; /* next code_blocks[] slot */
164 SSize_t maxlen; /* mininum possible number of chars in string to match */
165 #ifdef ADD_TO_REGEXEC
166 char *starttry; /* -Dr: where regtry was called. */
167 #define RExC_starttry (pRExC_state->starttry)
169 SV *runtime_code_qr; /* qr with the runtime code blocks */
171 const char *lastparse;
173 AV *paren_name_list; /* idx -> name */
174 #define RExC_lastparse (pRExC_state->lastparse)
175 #define RExC_lastnum (pRExC_state->lastnum)
176 #define RExC_paren_name_list (pRExC_state->paren_name_list)
180 #define RExC_flags (pRExC_state->flags)
181 #define RExC_pm_flags (pRExC_state->pm_flags)
182 #define RExC_precomp (pRExC_state->precomp)
183 #define RExC_rx_sv (pRExC_state->rx_sv)
184 #define RExC_rx (pRExC_state->rx)
185 #define RExC_rxi (pRExC_state->rxi)
186 #define RExC_start (pRExC_state->start)
187 #define RExC_end (pRExC_state->end)
188 #define RExC_parse (pRExC_state->parse)
189 #define RExC_whilem_seen (pRExC_state->whilem_seen)
190 #ifdef RE_TRACK_PATTERN_OFFSETS
191 #define RExC_offsets (pRExC_state->rxi->u.offsets) /* I am not like the
194 #define RExC_emit (pRExC_state->emit)
195 #define RExC_emit_dummy (pRExC_state->emit_dummy)
196 #define RExC_emit_start (pRExC_state->emit_start)
197 #define RExC_emit_bound (pRExC_state->emit_bound)
198 #define RExC_naughty (pRExC_state->naughty)
199 #define RExC_sawback (pRExC_state->sawback)
200 #define RExC_seen (pRExC_state->seen)
201 #define RExC_size (pRExC_state->size)
202 #define RExC_maxlen (pRExC_state->maxlen)
203 #define RExC_npar (pRExC_state->npar)
204 #define RExC_nestroot (pRExC_state->nestroot)
205 #define RExC_extralen (pRExC_state->extralen)
206 #define RExC_seen_zerolen (pRExC_state->seen_zerolen)
207 #define RExC_utf8 (pRExC_state->utf8)
208 #define RExC_uni_semantics (pRExC_state->uni_semantics)
209 #define RExC_orig_utf8 (pRExC_state->orig_utf8)
210 #define RExC_open_parens (pRExC_state->open_parens)
211 #define RExC_close_parens (pRExC_state->close_parens)
212 #define RExC_opend (pRExC_state->opend)
213 #define RExC_paren_names (pRExC_state->paren_names)
214 #define RExC_recurse (pRExC_state->recurse)
215 #define RExC_recurse_count (pRExC_state->recurse_count)
216 #define RExC_study_chunk_recursed (pRExC_state->study_chunk_recursed)
217 #define RExC_study_chunk_recursed_bytes \
218 (pRExC_state->study_chunk_recursed_bytes)
219 #define RExC_in_lookbehind (pRExC_state->in_lookbehind)
220 #define RExC_contains_locale (pRExC_state->contains_locale)
221 #define RExC_contains_i (pRExC_state->contains_i)
222 #define RExC_override_recoding (pRExC_state->override_recoding)
223 #define RExC_in_multi_char_class (pRExC_state->in_multi_char_class)
226 #define ISMULT1(c) ((c) == '*' || (c) == '+' || (c) == '?')
227 #define ISMULT2(s) ((*s) == '*' || (*s) == '+' || (*s) == '?' || \
228 ((*s) == '{' && regcurly(s)))
231 * Flags to be passed up and down.
233 #define WORST 0 /* Worst case. */
234 #define HASWIDTH 0x01 /* Known to match non-null strings. */
236 /* Simple enough to be STAR/PLUS operand; in an EXACTish node must be a single
237 * character. (There needs to be a case: in the switch statement in regexec.c
238 * for any node marked SIMPLE.) Note that this is not the same thing as
241 #define SPSTART 0x04 /* Starts with * or + */
242 #define POSTPONED 0x08 /* (?1),(?&name), (??{...}) or similar */
243 #define TRYAGAIN 0x10 /* Weeded out a declaration. */
244 #define RESTART_UTF8 0x20 /* Restart, need to calcuate sizes as UTF-8 */
246 #define REG_NODE_NUM(x) ((x) ? (int)((x)-RExC_emit_start) : -1)
248 /* whether trie related optimizations are enabled */
249 #if PERL_ENABLE_EXTENDED_TRIE_OPTIMISATION
250 #define TRIE_STUDY_OPT
251 #define FULL_TRIE_STUDY
257 #define PBYTE(u8str,paren) ((U8*)(u8str))[(paren) >> 3]
258 #define PBITVAL(paren) (1 << ((paren) & 7))
259 #define PAREN_TEST(u8str,paren) ( PBYTE(u8str,paren) & PBITVAL(paren))
260 #define PAREN_SET(u8str,paren) PBYTE(u8str,paren) |= PBITVAL(paren)
261 #define PAREN_UNSET(u8str,paren) PBYTE(u8str,paren) &= (~PBITVAL(paren))
263 #define REQUIRE_UTF8 STMT_START { \
265 *flagp = RESTART_UTF8; \
270 /* This converts the named class defined in regcomp.h to its equivalent class
271 * number defined in handy.h. */
272 #define namedclass_to_classnum(class) ((int) ((class) / 2))
273 #define classnum_to_namedclass(classnum) ((classnum) * 2)
275 #define _invlist_union_complement_2nd(a, b, output) \
276 _invlist_union_maybe_complement_2nd(a, b, TRUE, output)
277 #define _invlist_intersection_complement_2nd(a, b, output) \
278 _invlist_intersection_maybe_complement_2nd(a, b, TRUE, output)
280 /* About scan_data_t.
282 During optimisation we recurse through the regexp program performing
283 various inplace (keyhole style) optimisations. In addition study_chunk
284 and scan_commit populate this data structure with information about
285 what strings MUST appear in the pattern. We look for the longest
286 string that must appear at a fixed location, and we look for the
287 longest string that may appear at a floating location. So for instance
292 Both 'FOO' and 'A' are fixed strings. Both 'B' and 'BAR' are floating
293 strings (because they follow a .* construct). study_chunk will identify
294 both FOO and BAR as being the longest fixed and floating strings respectively.
296 The strings can be composites, for instance
300 will result in a composite fixed substring 'foo'.
302 For each string some basic information is maintained:
304 - offset or min_offset
305 This is the position the string must appear at, or not before.
306 It also implicitly (when combined with minlenp) tells us how many
307 characters must match before the string we are searching for.
308 Likewise when combined with minlenp and the length of the string it
309 tells us how many characters must appear after the string we have
313 Only used for floating strings. This is the rightmost point that
314 the string can appear at. If set to SSize_t_MAX it indicates that the
315 string can occur infinitely far to the right.
318 A pointer to the minimum number of characters of the pattern that the
319 string was found inside. This is important as in the case of positive
320 lookahead or positive lookbehind we can have multiple patterns
325 The minimum length of the pattern overall is 3, the minimum length
326 of the lookahead part is 3, but the minimum length of the part that
327 will actually match is 1. So 'FOO's minimum length is 3, but the
328 minimum length for the F is 1. This is important as the minimum length
329 is used to determine offsets in front of and behind the string being
330 looked for. Since strings can be composites this is the length of the
331 pattern at the time it was committed with a scan_commit. Note that
332 the length is calculated by study_chunk, so that the minimum lengths
333 are not known until the full pattern has been compiled, thus the
334 pointer to the value.
338 In the case of lookbehind the string being searched for can be
339 offset past the start point of the final matching string.
340 If this value was just blithely removed from the min_offset it would
341 invalidate some of the calculations for how many chars must match
342 before or after (as they are derived from min_offset and minlen and
343 the length of the string being searched for).
344 When the final pattern is compiled and the data is moved from the
345 scan_data_t structure into the regexp structure the information
346 about lookbehind is factored in, with the information that would
347 have been lost precalculated in the end_shift field for the
350 The fields pos_min and pos_delta are used to store the minimum offset
351 and the delta to the maximum offset at the current point in the pattern.
355 typedef struct scan_data_t {
356 /*I32 len_min; unused */
357 /*I32 len_delta; unused */
361 SSize_t last_end; /* min value, <0 unless valid. */
362 SSize_t last_start_min;
363 SSize_t last_start_max;
364 SV **longest; /* Either &l_fixed, or &l_float. */
365 SV *longest_fixed; /* longest fixed string found in pattern */
366 SSize_t offset_fixed; /* offset where it starts */
367 SSize_t *minlen_fixed; /* pointer to the minlen relevant to the string */
368 I32 lookbehind_fixed; /* is the position of the string modfied by LB */
369 SV *longest_float; /* longest floating string found in pattern */
370 SSize_t offset_float_min; /* earliest point in string it can appear */
371 SSize_t offset_float_max; /* latest point in string it can appear */
372 SSize_t *minlen_float; /* pointer to the minlen relevant to the string */
373 SSize_t lookbehind_float; /* is the pos of the string modified by LB */
376 SSize_t *last_closep;
377 regnode_ssc *start_class;
380 /* The below is perhaps overboard, but this allows us to save a test at the
381 * expense of a mask. This is because on both EBCDIC and ASCII machines, 'A'
382 * and 'a' differ by a single bit; the same with the upper and lower case of
383 * all other ASCII-range alphabetics. On ASCII platforms, they are 32 apart;
384 * on EBCDIC, they are 64. This uses an exclusive 'or' to find that bit and
385 * then inverts it to form a mask, with just a single 0, in the bit position
386 * where the upper- and lowercase differ. XXX There are about 40 other
387 * instances in the Perl core where this micro-optimization could be used.
388 * Should decide if maintenance cost is worse, before changing those
390 * Returns a boolean as to whether or not 'v' is either a lowercase or
391 * uppercase instance of 'c', where 'c' is in [A-Za-z]. If 'c' is a
392 * compile-time constant, the generated code is better than some optimizing
393 * compilers figure out, amounting to a mask and test. The results are
394 * meaningless if 'c' is not one of [A-Za-z] */
395 #define isARG2_lower_or_UPPER_ARG1(c, v) \
396 (((v) & ~('A' ^ 'a')) == ((c) & ~('A' ^ 'a')))
399 * Forward declarations for pregcomp()'s friends.
402 static const scan_data_t zero_scan_data =
403 { 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0 ,0};
405 #define SF_BEFORE_EOL (SF_BEFORE_SEOL|SF_BEFORE_MEOL)
406 #define SF_BEFORE_SEOL 0x0001
407 #define SF_BEFORE_MEOL 0x0002
408 #define SF_FIX_BEFORE_EOL (SF_FIX_BEFORE_SEOL|SF_FIX_BEFORE_MEOL)
409 #define SF_FL_BEFORE_EOL (SF_FL_BEFORE_SEOL|SF_FL_BEFORE_MEOL)
411 #define SF_FIX_SHIFT_EOL (+2)
412 #define SF_FL_SHIFT_EOL (+4)
414 #define SF_FIX_BEFORE_SEOL (SF_BEFORE_SEOL << SF_FIX_SHIFT_EOL)
415 #define SF_FIX_BEFORE_MEOL (SF_BEFORE_MEOL << SF_FIX_SHIFT_EOL)
417 #define SF_FL_BEFORE_SEOL (SF_BEFORE_SEOL << SF_FL_SHIFT_EOL)
418 #define SF_FL_BEFORE_MEOL (SF_BEFORE_MEOL << SF_FL_SHIFT_EOL) /* 0x20 */
419 #define SF_IS_INF 0x0040
420 #define SF_HAS_PAR 0x0080
421 #define SF_IN_PAR 0x0100
422 #define SF_HAS_EVAL 0x0200
423 #define SCF_DO_SUBSTR 0x0400
424 #define SCF_DO_STCLASS_AND 0x0800
425 #define SCF_DO_STCLASS_OR 0x1000
426 #define SCF_DO_STCLASS (SCF_DO_STCLASS_AND|SCF_DO_STCLASS_OR)
427 #define SCF_WHILEM_VISITED_POS 0x2000
429 #define SCF_TRIE_RESTUDY 0x4000 /* Do restudy? */
430 #define SCF_SEEN_ACCEPT 0x8000
431 #define SCF_TRIE_DOING_RESTUDY 0x10000
433 #define UTF cBOOL(RExC_utf8)
435 /* The enums for all these are ordered so things work out correctly */
436 #define LOC (get_regex_charset(RExC_flags) == REGEX_LOCALE_CHARSET)
437 #define DEPENDS_SEMANTICS (get_regex_charset(RExC_flags) \
438 == REGEX_DEPENDS_CHARSET)
439 #define UNI_SEMANTICS (get_regex_charset(RExC_flags) == REGEX_UNICODE_CHARSET)
440 #define AT_LEAST_UNI_SEMANTICS (get_regex_charset(RExC_flags) \
441 >= REGEX_UNICODE_CHARSET)
442 #define ASCII_RESTRICTED (get_regex_charset(RExC_flags) \
443 == REGEX_ASCII_RESTRICTED_CHARSET)
444 #define AT_LEAST_ASCII_RESTRICTED (get_regex_charset(RExC_flags) \
445 >= REGEX_ASCII_RESTRICTED_CHARSET)
446 #define ASCII_FOLD_RESTRICTED (get_regex_charset(RExC_flags) \
447 == REGEX_ASCII_MORE_RESTRICTED_CHARSET)
449 #define FOLD cBOOL(RExC_flags & RXf_PMf_FOLD)
451 /* For programs that want to be strictly Unicode compatible by dying if any
452 * attempt is made to match a non-Unicode code point against a Unicode
454 #define ALWAYS_WARN_SUPER ckDEAD(packWARN(WARN_NON_UNICODE))
456 #define OOB_NAMEDCLASS -1
458 /* There is no code point that is out-of-bounds, so this is problematic. But
459 * its only current use is to initialize a variable that is always set before
461 #define OOB_UNICODE 0xDEADBEEF
463 #define CHR_SVLEN(sv) (UTF ? sv_len_utf8(sv) : SvCUR(sv))
464 #define CHR_DIST(a,b) (UTF ? utf8_distance(a,b) : a - b)
467 /* length of regex to show in messages that don't mark a position within */
468 #define RegexLengthToShowInErrorMessages 127
471 * If MARKER[12] are adjusted, be sure to adjust the constants at the top
472 * of t/op/regmesg.t, the tests in t/op/re_tests, and those in
473 * op/pragma/warn/regcomp.
475 #define MARKER1 "<-- HERE" /* marker as it appears in the description */
476 #define MARKER2 " <-- HERE " /* marker as it appears within the regex */
478 #define REPORT_LOCATION " in regex; marked by " MARKER1 \
479 " in m/%"UTF8f MARKER2 "%"UTF8f"/"
481 #define REPORT_LOCATION_ARGS(offset) \
482 UTF8fARG(UTF, offset, RExC_precomp), \
483 UTF8fARG(UTF, RExC_end - RExC_precomp - offset, RExC_precomp + offset)
486 * Calls SAVEDESTRUCTOR_X if needed, then calls Perl_croak with the given
487 * arg. Show regex, up to a maximum length. If it's too long, chop and add
490 #define _FAIL(code) STMT_START { \
491 const char *ellipses = ""; \
492 IV len = RExC_end - RExC_precomp; \
495 SAVEFREESV(RExC_rx_sv); \
496 if (len > RegexLengthToShowInErrorMessages) { \
497 /* chop 10 shorter than the max, to ensure meaning of "..." */ \
498 len = RegexLengthToShowInErrorMessages - 10; \
504 #define FAIL(msg) _FAIL( \
505 Perl_croak(aTHX_ "%s in regex m/%"UTF8f"%s/", \
506 msg, UTF8fARG(UTF, len, RExC_precomp), ellipses))
508 #define FAIL2(msg,arg) _FAIL( \
509 Perl_croak(aTHX_ msg " in regex m/%"UTF8f"%s/", \
510 arg, UTF8fARG(UTF, len, RExC_precomp), ellipses))
513 * Simple_vFAIL -- like FAIL, but marks the current location in the scan
515 #define Simple_vFAIL(m) STMT_START { \
516 const IV offset = RExC_parse - RExC_precomp; \
517 Perl_croak(aTHX_ "%s" REPORT_LOCATION, \
518 m, REPORT_LOCATION_ARGS(offset)); \
522 * Calls SAVEDESTRUCTOR_X if needed, then Simple_vFAIL()
524 #define vFAIL(m) STMT_START { \
526 SAVEFREESV(RExC_rx_sv); \
531 * Like Simple_vFAIL(), but accepts two arguments.
533 #define Simple_vFAIL2(m,a1) STMT_START { \
534 const IV offset = RExC_parse - RExC_precomp; \
535 S_re_croak2(aTHX_ UTF, m, REPORT_LOCATION, a1, \
536 REPORT_LOCATION_ARGS(offset)); \
540 * Calls SAVEDESTRUCTOR_X if needed, then Simple_vFAIL2().
542 #define vFAIL2(m,a1) STMT_START { \
544 SAVEFREESV(RExC_rx_sv); \
545 Simple_vFAIL2(m, a1); \
550 * Like Simple_vFAIL(), but accepts three arguments.
552 #define Simple_vFAIL3(m, a1, a2) STMT_START { \
553 const IV offset = RExC_parse - RExC_precomp; \
554 S_re_croak2(aTHX_ UTF, m, REPORT_LOCATION, a1, a2, \
555 REPORT_LOCATION_ARGS(offset)); \
559 * Calls SAVEDESTRUCTOR_X if needed, then Simple_vFAIL3().
561 #define vFAIL3(m,a1,a2) STMT_START { \
563 SAVEFREESV(RExC_rx_sv); \
564 Simple_vFAIL3(m, a1, a2); \
568 * Like Simple_vFAIL(), but accepts four arguments.
570 #define Simple_vFAIL4(m, a1, a2, a3) STMT_START { \
571 const IV offset = RExC_parse - RExC_precomp; \
572 S_re_croak2(aTHX_ UTF, m, REPORT_LOCATION, a1, a2, a3, \
573 REPORT_LOCATION_ARGS(offset)); \
576 #define vFAIL4(m,a1,a2,a3) STMT_START { \
578 SAVEFREESV(RExC_rx_sv); \
579 Simple_vFAIL4(m, a1, a2, a3); \
582 /* A specialized version of vFAIL2 that works with UTF8f */
583 #define vFAIL2utf8f(m, a1) STMT_START { \
584 const IV offset = RExC_parse - RExC_precomp; \
586 SAVEFREESV(RExC_rx_sv); \
587 S_re_croak2(aTHX_ UTF, m, REPORT_LOCATION, a1, \
588 REPORT_LOCATION_ARGS(offset)); \
592 /* m is not necessarily a "literal string", in this macro */
593 #define reg_warn_non_literal_string(loc, m) STMT_START { \
594 const IV offset = loc - RExC_precomp; \
595 Perl_warner(aTHX_ packWARN(WARN_REGEXP), "%s" REPORT_LOCATION, \
596 m, REPORT_LOCATION_ARGS(offset)); \
599 #define ckWARNreg(loc,m) STMT_START { \
600 const IV offset = loc - RExC_precomp; \
601 Perl_ck_warner(aTHX_ packWARN(WARN_REGEXP), m REPORT_LOCATION, \
602 REPORT_LOCATION_ARGS(offset)); \
605 #define vWARN_dep(loc, m) STMT_START { \
606 const IV offset = loc - RExC_precomp; \
607 Perl_warner(aTHX_ packWARN(WARN_DEPRECATED), m REPORT_LOCATION, \
608 REPORT_LOCATION_ARGS(offset)); \
611 #define ckWARNdep(loc,m) STMT_START { \
612 const IV offset = loc - RExC_precomp; \
613 Perl_ck_warner_d(aTHX_ packWARN(WARN_DEPRECATED), \
615 REPORT_LOCATION_ARGS(offset)); \
618 #define ckWARNregdep(loc,m) STMT_START { \
619 const IV offset = loc - RExC_precomp; \
620 Perl_ck_warner_d(aTHX_ packWARN2(WARN_DEPRECATED, WARN_REGEXP), \
622 REPORT_LOCATION_ARGS(offset)); \
625 #define ckWARN2reg_d(loc,m, a1) STMT_START { \
626 const IV offset = loc - RExC_precomp; \
627 Perl_ck_warner_d(aTHX_ packWARN(WARN_REGEXP), \
629 a1, REPORT_LOCATION_ARGS(offset)); \
632 #define ckWARN2reg(loc, m, a1) STMT_START { \
633 const IV offset = loc - RExC_precomp; \
634 Perl_ck_warner(aTHX_ packWARN(WARN_REGEXP), m REPORT_LOCATION, \
635 a1, REPORT_LOCATION_ARGS(offset)); \
638 #define vWARN3(loc, m, a1, a2) STMT_START { \
639 const IV offset = loc - RExC_precomp; \
640 Perl_warner(aTHX_ packWARN(WARN_REGEXP), m REPORT_LOCATION, \
641 a1, a2, REPORT_LOCATION_ARGS(offset)); \
644 #define ckWARN3reg(loc, m, a1, a2) STMT_START { \
645 const IV offset = loc - RExC_precomp; \
646 Perl_ck_warner(aTHX_ packWARN(WARN_REGEXP), m REPORT_LOCATION, \
647 a1, a2, REPORT_LOCATION_ARGS(offset)); \
650 #define vWARN4(loc, m, a1, a2, a3) STMT_START { \
651 const IV offset = loc - RExC_precomp; \
652 Perl_warner(aTHX_ packWARN(WARN_REGEXP), m REPORT_LOCATION, \
653 a1, a2, a3, REPORT_LOCATION_ARGS(offset)); \
656 #define ckWARN4reg(loc, m, a1, a2, a3) STMT_START { \
657 const IV offset = loc - RExC_precomp; \
658 Perl_ck_warner(aTHX_ packWARN(WARN_REGEXP), m REPORT_LOCATION, \
659 a1, a2, a3, REPORT_LOCATION_ARGS(offset)); \
662 #define vWARN5(loc, m, a1, a2, a3, a4) STMT_START { \
663 const IV offset = loc - RExC_precomp; \
664 Perl_warner(aTHX_ packWARN(WARN_REGEXP), m REPORT_LOCATION, \
665 a1, a2, a3, a4, REPORT_LOCATION_ARGS(offset)); \
669 /* Allow for side effects in s */
670 #define REGC(c,s) STMT_START { \
671 if (!SIZE_ONLY) *(s) = (c); else (void)(s); \
674 /* Macros for recording node offsets. 20001227 mjd@plover.com
675 * Nodes are numbered 1, 2, 3, 4. Node #n's position is recorded in
676 * element 2*n-1 of the array. Element #2n holds the byte length node #n.
677 * Element 0 holds the number n.
678 * Position is 1 indexed.
680 #ifndef RE_TRACK_PATTERN_OFFSETS
681 #define Set_Node_Offset_To_R(node,byte)
682 #define Set_Node_Offset(node,byte)
683 #define Set_Cur_Node_Offset
684 #define Set_Node_Length_To_R(node,len)
685 #define Set_Node_Length(node,len)
686 #define Set_Node_Cur_Length(node,start)
687 #define Node_Offset(n)
688 #define Node_Length(n)
689 #define Set_Node_Offset_Length(node,offset,len)
690 #define ProgLen(ri) ri->u.proglen
691 #define SetProgLen(ri,x) ri->u.proglen = x
693 #define ProgLen(ri) ri->u.offsets[0]
694 #define SetProgLen(ri,x) ri->u.offsets[0] = x
695 #define Set_Node_Offset_To_R(node,byte) STMT_START { \
697 MJD_OFFSET_DEBUG(("** (%d) offset of node %d is %d.\n", \
698 __LINE__, (int)(node), (int)(byte))); \
700 Perl_croak(aTHX_ "value of node is %d in Offset macro", \
703 RExC_offsets[2*(node)-1] = (byte); \
708 #define Set_Node_Offset(node,byte) \
709 Set_Node_Offset_To_R((node)-RExC_emit_start, (byte)-RExC_start)
710 #define Set_Cur_Node_Offset Set_Node_Offset(RExC_emit, RExC_parse)
712 #define Set_Node_Length_To_R(node,len) STMT_START { \
714 MJD_OFFSET_DEBUG(("** (%d) size of node %d is %d.\n", \
715 __LINE__, (int)(node), (int)(len))); \
717 Perl_croak(aTHX_ "value of node is %d in Length macro", \
720 RExC_offsets[2*(node)] = (len); \
725 #define Set_Node_Length(node,len) \
726 Set_Node_Length_To_R((node)-RExC_emit_start, len)
727 #define Set_Node_Cur_Length(node, start) \
728 Set_Node_Length(node, RExC_parse - start)
730 /* Get offsets and lengths */
731 #define Node_Offset(n) (RExC_offsets[2*((n)-RExC_emit_start)-1])
732 #define Node_Length(n) (RExC_offsets[2*((n)-RExC_emit_start)])
734 #define Set_Node_Offset_Length(node,offset,len) STMT_START { \
735 Set_Node_Offset_To_R((node)-RExC_emit_start, (offset)); \
736 Set_Node_Length_To_R((node)-RExC_emit_start, (len)); \
740 #if PERL_ENABLE_EXPERIMENTAL_REGEX_OPTIMISATIONS
741 #define EXPERIMENTAL_INPLACESCAN
742 #endif /*PERL_ENABLE_EXPERIMENTAL_REGEX_OPTIMISATIONS*/
744 #define DEBUG_RExC_seen() \
745 DEBUG_OPTIMISE_MORE_r({ \
746 PerlIO_printf(Perl_debug_log,"RExC_seen: "); \
748 if (RExC_seen & REG_ZERO_LEN_SEEN) \
749 PerlIO_printf(Perl_debug_log,"REG_ZERO_LEN_SEEN "); \
751 if (RExC_seen & REG_LOOKBEHIND_SEEN) \
752 PerlIO_printf(Perl_debug_log,"REG_LOOKBEHIND_SEEN "); \
754 if (RExC_seen & REG_GPOS_SEEN) \
755 PerlIO_printf(Perl_debug_log,"REG_GPOS_SEEN "); \
757 if (RExC_seen & REG_CANY_SEEN) \
758 PerlIO_printf(Perl_debug_log,"REG_CANY_SEEN "); \
760 if (RExC_seen & REG_RECURSE_SEEN) \
761 PerlIO_printf(Perl_debug_log,"REG_RECURSE_SEEN "); \
763 if (RExC_seen & REG_TOP_LEVEL_BRANCHES_SEEN) \
764 PerlIO_printf(Perl_debug_log,"REG_TOP_LEVEL_BRANCHES_SEEN "); \
766 if (RExC_seen & REG_VERBARG_SEEN) \
767 PerlIO_printf(Perl_debug_log,"REG_VERBARG_SEEN "); \
769 if (RExC_seen & REG_CUTGROUP_SEEN) \
770 PerlIO_printf(Perl_debug_log,"REG_CUTGROUP_SEEN "); \
772 if (RExC_seen & REG_RUN_ON_COMMENT_SEEN) \
773 PerlIO_printf(Perl_debug_log,"REG_RUN_ON_COMMENT_SEEN "); \
775 if (RExC_seen & REG_UNFOLDED_MULTI_SEEN) \
776 PerlIO_printf(Perl_debug_log,"REG_UNFOLDED_MULTI_SEEN "); \
778 if (RExC_seen & REG_GOSTART_SEEN) \
779 PerlIO_printf(Perl_debug_log,"REG_GOSTART_SEEN "); \
781 if (RExC_seen & REG_UNBOUNDED_QUANTIFIER_SEEN) \
782 PerlIO_printf(Perl_debug_log,"REG_UNBOUNDED_QUANTIFIER_SEEN "); \
784 PerlIO_printf(Perl_debug_log,"\n"); \
787 #define DEBUG_STUDYDATA(str,data,depth) \
788 DEBUG_OPTIMISE_MORE_r(if(data){ \
789 PerlIO_printf(Perl_debug_log, \
790 "%*s" str "Pos:%"IVdf"/%"IVdf \
791 " Flags: 0x%"UVXf" Whilem_c: %"IVdf" Lcp: %"IVdf" %s", \
792 (int)(depth)*2, "", \
793 (IV)((data)->pos_min), \
794 (IV)((data)->pos_delta), \
795 (UV)((data)->flags), \
796 (IV)((data)->whilem_c), \
797 (IV)((data)->last_closep ? *((data)->last_closep) : -1), \
798 is_inf ? "INF " : "" \
800 if ((data)->last_found) \
801 PerlIO_printf(Perl_debug_log, \
802 "Last:'%s' %"IVdf":%"IVdf"/%"IVdf" %sFixed:'%s' @ %"IVdf \
803 " %sFloat: '%s' @ %"IVdf"/%"IVdf"", \
804 SvPVX_const((data)->last_found), \
805 (IV)((data)->last_end), \
806 (IV)((data)->last_start_min), \
807 (IV)((data)->last_start_max), \
808 ((data)->longest && \
809 (data)->longest==&((data)->longest_fixed)) ? "*" : "", \
810 SvPVX_const((data)->longest_fixed), \
811 (IV)((data)->offset_fixed), \
812 ((data)->longest && \
813 (data)->longest==&((data)->longest_float)) ? "*" : "", \
814 SvPVX_const((data)->longest_float), \
815 (IV)((data)->offset_float_min), \
816 (IV)((data)->offset_float_max) \
818 PerlIO_printf(Perl_debug_log,"\n"); \
821 /* Mark that we cannot extend a found fixed substring at this point.
822 Update the longest found anchored substring and the longest found
823 floating substrings if needed. */
826 S_scan_commit(pTHX_ const RExC_state_t *pRExC_state, scan_data_t *data,
827 SSize_t *minlenp, int is_inf)
829 const STRLEN l = CHR_SVLEN(data->last_found);
830 const STRLEN old_l = CHR_SVLEN(*data->longest);
831 GET_RE_DEBUG_FLAGS_DECL;
833 PERL_ARGS_ASSERT_SCAN_COMMIT;
835 if ((l >= old_l) && ((l > old_l) || (data->flags & SF_BEFORE_EOL))) {
836 SvSetMagicSV(*data->longest, data->last_found);
837 if (*data->longest == data->longest_fixed) {
838 data->offset_fixed = l ? data->last_start_min : data->pos_min;
839 if (data->flags & SF_BEFORE_EOL)
841 |= ((data->flags & SF_BEFORE_EOL) << SF_FIX_SHIFT_EOL);
843 data->flags &= ~SF_FIX_BEFORE_EOL;
844 data->minlen_fixed=minlenp;
845 data->lookbehind_fixed=0;
847 else { /* *data->longest == data->longest_float */
848 data->offset_float_min = l ? data->last_start_min : data->pos_min;
849 data->offset_float_max = (l
850 ? data->last_start_max
851 : (data->pos_delta == SSize_t_MAX
853 : data->pos_min + data->pos_delta));
855 || (STRLEN)data->offset_float_max > (STRLEN)SSize_t_MAX)
856 data->offset_float_max = SSize_t_MAX;
857 if (data->flags & SF_BEFORE_EOL)
859 |= ((data->flags & SF_BEFORE_EOL) << SF_FL_SHIFT_EOL);
861 data->flags &= ~SF_FL_BEFORE_EOL;
862 data->minlen_float=minlenp;
863 data->lookbehind_float=0;
866 SvCUR_set(data->last_found, 0);
868 SV * const sv = data->last_found;
869 if (SvUTF8(sv) && SvMAGICAL(sv)) {
870 MAGIC * const mg = mg_find(sv, PERL_MAGIC_utf8);
876 data->flags &= ~SF_BEFORE_EOL;
877 DEBUG_STUDYDATA("commit: ",data,0);
880 /* An SSC is just a regnode_charclass_posix with an extra field: the inversion
881 * list that describes which code points it matches */
884 S_ssc_anything(pTHX_ regnode_ssc *ssc)
886 /* Set the SSC 'ssc' to match an empty string or any code point */
888 PERL_ARGS_ASSERT_SSC_ANYTHING;
890 assert(is_ANYOF_SYNTHETIC(ssc));
892 ssc->invlist = sv_2mortal(_new_invlist(2)); /* mortalize so won't leak */
893 _append_range_to_invlist(ssc->invlist, 0, UV_MAX);
894 ANYOF_FLAGS(ssc) |= ANYOF_EMPTY_STRING; /* Plus match empty string */
898 S_ssc_is_anything(const regnode_ssc *ssc)
900 /* Returns TRUE if the SSC 'ssc' can match the empty string and any code
901 * point; FALSE otherwise. Thus, this is used to see if using 'ssc' buys
902 * us anything: if the function returns TRUE, 'ssc' hasn't been restricted
903 * in any way, so there's no point in using it */
908 PERL_ARGS_ASSERT_SSC_IS_ANYTHING;
910 assert(is_ANYOF_SYNTHETIC(ssc));
912 if (! (ANYOF_FLAGS(ssc) & ANYOF_EMPTY_STRING)) {
916 /* See if the list consists solely of the range 0 - Infinity */
917 invlist_iterinit(ssc->invlist);
918 ret = invlist_iternext(ssc->invlist, &start, &end)
922 invlist_iterfinish(ssc->invlist);
928 /* If e.g., both \w and \W are set, matches everything */
929 if (ANYOF_POSIXL_SSC_TEST_ANY_SET(ssc)) {
931 for (i = 0; i < ANYOF_POSIXL_MAX; i += 2) {
932 if (ANYOF_POSIXL_TEST(ssc, i) && ANYOF_POSIXL_TEST(ssc, i+1)) {
942 S_ssc_init(pTHX_ const RExC_state_t *pRExC_state, regnode_ssc *ssc)
944 /* Initializes the SSC 'ssc'. This includes setting it to match an empty
945 * string, any code point, or any posix class under locale */
947 PERL_ARGS_ASSERT_SSC_INIT;
949 Zero(ssc, 1, regnode_ssc);
950 set_ANYOF_SYNTHETIC(ssc);
951 ARG_SET(ssc, ANYOF_NONBITMAP_EMPTY);
954 /* If any portion of the regex is to operate under locale rules,
955 * initialization includes it. The reason this isn't done for all regexes
956 * is that the optimizer was written under the assumption that locale was
957 * all-or-nothing. Given the complexity and lack of documentation in the
958 * optimizer, and that there are inadequate test cases for locale, many
959 * parts of it may not work properly, it is safest to avoid locale unless
961 if (RExC_contains_locale) {
962 ANYOF_POSIXL_SETALL(ssc);
965 ANYOF_POSIXL_ZERO(ssc);
970 S_ssc_is_cp_posixl_init(const RExC_state_t *pRExC_state,
971 const regnode_ssc *ssc)
973 /* Returns TRUE if the SSC 'ssc' is in its initial state with regard only
974 * to the list of code points matched, and locale posix classes; hence does
975 * not check its flags) */
980 PERL_ARGS_ASSERT_SSC_IS_CP_POSIXL_INIT;
982 assert(is_ANYOF_SYNTHETIC(ssc));
984 invlist_iterinit(ssc->invlist);
985 ret = invlist_iternext(ssc->invlist, &start, &end)
989 invlist_iterfinish(ssc->invlist);
995 if (RExC_contains_locale && ! ANYOF_POSIXL_SSC_TEST_ALL_SET(ssc)) {
1003 S_get_ANYOF_cp_list_for_ssc(pTHX_ const RExC_state_t *pRExC_state,
1004 const regnode_charclass* const node)
1006 /* Returns a mortal inversion list defining which code points are matched
1007 * by 'node', which is of type ANYOF. Handles complementing the result if
1008 * appropriate. If some code points aren't knowable at this time, the
1009 * returned list must, and will, contain every code point that is a
1012 SV* invlist = sv_2mortal(_new_invlist(0));
1013 SV* only_utf8_locale_invlist = NULL;
1015 const U32 n = ARG(node);
1016 bool new_node_has_latin1 = FALSE;
1018 PERL_ARGS_ASSERT_GET_ANYOF_CP_LIST_FOR_SSC;
1020 /* Look at the data structure created by S_set_ANYOF_arg() */
1021 if (n != ANYOF_NONBITMAP_EMPTY) {
1022 SV * const rv = MUTABLE_SV(RExC_rxi->data->data[n]);
1023 AV * const av = MUTABLE_AV(SvRV(rv));
1024 SV **const ary = AvARRAY(av);
1025 assert(RExC_rxi->data->what[n] == 's');
1027 if (ary[1] && ary[1] != &PL_sv_undef) { /* Has compile-time swash */
1028 invlist = sv_2mortal(invlist_clone(_get_swash_invlist(ary[1])));
1030 else if (ary[0] && ary[0] != &PL_sv_undef) {
1032 /* Here, no compile-time swash, and there are things that won't be
1033 * known until runtime -- we have to assume it could be anything */
1034 return _add_range_to_invlist(invlist, 0, UV_MAX);
1036 else if (ary[3] && ary[3] != &PL_sv_undef) {
1038 /* Here no compile-time swash, and no run-time only data. Use the
1039 * node's inversion list */
1040 invlist = sv_2mortal(invlist_clone(ary[3]));
1043 /* Get the code points valid only under UTF-8 locales */
1044 if ((ANYOF_FLAGS(node) & ANYOF_LOC_FOLD)
1045 && ary[2] && ary[2] != &PL_sv_undef)
1047 only_utf8_locale_invlist = ary[2];
1051 /* An ANYOF node contains a bitmap for the first 256 code points, and an
1052 * inversion list for the others, but if there are code points that should
1053 * match only conditionally on the target string being UTF-8, those are
1054 * placed in the inversion list, and not the bitmap. Since there are
1055 * circumstances under which they could match, they are included in the
1056 * SSC. But if the ANYOF node is to be inverted, we have to exclude them
1057 * here, so that when we invert below, the end result actually does include
1058 * them. (Think about "\xe0" =~ /[^\xc0]/di;). We have to do this here
1059 * before we add the unconditionally matched code points */
1060 if (ANYOF_FLAGS(node) & ANYOF_INVERT) {
1061 _invlist_intersection_complement_2nd(invlist,
1066 /* Add in the points from the bit map */
1067 for (i = 0; i < 256; i++) {
1068 if (ANYOF_BITMAP_TEST(node, i)) {
1069 invlist = add_cp_to_invlist(invlist, i);
1070 new_node_has_latin1 = TRUE;
1074 /* If this can match all upper Latin1 code points, have to add them
1076 if (ANYOF_FLAGS(node) & ANYOF_NON_UTF8_NON_ASCII_ALL) {
1077 _invlist_union(invlist, PL_UpperLatin1, &invlist);
1080 /* Similarly for these */
1081 if (ANYOF_FLAGS(node) & ANYOF_ABOVE_LATIN1_ALL) {
1082 invlist = _add_range_to_invlist(invlist, 256, UV_MAX);
1085 if (ANYOF_FLAGS(node) & ANYOF_INVERT) {
1086 _invlist_invert(invlist);
1088 else if (new_node_has_latin1 && ANYOF_FLAGS(node) & ANYOF_LOC_FOLD) {
1090 /* Under /li, any 0-255 could fold to any other 0-255, depending on the
1091 * locale. We can skip this if there are no 0-255 at all. */
1092 _invlist_union(invlist, PL_Latin1, &invlist);
1095 /* Similarly add the UTF-8 locale possible matches. These have to be
1096 * deferred until after the non-UTF-8 locale ones are taken care of just
1097 * above, or it leads to wrong results under ANYOF_INVERT */
1098 if (only_utf8_locale_invlist) {
1099 _invlist_union_maybe_complement_2nd(invlist,
1100 only_utf8_locale_invlist,
1101 ANYOF_FLAGS(node) & ANYOF_INVERT,
1108 /* These two functions currently do the exact same thing */
1109 #define ssc_init_zero ssc_init
1111 #define ssc_add_cp(ssc, cp) ssc_add_range((ssc), (cp), (cp))
1112 #define ssc_match_all_cp(ssc) ssc_add_range(ssc, 0, UV_MAX)
1114 /* 'AND' a given class with another one. Can create false positives. 'ssc'
1115 * should not be inverted. 'and_with->flags & ANYOF_POSIXL' should be 0 if
1116 * 'and_with' is a regnode_charclass instead of a regnode_ssc. */
1119 S_ssc_and(pTHX_ const RExC_state_t *pRExC_state, regnode_ssc *ssc,
1120 const regnode_charclass *and_with)
1122 /* Accumulate into SSC 'ssc' its 'AND' with 'and_with', which is either
1123 * another SSC or a regular ANYOF class. Can create false positives. */
1128 PERL_ARGS_ASSERT_SSC_AND;
1130 assert(is_ANYOF_SYNTHETIC(ssc));
1132 /* 'and_with' is used as-is if it too is an SSC; otherwise have to extract
1133 * the code point inversion list and just the relevant flags */
1134 if (is_ANYOF_SYNTHETIC(and_with)) {
1135 anded_cp_list = ((regnode_ssc *)and_with)->invlist;
1136 anded_flags = ANYOF_FLAGS(and_with);
1138 /* XXX This is a kludge around what appears to be deficiencies in the
1139 * optimizer. If we make S_ssc_anything() add in the WARN_SUPER flag,
1140 * there are paths through the optimizer where it doesn't get weeded
1141 * out when it should. And if we don't make some extra provision for
1142 * it like the code just below, it doesn't get added when it should.
1143 * This solution is to add it only when AND'ing, which is here, and
1144 * only when what is being AND'ed is the pristine, original node
1145 * matching anything. Thus it is like adding it to ssc_anything() but
1146 * only when the result is to be AND'ed. Probably the same solution
1147 * could be adopted for the same problem we have with /l matching,
1148 * which is solved differently in S_ssc_init(), and that would lead to
1149 * fewer false positives than that solution has. But if this solution
1150 * creates bugs, the consequences are only that a warning isn't raised
1151 * that should be; while the consequences for having /l bugs is
1152 * incorrect matches */
1153 if (ssc_is_anything((regnode_ssc *)and_with)) {
1154 anded_flags |= ANYOF_WARN_SUPER;
1158 anded_cp_list = get_ANYOF_cp_list_for_ssc(pRExC_state, and_with);
1159 anded_flags = ANYOF_FLAGS(and_with) & ANYOF_COMMON_FLAGS;
1162 ANYOF_FLAGS(ssc) &= anded_flags;
1164 /* Below, C1 is the list of code points in 'ssc'; P1, its posix classes.
1165 * C2 is the list of code points in 'and-with'; P2, its posix classes.
1166 * 'and_with' may be inverted. When not inverted, we have the situation of
1168 * (C1 | P1) & (C2 | P2)
1169 * = (C1 & (C2 | P2)) | (P1 & (C2 | P2))
1170 * = ((C1 & C2) | (C1 & P2)) | ((P1 & C2) | (P1 & P2))
1171 * <= ((C1 & C2) | P2)) | ( P1 | (P1 & P2))
1172 * <= ((C1 & C2) | P1 | P2)
1173 * Alternatively, the last few steps could be:
1174 * = ((C1 & C2) | (C1 & P2)) | ((P1 & C2) | (P1 & P2))
1175 * <= ((C1 & C2) | C1 ) | ( C2 | (P1 & P2))
1176 * <= (C1 | C2 | (P1 & P2))
1177 * We favor the second approach if either P1 or P2 is non-empty. This is
1178 * because these components are a barrier to doing optimizations, as what
1179 * they match cannot be known until the moment of matching as they are
1180 * dependent on the current locale, 'AND"ing them likely will reduce or
1182 * But we can do better if we know that C1,P1 are in their initial state (a
1183 * frequent occurrence), each matching everything:
1184 * (<everything>) & (C2 | P2) = C2 | P2
1185 * Similarly, if C2,P2 are in their initial state (again a frequent
1186 * occurrence), the result is a no-op
1187 * (C1 | P1) & (<everything>) = C1 | P1
1190 * (C1 | P1) & ~(C2 | P2) = (C1 | P1) & (~C2 & ~P2)
1191 * = (C1 & (~C2 & ~P2)) | (P1 & (~C2 & ~P2))
1192 * <= (C1 & ~C2) | (P1 & ~P2)
1195 if ((ANYOF_FLAGS(and_with) & ANYOF_INVERT)
1196 && ! is_ANYOF_SYNTHETIC(and_with))
1200 ssc_intersection(ssc,
1202 FALSE /* Has already been inverted */
1205 /* If either P1 or P2 is empty, the intersection will be also; can skip
1207 if (! (ANYOF_FLAGS(and_with) & ANYOF_POSIXL)) {
1208 ANYOF_POSIXL_ZERO(ssc);
1210 else if (ANYOF_POSIXL_SSC_TEST_ANY_SET(ssc)) {
1212 /* Note that the Posix class component P from 'and_with' actually
1214 * P = Pa | Pb | ... | Pn
1215 * where each component is one posix class, such as in [\w\s].
1217 * ~P = ~(Pa | Pb | ... | Pn)
1218 * = ~Pa & ~Pb & ... & ~Pn
1219 * <= ~Pa | ~Pb | ... | ~Pn
1220 * The last is something we can easily calculate, but unfortunately
1221 * is likely to have many false positives. We could do better
1222 * in some (but certainly not all) instances if two classes in
1223 * P have known relationships. For example
1224 * :lower: <= :alpha: <= :alnum: <= \w <= :graph: <= :print:
1226 * :lower: & :print: = :lower:
1227 * And similarly for classes that must be disjoint. For example,
1228 * since \s and \w can have no elements in common based on rules in
1229 * the POSIX standard,
1230 * \w & ^\S = nothing
1231 * Unfortunately, some vendor locales do not meet the Posix
1232 * standard, in particular almost everything by Microsoft.
1233 * The loop below just changes e.g., \w into \W and vice versa */
1235 regnode_charclass_posixl temp;
1236 int add = 1; /* To calculate the index of the complement */
1238 ANYOF_POSIXL_ZERO(&temp);
1239 for (i = 0; i < ANYOF_MAX; i++) {
1241 || ! ANYOF_POSIXL_TEST((regnode_charclass_posixl*) and_with, i)
1242 || ! ANYOF_POSIXL_TEST((regnode_charclass_posixl*) and_with, i + 1));
1244 if (ANYOF_POSIXL_TEST((regnode_charclass_posixl*) and_with, i)) {
1245 ANYOF_POSIXL_SET(&temp, i + add);
1247 add = 0 - add; /* 1 goes to -1; -1 goes to 1 */
1249 ANYOF_POSIXL_AND(&temp, ssc);
1251 } /* else ssc already has no posixes */
1252 } /* else: Not inverted. This routine is a no-op if 'and_with' is an SSC
1253 in its initial state */
1254 else if (! is_ANYOF_SYNTHETIC(and_with)
1255 || ! ssc_is_cp_posixl_init(pRExC_state, (regnode_ssc *)and_with))
1257 /* But if 'ssc' is in its initial state, the result is just 'and_with';
1258 * copy it over 'ssc' */
1259 if (ssc_is_cp_posixl_init(pRExC_state, ssc)) {
1260 if (is_ANYOF_SYNTHETIC(and_with)) {
1261 StructCopy(and_with, ssc, regnode_ssc);
1264 ssc->invlist = anded_cp_list;
1265 ANYOF_POSIXL_ZERO(ssc);
1266 if (ANYOF_FLAGS(and_with) & ANYOF_POSIXL) {
1267 ANYOF_POSIXL_OR((regnode_charclass_posixl*) and_with, ssc);
1271 else if (ANYOF_POSIXL_SSC_TEST_ANY_SET(ssc)
1272 || (ANYOF_FLAGS(and_with) & ANYOF_POSIXL))
1274 /* One or the other of P1, P2 is non-empty. */
1275 if (ANYOF_FLAGS(and_with) & ANYOF_POSIXL) {
1276 ANYOF_POSIXL_AND((regnode_charclass_posixl*) and_with, ssc);
1278 ssc_union(ssc, anded_cp_list, FALSE);
1280 else { /* P1 = P2 = empty */
1281 ssc_intersection(ssc, anded_cp_list, FALSE);
1287 S_ssc_or(pTHX_ const RExC_state_t *pRExC_state, regnode_ssc *ssc,
1288 const regnode_charclass *or_with)
1290 /* Accumulate into SSC 'ssc' its 'OR' with 'or_with', which is either
1291 * another SSC or a regular ANYOF class. Can create false positives if
1292 * 'or_with' is to be inverted. */
1297 PERL_ARGS_ASSERT_SSC_OR;
1299 assert(is_ANYOF_SYNTHETIC(ssc));
1301 /* 'or_with' is used as-is if it too is an SSC; otherwise have to extract
1302 * the code point inversion list and just the relevant flags */
1303 if (is_ANYOF_SYNTHETIC(or_with)) {
1304 ored_cp_list = ((regnode_ssc*) or_with)->invlist;
1305 ored_flags = ANYOF_FLAGS(or_with);
1308 ored_cp_list = get_ANYOF_cp_list_for_ssc(pRExC_state, or_with);
1309 ored_flags = ANYOF_FLAGS(or_with) & ANYOF_COMMON_FLAGS;
1312 ANYOF_FLAGS(ssc) |= ored_flags;
1314 /* Below, C1 is the list of code points in 'ssc'; P1, its posix classes.
1315 * C2 is the list of code points in 'or-with'; P2, its posix classes.
1316 * 'or_with' may be inverted. When not inverted, we have the simple
1317 * situation of computing:
1318 * (C1 | P1) | (C2 | P2) = (C1 | C2) | (P1 | P2)
1319 * If P1|P2 yields a situation with both a class and its complement are
1320 * set, like having both \w and \W, this matches all code points, and we
1321 * can delete these from the P component of the ssc going forward. XXX We
1322 * might be able to delete all the P components, but I (khw) am not certain
1323 * about this, and it is better to be safe.
1326 * (C1 | P1) | ~(C2 | P2) = (C1 | P1) | (~C2 & ~P2)
1327 * <= (C1 | P1) | ~C2
1328 * <= (C1 | ~C2) | P1
1329 * (which results in actually simpler code than the non-inverted case)
1332 if ((ANYOF_FLAGS(or_with) & ANYOF_INVERT)
1333 && ! is_ANYOF_SYNTHETIC(or_with))
1335 /* We ignore P2, leaving P1 going forward */
1336 } /* else Not inverted */
1337 else if (ANYOF_FLAGS(or_with) & ANYOF_POSIXL) {
1338 ANYOF_POSIXL_OR((regnode_charclass_posixl*)or_with, ssc);
1339 if (ANYOF_POSIXL_SSC_TEST_ANY_SET(ssc)) {
1341 for (i = 0; i < ANYOF_MAX; i += 2) {
1342 if (ANYOF_POSIXL_TEST(ssc, i) && ANYOF_POSIXL_TEST(ssc, i + 1))
1344 ssc_match_all_cp(ssc);
1345 ANYOF_POSIXL_CLEAR(ssc, i);
1346 ANYOF_POSIXL_CLEAR(ssc, i+1);
1354 FALSE /* Already has been inverted */
1358 PERL_STATIC_INLINE void
1359 S_ssc_union(pTHX_ regnode_ssc *ssc, SV* const invlist, const bool invert2nd)
1361 PERL_ARGS_ASSERT_SSC_UNION;
1363 assert(is_ANYOF_SYNTHETIC(ssc));
1365 _invlist_union_maybe_complement_2nd(ssc->invlist,
1371 PERL_STATIC_INLINE void
1372 S_ssc_intersection(pTHX_ regnode_ssc *ssc,
1374 const bool invert2nd)
1376 PERL_ARGS_ASSERT_SSC_INTERSECTION;
1378 assert(is_ANYOF_SYNTHETIC(ssc));
1380 _invlist_intersection_maybe_complement_2nd(ssc->invlist,
1386 PERL_STATIC_INLINE void
1387 S_ssc_add_range(pTHX_ regnode_ssc *ssc, const UV start, const UV end)
1389 PERL_ARGS_ASSERT_SSC_ADD_RANGE;
1391 assert(is_ANYOF_SYNTHETIC(ssc));
1393 ssc->invlist = _add_range_to_invlist(ssc->invlist, start, end);
1396 PERL_STATIC_INLINE void
1397 S_ssc_cp_and(pTHX_ regnode_ssc *ssc, const UV cp)
1399 /* AND just the single code point 'cp' into the SSC 'ssc' */
1401 SV* cp_list = _new_invlist(2);
1403 PERL_ARGS_ASSERT_SSC_CP_AND;
1405 assert(is_ANYOF_SYNTHETIC(ssc));
1407 cp_list = add_cp_to_invlist(cp_list, cp);
1408 ssc_intersection(ssc, cp_list,
1409 FALSE /* Not inverted */
1411 SvREFCNT_dec_NN(cp_list);
1414 PERL_STATIC_INLINE void
1415 S_ssc_clear_locale(regnode_ssc *ssc)
1417 /* Set the SSC 'ssc' to not match any locale things */
1418 PERL_ARGS_ASSERT_SSC_CLEAR_LOCALE;
1420 assert(is_ANYOF_SYNTHETIC(ssc));
1422 ANYOF_POSIXL_ZERO(ssc);
1423 ANYOF_FLAGS(ssc) &= ~ANYOF_LOCALE_FLAGS;
1427 S_ssc_finalize(pTHX_ RExC_state_t *pRExC_state, regnode_ssc *ssc)
1429 /* The inversion list in the SSC is marked mortal; now we need a more
1430 * permanent copy, which is stored the same way that is done in a regular
1431 * ANYOF node, with the first 256 code points in a bit map */
1433 SV* invlist = invlist_clone(ssc->invlist);
1435 PERL_ARGS_ASSERT_SSC_FINALIZE;
1437 assert(is_ANYOF_SYNTHETIC(ssc));
1439 /* The code in this file assumes that all but these flags aren't relevant
1440 * to the SSC, except ANYOF_EMPTY_STRING, which should be cleared by the
1441 * time we reach here */
1442 assert(! (ANYOF_FLAGS(ssc) & ~ANYOF_COMMON_FLAGS));
1444 populate_ANYOF_from_invlist( (regnode *) ssc, &invlist);
1446 set_ANYOF_arg(pRExC_state, (regnode *) ssc, invlist,
1447 NULL, NULL, NULL, FALSE);
1449 /* Make sure is clone-safe */
1450 ssc->invlist = NULL;
1452 if (ANYOF_POSIXL_SSC_TEST_ANY_SET(ssc)) {
1453 ANYOF_FLAGS(ssc) |= ANYOF_POSIXL;
1456 assert(! (ANYOF_FLAGS(ssc) & ANYOF_LOCALE_FLAGS) || RExC_contains_locale);
1459 #define TRIE_LIST_ITEM(state,idx) (trie->states[state].trans.list)[ idx ]
1460 #define TRIE_LIST_CUR(state) ( TRIE_LIST_ITEM( state, 0 ).forid )
1461 #define TRIE_LIST_LEN(state) ( TRIE_LIST_ITEM( state, 0 ).newstate )
1462 #define TRIE_LIST_USED(idx) ( trie->states[state].trans.list \
1463 ? (TRIE_LIST_CUR( idx ) - 1) \
1469 dump_trie(trie,widecharmap,revcharmap)
1470 dump_trie_interim_list(trie,widecharmap,revcharmap,next_alloc)
1471 dump_trie_interim_table(trie,widecharmap,revcharmap,next_alloc)
1473 These routines dump out a trie in a somewhat readable format.
1474 The _interim_ variants are used for debugging the interim
1475 tables that are used to generate the final compressed
1476 representation which is what dump_trie expects.
1478 Part of the reason for their existence is to provide a form
1479 of documentation as to how the different representations function.
1484 Dumps the final compressed table form of the trie to Perl_debug_log.
1485 Used for debugging make_trie().
1489 S_dump_trie(pTHX_ const struct _reg_trie_data *trie, HV *widecharmap,
1490 AV *revcharmap, U32 depth)
1493 SV *sv=sv_newmortal();
1494 int colwidth= widecharmap ? 6 : 4;
1496 GET_RE_DEBUG_FLAGS_DECL;
1498 PERL_ARGS_ASSERT_DUMP_TRIE;
1500 PerlIO_printf( Perl_debug_log, "%*sChar : %-6s%-6s%-4s ",
1501 (int)depth * 2 + 2,"",
1502 "Match","Base","Ofs" );
1504 for( state = 0 ; state < trie->uniquecharcount ; state++ ) {
1505 SV ** const tmp = av_fetch( revcharmap, state, 0);
1507 PerlIO_printf( Perl_debug_log, "%*s",
1509 pv_pretty(sv, SvPV_nolen_const(*tmp), SvCUR(*tmp), colwidth,
1510 PL_colors[0], PL_colors[1],
1511 (SvUTF8(*tmp) ? PERL_PV_ESCAPE_UNI : 0) |
1512 PERL_PV_ESCAPE_FIRSTCHAR
1517 PerlIO_printf( Perl_debug_log, "\n%*sState|-----------------------",
1518 (int)depth * 2 + 2,"");
1520 for( state = 0 ; state < trie->uniquecharcount ; state++ )
1521 PerlIO_printf( Perl_debug_log, "%.*s", colwidth, "--------");
1522 PerlIO_printf( Perl_debug_log, "\n");
1524 for( state = 1 ; state < trie->statecount ; state++ ) {
1525 const U32 base = trie->states[ state ].trans.base;
1527 PerlIO_printf( Perl_debug_log, "%*s#%4"UVXf"|",
1528 (int)depth * 2 + 2,"", (UV)state);
1530 if ( trie->states[ state ].wordnum ) {
1531 PerlIO_printf( Perl_debug_log, " W%4X",
1532 trie->states[ state ].wordnum );
1534 PerlIO_printf( Perl_debug_log, "%6s", "" );
1537 PerlIO_printf( Perl_debug_log, " @%4"UVXf" ", (UV)base );
1542 while( ( base + ofs < trie->uniquecharcount ) ||
1543 ( base + ofs - trie->uniquecharcount < trie->lasttrans
1544 && trie->trans[ base + ofs - trie->uniquecharcount ].check
1548 PerlIO_printf( Perl_debug_log, "+%2"UVXf"[ ", (UV)ofs);
1550 for ( ofs = 0 ; ofs < trie->uniquecharcount ; ofs++ ) {
1551 if ( ( base + ofs >= trie->uniquecharcount )
1552 && ( base + ofs - trie->uniquecharcount
1554 && trie->trans[ base + ofs
1555 - trie->uniquecharcount ].check == state )
1557 PerlIO_printf( Perl_debug_log, "%*"UVXf,
1559 (UV)trie->trans[ base + ofs
1560 - trie->uniquecharcount ].next );
1562 PerlIO_printf( Perl_debug_log, "%*s",colwidth," ." );
1566 PerlIO_printf( Perl_debug_log, "]");
1569 PerlIO_printf( Perl_debug_log, "\n" );
1571 PerlIO_printf(Perl_debug_log, "%*sword_info N:(prev,len)=",
1573 for (word=1; word <= trie->wordcount; word++) {
1574 PerlIO_printf(Perl_debug_log, " %d:(%d,%d)",
1575 (int)word, (int)(trie->wordinfo[word].prev),
1576 (int)(trie->wordinfo[word].len));
1578 PerlIO_printf(Perl_debug_log, "\n" );
1581 Dumps a fully constructed but uncompressed trie in list form.
1582 List tries normally only are used for construction when the number of
1583 possible chars (trie->uniquecharcount) is very high.
1584 Used for debugging make_trie().
1587 S_dump_trie_interim_list(pTHX_ const struct _reg_trie_data *trie,
1588 HV *widecharmap, AV *revcharmap, U32 next_alloc,
1592 SV *sv=sv_newmortal();
1593 int colwidth= widecharmap ? 6 : 4;
1594 GET_RE_DEBUG_FLAGS_DECL;
1596 PERL_ARGS_ASSERT_DUMP_TRIE_INTERIM_LIST;
1598 /* print out the table precompression. */
1599 PerlIO_printf( Perl_debug_log, "%*sState :Word | Transition Data\n%*s%s",
1600 (int)depth * 2 + 2,"", (int)depth * 2 + 2,"",
1601 "------:-----+-----------------\n" );
1603 for( state=1 ; state < next_alloc ; state ++ ) {
1606 PerlIO_printf( Perl_debug_log, "%*s %4"UVXf" :",
1607 (int)depth * 2 + 2,"", (UV)state );
1608 if ( ! trie->states[ state ].wordnum ) {
1609 PerlIO_printf( Perl_debug_log, "%5s| ","");
1611 PerlIO_printf( Perl_debug_log, "W%4x| ",
1612 trie->states[ state ].wordnum
1615 for( charid = 1 ; charid <= TRIE_LIST_USED( state ) ; charid++ ) {
1616 SV ** const tmp = av_fetch( revcharmap,
1617 TRIE_LIST_ITEM(state,charid).forid, 0);
1619 PerlIO_printf( Perl_debug_log, "%*s:%3X=%4"UVXf" | ",
1621 pv_pretty(sv, SvPV_nolen_const(*tmp), SvCUR(*tmp),
1623 PL_colors[0], PL_colors[1],
1624 (SvUTF8(*tmp) ? PERL_PV_ESCAPE_UNI : 0)
1625 | PERL_PV_ESCAPE_FIRSTCHAR
1627 TRIE_LIST_ITEM(state,charid).forid,
1628 (UV)TRIE_LIST_ITEM(state,charid).newstate
1631 PerlIO_printf(Perl_debug_log, "\n%*s| ",
1632 (int)((depth * 2) + 14), "");
1635 PerlIO_printf( Perl_debug_log, "\n");
1640 Dumps a fully constructed but uncompressed trie in table form.
1641 This is the normal DFA style state transition table, with a few
1642 twists to facilitate compression later.
1643 Used for debugging make_trie().
1646 S_dump_trie_interim_table(pTHX_ const struct _reg_trie_data *trie,
1647 HV *widecharmap, AV *revcharmap, U32 next_alloc,
1652 SV *sv=sv_newmortal();
1653 int colwidth= widecharmap ? 6 : 4;
1654 GET_RE_DEBUG_FLAGS_DECL;
1656 PERL_ARGS_ASSERT_DUMP_TRIE_INTERIM_TABLE;
1659 print out the table precompression so that we can do a visual check
1660 that they are identical.
1663 PerlIO_printf( Perl_debug_log, "%*sChar : ",(int)depth * 2 + 2,"" );
1665 for( charid = 0 ; charid < trie->uniquecharcount ; charid++ ) {
1666 SV ** const tmp = av_fetch( revcharmap, charid, 0);
1668 PerlIO_printf( Perl_debug_log, "%*s",
1670 pv_pretty(sv, SvPV_nolen_const(*tmp), SvCUR(*tmp), colwidth,
1671 PL_colors[0], PL_colors[1],
1672 (SvUTF8(*tmp) ? PERL_PV_ESCAPE_UNI : 0) |
1673 PERL_PV_ESCAPE_FIRSTCHAR
1679 PerlIO_printf( Perl_debug_log, "\n%*sState+-",(int)depth * 2 + 2,"" );
1681 for( charid=0 ; charid < trie->uniquecharcount ; charid++ ) {
1682 PerlIO_printf( Perl_debug_log, "%.*s", colwidth,"--------");
1685 PerlIO_printf( Perl_debug_log, "\n" );
1687 for( state=1 ; state < next_alloc ; state += trie->uniquecharcount ) {
1689 PerlIO_printf( Perl_debug_log, "%*s%4"UVXf" : ",
1690 (int)depth * 2 + 2,"",
1691 (UV)TRIE_NODENUM( state ) );
1693 for( charid = 0 ; charid < trie->uniquecharcount ; charid++ ) {
1694 UV v=(UV)SAFE_TRIE_NODENUM( trie->trans[ state + charid ].next );
1696 PerlIO_printf( Perl_debug_log, "%*"UVXf, colwidth, v );
1698 PerlIO_printf( Perl_debug_log, "%*s", colwidth, "." );
1700 if ( ! trie->states[ TRIE_NODENUM( state ) ].wordnum ) {
1701 PerlIO_printf( Perl_debug_log, " (%4"UVXf")\n",
1702 (UV)trie->trans[ state ].check );
1704 PerlIO_printf( Perl_debug_log, " (%4"UVXf") W%4X\n",
1705 (UV)trie->trans[ state ].check,
1706 trie->states[ TRIE_NODENUM( state ) ].wordnum );
1714 /* make_trie(startbranch,first,last,tail,word_count,flags,depth)
1715 startbranch: the first branch in the whole branch sequence
1716 first : start branch of sequence of branch-exact nodes.
1717 May be the same as startbranch
1718 last : Thing following the last branch.
1719 May be the same as tail.
1720 tail : item following the branch sequence
1721 count : words in the sequence
1722 flags : currently the OP() type we will be building one of /EXACT(|F|FA|FU|FU_SS)/
1723 depth : indent depth
1725 Inplace optimizes a sequence of 2 or more Branch-Exact nodes into a TRIE node.
1727 A trie is an N'ary tree where the branches are determined by digital
1728 decomposition of the key. IE, at the root node you look up the 1st character and
1729 follow that branch repeat until you find the end of the branches. Nodes can be
1730 marked as "accepting" meaning they represent a complete word. Eg:
1734 would convert into the following structure. Numbers represent states, letters
1735 following numbers represent valid transitions on the letter from that state, if
1736 the number is in square brackets it represents an accepting state, otherwise it
1737 will be in parenthesis.
1739 +-h->+-e->[3]-+-r->(8)-+-s->[9]
1743 (1) +-i->(6)-+-s->[7]
1745 +-s->(3)-+-h->(4)-+-e->[5]
1747 Accept Word Mapping: 3=>1 (he),5=>2 (she), 7=>3 (his), 9=>4 (hers)
1749 This shows that when matching against the string 'hers' we will begin at state 1
1750 read 'h' and move to state 2, read 'e' and move to state 3 which is accepting,
1751 then read 'r' and go to state 8 followed by 's' which takes us to state 9 which
1752 is also accepting. Thus we know that we can match both 'he' and 'hers' with a
1753 single traverse. We store a mapping from accepting to state to which word was
1754 matched, and then when we have multiple possibilities we try to complete the
1755 rest of the regex in the order in which they occured in the alternation.
1757 The only prior NFA like behaviour that would be changed by the TRIE support is
1758 the silent ignoring of duplicate alternations which are of the form:
1760 / (DUPE|DUPE) X? (?{ ... }) Y /x
1762 Thus EVAL blocks following a trie may be called a different number of times with
1763 and without the optimisation. With the optimisations dupes will be silently
1764 ignored. This inconsistent behaviour of EVAL type nodes is well established as
1765 the following demonstrates:
1767 'words'=~/(word|word|word)(?{ print $1 })[xyz]/
1769 which prints out 'word' three times, but
1771 'words'=~/(word|word|word)(?{ print $1 })S/
1773 which doesnt print it out at all. This is due to other optimisations kicking in.
1775 Example of what happens on a structural level:
1777 The regexp /(ac|ad|ab)+/ will produce the following debug output:
1779 1: CURLYM[1] {1,32767}(18)
1790 This would be optimizable with startbranch=5, first=5, last=16, tail=16
1791 and should turn into:
1793 1: CURLYM[1] {1,32767}(18)
1795 [Words:3 Chars Stored:6 Unique Chars:4 States:5 NCP:1]
1803 Cases where tail != last would be like /(?foo|bar)baz/:
1813 which would be optimizable with startbranch=1, first=1, last=7, tail=8
1814 and would end up looking like:
1817 [Words:2 Chars Stored:6 Unique Chars:5 States:7 NCP:1]
1824 d = uvchr_to_utf8_flags(d, uv, 0);
1826 is the recommended Unicode-aware way of saying
1831 #define TRIE_STORE_REVCHAR(val) \
1834 SV *zlopp = newSV(7); /* XXX: optimize me */ \
1835 unsigned char *flrbbbbb = (unsigned char *) SvPVX(zlopp); \
1836 unsigned const char *const kapow = uvchr_to_utf8(flrbbbbb, val); \
1837 SvCUR_set(zlopp, kapow - flrbbbbb); \
1840 av_push(revcharmap, zlopp); \
1842 char ooooff = (char)val; \
1843 av_push(revcharmap, newSVpvn(&ooooff, 1)); \
1847 /* This gets the next character from the input, folding it if not already
1849 #define TRIE_READ_CHAR STMT_START { \
1852 /* if it is UTF then it is either already folded, or does not need \
1854 uvc = valid_utf8_to_uvchr( (const U8*) uc, &len); \
1856 else if (folder == PL_fold_latin1) { \
1857 /* This folder implies Unicode rules, which in the range expressible \
1858 * by not UTF is the lower case, with the two exceptions, one of \
1859 * which should have been taken care of before calling this */ \
1860 assert(*uc != LATIN_SMALL_LETTER_SHARP_S); \
1861 uvc = toLOWER_L1(*uc); \
1862 if (UNLIKELY(uvc == MICRO_SIGN)) uvc = GREEK_SMALL_LETTER_MU; \
1865 /* raw data, will be folded later if needed */ \
1873 #define TRIE_LIST_PUSH(state,fid,ns) STMT_START { \
1874 if ( TRIE_LIST_CUR( state ) >=TRIE_LIST_LEN( state ) ) { \
1875 U32 ging = TRIE_LIST_LEN( state ) *= 2; \
1876 Renew( trie->states[ state ].trans.list, ging, reg_trie_trans_le ); \
1878 TRIE_LIST_ITEM( state, TRIE_LIST_CUR( state ) ).forid = fid; \
1879 TRIE_LIST_ITEM( state, TRIE_LIST_CUR( state ) ).newstate = ns; \
1880 TRIE_LIST_CUR( state )++; \
1883 #define TRIE_LIST_NEW(state) STMT_START { \
1884 Newxz( trie->states[ state ].trans.list, \
1885 4, reg_trie_trans_le ); \
1886 TRIE_LIST_CUR( state ) = 1; \
1887 TRIE_LIST_LEN( state ) = 4; \
1890 #define TRIE_HANDLE_WORD(state) STMT_START { \
1891 U16 dupe= trie->states[ state ].wordnum; \
1892 regnode * const noper_next = regnext( noper ); \
1895 /* store the word for dumping */ \
1897 if (OP(noper) != NOTHING) \
1898 tmp = newSVpvn_utf8(STRING(noper), STR_LEN(noper), UTF); \
1900 tmp = newSVpvn_utf8( "", 0, UTF ); \
1901 av_push( trie_words, tmp ); \
1905 trie->wordinfo[curword].prev = 0; \
1906 trie->wordinfo[curword].len = wordlen; \
1907 trie->wordinfo[curword].accept = state; \
1909 if ( noper_next < tail ) { \
1911 trie->jump = (U16 *) PerlMemShared_calloc( word_count + 1, \
1913 trie->jump[curword] = (U16)(noper_next - convert); \
1915 jumper = noper_next; \
1917 nextbranch= regnext(cur); \
1921 /* It's a dupe. Pre-insert into the wordinfo[].prev */\
1922 /* chain, so that when the bits of chain are later */\
1923 /* linked together, the dups appear in the chain */\
1924 trie->wordinfo[curword].prev = trie->wordinfo[dupe].prev; \
1925 trie->wordinfo[dupe].prev = curword; \
1927 /* we haven't inserted this word yet. */ \
1928 trie->states[ state ].wordnum = curword; \
1933 #define TRIE_TRANS_STATE(state,base,ucharcount,charid,special) \
1934 ( ( base + charid >= ucharcount \
1935 && base + charid < ubound \
1936 && state == trie->trans[ base - ucharcount + charid ].check \
1937 && trie->trans[ base - ucharcount + charid ].next ) \
1938 ? trie->trans[ base - ucharcount + charid ].next \
1939 : ( state==1 ? special : 0 ) \
1943 #define MADE_JUMP_TRIE 2
1944 #define MADE_EXACT_TRIE 4
1947 S_make_trie(pTHX_ RExC_state_t *pRExC_state, regnode *startbranch,
1948 regnode *first, regnode *last, regnode *tail,
1949 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 assert(re_trie_maxbuff);
2011 if (!SvIOK(re_trie_maxbuff)) {
2012 sv_setiv(re_trie_maxbuff, RE_TRIE_MAXBUF_INIT);
2014 DEBUG_TRIE_COMPILE_r({
2015 PerlIO_printf( Perl_debug_log,
2016 "%*smake_trie start==%d, first==%d, last==%d, tail==%d depth=%d\n",
2017 (int)depth * 2 + 2, "",
2018 REG_NODE_NUM(startbranch),REG_NODE_NUM(first),
2019 REG_NODE_NUM(last), REG_NODE_NUM(tail), (int)depth);
2022 /* Find the node we are going to overwrite */
2023 if ( first == startbranch && OP( last ) != BRANCH ) {
2024 /* whole branch chain */
2027 /* branch sub-chain */
2028 convert = NEXTOPER( first );
2031 /* -- First loop and Setup --
2033 We first traverse the branches and scan each word to determine if it
2034 contains widechars, and how many unique chars there are, this is
2035 important as we have to build a table with at least as many columns as we
2038 We use an array of integers to represent the character codes 0..255
2039 (trie->charmap) and we use a an HV* to store Unicode characters. We use
2040 the native representation of the character value as the key and IV's for
2043 *TODO* If we keep track of how many times each character is used we can
2044 remap the columns so that the table compression later on is more
2045 efficient in terms of memory by ensuring the most common value is in the
2046 middle and the least common are on the outside. IMO this would be better
2047 than a most to least common mapping as theres a decent chance the most
2048 common letter will share a node with the least common, meaning the node
2049 will not be compressible. With a middle is most common approach the worst
2050 case is when we have the least common nodes twice.
2054 for ( cur = first ; cur < last ; cur = regnext( cur ) ) {
2055 regnode *noper = NEXTOPER( cur );
2056 const U8 *uc = (U8*)STRING( noper );
2057 const U8 *e = uc + STR_LEN( noper );
2059 U32 wordlen = 0; /* required init */
2060 STRLEN minchars = 0;
2061 STRLEN maxchars = 0;
2062 bool set_bit = trie->bitmap ? 1 : 0; /*store the first char in the
2065 if (OP(noper) == NOTHING) {
2066 regnode *noper_next= regnext(noper);
2067 if (noper_next != tail && OP(noper_next) == flags) {
2069 uc= (U8*)STRING(noper);
2070 e= uc + STR_LEN(noper);
2071 trie->minlen= STR_LEN(noper);
2078 if ( set_bit ) { /* bitmap only alloced when !(UTF&&Folding) */
2079 TRIE_BITMAP_SET(trie,*uc); /* store the raw first byte
2080 regardless of encoding */
2081 if (OP( noper ) == EXACTFU_SS) {
2082 /* false positives are ok, so just set this */
2083 TRIE_BITMAP_SET(trie, LATIN_SMALL_LETTER_SHARP_S);
2086 for ( ; uc < e ; uc += len ) { /* Look at each char in the current
2088 TRIE_CHARCOUNT(trie)++;
2091 /* TRIE_READ_CHAR returns the current character, or its fold if /i
2092 * is in effect. Under /i, this character can match itself, or
2093 * anything that folds to it. If not under /i, it can match just
2094 * itself. Most folds are 1-1, for example k, K, and KELVIN SIGN
2095 * all fold to k, and all are single characters. But some folds
2096 * expand to more than one character, so for example LATIN SMALL
2097 * LIGATURE FFI folds to the three character sequence 'ffi'. If
2098 * the string beginning at 'uc' is 'ffi', it could be matched by
2099 * three characters, or just by the one ligature character. (It
2100 * could also be matched by two characters: LATIN SMALL LIGATURE FF
2101 * followed by 'i', or by 'f' followed by LATIN SMALL LIGATURE FI).
2102 * (Of course 'I' and/or 'F' instead of 'i' and 'f' can also
2103 * match.) The trie needs to know the minimum and maximum number
2104 * of characters that could match so that it can use size alone to
2105 * quickly reject many match attempts. The max is simple: it is
2106 * the number of folded characters in this branch (since a fold is
2107 * never shorter than what folds to it. */
2111 /* And the min is equal to the max if not under /i (indicated by
2112 * 'folder' being NULL), or there are no multi-character folds. If
2113 * there is a multi-character fold, the min is incremented just
2114 * once, for the character that folds to the sequence. Each
2115 * character in the sequence needs to be added to the list below of
2116 * characters in the trie, but we count only the first towards the
2117 * min number of characters needed. This is done through the
2118 * variable 'foldlen', which is returned by the macros that look
2119 * for these sequences as the number of bytes the sequence
2120 * occupies. Each time through the loop, we decrement 'foldlen' by
2121 * how many bytes the current char occupies. Only when it reaches
2122 * 0 do we increment 'minchars' or look for another multi-character
2124 if (folder == NULL) {
2127 else if (foldlen > 0) {
2128 foldlen -= (UTF) ? UTF8SKIP(uc) : 1;
2133 /* See if *uc is the beginning of a multi-character fold. If
2134 * so, we decrement the length remaining to look at, to account
2135 * for the current character this iteration. (We can use 'uc'
2136 * instead of the fold returned by TRIE_READ_CHAR because for
2137 * non-UTF, the latin1_safe macro is smart enough to account
2138 * for all the unfolded characters, and because for UTF, the
2139 * string will already have been folded earlier in the
2140 * compilation process */
2142 if ((foldlen = is_MULTI_CHAR_FOLD_utf8_safe(uc, e))) {
2143 foldlen -= UTF8SKIP(uc);
2146 else if ((foldlen = is_MULTI_CHAR_FOLD_latin1_safe(uc, e))) {
2151 /* The current character (and any potential folds) should be added
2152 * to the possible matching characters for this position in this
2156 U8 folded= folder[ (U8) uvc ];
2157 if ( !trie->charmap[ folded ] ) {
2158 trie->charmap[ folded ]=( ++trie->uniquecharcount );
2159 TRIE_STORE_REVCHAR( folded );
2162 if ( !trie->charmap[ uvc ] ) {
2163 trie->charmap[ uvc ]=( ++trie->uniquecharcount );
2164 TRIE_STORE_REVCHAR( uvc );
2167 /* store the codepoint in the bitmap, and its folded
2169 TRIE_BITMAP_SET(trie, uvc);
2171 /* store the folded codepoint */
2172 if ( folder ) TRIE_BITMAP_SET(trie, folder[(U8) uvc ]);
2175 /* store first byte of utf8 representation of
2176 variant codepoints */
2177 if (! UVCHR_IS_INVARIANT(uvc)) {
2178 TRIE_BITMAP_SET(trie, UTF8_TWO_BYTE_HI(uvc));
2181 set_bit = 0; /* We've done our bit :-) */
2185 /* XXX We could come up with the list of code points that fold
2186 * to this using PL_utf8_foldclosures, except not for
2187 * multi-char folds, as there may be multiple combinations
2188 * there that could work, which needs to wait until runtime to
2189 * resolve (The comment about LIGATURE FFI above is such an
2194 widecharmap = newHV();
2196 svpp = hv_fetch( widecharmap, (char*)&uvc, sizeof( UV ), 1 );
2199 Perl_croak( aTHX_ "error creating/fetching widecharmap entry for 0x%"UVXf, uvc );
2201 if ( !SvTRUE( *svpp ) ) {
2202 sv_setiv( *svpp, ++trie->uniquecharcount );
2203 TRIE_STORE_REVCHAR(uvc);
2206 } /* end loop through characters in this branch of the trie */
2208 /* We take the min and max for this branch and combine to find the min
2209 * and max for all branches processed so far */
2210 if( cur == first ) {
2211 trie->minlen = minchars;
2212 trie->maxlen = maxchars;
2213 } else if (minchars < trie->minlen) {
2214 trie->minlen = minchars;
2215 } else if (maxchars > trie->maxlen) {
2216 trie->maxlen = maxchars;
2218 } /* end first pass */
2219 DEBUG_TRIE_COMPILE_r(
2220 PerlIO_printf( Perl_debug_log,
2221 "%*sTRIE(%s): W:%d C:%d Uq:%d Min:%d Max:%d\n",
2222 (int)depth * 2 + 2,"",
2223 ( widecharmap ? "UTF8" : "NATIVE" ), (int)word_count,
2224 (int)TRIE_CHARCOUNT(trie), trie->uniquecharcount,
2225 (int)trie->minlen, (int)trie->maxlen )
2229 We now know what we are dealing with in terms of unique chars and
2230 string sizes so we can calculate how much memory a naive
2231 representation using a flat table will take. If it's over a reasonable
2232 limit (as specified by ${^RE_TRIE_MAXBUF}) we use a more memory
2233 conservative but potentially much slower representation using an array
2236 At the end we convert both representations into the same compressed
2237 form that will be used in regexec.c for matching with. The latter
2238 is a form that cannot be used to construct with but has memory
2239 properties similar to the list form and access properties similar
2240 to the table form making it both suitable for fast searches and
2241 small enough that its feasable to store for the duration of a program.
2243 See the comment in the code where the compressed table is produced
2244 inplace from the flat tabe representation for an explanation of how
2245 the compression works.
2250 Newx(prev_states, TRIE_CHARCOUNT(trie) + 2, U32);
2253 if ( (IV)( ( TRIE_CHARCOUNT(trie) + 1 ) * trie->uniquecharcount + 1)
2254 > SvIV(re_trie_maxbuff) )
2257 Second Pass -- Array Of Lists Representation
2259 Each state will be represented by a list of charid:state records
2260 (reg_trie_trans_le) the first such element holds the CUR and LEN
2261 points of the allocated array. (See defines above).
2263 We build the initial structure using the lists, and then convert
2264 it into the compressed table form which allows faster lookups
2265 (but cant be modified once converted).
2268 STRLEN transcount = 1;
2270 DEBUG_TRIE_COMPILE_MORE_r( PerlIO_printf( Perl_debug_log,
2271 "%*sCompiling trie using list compiler\n",
2272 (int)depth * 2 + 2, ""));
2274 trie->states = (reg_trie_state *)
2275 PerlMemShared_calloc( TRIE_CHARCOUNT(trie) + 2,
2276 sizeof(reg_trie_state) );
2280 for ( cur = first ; cur < last ; cur = regnext( cur ) ) {
2282 regnode *noper = NEXTOPER( cur );
2283 U8 *uc = (U8*)STRING( noper );
2284 const U8 *e = uc + STR_LEN( noper );
2285 U32 state = 1; /* required init */
2286 U16 charid = 0; /* sanity init */
2287 U32 wordlen = 0; /* required init */
2289 if (OP(noper) == NOTHING) {
2290 regnode *noper_next= regnext(noper);
2291 if (noper_next != tail && OP(noper_next) == flags) {
2293 uc= (U8*)STRING(noper);
2294 e= uc + STR_LEN(noper);
2298 if (OP(noper) != NOTHING) {
2299 for ( ; uc < e ; uc += len ) {
2304 charid = trie->charmap[ uvc ];
2306 SV** const svpp = hv_fetch( widecharmap,
2313 charid=(U16)SvIV( *svpp );
2316 /* charid is now 0 if we dont know the char read, or
2317 * nonzero if we do */
2324 if ( !trie->states[ state ].trans.list ) {
2325 TRIE_LIST_NEW( state );
2328 check <= TRIE_LIST_USED( state );
2331 if ( TRIE_LIST_ITEM( state, check ).forid
2334 newstate = TRIE_LIST_ITEM( state, check ).newstate;
2339 newstate = next_alloc++;
2340 prev_states[newstate] = state;
2341 TRIE_LIST_PUSH( state, charid, newstate );
2346 Perl_croak( aTHX_ "panic! In trie construction, no char mapping for %"IVdf, uvc );
2350 TRIE_HANDLE_WORD(state);
2352 } /* end second pass */
2354 /* next alloc is the NEXT state to be allocated */
2355 trie->statecount = next_alloc;
2356 trie->states = (reg_trie_state *)
2357 PerlMemShared_realloc( trie->states,
2359 * sizeof(reg_trie_state) );
2361 /* and now dump it out before we compress it */
2362 DEBUG_TRIE_COMPILE_MORE_r(dump_trie_interim_list(trie, widecharmap,
2363 revcharmap, next_alloc,
2367 trie->trans = (reg_trie_trans *)
2368 PerlMemShared_calloc( transcount, sizeof(reg_trie_trans) );
2375 for( state=1 ; state < next_alloc ; state ++ ) {
2379 DEBUG_TRIE_COMPILE_MORE_r(
2380 PerlIO_printf( Perl_debug_log, "tp: %d zp: %d ",tp,zp)
2384 if (trie->states[state].trans.list) {
2385 U16 minid=TRIE_LIST_ITEM( state, 1).forid;
2389 for( idx = 2 ; idx <= TRIE_LIST_USED( state ) ; idx++ ) {
2390 const U16 forid = TRIE_LIST_ITEM( state, idx).forid;
2391 if ( forid < minid ) {
2393 } else if ( forid > maxid ) {
2397 if ( transcount < tp + maxid - minid + 1) {
2399 trie->trans = (reg_trie_trans *)
2400 PerlMemShared_realloc( trie->trans,
2402 * sizeof(reg_trie_trans) );
2403 Zero( trie->trans + (transcount / 2),
2407 base = trie->uniquecharcount + tp - minid;
2408 if ( maxid == minid ) {
2410 for ( ; zp < tp ; zp++ ) {
2411 if ( ! trie->trans[ zp ].next ) {
2412 base = trie->uniquecharcount + zp - minid;
2413 trie->trans[ zp ].next = TRIE_LIST_ITEM( state,
2415 trie->trans[ zp ].check = state;
2421 trie->trans[ tp ].next = TRIE_LIST_ITEM( state,
2423 trie->trans[ tp ].check = state;
2428 for ( idx=1; idx <= TRIE_LIST_USED( state ) ; idx++ ) {
2429 const U32 tid = base
2430 - trie->uniquecharcount
2431 + TRIE_LIST_ITEM( state, idx ).forid;
2432 trie->trans[ tid ].next = TRIE_LIST_ITEM( state,
2434 trie->trans[ tid ].check = state;
2436 tp += ( maxid - minid + 1 );
2438 Safefree(trie->states[ state ].trans.list);
2441 DEBUG_TRIE_COMPILE_MORE_r(
2442 PerlIO_printf( Perl_debug_log, " base: %d\n",base);
2445 trie->states[ state ].trans.base=base;
2447 trie->lasttrans = tp + 1;
2451 Second Pass -- Flat Table Representation.
2453 we dont use the 0 slot of either trans[] or states[] so we add 1 to
2454 each. We know that we will need Charcount+1 trans at most to store
2455 the data (one row per char at worst case) So we preallocate both
2456 structures assuming worst case.
2458 We then construct the trie using only the .next slots of the entry
2461 We use the .check field of the first entry of the node temporarily
2462 to make compression both faster and easier by keeping track of how
2463 many non zero fields are in the node.
2465 Since trans are numbered from 1 any 0 pointer in the table is a FAIL
2468 There are two terms at use here: state as a TRIE_NODEIDX() which is
2469 a number representing the first entry of the node, and state as a
2470 TRIE_NODENUM() which is the trans number. state 1 is TRIE_NODEIDX(1)
2471 and TRIE_NODENUM(1), state 2 is TRIE_NODEIDX(2) and TRIE_NODENUM(3)
2472 if there are 2 entrys per node. eg:
2480 The table is internally in the right hand, idx form. However as we
2481 also have to deal with the states array which is indexed by nodenum
2482 we have to use TRIE_NODENUM() to convert.
2485 DEBUG_TRIE_COMPILE_MORE_r( PerlIO_printf( Perl_debug_log,
2486 "%*sCompiling trie using table compiler\n",
2487 (int)depth * 2 + 2, ""));
2489 trie->trans = (reg_trie_trans *)
2490 PerlMemShared_calloc( ( TRIE_CHARCOUNT(trie) + 1 )
2491 * trie->uniquecharcount + 1,
2492 sizeof(reg_trie_trans) );
2493 trie->states = (reg_trie_state *)
2494 PerlMemShared_calloc( TRIE_CHARCOUNT(trie) + 2,
2495 sizeof(reg_trie_state) );
2496 next_alloc = trie->uniquecharcount + 1;
2499 for ( cur = first ; cur < last ; cur = regnext( cur ) ) {
2501 regnode *noper = NEXTOPER( cur );
2502 const U8 *uc = (U8*)STRING( noper );
2503 const U8 *e = uc + STR_LEN( noper );
2505 U32 state = 1; /* required init */
2507 U16 charid = 0; /* sanity init */
2508 U32 accept_state = 0; /* sanity init */
2510 U32 wordlen = 0; /* required init */
2512 if (OP(noper) == NOTHING) {
2513 regnode *noper_next= regnext(noper);
2514 if (noper_next != tail && OP(noper_next) == flags) {
2516 uc= (U8*)STRING(noper);
2517 e= uc + STR_LEN(noper);
2521 if ( OP(noper) != NOTHING ) {
2522 for ( ; uc < e ; uc += len ) {
2527 charid = trie->charmap[ uvc ];
2529 SV* const * const svpp = hv_fetch( widecharmap,
2533 charid = svpp ? (U16)SvIV(*svpp) : 0;
2537 if ( !trie->trans[ state + charid ].next ) {
2538 trie->trans[ state + charid ].next = next_alloc;
2539 trie->trans[ state ].check++;
2540 prev_states[TRIE_NODENUM(next_alloc)]
2541 = TRIE_NODENUM(state);
2542 next_alloc += trie->uniquecharcount;
2544 state = trie->trans[ state + charid ].next;
2546 Perl_croak( aTHX_ "panic! In trie construction, no char mapping for %"IVdf, uvc );
2548 /* charid is now 0 if we dont know the char read, or
2549 * nonzero if we do */
2552 accept_state = TRIE_NODENUM( state );
2553 TRIE_HANDLE_WORD(accept_state);
2555 } /* end second pass */
2557 /* and now dump it out before we compress it */
2558 DEBUG_TRIE_COMPILE_MORE_r(dump_trie_interim_table(trie, widecharmap,
2560 next_alloc, depth+1));
2564 * Inplace compress the table.*
2566 For sparse data sets the table constructed by the trie algorithm will
2567 be mostly 0/FAIL transitions or to put it another way mostly empty.
2568 (Note that leaf nodes will not contain any transitions.)
2570 This algorithm compresses the tables by eliminating most such
2571 transitions, at the cost of a modest bit of extra work during lookup:
2573 - Each states[] entry contains a .base field which indicates the
2574 index in the state[] array wheres its transition data is stored.
2576 - If .base is 0 there are no valid transitions from that node.
2578 - If .base is nonzero then charid is added to it to find an entry in
2581 -If trans[states[state].base+charid].check!=state then the
2582 transition is taken to be a 0/Fail transition. Thus if there are fail
2583 transitions at the front of the node then the .base offset will point
2584 somewhere inside the previous nodes data (or maybe even into a node
2585 even earlier), but the .check field determines if the transition is
2589 The following process inplace converts the table to the compressed
2590 table: We first do not compress the root node 1,and mark all its
2591 .check pointers as 1 and set its .base pointer as 1 as well. This
2592 allows us to do a DFA construction from the compressed table later,
2593 and ensures that any .base pointers we calculate later are greater
2596 - We set 'pos' to indicate the first entry of the second node.
2598 - We then iterate over the columns of the node, finding the first and
2599 last used entry at l and m. We then copy l..m into pos..(pos+m-l),
2600 and set the .check pointers accordingly, and advance pos
2601 appropriately and repreat for the next node. Note that when we copy
2602 the next pointers we have to convert them from the original
2603 NODEIDX form to NODENUM form as the former is not valid post
2606 - If a node has no transitions used we mark its base as 0 and do not
2607 advance the pos pointer.
2609 - If a node only has one transition we use a second pointer into the
2610 structure to fill in allocated fail transitions from other states.
2611 This pointer is independent of the main pointer and scans forward
2612 looking for null transitions that are allocated to a state. When it
2613 finds one it writes the single transition into the "hole". If the
2614 pointer doesnt find one the single transition is appended as normal.
2616 - Once compressed we can Renew/realloc the structures to release the
2619 See "Table-Compression Methods" in sec 3.9 of the Red Dragon,
2620 specifically Fig 3.47 and the associated pseudocode.
2624 const U32 laststate = TRIE_NODENUM( next_alloc );
2627 trie->statecount = laststate;
2629 for ( state = 1 ; state < laststate ; state++ ) {
2631 const U32 stateidx = TRIE_NODEIDX( state );
2632 const U32 o_used = trie->trans[ stateidx ].check;
2633 U32 used = trie->trans[ stateidx ].check;
2634 trie->trans[ stateidx ].check = 0;
2637 used && charid < trie->uniquecharcount;
2640 if ( flag || trie->trans[ stateidx + charid ].next ) {
2641 if ( trie->trans[ stateidx + charid ].next ) {
2643 for ( ; zp < pos ; zp++ ) {
2644 if ( ! trie->trans[ zp ].next ) {
2648 trie->states[ state ].trans.base
2650 + trie->uniquecharcount
2652 trie->trans[ zp ].next
2653 = SAFE_TRIE_NODENUM( trie->trans[ stateidx
2655 trie->trans[ zp ].check = state;
2656 if ( ++zp > pos ) pos = zp;
2663 trie->states[ state ].trans.base
2664 = pos + trie->uniquecharcount - charid ;
2666 trie->trans[ pos ].next
2667 = SAFE_TRIE_NODENUM(
2668 trie->trans[ stateidx + charid ].next );
2669 trie->trans[ pos ].check = state;
2674 trie->lasttrans = pos + 1;
2675 trie->states = (reg_trie_state *)
2676 PerlMemShared_realloc( trie->states, laststate
2677 * sizeof(reg_trie_state) );
2678 DEBUG_TRIE_COMPILE_MORE_r(
2679 PerlIO_printf( Perl_debug_log,
2680 "%*sAlloc: %d Orig: %"IVdf" elements, Final:%"IVdf". Savings of %%%5.2f\n",
2681 (int)depth * 2 + 2,"",
2682 (int)( ( TRIE_CHARCOUNT(trie) + 1 ) * trie->uniquecharcount
2686 ( ( next_alloc - pos ) * 100 ) / (double)next_alloc );
2689 } /* end table compress */
2691 DEBUG_TRIE_COMPILE_MORE_r(
2692 PerlIO_printf(Perl_debug_log,
2693 "%*sStatecount:%"UVxf" Lasttrans:%"UVxf"\n",
2694 (int)depth * 2 + 2, "",
2695 (UV)trie->statecount,
2696 (UV)trie->lasttrans)
2698 /* resize the trans array to remove unused space */
2699 trie->trans = (reg_trie_trans *)
2700 PerlMemShared_realloc( trie->trans, trie->lasttrans
2701 * sizeof(reg_trie_trans) );
2703 { /* Modify the program and insert the new TRIE node */
2704 U8 nodetype =(U8)(flags & 0xFF);
2708 regnode *optimize = NULL;
2709 #ifdef RE_TRACK_PATTERN_OFFSETS
2712 U32 mjd_nodelen = 0;
2713 #endif /* RE_TRACK_PATTERN_OFFSETS */
2714 #endif /* DEBUGGING */
2716 This means we convert either the first branch or the first Exact,
2717 depending on whether the thing following (in 'last') is a branch
2718 or not and whther first is the startbranch (ie is it a sub part of
2719 the alternation or is it the whole thing.)
2720 Assuming its a sub part we convert the EXACT otherwise we convert
2721 the whole branch sequence, including the first.
2723 /* Find the node we are going to overwrite */
2724 if ( first != startbranch || OP( last ) == BRANCH ) {
2725 /* branch sub-chain */
2726 NEXT_OFF( first ) = (U16)(last - first);
2727 #ifdef RE_TRACK_PATTERN_OFFSETS
2729 mjd_offset= Node_Offset((convert));
2730 mjd_nodelen= Node_Length((convert));
2733 /* whole branch chain */
2735 #ifdef RE_TRACK_PATTERN_OFFSETS
2738 const regnode *nop = NEXTOPER( convert );
2739 mjd_offset= Node_Offset((nop));
2740 mjd_nodelen= Node_Length((nop));
2744 PerlIO_printf(Perl_debug_log,
2745 "%*sMJD offset:%"UVuf" MJD length:%"UVuf"\n",
2746 (int)depth * 2 + 2, "",
2747 (UV)mjd_offset, (UV)mjd_nodelen)
2750 /* But first we check to see if there is a common prefix we can
2751 split out as an EXACT and put in front of the TRIE node. */
2752 trie->startstate= 1;
2753 if ( trie->bitmap && !widecharmap && !trie->jump ) {
2755 for ( state = 1 ; state < trie->statecount-1 ; state++ ) {
2759 const U32 base = trie->states[ state ].trans.base;
2761 if ( trie->states[state].wordnum )
2764 for ( ofs = 0 ; ofs < trie->uniquecharcount ; ofs++ ) {
2765 if ( ( base + ofs >= trie->uniquecharcount ) &&
2766 ( base + ofs - trie->uniquecharcount < trie->lasttrans ) &&
2767 trie->trans[ base + ofs - trie->uniquecharcount ].check == state )
2769 if ( ++count > 1 ) {
2770 SV **tmp = av_fetch( revcharmap, ofs, 0);
2771 const U8 *ch = (U8*)SvPV_nolen_const( *tmp );
2772 if ( state == 1 ) break;
2774 Zero(trie->bitmap, ANYOF_BITMAP_SIZE, char);
2776 PerlIO_printf(Perl_debug_log,
2777 "%*sNew Start State=%"UVuf" Class: [",
2778 (int)depth * 2 + 2, "",
2781 SV ** const tmp = av_fetch( revcharmap, idx, 0);
2782 const U8 * const ch = (U8*)SvPV_nolen_const( *tmp );
2784 TRIE_BITMAP_SET(trie,*ch);
2786 TRIE_BITMAP_SET(trie, folder[ *ch ]);
2788 PerlIO_printf(Perl_debug_log, "%s", (char*)ch)
2792 TRIE_BITMAP_SET(trie,*ch);
2794 TRIE_BITMAP_SET(trie,folder[ *ch ]);
2795 DEBUG_OPTIMISE_r(PerlIO_printf( Perl_debug_log,"%s", ch));
2801 SV **tmp = av_fetch( revcharmap, idx, 0);
2803 char *ch = SvPV( *tmp, len );
2805 SV *sv=sv_newmortal();
2806 PerlIO_printf( Perl_debug_log,
2807 "%*sPrefix State: %"UVuf" Idx:%"UVuf" Char='%s'\n",
2808 (int)depth * 2 + 2, "",
2810 pv_pretty(sv, SvPV_nolen_const(*tmp), SvCUR(*tmp), 6,
2811 PL_colors[0], PL_colors[1],
2812 (SvUTF8(*tmp) ? PERL_PV_ESCAPE_UNI : 0) |
2813 PERL_PV_ESCAPE_FIRSTCHAR
2818 OP( convert ) = nodetype;
2819 str=STRING(convert);
2822 STR_LEN(convert) += len;
2828 DEBUG_OPTIMISE_r(PerlIO_printf( Perl_debug_log,"]\n"));
2833 trie->prefixlen = (state-1);
2835 regnode *n = convert+NODE_SZ_STR(convert);
2836 NEXT_OFF(convert) = NODE_SZ_STR(convert);
2837 trie->startstate = state;
2838 trie->minlen -= (state - 1);
2839 trie->maxlen -= (state - 1);
2841 /* At least the UNICOS C compiler choked on this
2842 * being argument to DEBUG_r(), so let's just have
2845 #ifdef PERL_EXT_RE_BUILD
2851 regnode *fix = convert;
2852 U32 word = trie->wordcount;
2854 Set_Node_Offset_Length(convert, mjd_offset, state - 1);
2855 while( ++fix < n ) {
2856 Set_Node_Offset_Length(fix, 0, 0);
2859 SV ** const tmp = av_fetch( trie_words, word, 0 );
2861 if ( STR_LEN(convert) <= SvCUR(*tmp) )
2862 sv_chop(*tmp, SvPV_nolen(*tmp) + STR_LEN(convert));
2864 sv_chop(*tmp, SvPV_nolen(*tmp) + SvCUR(*tmp));
2872 NEXT_OFF(convert) = (U16)(tail - convert);
2873 DEBUG_r(optimize= n);
2879 if ( trie->maxlen ) {
2880 NEXT_OFF( convert ) = (U16)(tail - convert);
2881 ARG_SET( convert, data_slot );
2882 /* Store the offset to the first unabsorbed branch in
2883 jump[0], which is otherwise unused by the jump logic.
2884 We use this when dumping a trie and during optimisation. */
2886 trie->jump[0] = (U16)(nextbranch - convert);
2888 /* If the start state is not accepting (meaning there is no empty string/NOTHING)
2889 * and there is a bitmap
2890 * and the first "jump target" node we found leaves enough room
2891 * then convert the TRIE node into a TRIEC node, with the bitmap
2892 * embedded inline in the opcode - this is hypothetically faster.
2894 if ( !trie->states[trie->startstate].wordnum
2896 && ( (char *)jumper - (char *)convert) >= (int)sizeof(struct regnode_charclass) )
2898 OP( convert ) = TRIEC;
2899 Copy(trie->bitmap, ((struct regnode_charclass *)convert)->bitmap, ANYOF_BITMAP_SIZE, char);
2900 PerlMemShared_free(trie->bitmap);
2903 OP( convert ) = TRIE;
2905 /* store the type in the flags */
2906 convert->flags = nodetype;
2910 + regarglen[ OP( convert ) ];
2912 /* XXX We really should free up the resource in trie now,
2913 as we won't use them - (which resources?) dmq */
2915 /* needed for dumping*/
2916 DEBUG_r(if (optimize) {
2917 regnode *opt = convert;
2919 while ( ++opt < optimize) {
2920 Set_Node_Offset_Length(opt,0,0);
2923 Try to clean up some of the debris left after the
2926 while( optimize < jumper ) {
2927 mjd_nodelen += Node_Length((optimize));
2928 OP( optimize ) = OPTIMIZED;
2929 Set_Node_Offset_Length(optimize,0,0);
2932 Set_Node_Offset_Length(convert,mjd_offset,mjd_nodelen);
2934 } /* end node insert */
2935 REH_CALL_COMP_NODE_HOOK(pRExC_state->rx, convert);
2937 /* Finish populating the prev field of the wordinfo array. Walk back
2938 * from each accept state until we find another accept state, and if
2939 * so, point the first word's .prev field at the second word. If the
2940 * second already has a .prev field set, stop now. This will be the
2941 * case either if we've already processed that word's accept state,
2942 * or that state had multiple words, and the overspill words were
2943 * already linked up earlier.
2950 for (word=1; word <= trie->wordcount; word++) {
2952 if (trie->wordinfo[word].prev)
2954 state = trie->wordinfo[word].accept;
2956 state = prev_states[state];
2959 prev = trie->states[state].wordnum;
2963 trie->wordinfo[word].prev = prev;
2965 Safefree(prev_states);
2969 /* and now dump out the compressed format */
2970 DEBUG_TRIE_COMPILE_r(dump_trie(trie, widecharmap, revcharmap, depth+1));
2972 RExC_rxi->data->data[ data_slot + 1 ] = (void*)widecharmap;
2974 RExC_rxi->data->data[ data_slot + TRIE_WORDS_OFFSET ] = (void*)trie_words;
2975 RExC_rxi->data->data[ data_slot + 3 ] = (void*)revcharmap;
2977 SvREFCNT_dec_NN(revcharmap);
2981 : trie->startstate>1
2987 S_construct_ahocorasick_from_trie(pTHX_ RExC_state_t *pRExC_state, regnode *source, U32 depth)
2989 /* The Trie is constructed and compressed now so we can build a fail array if
2992 This is basically the Aho-Corasick algorithm. Its from exercise 3.31 and
2994 "Red Dragon" -- Compilers, principles, techniques, and tools. Aho, Sethi,
2998 We find the fail state for each state in the trie, this state is the longest
2999 proper suffix of the current state's 'word' that is also a proper prefix of
3000 another word in our trie. State 1 represents the word '' and is thus the
3001 default fail state. This allows the DFA not to have to restart after its
3002 tried and failed a word at a given point, it simply continues as though it
3003 had been matching the other word in the first place.
3005 'abcdgu'=~/abcdefg|cdgu/
3006 When we get to 'd' we are still matching the first word, we would encounter
3007 'g' which would fail, which would bring us to the state representing 'd' in
3008 the second word where we would try 'g' and succeed, proceeding to match
3011 /* add a fail transition */
3012 const U32 trie_offset = ARG(source);
3013 reg_trie_data *trie=(reg_trie_data *)RExC_rxi->data->data[trie_offset];
3015 const U32 ucharcount = trie->uniquecharcount;
3016 const U32 numstates = trie->statecount;
3017 const U32 ubound = trie->lasttrans + ucharcount;
3021 U32 base = trie->states[ 1 ].trans.base;
3024 const U32 data_slot = add_data( pRExC_state, STR_WITH_LEN("T"));
3026 GET_RE_DEBUG_FLAGS_DECL;
3028 PERL_ARGS_ASSERT_CONSTRUCT_AHOCORASICK_FROM_TRIE;
3029 PERL_UNUSED_CONTEXT;
3031 PERL_UNUSED_ARG(depth);
3034 if ( OP(source) == TRIE ) {
3035 struct regnode_1 *op = (struct regnode_1 *)
3036 PerlMemShared_calloc(1, sizeof(struct regnode_1));
3037 StructCopy(source,op,struct regnode_1);
3038 stclass = (regnode *)op;
3040 struct regnode_charclass *op = (struct regnode_charclass *)
3041 PerlMemShared_calloc(1, sizeof(struct regnode_charclass));
3042 StructCopy(source,op,struct regnode_charclass);
3043 stclass = (regnode *)op;
3045 OP(stclass)+=2; /* covert the TRIE type to its AHO-CORASICK equivalent */
3047 ARG_SET( stclass, data_slot );
3048 aho = (reg_ac_data *) PerlMemShared_calloc( 1, sizeof(reg_ac_data) );
3049 RExC_rxi->data->data[ data_slot ] = (void*)aho;
3050 aho->trie=trie_offset;
3051 aho->states=(reg_trie_state *)PerlMemShared_malloc( numstates * sizeof(reg_trie_state) );
3052 Copy( trie->states, aho->states, numstates, reg_trie_state );
3053 Newxz( q, numstates, U32);
3054 aho->fail = (U32 *) PerlMemShared_calloc( numstates, sizeof(U32) );
3057 /* initialize fail[0..1] to be 1 so that we always have
3058 a valid final fail state */
3059 fail[ 0 ] = fail[ 1 ] = 1;
3061 for ( charid = 0; charid < ucharcount ; charid++ ) {
3062 const U32 newstate = TRIE_TRANS_STATE( 1, base, ucharcount, charid, 0 );
3064 q[ q_write ] = newstate;
3065 /* set to point at the root */
3066 fail[ q[ q_write++ ] ]=1;
3069 while ( q_read < q_write) {
3070 const U32 cur = q[ q_read++ % numstates ];
3071 base = trie->states[ cur ].trans.base;
3073 for ( charid = 0 ; charid < ucharcount ; charid++ ) {
3074 const U32 ch_state = TRIE_TRANS_STATE( cur, base, ucharcount, charid, 1 );
3076 U32 fail_state = cur;
3079 fail_state = fail[ fail_state ];
3080 fail_base = aho->states[ fail_state ].trans.base;
3081 } while ( !TRIE_TRANS_STATE( fail_state, fail_base, ucharcount, charid, 1 ) );
3083 fail_state = TRIE_TRANS_STATE( fail_state, fail_base, ucharcount, charid, 1 );
3084 fail[ ch_state ] = fail_state;
3085 if ( !aho->states[ ch_state ].wordnum && aho->states[ fail_state ].wordnum )
3087 aho->states[ ch_state ].wordnum = aho->states[ fail_state ].wordnum;
3089 q[ q_write++ % numstates] = ch_state;
3093 /* restore fail[0..1] to 0 so that we "fall out" of the AC loop
3094 when we fail in state 1, this allows us to use the
3095 charclass scan to find a valid start char. This is based on the principle
3096 that theres a good chance the string being searched contains lots of stuff
3097 that cant be a start char.
3099 fail[ 0 ] = fail[ 1 ] = 0;
3100 DEBUG_TRIE_COMPILE_r({
3101 PerlIO_printf(Perl_debug_log,
3102 "%*sStclass Failtable (%"UVuf" states): 0",
3103 (int)(depth * 2), "", (UV)numstates
3105 for( q_read=1; q_read<numstates; q_read++ ) {
3106 PerlIO_printf(Perl_debug_log, ", %"UVuf, (UV)fail[q_read]);
3108 PerlIO_printf(Perl_debug_log, "\n");
3111 /*RExC_seen |= REG_TRIEDFA_SEEN;*/
3116 #define DEBUG_PEEP(str,scan,depth) \
3117 DEBUG_OPTIMISE_r({if (scan){ \
3118 SV * const mysv=sv_newmortal(); \
3119 regnode *Next = regnext(scan); \
3120 regprop(RExC_rx, mysv, scan, NULL); \
3121 PerlIO_printf(Perl_debug_log, "%*s" str ">%3d: %s (%d)\n", \
3122 (int)depth*2, "", REG_NODE_NUM(scan), SvPV_nolen_const(mysv),\
3123 Next ? (REG_NODE_NUM(Next)) : 0 ); \
3127 /* The below joins as many adjacent EXACTish nodes as possible into a single
3128 * one. The regop may be changed if the node(s) contain certain sequences that
3129 * require special handling. The joining is only done if:
3130 * 1) there is room in the current conglomerated node to entirely contain the
3132 * 2) they are the exact same node type
3134 * The adjacent nodes actually may be separated by NOTHING-kind nodes, and
3135 * these get optimized out
3137 * If a node is to match under /i (folded), the number of characters it matches
3138 * can be different than its character length if it contains a multi-character
3139 * fold. *min_subtract is set to the total delta number of characters of the
3142 * And *unfolded_multi_char is set to indicate whether or not the node contains
3143 * an unfolded multi-char fold. This happens when whether the fold is valid or
3144 * not won't be known until runtime; namely for EXACTF nodes that contain LATIN
3145 * SMALL LETTER SHARP S, as only if the target string being matched against
3146 * turns out to be UTF-8 is that fold valid; and also for EXACTFL nodes whose
3147 * folding rules depend on the locale in force at runtime. (Multi-char folds
3148 * whose components are all above the Latin1 range are not run-time locale
3149 * dependent, and have already been folded by the time this function is
3152 * This is as good a place as any to discuss the design of handling these
3153 * multi-character fold sequences. It's been wrong in Perl for a very long
3154 * time. There are three code points in Unicode whose multi-character folds
3155 * were long ago discovered to mess things up. The previous designs for
3156 * dealing with these involved assigning a special node for them. This
3157 * approach doesn't always work, as evidenced by this example:
3158 * "\xDFs" =~ /s\xDF/ui # Used to fail before these patches
3159 * Both sides fold to "sss", but if the pattern is parsed to create a node that
3160 * would match just the \xDF, it won't be able to handle the case where a
3161 * successful match would have to cross the node's boundary. The new approach
3162 * that hopefully generally solves the problem generates an EXACTFU_SS node
3163 * that is "sss" in this case.
3165 * It turns out that there are problems with all multi-character folds, and not
3166 * just these three. Now the code is general, for all such cases. The
3167 * approach taken is:
3168 * 1) This routine examines each EXACTFish node that could contain multi-
3169 * character folded sequences. Since a single character can fold into
3170 * such a sequence, the minimum match length for this node is less than
3171 * the number of characters in the node. This routine returns in
3172 * *min_subtract how many characters to subtract from the the actual
3173 * length of the string to get a real minimum match length; it is 0 if
3174 * there are no multi-char foldeds. This delta is used by the caller to
3175 * adjust the min length of the match, and the delta between min and max,
3176 * so that the optimizer doesn't reject these possibilities based on size
3178 * 2) For the sequence involving the Sharp s (\xDF), the node type EXACTFU_SS
3179 * is used for an EXACTFU node that contains at least one "ss" sequence in
3180 * it. For non-UTF-8 patterns and strings, this is the only case where
3181 * there is a possible fold length change. That means that a regular
3182 * EXACTFU node without UTF-8 involvement doesn't have to concern itself
3183 * with length changes, and so can be processed faster. regexec.c takes
3184 * advantage of this. Generally, an EXACTFish node that is in UTF-8 is
3185 * pre-folded by regcomp.c (except EXACTFL, some of whose folds aren't
3186 * known until runtime). This saves effort in regex matching. However,
3187 * the pre-folding isn't done for non-UTF8 patterns because the fold of
3188 * the MICRO SIGN requires UTF-8, and we don't want to slow things down by
3189 * forcing the pattern into UTF8 unless necessary. Also what EXACTF (and,
3190 * again, EXACTFL) nodes fold to isn't known until runtime. The fold
3191 * possibilities for the non-UTF8 patterns are quite simple, except for
3192 * the sharp s. All the ones that don't involve a UTF-8 target string are
3193 * members of a fold-pair, and arrays are set up for all of them so that
3194 * the other member of the pair can be found quickly. Code elsewhere in
3195 * this file makes sure that in EXACTFU nodes, the sharp s gets folded to
3196 * 'ss', even if the pattern isn't UTF-8. This avoids the issues
3197 * described in the next item.
3198 * 3) A problem remains for unfolded multi-char folds. (These occur when the
3199 * validity of the fold won't be known until runtime, and so must remain
3200 * unfolded for now. This happens for the sharp s in EXACTF and EXACTFA
3201 * nodes when the pattern isn't in UTF-8. (Note, BTW, that there cannot
3202 * be an EXACTF node with a UTF-8 pattern.) They also occur for various
3203 * folds in EXACTFL nodes, regardless of the UTF-ness of the pattern.)
3204 * The reason this is a problem is that the optimizer part of regexec.c
3205 * (probably unwittingly, in Perl_regexec_flags()) makes an assumption
3206 * that a character in the pattern corresponds to at most a single
3207 * character in the target string. (And I do mean character, and not byte
3208 * here, unlike other parts of the documentation that have never been
3209 * updated to account for multibyte Unicode.) sharp s in EXACTF and
3210 * EXACTFL nodes can match the two character string 'ss'; in EXACTFA nodes
3211 * it can match "\x{17F}\x{17F}". These, along with other ones in EXACTFL
3212 * nodes, violate the assumption, and they are the only instances where it
3213 * is violated. I'm reluctant to try to change the assumption, as the
3214 * code involved is impenetrable to me (khw), so instead the code here
3215 * punts. This routine examines EXACTFL nodes, and (when the pattern
3216 * isn't UTF-8) EXACTF and EXACTFA for such unfolded folds, and returns a
3217 * boolean indicating whether or not the node contains such a fold. When
3218 * it is true, the caller sets a flag that later causes the optimizer in
3219 * this file to not set values for the floating and fixed string lengths,
3220 * and thus avoids the optimizer code in regexec.c that makes the invalid
3221 * assumption. Thus, there is no optimization based on string lengths for
3222 * EXACTFL nodes that contain these few folds, nor for non-UTF8-pattern
3223 * EXACTF and EXACTFA nodes that contain the sharp s. (The reason the
3224 * assumption is wrong only in these cases is that all other non-UTF-8
3225 * folds are 1-1; and, for UTF-8 patterns, we pre-fold all other folds to
3226 * their expanded versions. (Again, we can't prefold sharp s to 'ss' in
3227 * EXACTF nodes because we don't know at compile time if it actually
3228 * matches 'ss' or not. For EXACTF nodes it will match iff the target
3229 * string is in UTF-8. This is in contrast to EXACTFU nodes, where it
3230 * always matches; and EXACTFA where it never does. In an EXACTFA node in
3231 * a UTF-8 pattern, sharp s is folded to "\x{17F}\x{17F}, avoiding the
3232 * problem; but in a non-UTF8 pattern, folding it to that above-Latin1
3233 * string would require the pattern to be forced into UTF-8, the overhead
3234 * of which we want to avoid. Similarly the unfolded multi-char folds in
3235 * EXACTFL nodes will match iff the locale at the time of match is a UTF-8
3238 * Similarly, the code that generates tries doesn't currently handle
3239 * not-already-folded multi-char folds, and it looks like a pain to change
3240 * that. Therefore, trie generation of EXACTFA nodes with the sharp s
3241 * doesn't work. Instead, such an EXACTFA is turned into a new regnode,
3242 * EXACTFA_NO_TRIE, which the trie code knows not to handle. Most people
3243 * using /iaa matching will be doing so almost entirely with ASCII
3244 * strings, so this should rarely be encountered in practice */
3246 #define JOIN_EXACT(scan,min_subtract,unfolded_multi_char, flags) \
3247 if (PL_regkind[OP(scan)] == EXACT) \
3248 join_exact(pRExC_state,(scan),(min_subtract),unfolded_multi_char, (flags),NULL,depth+1)
3251 S_join_exact(pTHX_ RExC_state_t *pRExC_state, regnode *scan,
3252 UV *min_subtract, bool *unfolded_multi_char,
3253 U32 flags,regnode *val, U32 depth)
3255 /* Merge several consecutive EXACTish nodes into one. */
3256 regnode *n = regnext(scan);
3258 regnode *next = scan + NODE_SZ_STR(scan);
3262 regnode *stop = scan;
3263 GET_RE_DEBUG_FLAGS_DECL;
3265 PERL_UNUSED_ARG(depth);
3268 PERL_ARGS_ASSERT_JOIN_EXACT;
3269 #ifndef EXPERIMENTAL_INPLACESCAN
3270 PERL_UNUSED_ARG(flags);
3271 PERL_UNUSED_ARG(val);
3273 DEBUG_PEEP("join",scan,depth);
3275 /* Look through the subsequent nodes in the chain. Skip NOTHING, merge
3276 * EXACT ones that are mergeable to the current one. */
3278 && (PL_regkind[OP(n)] == NOTHING
3279 || (stringok && OP(n) == OP(scan)))
3281 && NEXT_OFF(scan) + NEXT_OFF(n) < I16_MAX)
3284 if (OP(n) == TAIL || n > next)
3286 if (PL_regkind[OP(n)] == NOTHING) {
3287 DEBUG_PEEP("skip:",n,depth);
3288 NEXT_OFF(scan) += NEXT_OFF(n);
3289 next = n + NODE_STEP_REGNODE;
3296 else if (stringok) {
3297 const unsigned int oldl = STR_LEN(scan);
3298 regnode * const nnext = regnext(n);
3300 /* XXX I (khw) kind of doubt that this works on platforms (should
3301 * Perl ever run on one) where U8_MAX is above 255 because of lots
3302 * of other assumptions */
3303 /* Don't join if the sum can't fit into a single node */
3304 if (oldl + STR_LEN(n) > U8_MAX)
3307 DEBUG_PEEP("merg",n,depth);
3310 NEXT_OFF(scan) += NEXT_OFF(n);
3311 STR_LEN(scan) += STR_LEN(n);
3312 next = n + NODE_SZ_STR(n);
3313 /* Now we can overwrite *n : */
3314 Move(STRING(n), STRING(scan) + oldl, STR_LEN(n), char);
3322 #ifdef EXPERIMENTAL_INPLACESCAN
3323 if (flags && !NEXT_OFF(n)) {
3324 DEBUG_PEEP("atch", val, depth);
3325 if (reg_off_by_arg[OP(n)]) {
3326 ARG_SET(n, val - n);
3329 NEXT_OFF(n) = val - n;
3337 *unfolded_multi_char = FALSE;
3339 /* Here, all the adjacent mergeable EXACTish nodes have been merged. We
3340 * can now analyze for sequences of problematic code points. (Prior to
3341 * this final joining, sequences could have been split over boundaries, and
3342 * hence missed). The sequences only happen in folding, hence for any
3343 * non-EXACT EXACTish node */
3344 if (OP(scan) != EXACT) {
3345 U8* s0 = (U8*) STRING(scan);
3347 U8* s_end = s0 + STR_LEN(scan);
3349 int total_count_delta = 0; /* Total delta number of characters that
3350 multi-char folds expand to */
3352 /* One pass is made over the node's string looking for all the
3353 * possibilities. To avoid some tests in the loop, there are two main
3354 * cases, for UTF-8 patterns (which can't have EXACTF nodes) and
3359 if (OP(scan) == EXACTFL) {
3362 /* An EXACTFL node would already have been changed to another
3363 * node type unless there is at least one character in it that
3364 * is problematic; likely a character whose fold definition
3365 * won't be known until runtime, and so has yet to be folded.
3366 * For all but the UTF-8 locale, folds are 1-1 in length, but
3367 * to handle the UTF-8 case, we need to create a temporary
3368 * folded copy using UTF-8 locale rules in order to analyze it.
3369 * This is because our macros that look to see if a sequence is
3370 * a multi-char fold assume everything is folded (otherwise the
3371 * tests in those macros would be too complicated and slow).
3372 * Note that here, the non-problematic folds will have already
3373 * been done, so we can just copy such characters. We actually
3374 * don't completely fold the EXACTFL string. We skip the
3375 * unfolded multi-char folds, as that would just create work
3376 * below to figure out the size they already are */
3378 Newx(folded, UTF8_MAX_FOLD_CHAR_EXPAND * STR_LEN(scan) + 1, U8);
3381 STRLEN s_len = UTF8SKIP(s);
3382 if (! is_PROBLEMATIC_LOCALE_FOLD_utf8(s)) {
3383 Copy(s, d, s_len, U8);
3386 else if (is_FOLDS_TO_MULTI_utf8(s)) {
3387 *unfolded_multi_char = TRUE;
3388 Copy(s, d, s_len, U8);
3391 else if (isASCII(*s)) {
3392 *(d++) = toFOLD(*s);
3396 _to_utf8_fold_flags(s, d, &len, FOLD_FLAGS_FULL);
3402 /* Point the remainder of the routine to look at our temporary
3406 } /* End of creating folded copy of EXACTFL string */
3408 /* Examine the string for a multi-character fold sequence. UTF-8
3409 * patterns have all characters pre-folded by the time this code is
3411 while (s < s_end - 1) /* Can stop 1 before the end, as minimum
3412 length sequence we are looking for is 2 */
3414 int count = 0; /* How many characters in a multi-char fold */
3415 int len = is_MULTI_CHAR_FOLD_utf8_safe(s, s_end);
3416 if (! len) { /* Not a multi-char fold: get next char */
3421 /* Nodes with 'ss' require special handling, except for
3422 * EXACTFA-ish for which there is no multi-char fold to this */
3423 if (len == 2 && *s == 's' && *(s+1) == 's'
3424 && OP(scan) != EXACTFA
3425 && OP(scan) != EXACTFA_NO_TRIE)
3428 if (OP(scan) != EXACTFL) {
3429 OP(scan) = EXACTFU_SS;
3433 else { /* Here is a generic multi-char fold. */
3434 U8* multi_end = s + len;
3436 /* Count how many characters are in it. In the case of
3437 * /aa, no folds which contain ASCII code points are
3438 * allowed, so check for those, and skip if found. */
3439 if (OP(scan) != EXACTFA && OP(scan) != EXACTFA_NO_TRIE) {
3440 count = utf8_length(s, multi_end);
3444 while (s < multi_end) {
3447 goto next_iteration;
3457 /* The delta is how long the sequence is minus 1 (1 is how long
3458 * the character that folds to the sequence is) */
3459 total_count_delta += count - 1;
3463 /* We created a temporary folded copy of the string in EXACTFL
3464 * nodes. Therefore we need to be sure it doesn't go below zero,
3465 * as the real string could be shorter */
3466 if (OP(scan) == EXACTFL) {
3467 int total_chars = utf8_length((U8*) STRING(scan),
3468 (U8*) STRING(scan) + STR_LEN(scan));
3469 if (total_count_delta > total_chars) {
3470 total_count_delta = total_chars;
3474 *min_subtract += total_count_delta;
3477 else if (OP(scan) == EXACTFA) {
3479 /* Non-UTF-8 pattern, EXACTFA node. There can't be a multi-char
3480 * fold to the ASCII range (and there are no existing ones in the
3481 * upper latin1 range). But, as outlined in the comments preceding
3482 * this function, we need to flag any occurrences of the sharp s.
3483 * This character forbids trie formation (because of added
3486 if (*s == LATIN_SMALL_LETTER_SHARP_S) {
3487 OP(scan) = EXACTFA_NO_TRIE;
3488 *unfolded_multi_char = TRUE;
3497 /* Non-UTF-8 pattern, not EXACTFA node. Look for the multi-char
3498 * folds that are all Latin1. As explained in the comments
3499 * preceding this function, we look also for the sharp s in EXACTF
3500 * and EXACTFL nodes; it can be in the final position. Otherwise
3501 * we can stop looking 1 byte earlier because have to find at least
3502 * two characters for a multi-fold */
3503 const U8* upper = (OP(scan) == EXACTF || OP(scan) == EXACTFL)
3508 int len = is_MULTI_CHAR_FOLD_latin1_safe(s, s_end);
3509 if (! len) { /* Not a multi-char fold. */
3510 if (*s == LATIN_SMALL_LETTER_SHARP_S
3511 && (OP(scan) == EXACTF || OP(scan) == EXACTFL))
3513 *unfolded_multi_char = TRUE;
3520 && isARG2_lower_or_UPPER_ARG1('s', *s)
3521 && isARG2_lower_or_UPPER_ARG1('s', *(s+1)))
3524 /* EXACTF nodes need to know that the minimum length
3525 * changed so that a sharp s in the string can match this
3526 * ss in the pattern, but they remain EXACTF nodes, as they
3527 * won't match this unless the target string is is UTF-8,
3528 * which we don't know until runtime. EXACTFL nodes can't
3529 * transform into EXACTFU nodes */
3530 if (OP(scan) != EXACTF && OP(scan) != EXACTFL) {
3531 OP(scan) = EXACTFU_SS;
3535 *min_subtract += len - 1;
3542 /* Allow dumping but overwriting the collection of skipped
3543 * ops and/or strings with fake optimized ops */
3544 n = scan + NODE_SZ_STR(scan);
3552 DEBUG_OPTIMISE_r(if (merged){DEBUG_PEEP("finl",scan,depth)});
3556 /* REx optimizer. Converts nodes into quicker variants "in place".
3557 Finds fixed substrings. */
3559 /* Stops at toplevel WHILEM as well as at "last". At end *scanp is set
3560 to the position after last scanned or to NULL. */
3562 #define INIT_AND_WITHP \
3563 assert(!and_withp); \
3564 Newx(and_withp,1, regnode_ssc); \
3565 SAVEFREEPV(and_withp)
3567 /* this is a chain of data about sub patterns we are processing that
3568 need to be handled separately/specially in study_chunk. Its so
3569 we can simulate recursion without losing state. */
3571 typedef struct scan_frame {
3572 regnode *last; /* last node to process in this frame */
3573 regnode *next; /* next node to process when last is reached */
3574 struct scan_frame *prev; /*previous frame*/
3575 U32 prev_recursed_depth;
3576 I32 stop; /* what stopparen do we use */
3581 S_study_chunk(pTHX_ RExC_state_t *pRExC_state, regnode **scanp,
3582 SSize_t *minlenp, SSize_t *deltap,
3587 regnode_ssc *and_withp,
3588 U32 flags, U32 depth)
3589 /* scanp: Start here (read-write). */
3590 /* deltap: Write maxlen-minlen here. */
3591 /* last: Stop before this one. */
3592 /* data: string data about the pattern */
3593 /* stopparen: treat close N as END */
3594 /* recursed: which subroutines have we recursed into */
3595 /* and_withp: Valid if flags & SCF_DO_STCLASS_OR */
3597 /* There must be at least this number of characters to match */
3600 regnode *scan = *scanp, *next;
3602 int is_inf = (flags & SCF_DO_SUBSTR) && (data->flags & SF_IS_INF);
3603 int is_inf_internal = 0; /* The studied chunk is infinite */
3604 I32 is_par = OP(scan) == OPEN ? ARG(scan) : 0;
3605 scan_data_t data_fake;
3606 SV *re_trie_maxbuff = NULL;
3607 regnode *first_non_open = scan;
3608 SSize_t stopmin = SSize_t_MAX;
3609 scan_frame *frame = NULL;
3610 GET_RE_DEBUG_FLAGS_DECL;
3612 PERL_ARGS_ASSERT_STUDY_CHUNK;
3615 StructCopy(&zero_scan_data, &data_fake, scan_data_t);
3618 while (first_non_open && OP(first_non_open) == OPEN)
3619 first_non_open=regnext(first_non_open);
3624 while ( scan && OP(scan) != END && scan < last ){
3625 UV min_subtract = 0; /* How mmany chars to subtract from the minimum
3626 node length to get a real minimum (because
3627 the folded version may be shorter) */
3628 bool unfolded_multi_char = FALSE;
3629 /* Peephole optimizer: */
3630 DEBUG_OPTIMISE_MORE_r(
3632 PerlIO_printf(Perl_debug_log,
3633 "%*sstudy_chunk stopparen=%ld depth=%lu recursed_depth=%lu ",
3634 ((int) depth*2), "", (long)stopparen,
3635 (unsigned long)depth, (unsigned long)recursed_depth);
3636 if (recursed_depth) {
3639 for ( j = 0 ; j < recursed_depth ; j++ ) {
3640 PerlIO_printf(Perl_debug_log,"[");
3641 for ( i = 0 ; i < (U32)RExC_npar ; i++ )
3642 PerlIO_printf(Perl_debug_log,"%d",
3643 PAREN_TEST(RExC_study_chunk_recursed +
3644 (j * RExC_study_chunk_recursed_bytes), i)
3647 PerlIO_printf(Perl_debug_log,"]");
3650 PerlIO_printf(Perl_debug_log,"\n");
3653 DEBUG_STUDYDATA("Peep:", data, depth);
3654 DEBUG_PEEP("Peep", scan, depth);
3657 /* The reason we do this here we need to deal with things like /(?:f)(?:o)(?:o)/
3658 * which cant be dealt with by the normal EXACT parsing code, as each (?:..) is handled
3659 * by a different invocation of reg() -- Yves
3661 JOIN_EXACT(scan,&min_subtract, &unfolded_multi_char, 0);
3663 /* Follow the next-chain of the current node and optimize
3664 away all the NOTHINGs from it. */
3665 if (OP(scan) != CURLYX) {
3666 const int max = (reg_off_by_arg[OP(scan)]
3668 /* I32 may be smaller than U16 on CRAYs! */
3669 : (I32_MAX < U16_MAX ? I32_MAX : U16_MAX));
3670 int off = (reg_off_by_arg[OP(scan)] ? ARG(scan) : NEXT_OFF(scan));
3674 /* Skip NOTHING and LONGJMP. */
3675 while ((n = regnext(n))
3676 && ((PL_regkind[OP(n)] == NOTHING && (noff = NEXT_OFF(n)))
3677 || ((OP(n) == LONGJMP) && (noff = ARG(n))))
3678 && off + noff < max)
3680 if (reg_off_by_arg[OP(scan)])
3683 NEXT_OFF(scan) = off;
3688 /* The principal pseudo-switch. Cannot be a switch, since we
3689 look into several different things. */
3690 if (OP(scan) == BRANCH || OP(scan) == BRANCHJ
3691 || OP(scan) == IFTHEN) {
3692 next = regnext(scan);
3694 /* demq: the op(next)==code check is to see if we have
3695 * "branch-branch" AFAICT */
3697 if (OP(next) == code || code == IFTHEN) {
3698 /* NOTE - There is similar code to this block below for
3699 * handling TRIE nodes on a re-study. If you change stuff here
3700 * check there too. */
3701 SSize_t max1 = 0, min1 = SSize_t_MAX, num = 0;
3703 regnode * const startbranch=scan;
3705 if (flags & SCF_DO_SUBSTR) {
3706 /* Cannot merge strings after this. */
3707 scan_commit(pRExC_state, data, minlenp, is_inf);
3710 if (flags & SCF_DO_STCLASS)
3711 ssc_init_zero(pRExC_state, &accum);
3713 while (OP(scan) == code) {
3714 SSize_t deltanext, minnext, fake;
3716 regnode_ssc this_class;
3719 data_fake.flags = 0;
3721 data_fake.whilem_c = data->whilem_c;
3722 data_fake.last_closep = data->last_closep;
3725 data_fake.last_closep = &fake;
3727 data_fake.pos_delta = delta;
3728 next = regnext(scan);
3729 scan = NEXTOPER(scan);
3731 scan = NEXTOPER(scan);
3732 if (flags & SCF_DO_STCLASS) {
3733 ssc_init(pRExC_state, &this_class);
3734 data_fake.start_class = &this_class;
3735 f = SCF_DO_STCLASS_AND;
3737 if (flags & SCF_WHILEM_VISITED_POS)
3738 f |= SCF_WHILEM_VISITED_POS;
3740 /* we suppose the run is continuous, last=next...*/
3741 minnext = study_chunk(pRExC_state, &scan, minlenp,
3742 &deltanext, next, &data_fake, stopparen,
3743 recursed_depth, NULL, f,depth+1);
3746 if (deltanext == SSize_t_MAX) {
3747 is_inf = is_inf_internal = 1;
3749 } else if (max1 < minnext + deltanext)
3750 max1 = minnext + deltanext;
3752 if (data_fake.flags & (SF_HAS_PAR|SF_IN_PAR))
3754 if (data_fake.flags & SCF_SEEN_ACCEPT) {
3755 if ( stopmin > minnext)
3756 stopmin = min + min1;
3757 flags &= ~SCF_DO_SUBSTR;
3759 data->flags |= SCF_SEEN_ACCEPT;
3762 if (data_fake.flags & SF_HAS_EVAL)
3763 data->flags |= SF_HAS_EVAL;
3764 data->whilem_c = data_fake.whilem_c;
3766 if (flags & SCF_DO_STCLASS)
3767 ssc_or(pRExC_state, &accum, (regnode_charclass*)&this_class);
3769 if (code == IFTHEN && num < 2) /* Empty ELSE branch */
3771 if (flags & SCF_DO_SUBSTR) {
3772 data->pos_min += min1;
3773 if (data->pos_delta >= SSize_t_MAX - (max1 - min1))
3774 data->pos_delta = SSize_t_MAX;
3776 data->pos_delta += max1 - min1;
3777 if (max1 != min1 || is_inf)
3778 data->longest = &(data->longest_float);
3781 if (delta == SSize_t_MAX
3782 || SSize_t_MAX - delta - (max1 - min1) < 0)
3783 delta = SSize_t_MAX;
3785 delta += max1 - min1;
3786 if (flags & SCF_DO_STCLASS_OR) {
3787 ssc_or(pRExC_state, data->start_class, (regnode_charclass*) &accum);
3789 ssc_and(pRExC_state, data->start_class, (regnode_charclass *) and_withp);
3790 flags &= ~SCF_DO_STCLASS;
3793 else if (flags & SCF_DO_STCLASS_AND) {
3795 ssc_and(pRExC_state, data->start_class, (regnode_charclass *) &accum);
3796 flags &= ~SCF_DO_STCLASS;
3799 /* Switch to OR mode: cache the old value of
3800 * data->start_class */
3802 StructCopy(data->start_class, and_withp, regnode_ssc);
3803 flags &= ~SCF_DO_STCLASS_AND;
3804 StructCopy(&accum, data->start_class, regnode_ssc);
3805 flags |= SCF_DO_STCLASS_OR;
3809 if (PERL_ENABLE_TRIE_OPTIMISATION &&
3810 OP( startbranch ) == BRANCH )
3814 Assuming this was/is a branch we are dealing with: 'scan'
3815 now points at the item that follows the branch sequence,
3816 whatever it is. We now start at the beginning of the
3817 sequence and look for subsequences of
3823 which would be constructed from a pattern like
3826 If we can find such a subsequence we need to turn the first
3827 element into a trie and then add the subsequent branch exact
3828 strings to the trie.
3832 1. patterns where the whole set of branches can be
3835 2. patterns where only a subset can be converted.
3837 In case 1 we can replace the whole set with a single regop
3838 for the trie. In case 2 we need to keep the start and end
3841 'BRANCH EXACT; BRANCH EXACT; BRANCH X'
3842 becomes BRANCH TRIE; BRANCH X;
3844 There is an additional case, that being where there is a
3845 common prefix, which gets split out into an EXACT like node
3846 preceding the TRIE node.
3848 If x(1..n)==tail then we can do a simple trie, if not we make
3849 a "jump" trie, such that when we match the appropriate word
3850 we "jump" to the appropriate tail node. Essentially we turn
3851 a nested if into a case structure of sorts.
3856 if (!re_trie_maxbuff) {
3857 re_trie_maxbuff = get_sv(RE_TRIE_MAXBUF_NAME, 1);
3858 if (!SvIOK(re_trie_maxbuff))
3859 sv_setiv(re_trie_maxbuff, RE_TRIE_MAXBUF_INIT);
3861 if ( SvIV(re_trie_maxbuff)>=0 ) {
3863 regnode *first = (regnode *)NULL;
3864 regnode *last = (regnode *)NULL;
3865 regnode *tail = scan;
3870 SV * const mysv = sv_newmortal(); /* for dumping */
3872 /* var tail is used because there may be a TAIL
3873 regop in the way. Ie, the exacts will point to the
3874 thing following the TAIL, but the last branch will
3875 point at the TAIL. So we advance tail. If we
3876 have nested (?:) we may have to move through several
3880 while ( OP( tail ) == TAIL ) {
3881 /* this is the TAIL generated by (?:) */
3882 tail = regnext( tail );
3886 DEBUG_TRIE_COMPILE_r({
3887 regprop(RExC_rx, mysv, tail, NULL);
3888 PerlIO_printf( Perl_debug_log, "%*s%s%s\n",
3889 (int)depth * 2 + 2, "",
3890 "Looking for TRIE'able sequences. Tail node is: ",
3891 SvPV_nolen_const( mysv )
3897 Step through the branches
3898 cur represents each branch,
3899 noper is the first thing to be matched as part
3901 noper_next is the regnext() of that node.
3903 We normally handle a case like this
3904 /FOO[xyz]|BAR[pqr]/ via a "jump trie" but we also
3905 support building with NOJUMPTRIE, which restricts
3906 the trie logic to structures like /FOO|BAR/.
3908 If noper is a trieable nodetype then the branch is
3909 a possible optimization target. If we are building
3910 under NOJUMPTRIE then we require that noper_next is
3911 the same as scan (our current position in the regex
3914 Once we have two or more consecutive such branches
3915 we can create a trie of the EXACT's contents and
3916 stitch it in place into the program.
3918 If the sequence represents all of the branches in
3919 the alternation we replace the entire thing with a
3922 Otherwise when it is a subsequence we need to
3923 stitch it in place and replace only the relevant
3924 branches. This means the first branch has to remain
3925 as it is used by the alternation logic, and its
3926 next pointer, and needs to be repointed at the item
3927 on the branch chain following the last branch we
3928 have optimized away.
3930 This could be either a BRANCH, in which case the
3931 subsequence is internal, or it could be the item
3932 following the branch sequence in which case the
3933 subsequence is at the end (which does not
3934 necessarily mean the first node is the start of the
3937 TRIE_TYPE(X) is a define which maps the optype to a
3941 ----------------+-----------
3945 EXACTFU_SS | EXACTFU
3950 #define TRIE_TYPE(X) ( ( NOTHING == (X) ) ? NOTHING : \
3951 ( EXACT == (X) ) ? EXACT : \
3952 ( EXACTFU == (X) || EXACTFU_SS == (X) ) ? EXACTFU : \
3953 ( EXACTFA == (X) ) ? EXACTFA : \
3956 /* dont use tail as the end marker for this traverse */
3957 for ( cur = startbranch ; cur != scan ; cur = regnext( cur ) ) {
3958 regnode * const noper = NEXTOPER( cur );
3959 U8 noper_type = OP( noper );
3960 U8 noper_trietype = TRIE_TYPE( noper_type );
3961 #if defined(DEBUGGING) || defined(NOJUMPTRIE)
3962 regnode * const noper_next = regnext( noper );
3963 U8 noper_next_type = (noper_next && noper_next != tail) ? OP(noper_next) : 0;
3964 U8 noper_next_trietype = (noper_next && noper_next != tail) ? TRIE_TYPE( noper_next_type ) :0;
3967 DEBUG_TRIE_COMPILE_r({
3968 regprop(RExC_rx, mysv, cur, NULL);
3969 PerlIO_printf( Perl_debug_log, "%*s- %s (%d)",
3970 (int)depth * 2 + 2,"", SvPV_nolen_const( mysv ), REG_NODE_NUM(cur) );
3972 regprop(RExC_rx, mysv, noper, NULL);
3973 PerlIO_printf( Perl_debug_log, " -> %s",
3974 SvPV_nolen_const(mysv));
3977 regprop(RExC_rx, mysv, noper_next, NULL);
3978 PerlIO_printf( Perl_debug_log,"\t=> %s\t",
3979 SvPV_nolen_const(mysv));
3981 PerlIO_printf( Perl_debug_log, "(First==%d,Last==%d,Cur==%d,tt==%s,nt==%s,nnt==%s)\n",
3982 REG_NODE_NUM(first), REG_NODE_NUM(last), REG_NODE_NUM(cur),
3983 PL_reg_name[trietype], PL_reg_name[noper_trietype], PL_reg_name[noper_next_trietype]
3987 /* Is noper a trieable nodetype that can be merged
3988 * with the current trie (if there is one)? */
3992 ( noper_trietype == NOTHING)
3993 || ( trietype == NOTHING )
3994 || ( trietype == noper_trietype )
3997 && noper_next == tail
4001 /* Handle mergable triable node Either we are
4002 * the first node in a new trieable sequence,
4003 * in which case we do some bookkeeping,
4004 * otherwise we update the end pointer. */
4007 if ( noper_trietype == NOTHING ) {
4008 #if !defined(DEBUGGING) && !defined(NOJUMPTRIE)
4009 regnode * const noper_next = regnext( noper );
4010 U8 noper_next_type = (noper_next && noper_next!=tail) ? OP(noper_next) : 0;
4011 U8 noper_next_trietype = noper_next_type ? TRIE_TYPE( noper_next_type ) :0;
4014 if ( noper_next_trietype ) {
4015 trietype = noper_next_trietype;
4016 } else if (noper_next_type) {
4017 /* a NOTHING regop is 1 regop wide.
4018 * We need at least two for a trie
4019 * so we can't merge this in */
4023 trietype = noper_trietype;
4026 if ( trietype == NOTHING )
4027 trietype = noper_trietype;
4032 } /* end handle mergable triable node */
4034 /* handle unmergable node -
4035 * noper may either be a triable node which can
4036 * not be tried together with the current trie,
4037 * or a non triable node */
4039 /* If last is set and trietype is not
4040 * NOTHING then we have found at least two
4041 * triable branch sequences in a row of a
4042 * similar trietype so we can turn them
4043 * into a trie. If/when we allow NOTHING to
4044 * start a trie sequence this condition
4045 * will be required, and it isn't expensive
4046 * so we leave it in for now. */
4047 if ( trietype && trietype != NOTHING )
4048 make_trie( pRExC_state,
4049 startbranch, first, cur, tail,
4050 count, trietype, depth+1 );
4051 last = NULL; /* note: we clear/update
4052 first, trietype etc below,
4053 so we dont do it here */
4057 && noper_next == tail
4060 /* noper is triable, so we can start a new
4064 trietype = noper_trietype;
4066 /* if we already saw a first but the
4067 * current node is not triable then we have
4068 * to reset the first information. */
4073 } /* end handle unmergable node */
4074 } /* loop over branches */
4075 DEBUG_TRIE_COMPILE_r({
4076 regprop(RExC_rx, mysv, cur, NULL);
4077 PerlIO_printf( Perl_debug_log,
4078 "%*s- %s (%d) <SCAN FINISHED>\n",
4080 "", SvPV_nolen_const( mysv ),REG_NODE_NUM(cur));
4083 if ( last && trietype ) {
4084 if ( trietype != NOTHING ) {
4085 /* the last branch of the sequence was part of
4086 * a trie, so we have to construct it here
4087 * outside of the loop */
4088 made= make_trie( pRExC_state, startbranch,
4089 first, scan, tail, count,
4090 trietype, depth+1 );
4091 #ifdef TRIE_STUDY_OPT
4092 if ( ((made == MADE_EXACT_TRIE &&
4093 startbranch == first)
4094 || ( first_non_open == first )) &&
4096 flags |= SCF_TRIE_RESTUDY;
4097 if ( startbranch == first
4100 RExC_seen &=~REG_TOP_LEVEL_BRANCHES_SEEN;
4105 /* at this point we know whatever we have is a
4106 * NOTHING sequence/branch AND if 'startbranch'
4107 * is 'first' then we can turn the whole thing
4110 if ( startbranch == first ) {
4112 /* the entire thing is a NOTHING sequence,
4113 * something like this: (?:|) So we can
4114 * turn it into a plain NOTHING op. */
4115 DEBUG_TRIE_COMPILE_r({
4116 regprop(RExC_rx, mysv, cur, NULL);
4117 PerlIO_printf( Perl_debug_log,
4118 "%*s- %s (%d) <NOTHING BRANCH SEQUENCE>\n", (int)depth * 2 + 2,
4119 "", SvPV_nolen_const( mysv ),REG_NODE_NUM(cur));
4122 OP(startbranch)= NOTHING;
4123 NEXT_OFF(startbranch)= tail - startbranch;
4124 for ( opt= startbranch + 1; opt < tail ; opt++ )
4128 } /* end if ( last) */
4129 } /* TRIE_MAXBUF is non zero */
4134 else if ( code == BRANCHJ ) { /* single branch is optimized. */
4135 scan = NEXTOPER(NEXTOPER(scan));
4136 } else /* single branch is optimized. */
4137 scan = NEXTOPER(scan);
4139 } else if (OP(scan) == SUSPEND || OP(scan) == GOSUB || OP(scan) == GOSTART) {
4140 scan_frame *newframe = NULL;
4144 U32 my_recursed_depth= recursed_depth;
4146 if (OP(scan) != SUSPEND) {
4147 /* set the pointer */
4148 if (OP(scan) == GOSUB) {
4150 RExC_recurse[ARG2L(scan)] = scan;
4151 start = RExC_open_parens[paren-1];
4152 end = RExC_close_parens[paren-1];
4155 start = RExC_rxi->program + 1;
4160 !PAREN_TEST(RExC_study_chunk_recursed + ((recursed_depth-1) * RExC_study_chunk_recursed_bytes), paren)
4162 if (!recursed_depth) {
4163 Zero(RExC_study_chunk_recursed, RExC_study_chunk_recursed_bytes, U8);
4165 Copy(RExC_study_chunk_recursed + ((recursed_depth-1) * RExC_study_chunk_recursed_bytes),
4166 RExC_study_chunk_recursed + (recursed_depth * RExC_study_chunk_recursed_bytes),
4167 RExC_study_chunk_recursed_bytes, U8);
4169 /* we havent recursed into this paren yet, so recurse into it */
4170 DEBUG_STUDYDATA("set:", data,depth);
4171 PAREN_SET(RExC_study_chunk_recursed + (recursed_depth * RExC_study_chunk_recursed_bytes), paren);
4172 my_recursed_depth= recursed_depth + 1;
4173 Newx(newframe,1,scan_frame);
4175 DEBUG_STUDYDATA("inf:", data,depth);
4176 /* some form of infinite recursion, assume infinite length
4178 if (flags & SCF_DO_SUBSTR) {
4179 scan_commit(pRExC_state, data, minlenp, is_inf);
4180 data->longest = &(data->longest_float);
4182 is_inf = is_inf_internal = 1;
4183 if (flags & SCF_DO_STCLASS_OR) /* Allow everything */
4184 ssc_anything(data->start_class);
4185 flags &= ~SCF_DO_STCLASS;
4188 Newx(newframe,1,scan_frame);
4191 end = regnext(scan);
4196 SAVEFREEPV(newframe);
4197 newframe->next = regnext(scan);
4198 newframe->last = last;
4199 newframe->stop = stopparen;
4200 newframe->prev = frame;
4201 newframe->prev_recursed_depth = recursed_depth;
4203 DEBUG_STUDYDATA("frame-new:",data,depth);
4204 DEBUG_PEEP("fnew", scan, depth);
4211 recursed_depth= my_recursed_depth;
4216 else if (OP(scan) == EXACT) {
4217 SSize_t l = STR_LEN(scan);
4220 const U8 * const s = (U8*)STRING(scan);
4221 uc = utf8_to_uvchr_buf(s, s + l, NULL);
4222 l = utf8_length(s, s + l);
4224 uc = *((U8*)STRING(scan));
4227 if (flags & SCF_DO_SUBSTR) { /* Update longest substr. */
4228 /* The code below prefers earlier match for fixed
4229 offset, later match for variable offset. */
4230 if (data->last_end == -1) { /* Update the start info. */
4231 data->last_start_min = data->pos_min;
4232 data->last_start_max = is_inf
4233 ? SSize_t_MAX : data->pos_min + data->pos_delta;
4235 sv_catpvn(data->last_found, STRING(scan), STR_LEN(scan));
4237 SvUTF8_on(data->last_found);
4239 SV * const sv = data->last_found;
4240 MAGIC * const mg = SvUTF8(sv) && SvMAGICAL(sv) ?
4241 mg_find(sv, PERL_MAGIC_utf8) : NULL;
4242 if (mg && mg->mg_len >= 0)
4243 mg->mg_len += utf8_length((U8*)STRING(scan),
4244 (U8*)STRING(scan)+STR_LEN(scan));
4246 data->last_end = data->pos_min + l;
4247 data->pos_min += l; /* As in the first entry. */
4248 data->flags &= ~SF_BEFORE_EOL;
4251 /* ANDing the code point leaves at most it, and not in locale, and
4252 * can't match null string */
4253 if (flags & SCF_DO_STCLASS_AND) {
4254 ssc_cp_and(data->start_class, uc);
4255 ANYOF_FLAGS(data->start_class) &= ~ANYOF_EMPTY_STRING;
4256 ssc_clear_locale(data->start_class);
4258 else if (flags & SCF_DO_STCLASS_OR) {
4259 ssc_add_cp(data->start_class, uc);
4260 ssc_and(pRExC_state, data->start_class, (regnode_charclass *) and_withp);
4262 /* See commit msg 749e076fceedeb708a624933726e7989f2302f6a */
4263 ANYOF_FLAGS(data->start_class) &= ~ANYOF_EMPTY_STRING;
4265 flags &= ~SCF_DO_STCLASS;
4267 else if (PL_regkind[OP(scan)] == EXACT) { /* But OP != EXACT!, so is
4269 SSize_t l = STR_LEN(scan);
4270 UV uc = *((U8*)STRING(scan));
4271 SV* EXACTF_invlist = _new_invlist(4); /* Start out big enough for 2
4272 separate code points */
4273 const U8 * s = (U8*)STRING(scan);
4275 /* Search for fixed substrings supports EXACT only. */
4276 if (flags & SCF_DO_SUBSTR) {
4278 scan_commit(pRExC_state, data, minlenp, is_inf);
4281 uc = utf8_to_uvchr_buf(s, s + l, NULL);
4282 l = utf8_length(s, s + l);
4284 if (unfolded_multi_char) {
4285 RExC_seen |= REG_UNFOLDED_MULTI_SEEN;
4287 min += l - min_subtract;
4289 delta += min_subtract;
4290 if (flags & SCF_DO_SUBSTR) {
4291 data->pos_min += l - min_subtract;
4292 if (data->pos_min < 0) {
4295 data->pos_delta += min_subtract;
4297 data->longest = &(data->longest_float);
4301 if (OP(scan) != EXACTFL && flags & SCF_DO_STCLASS_AND) {
4302 ssc_clear_locale(data->start_class);
4307 /* We punt and assume can match anything if the node begins
4308 * with a multi-character fold. Things are complicated. For
4309 * example, /ffi/i could match any of:
4310 * "\N{LATIN SMALL LIGATURE FFI}"
4311 * "\N{LATIN SMALL LIGATURE FF}I"
4312 * "F\N{LATIN SMALL LIGATURE FI}"
4313 * plus several other things; and making sure we have all the
4314 * possibilities is hard. */
4315 if (is_MULTI_CHAR_FOLD_latin1_safe(s, s + STR_LEN(scan))) {
4317 _add_range_to_invlist(EXACTF_invlist, 0, UV_MAX);
4321 /* Any Latin1 range character can potentially match any
4322 * other depending on the locale */
4323 if (OP(scan) == EXACTFL) {
4324 _invlist_union(EXACTF_invlist, PL_Latin1,
4328 /* But otherwise, it matches at least itself. We can
4329 * quickly tell if it has a distinct fold, and if so,
4330 * it matches that as well */
4331 EXACTF_invlist = add_cp_to_invlist(EXACTF_invlist, uc);
4332 if (IS_IN_SOME_FOLD_L1(uc)) {
4333 EXACTF_invlist = add_cp_to_invlist(EXACTF_invlist,
4334 PL_fold_latin1[uc]);
4338 /* Some characters match above-Latin1 ones under /i. This
4339 * is true of EXACTFL ones when the locale is UTF-8 */
4340 if (HAS_NONLATIN1_SIMPLE_FOLD_CLOSURE(uc)
4341 && (! isASCII(uc) || (OP(scan) != EXACTFA
4342 && OP(scan) != EXACTFA_NO_TRIE)))
4344 add_above_Latin1_folds(pRExC_state,
4350 else { /* Pattern is UTF-8 */
4351 U8 folded[UTF8_MAX_FOLD_CHAR_EXPAND * UTF8_MAXBYTES_CASE + 1] = { '\0' };
4352 STRLEN foldlen = UTF8SKIP(s);
4353 const U8* e = s + STR_LEN(scan);
4356 /* The only code points that aren't folded in a UTF EXACTFish
4357 * node are are the problematic ones in EXACTFL nodes */
4358 if (OP(scan) == EXACTFL
4359 && is_PROBLEMATIC_LOCALE_FOLDEDS_START_cp(uc))
4361 /* We need to check for the possibility that this EXACTFL
4362 * node begins with a multi-char fold. Therefore we fold
4363 * the first few characters of it so that we can make that
4368 for (i = 0; i < UTF8_MAX_FOLD_CHAR_EXPAND && s < e; i++) {
4370 *(d++) = (U8) toFOLD(*s);
4375 to_utf8_fold(s, d, &len);
4381 /* And set up so the code below that looks in this folded
4382 * buffer instead of the node's string */
4384 foldlen = UTF8SKIP(folded);
4388 /* When we reach here 's' points to the fold of the first
4389 * character(s) of the node; and 'e' points to far enough along
4390 * the folded string to be just past any possible multi-char
4391 * fold. 'foldlen' is the length in bytes of the first
4394 * Unlike the non-UTF-8 case, the macro for determining if a
4395 * string is a multi-char fold requires all the characters to
4396 * already be folded. This is because of all the complications
4397 * if not. Note that they are folded anyway, except in EXACTFL
4398 * nodes. Like the non-UTF case above, we punt if the node
4399 * begins with a multi-char fold */
4401 if (is_MULTI_CHAR_FOLD_utf8_safe(s, e)) {
4403 _add_range_to_invlist(EXACTF_invlist, 0, UV_MAX);
4405 else { /* Single char fold */
4407 /* It matches all the things that fold to it, which are
4408 * found in PL_utf8_foldclosures (including itself) */
4409 EXACTF_invlist = add_cp_to_invlist(EXACTF_invlist, uc);
4410 if (! PL_utf8_foldclosures) {
4411 _load_PL_utf8_foldclosures();
4413 if ((listp = hv_fetch(PL_utf8_foldclosures,
4414 (char *) s, foldlen, FALSE)))
4416 AV* list = (AV*) *listp;
4418 for (k = 0; k <= av_tindex(list); k++) {
4419 SV** c_p = av_fetch(list, k, FALSE);
4425 /* /aa doesn't allow folds between ASCII and non- */
4426 if ((OP(scan) == EXACTFA || OP(scan) == EXACTFA_NO_TRIE)
4427 && isASCII(c) != isASCII(uc))
4432 EXACTF_invlist = add_cp_to_invlist(EXACTF_invlist, c);
4437 if (flags & SCF_DO_STCLASS_AND) {
4438 ANYOF_FLAGS(data->start_class) &= ~ANYOF_EMPTY_STRING;
4439 ANYOF_POSIXL_ZERO(data->start_class);
4440 ssc_intersection(data->start_class, EXACTF_invlist, FALSE);
4442 else if (flags & SCF_DO_STCLASS_OR) {
4443 ssc_union(data->start_class, EXACTF_invlist, FALSE);
4444 ssc_and(pRExC_state, data->start_class, (regnode_charclass *) and_withp);
4446 /* See commit msg 749e076fceedeb708a624933726e7989f2302f6a */
4447 ANYOF_FLAGS(data->start_class) &= ~ANYOF_EMPTY_STRING;
4449 flags &= ~SCF_DO_STCLASS;
4450 SvREFCNT_dec(EXACTF_invlist);
4452 else if (REGNODE_VARIES(OP(scan))) {
4453 SSize_t mincount, maxcount, minnext, deltanext, pos_before = 0;
4454 I32 fl = 0, f = flags;
4455 regnode * const oscan = scan;
4456 regnode_ssc this_class;
4457 regnode_ssc *oclass = NULL;
4458 I32 next_is_eval = 0;
4460 switch (PL_regkind[OP(scan)]) {
4461 case WHILEM: /* End of (?:...)* . */
4462 scan = NEXTOPER(scan);
4465 if (flags & (SCF_DO_SUBSTR | SCF_DO_STCLASS)) {
4466 next = NEXTOPER(scan);
4467 if (OP(next) == EXACT || (flags & SCF_DO_STCLASS)) {
4469 maxcount = REG_INFTY;
4470 next = regnext(scan);
4471 scan = NEXTOPER(scan);
4475 if (flags & SCF_DO_SUBSTR)
4480 if (flags & SCF_DO_STCLASS) {
4482 maxcount = REG_INFTY;
4483 next = regnext(scan);
4484 scan = NEXTOPER(scan);
4487 if (flags & SCF_DO_SUBSTR) {
4488 scan_commit(pRExC_state, data, minlenp, is_inf);
4489 /* Cannot extend fixed substrings */
4490 data->longest = &(data->longest_float);
4492 is_inf = is_inf_internal = 1;
4493 scan = regnext(scan);
4494 goto optimize_curly_tail;
4496 if (stopparen>0 && (OP(scan)==CURLYN || OP(scan)==CURLYM)
4497 && (scan->flags == stopparen))
4502 mincount = ARG1(scan);
4503 maxcount = ARG2(scan);
4505 next = regnext(scan);
4506 if (OP(scan) == CURLYX) {
4507 I32 lp = (data ? *(data->last_closep) : 0);
4508 scan->flags = ((lp <= (I32)U8_MAX) ? (U8)lp : U8_MAX);
4510 scan = NEXTOPER(scan) + EXTRA_STEP_2ARGS;
4511 next_is_eval = (OP(scan) == EVAL);
4513 if (flags & SCF_DO_SUBSTR) {
4515 scan_commit(pRExC_state, data, minlenp, is_inf);
4516 /* Cannot extend fixed substrings */
4517 pos_before = data->pos_min;
4521 data->flags &= ~(SF_HAS_PAR|SF_IN_PAR|SF_HAS_EVAL);
4523 data->flags |= SF_IS_INF;
4525 if (flags & SCF_DO_STCLASS) {
4526 ssc_init(pRExC_state, &this_class);
4527 oclass = data->start_class;
4528 data->start_class = &this_class;
4529 f |= SCF_DO_STCLASS_AND;
4530 f &= ~SCF_DO_STCLASS_OR;
4532 /* Exclude from super-linear cache processing any {n,m}
4533 regops for which the combination of input pos and regex
4534 pos is not enough information to determine if a match
4537 For example, in the regex /foo(bar\s*){4,8}baz/ with the
4538 regex pos at the \s*, the prospects for a match depend not
4539 only on the input position but also on how many (bar\s*)
4540 repeats into the {4,8} we are. */
4541 if ((mincount > 1) || (maxcount > 1 && maxcount != REG_INFTY))
4542 f &= ~SCF_WHILEM_VISITED_POS;
4544 /* This will finish on WHILEM, setting scan, or on NULL: */
4545 minnext = study_chunk(pRExC_state, &scan, minlenp, &deltanext,
4546 last, data, stopparen, recursed_depth, NULL,
4548 ? (f & ~SCF_DO_SUBSTR)
4552 if (flags & SCF_DO_STCLASS)
4553 data->start_class = oclass;
4554 if (mincount == 0 || minnext == 0) {
4555 if (flags & SCF_DO_STCLASS_OR) {
4556 ssc_or(pRExC_state, data->start_class, (regnode_charclass *) &this_class);
4558 else if (flags & SCF_DO_STCLASS_AND) {
4559 /* Switch to OR mode: cache the old value of
4560 * data->start_class */
4562 StructCopy(data->start_class, and_withp, regnode_ssc);
4563 flags &= ~SCF_DO_STCLASS_AND;
4564 StructCopy(&this_class, data->start_class, regnode_ssc);
4565 flags |= SCF_DO_STCLASS_OR;
4566 ANYOF_FLAGS(data->start_class) |= ANYOF_EMPTY_STRING;
4568 } else { /* Non-zero len */
4569 if (flags & SCF_DO_STCLASS_OR) {
4570 ssc_or(pRExC_state, data->start_class, (regnode_charclass *) &this_class);
4571 ssc_and(pRExC_state, data->start_class, (regnode_charclass *) and_withp);
4573 else if (flags & SCF_DO_STCLASS_AND)
4574 ssc_and(pRExC_state, data->start_class, (regnode_charclass *) &this_class);
4575 flags &= ~SCF_DO_STCLASS;
4577 if (!scan) /* It was not CURLYX, but CURLY. */
4579 if (!(flags & SCF_TRIE_DOING_RESTUDY)
4580 /* ? quantifier ok, except for (?{ ... }) */
4581 && (next_is_eval || !(mincount == 0 && maxcount == 1))
4582 && (minnext == 0) && (deltanext == 0)
4583 && data && !(data->flags & (SF_HAS_PAR|SF_IN_PAR))
4584 && maxcount <= REG_INFTY/3) /* Complement check for big
4587 /* Fatal warnings may leak the regexp without this: */
4588 SAVEFREESV(RExC_rx_sv);
4589 ckWARNreg(RExC_parse,
4590 "Quantifier unexpected on zero-length expression");
4591 (void)ReREFCNT_inc(RExC_rx_sv);
4594 min += minnext * mincount;
4595 is_inf_internal |= deltanext == SSize_t_MAX
4596 || (maxcount == REG_INFTY && minnext + deltanext > 0);
4597 is_inf |= is_inf_internal;
4599 delta = SSize_t_MAX;
4601 delta += (minnext + deltanext) * maxcount
4602 - minnext * mincount;
4604 /* Try powerful optimization CURLYX => CURLYN. */
4605 if ( OP(oscan) == CURLYX && data
4606 && data->flags & SF_IN_PAR
4607 && !(data->flags & SF_HAS_EVAL)
4608 && !deltanext && minnext == 1 ) {
4609 /* Try to optimize to CURLYN. */
4610 regnode *nxt = NEXTOPER(oscan) + EXTRA_STEP_2ARGS;
4611 regnode * const nxt1 = nxt;
4618 if (!REGNODE_SIMPLE(OP(nxt))
4619 && !(PL_regkind[OP(nxt)] == EXACT
4620 && STR_LEN(nxt) == 1))
4626 if (OP(nxt) != CLOSE)
4628 if (RExC_open_parens) {
4629 RExC_open_parens[ARG(nxt1)-1]=oscan; /*open->CURLYM*/
4630 RExC_close_parens[ARG(nxt1)-1]=nxt+2; /*close->while*/
4632 /* Now we know that nxt2 is the only contents: */
4633 oscan->flags = (U8)ARG(nxt);
4635 OP(nxt1) = NOTHING; /* was OPEN. */
4638 OP(nxt1 + 1) = OPTIMIZED; /* was count. */
4639 NEXT_OFF(nxt1+ 1) = 0; /* just for consistency. */
4640 NEXT_OFF(nxt2) = 0; /* just for consistency with CURLY. */
4641 OP(nxt) = OPTIMIZED; /* was CLOSE. */
4642 OP(nxt + 1) = OPTIMIZED; /* was count. */
4643 NEXT_OFF(nxt+ 1) = 0; /* just for consistency. */
4648 /* Try optimization CURLYX => CURLYM. */
4649 if ( OP(oscan) == CURLYX && data
4650 && !(data->flags & SF_HAS_PAR)
4651 && !(data->flags & SF_HAS_EVAL)
4652 && !deltanext /* atom is fixed width */
4653 && minnext != 0 /* CURLYM can't handle zero width */
4655 /* Nor characters whose fold at run-time may be
4656 * multi-character */
4657 && ! (RExC_seen & REG_UNFOLDED_MULTI_SEEN)
4659 /* XXXX How to optimize if data == 0? */
4660 /* Optimize to a simpler form. */
4661 regnode *nxt = NEXTOPER(oscan) + EXTRA_STEP_2ARGS; /* OPEN */
4665 while ( (nxt2 = regnext(nxt)) /* skip over embedded stuff*/
4666 && (OP(nxt2) != WHILEM))
4668 OP(nxt2) = SUCCEED; /* Whas WHILEM */
4669 /* Need to optimize away parenths. */
4670 if ((data->flags & SF_IN_PAR) && OP(nxt) == CLOSE) {
4671 /* Set the parenth number. */
4672 regnode *nxt1 = NEXTOPER(oscan) + EXTRA_STEP_2ARGS; /* OPEN*/
4674 oscan->flags = (U8)ARG(nxt);
4675 if (RExC_open_parens) {
4676 RExC_open_parens[ARG(nxt1)-1]=oscan; /*open->CURLYM*/
4677 RExC_close_parens[ARG(nxt1)-1]=nxt2+1; /*close->NOTHING*/
4679 OP(nxt1) = OPTIMIZED; /* was OPEN. */
4680 OP(nxt) = OPTIMIZED; /* was CLOSE. */
4683 OP(nxt1 + 1) = OPTIMIZED; /* was count. */
4684 OP(nxt + 1) = OPTIMIZED; /* was count. */
4685 NEXT_OFF(nxt1 + 1) = 0; /* just for consistency. */
4686 NEXT_OFF(nxt + 1) = 0; /* just for consistency. */
4689 while ( nxt1 && (OP(nxt1) != WHILEM)) {
4690 regnode *nnxt = regnext(nxt1);
4692 if (reg_off_by_arg[OP(nxt1)])
4693 ARG_SET(nxt1, nxt2 - nxt1);
4694 else if (nxt2 - nxt1 < U16_MAX)
4695 NEXT_OFF(nxt1) = nxt2 - nxt1;
4697 OP(nxt) = NOTHING; /* Cannot beautify */
4702 /* Optimize again: */
4703 study_chunk(pRExC_state, &nxt1, minlenp, &deltanext, nxt,
4704 NULL, stopparen, recursed_depth, NULL, 0,depth+1);
4709 else if ((OP(oscan) == CURLYX)
4710 && (flags & SCF_WHILEM_VISITED_POS)
4711 /* See the comment on a similar expression above.
4712 However, this time it's not a subexpression
4713 we care about, but the expression itself. */
4714 && (maxcount == REG_INFTY)
4715 && data && ++data->whilem_c < 16) {
4716 /* This stays as CURLYX, we can put the count/of pair. */
4717 /* Find WHILEM (as in regexec.c) */
4718 regnode *nxt = oscan + NEXT_OFF(oscan);
4720 if (OP(PREVOPER(nxt)) == NOTHING) /* LONGJMP */
4722 PREVOPER(nxt)->flags = (U8)(data->whilem_c
4723 | (RExC_whilem_seen << 4)); /* On WHILEM */
4725 if (data && fl & (SF_HAS_PAR|SF_IN_PAR))
4727 if (flags & SCF_DO_SUBSTR) {
4728 SV *last_str = NULL;
4729 STRLEN last_chrs = 0;
4730 int counted = mincount != 0;
4732 if (data->last_end > 0 && mincount != 0) { /* Ends with a
4734 SSize_t b = pos_before >= data->last_start_min
4735 ? pos_before : data->last_start_min;
4737 const char * const s = SvPV_const(data->last_found, l);
4738 SSize_t old = b - data->last_start_min;
4741 old = utf8_hop((U8*)s, old) - (U8*)s;
4743 /* Get the added string: */
4744 last_str = newSVpvn_utf8(s + old, l, UTF);
4745 last_chrs = UTF ? utf8_length((U8*)(s + old),
4746 (U8*)(s + old + l)) : l;
4747 if (deltanext == 0 && pos_before == b) {
4748 /* What was added is a constant string */
4751 SvGROW(last_str, (mincount * l) + 1);
4752 repeatcpy(SvPVX(last_str) + l,
4753 SvPVX_const(last_str), l,
4755 SvCUR_set(last_str, SvCUR(last_str) * mincount);
4756 /* Add additional parts. */
4757 SvCUR_set(data->last_found,
4758 SvCUR(data->last_found) - l);
4759 sv_catsv(data->last_found, last_str);
4761 SV * sv = data->last_found;
4763 SvUTF8(sv) && SvMAGICAL(sv) ?
4764 mg_find(sv, PERL_MAGIC_utf8) : NULL;
4765 if (mg && mg->mg_len >= 0)
4766 mg->mg_len += last_chrs * (mincount-1);
4768 last_chrs *= mincount;
4769 data->last_end += l * (mincount - 1);
4772 /* start offset must point into the last copy */
4773 data->last_start_min += minnext * (mincount - 1);
4774 data->last_start_max += is_inf ? SSize_t_MAX
4775 : (maxcount - 1) * (minnext + data->pos_delta);
4778 /* It is counted once already... */
4779 data->pos_min += minnext * (mincount - counted);
4781 PerlIO_printf(Perl_debug_log, "counted=%"UVuf" deltanext=%"UVuf
4782 " SSize_t_MAX=%"UVuf" minnext=%"UVuf
4783 " maxcount=%"UVuf" mincount=%"UVuf"\n",
4784 (UV)counted, (UV)deltanext, (UV)SSize_t_MAX, (UV)minnext, (UV)maxcount,
4786 if (deltanext != SSize_t_MAX)
4787 PerlIO_printf(Perl_debug_log, "LHS=%"UVuf" RHS=%"UVuf"\n",
4788 (UV)(-counted * deltanext + (minnext + deltanext) * maxcount
4789 - minnext * mincount), (UV)(SSize_t_MAX - data->pos_delta));
4791 if (deltanext == SSize_t_MAX
4792 || -counted * deltanext + (minnext + deltanext) * maxcount - minnext * mincount >= SSize_t_MAX - data->pos_delta)
4793 data->pos_delta = SSize_t_MAX;
4795 data->pos_delta += - counted * deltanext +
4796 (minnext + deltanext) * maxcount - minnext * mincount;
4797 if (mincount != maxcount) {
4798 /* Cannot extend fixed substrings found inside
4800 scan_commit(pRExC_state, data, minlenp, is_inf);
4801 if (mincount && last_str) {
4802 SV * const sv = data->last_found;
4803 MAGIC * const mg = SvUTF8(sv) && SvMAGICAL(sv) ?
4804 mg_find(sv, PERL_MAGIC_utf8) : NULL;
4808 sv_setsv(sv, last_str);
4809 data->last_end = data->pos_min;
4810 data->last_start_min = data->pos_min - last_chrs;
4811 data->last_start_max = is_inf
4813 : data->pos_min + data->pos_delta - last_chrs;
4815 data->longest = &(data->longest_float);
4817 SvREFCNT_dec(last_str);
4819 if (data && (fl & SF_HAS_EVAL))
4820 data->flags |= SF_HAS_EVAL;
4821 optimize_curly_tail:
4822 if (OP(oscan) != CURLYX) {
4823 while (PL_regkind[OP(next = regnext(oscan))] == NOTHING
4825 NEXT_OFF(oscan) += NEXT_OFF(next);
4831 Perl_croak(aTHX_ "panic: unexpected varying REx opcode %d",
4836 if (flags & SCF_DO_SUBSTR) {
4837 /* Cannot expect anything... */
4838 scan_commit(pRExC_state, data, minlenp, is_inf);
4839 data->longest = &(data->longest_float);
4841 is_inf = is_inf_internal = 1;
4842 if (flags & SCF_DO_STCLASS_OR) {
4843 if (OP(scan) == CLUMP) {
4844 /* Actually is any start char, but very few code points
4845 * aren't start characters */
4846 ssc_match_all_cp(data->start_class);
4849 ssc_anything(data->start_class);
4852 flags &= ~SCF_DO_STCLASS;
4856 else if (OP(scan) == LNBREAK) {
4857 if (flags & SCF_DO_STCLASS) {
4858 if (flags & SCF_DO_STCLASS_AND) {
4859 ssc_intersection(data->start_class,
4860 PL_XPosix_ptrs[_CC_VERTSPACE], FALSE);
4861 ssc_clear_locale(data->start_class);
4862 ANYOF_FLAGS(data->start_class) &= ~ANYOF_EMPTY_STRING;
4864 else if (flags & SCF_DO_STCLASS_OR) {
4865 ssc_union(data->start_class,
4866 PL_XPosix_ptrs[_CC_VERTSPACE],
4868 ssc_and(pRExC_state, data->start_class, (regnode_charclass *) and_withp);
4870 /* See commit msg for
4871 * 749e076fceedeb708a624933726e7989f2302f6a */
4872 ANYOF_FLAGS(data->start_class) &= ~ANYOF_EMPTY_STRING;
4874 flags &= ~SCF_DO_STCLASS;
4877 delta++; /* Because of the 2 char string cr-lf */
4878 if (flags & SCF_DO_SUBSTR) {
4879 /* Cannot expect anything... */
4880 scan_commit(pRExC_state, data, minlenp, is_inf);
4882 data->pos_delta += 1;
4883 data->longest = &(data->longest_float);
4886 else if (REGNODE_SIMPLE(OP(scan))) {
4888 if (flags & SCF_DO_SUBSTR) {
4889 scan_commit(pRExC_state, data, minlenp, is_inf);
4893 if (flags & SCF_DO_STCLASS) {
4895 SV* my_invlist = sv_2mortal(_new_invlist(0));
4898 /* See commit msg 749e076fceedeb708a624933726e7989f2302f6a */
4899 ANYOF_FLAGS(data->start_class) &= ~ANYOF_EMPTY_STRING;
4901 /* Some of the logic below assumes that switching
4902 locale on will only add false positives. */
4907 Perl_croak(aTHX_ "panic: unexpected simple REx opcode %d",
4912 if (flags & SCF_DO_STCLASS_OR) /* Allow everything */
4913 ssc_match_all_cp(data->start_class);
4918 SV* REG_ANY_invlist = _new_invlist(2);
4919 REG_ANY_invlist = add_cp_to_invlist(REG_ANY_invlist,
4921 if (flags & SCF_DO_STCLASS_OR) {
4922 ssc_union(data->start_class,
4924 TRUE /* TRUE => invert, hence all but \n
4928 else if (flags & SCF_DO_STCLASS_AND) {
4929 ssc_intersection(data->start_class,
4931 TRUE /* TRUE => invert */
4933 ssc_clear_locale(data->start_class);
4935 SvREFCNT_dec_NN(REG_ANY_invlist);
4940 if (flags & SCF_DO_STCLASS_AND)
4941 ssc_and(pRExC_state, data->start_class,
4942 (regnode_charclass *) scan);
4944 ssc_or(pRExC_state, data->start_class,
4945 (regnode_charclass *) scan);
4953 namedclass = classnum_to_namedclass(FLAGS(scan)) + invert;
4954 if (flags & SCF_DO_STCLASS_AND) {
4955 bool was_there = cBOOL(
4956 ANYOF_POSIXL_TEST(data->start_class,
4958 ANYOF_POSIXL_ZERO(data->start_class);
4959 if (was_there) { /* Do an AND */
4960 ANYOF_POSIXL_SET(data->start_class, namedclass);
4962 /* No individual code points can now match */
4963 data->start_class->invlist
4964 = sv_2mortal(_new_invlist(0));
4967 int complement = namedclass + ((invert) ? -1 : 1);
4969 assert(flags & SCF_DO_STCLASS_OR);
4971 /* If the complement of this class was already there,
4972 * the result is that they match all code points,
4973 * (\d + \D == everything). Remove the classes from
4974 * future consideration. Locale is not relevant in
4976 if (ANYOF_POSIXL_TEST(data->start_class, complement)) {
4977 ssc_match_all_cp(data->start_class);
4978 ANYOF_POSIXL_CLEAR(data->start_class, namedclass);
4979 ANYOF_POSIXL_CLEAR(data->start_class, complement);
4981 else { /* The usual case; just add this class to the
4983 ANYOF_POSIXL_SET(data->start_class, namedclass);
4988 case NPOSIXA: /* For these, we always know the exact set of
4993 if (FLAGS(scan) == _CC_ASCII) {
4994 my_invlist = PL_XPosix_ptrs[_CC_ASCII];
4997 _invlist_intersection(PL_XPosix_ptrs[FLAGS(scan)],
4998 PL_XPosix_ptrs[_CC_ASCII],
5009 my_invlist = invlist_clone(PL_XPosix_ptrs[FLAGS(scan)]);
5011 /* NPOSIXD matches all upper Latin1 code points unless the
5012 * target string being matched is UTF-8, which is
5013 * unknowable until match time. Since we are going to
5014 * invert, we want to get rid of all of them so that the
5015 * inversion will match all */
5016 if (OP(scan) == NPOSIXD) {
5017 _invlist_subtract(my_invlist, PL_UpperLatin1,
5023 if (flags & SCF_DO_STCLASS_AND) {
5024 ssc_intersection(data->start_class, my_invlist, invert);
5025 ssc_clear_locale(data->start_class);
5028 assert(flags & SCF_DO_STCLASS_OR);
5029 ssc_union(data->start_class, my_invlist, invert);
5032 if (flags & SCF_DO_STCLASS_OR)
5033 ssc_and(pRExC_state, data->start_class, (regnode_charclass *) and_withp);
5034 flags &= ~SCF_DO_STCLASS;
5037 else if (PL_regkind[OP(scan)] == EOL && flags & SCF_DO_SUBSTR) {
5038 data->flags |= (OP(scan) == MEOL
5041 scan_commit(pRExC_state, data, minlenp, is_inf);
5044 else if ( PL_regkind[OP(scan)] == BRANCHJ
5045 /* Lookbehind, or need to calculate parens/evals/stclass: */
5046 && (scan->flags || data || (flags & SCF_DO_STCLASS))
5047 && (OP(scan) == IFMATCH || OP(scan) == UNLESSM)) {
5048 if ( OP(scan) == UNLESSM &&
5050 OP(NEXTOPER(NEXTOPER(scan))) == NOTHING &&
5051 OP(regnext(NEXTOPER(NEXTOPER(scan)))) == SUCCEED
5054 regnode *upto= regnext(scan);
5056 SV * const mysv_val=sv_newmortal();
5057 DEBUG_STUDYDATA("OPFAIL",data,depth);
5059 /*DEBUG_PARSE_MSG("opfail");*/
5060 regprop(RExC_rx, mysv_val, upto, NULL);
5061 PerlIO_printf(Perl_debug_log,
5062 "~ replace with OPFAIL pointed at %s (%"IVdf") offset %"IVdf"\n",
5063 SvPV_nolen_const(mysv_val),
5064 (IV)REG_NODE_NUM(upto),
5069 NEXT_OFF(scan) = upto - scan;
5070 for (opt= scan + 1; opt < upto ; opt++)
5071 OP(opt) = OPTIMIZED;
5075 if ( !PERL_ENABLE_POSITIVE_ASSERTION_STUDY
5076 || OP(scan) == UNLESSM )
5078 /* Negative Lookahead/lookbehind
5079 In this case we can't do fixed string optimisation.
5082 SSize_t deltanext, minnext, fake = 0;
5087 data_fake.flags = 0;
5089 data_fake.whilem_c = data->whilem_c;
5090 data_fake.last_closep = data->last_closep;
5093 data_fake.last_closep = &fake;
5094 data_fake.pos_delta = delta;
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));
5105 minnext = study_chunk(pRExC_state, &nscan, minlenp, &deltanext,
5106 last, &data_fake, stopparen,
5107 recursed_depth, NULL, f, depth+1);
5110 FAIL("Variable length lookbehind not implemented");
5112 else if (minnext > (I32)U8_MAX) {
5113 FAIL2("Lookbehind longer than %"UVuf" not implemented",
5116 scan->flags = (U8)minnext;
5119 if (data_fake.flags & (SF_HAS_PAR|SF_IN_PAR))
5121 if (data_fake.flags & SF_HAS_EVAL)
5122 data->flags |= SF_HAS_EVAL;
5123 data->whilem_c = data_fake.whilem_c;
5125 if (f & SCF_DO_STCLASS_AND) {
5126 if (flags & SCF_DO_STCLASS_OR) {
5127 /* OR before, AND after: ideally we would recurse with
5128 * data_fake to get the AND applied by study of the
5129 * remainder of the pattern, and then derecurse;
5130 * *** HACK *** for now just treat as "no information".
5131 * See [perl #56690].
5133 ssc_init(pRExC_state, data->start_class);
5135 /* AND before and after: combine and continue */
5136 ssc_and(pRExC_state, data->start_class, (regnode_charclass *) &intrnl);
5140 #if PERL_ENABLE_POSITIVE_ASSERTION_STUDY
5142 /* Positive Lookahead/lookbehind
5143 In this case we can do fixed string optimisation,
5144 but we must be careful about it. Note in the case of
5145 lookbehind the positions will be offset by the minimum
5146 length of the pattern, something we won't know about
5147 until after the recurse.
5149 SSize_t deltanext, fake = 0;
5153 /* We use SAVEFREEPV so that when the full compile
5154 is finished perl will clean up the allocated
5155 minlens when it's all done. This way we don't
5156 have to worry about freeing them when we know
5157 they wont be used, which would be a pain.
5160 Newx( minnextp, 1, SSize_t );
5161 SAVEFREEPV(minnextp);
5164 StructCopy(data, &data_fake, scan_data_t);
5165 if ((flags & SCF_DO_SUBSTR) && data->last_found) {
5168 scan_commit(pRExC_state, &data_fake, minlenp, is_inf);
5169 data_fake.last_found=newSVsv(data->last_found);
5173 data_fake.last_closep = &fake;
5174 data_fake.flags = 0;
5175 data_fake.pos_delta = delta;
5177 data_fake.flags |= SF_IS_INF;
5178 if ( flags & SCF_DO_STCLASS && !scan->flags
5179 && OP(scan) == IFMATCH ) { /* Lookahead */
5180 ssc_init(pRExC_state, &intrnl);
5181 data_fake.start_class = &intrnl;
5182 f |= SCF_DO_STCLASS_AND;
5184 if (flags & SCF_WHILEM_VISITED_POS)
5185 f |= SCF_WHILEM_VISITED_POS;
5186 next = regnext(scan);
5187 nscan = NEXTOPER(NEXTOPER(scan));
5189 *minnextp = study_chunk(pRExC_state, &nscan, minnextp,
5190 &deltanext, last, &data_fake,
5191 stopparen, recursed_depth, NULL,
5195 FAIL("Variable length lookbehind not implemented");
5197 else if (*minnextp > (I32)U8_MAX) {
5198 FAIL2("Lookbehind longer than %"UVuf" not implemented",
5201 scan->flags = (U8)*minnextp;
5206 if (f & SCF_DO_STCLASS_AND) {
5207 ssc_and(pRExC_state, data->start_class, (regnode_charclass *) &intrnl);
5210 if (data_fake.flags & (SF_HAS_PAR|SF_IN_PAR))
5212 if (data_fake.flags & SF_HAS_EVAL)
5213 data->flags |= SF_HAS_EVAL;
5214 data->whilem_c = data_fake.whilem_c;
5215 if ((flags & SCF_DO_SUBSTR) && data_fake.last_found) {
5216 if (RExC_rx->minlen<*minnextp)
5217 RExC_rx->minlen=*minnextp;
5218 scan_commit(pRExC_state, &data_fake, minnextp, is_inf);
5219 SvREFCNT_dec_NN(data_fake.last_found);
5221 if ( data_fake.minlen_fixed != minlenp )
5223 data->offset_fixed= data_fake.offset_fixed;
5224 data->minlen_fixed= data_fake.minlen_fixed;
5225 data->lookbehind_fixed+= scan->flags;
5227 if ( data_fake.minlen_float != minlenp )
5229 data->minlen_float= data_fake.minlen_float;
5230 data->offset_float_min=data_fake.offset_float_min;
5231 data->offset_float_max=data_fake.offset_float_max;
5232 data->lookbehind_float+= scan->flags;
5239 else if (OP(scan) == OPEN) {
5240 if (stopparen != (I32)ARG(scan))
5243 else if (OP(scan) == CLOSE) {
5244 if (stopparen == (I32)ARG(scan)) {
5247 if ((I32)ARG(scan) == is_par) {
5248 next = regnext(scan);
5250 if ( next && (OP(next) != WHILEM) && next < last)
5251 is_par = 0; /* Disable optimization */
5254 *(data->last_closep) = ARG(scan);
5256 else if (OP(scan) == EVAL) {
5258 data->flags |= SF_HAS_EVAL;
5260 else if ( PL_regkind[OP(scan)] == ENDLIKE ) {
5261 if (flags & SCF_DO_SUBSTR) {
5262 scan_commit(pRExC_state, data, minlenp, is_inf);
5263 flags &= ~SCF_DO_SUBSTR;
5265 if (data && OP(scan)==ACCEPT) {
5266 data->flags |= SCF_SEEN_ACCEPT;
5271 else if (OP(scan) == LOGICAL && scan->flags == 2) /* Embedded follows */
5273 if (flags & SCF_DO_SUBSTR) {
5274 scan_commit(pRExC_state, data, minlenp, is_inf);
5275 data->longest = &(data->longest_float);
5277 is_inf = is_inf_internal = 1;
5278 if (flags & SCF_DO_STCLASS_OR) /* Allow everything */
5279 ssc_anything(data->start_class);
5280 flags &= ~SCF_DO_STCLASS;
5282 else if (OP(scan) == GPOS) {
5283 if (!(RExC_rx->intflags & PREGf_GPOS_FLOAT) &&
5284 !(delta || is_inf || (data && data->pos_delta)))
5286 if (!(RExC_rx->intflags & PREGf_ANCH) && (flags & SCF_DO_SUBSTR))
5287 RExC_rx->intflags |= PREGf_ANCH_GPOS;
5288 if (RExC_rx->gofs < (STRLEN)min)
5289 RExC_rx->gofs = min;
5291 RExC_rx->intflags |= PREGf_GPOS_FLOAT;
5295 #ifdef TRIE_STUDY_OPT
5296 #ifdef FULL_TRIE_STUDY
5297 else if (PL_regkind[OP(scan)] == TRIE) {
5298 /* NOTE - There is similar code to this block above for handling
5299 BRANCH nodes on the initial study. If you change stuff here
5301 regnode *trie_node= scan;
5302 regnode *tail= regnext(scan);
5303 reg_trie_data *trie = (reg_trie_data*)RExC_rxi->data->data[ ARG(scan) ];
5304 SSize_t max1 = 0, min1 = SSize_t_MAX;
5307 if (flags & SCF_DO_SUBSTR) { /* XXXX Add !SUSPEND? */
5308 /* Cannot merge strings after this. */
5309 scan_commit(pRExC_state, data, minlenp, is_inf);
5311 if (flags & SCF_DO_STCLASS)
5312 ssc_init_zero(pRExC_state, &accum);
5318 const regnode *nextbranch= NULL;
5321 for ( word=1 ; word <= trie->wordcount ; word++)
5323 SSize_t deltanext=0, minnext=0, f = 0, fake;
5324 regnode_ssc this_class;
5326 data_fake.flags = 0;
5328 data_fake.whilem_c = data->whilem_c;
5329 data_fake.last_closep = data->last_closep;
5332 data_fake.last_closep = &fake;
5333 data_fake.pos_delta = delta;
5334 if (flags & SCF_DO_STCLASS) {
5335 ssc_init(pRExC_state, &this_class);
5336 data_fake.start_class = &this_class;
5337 f = SCF_DO_STCLASS_AND;
5339 if (flags & SCF_WHILEM_VISITED_POS)
5340 f |= SCF_WHILEM_VISITED_POS;
5342 if (trie->jump[word]) {
5344 nextbranch = trie_node + trie->jump[0];
5345 scan= trie_node + trie->jump[word];
5346 /* We go from the jump point to the branch that follows
5347 it. Note this means we need the vestigal unused
5348 branches even though they arent otherwise used. */
5349 minnext = study_chunk(pRExC_state, &scan, minlenp,
5350 &deltanext, (regnode *)nextbranch, &data_fake,
5351 stopparen, recursed_depth, NULL, f,depth+1);
5353 if (nextbranch && PL_regkind[OP(nextbranch)]==BRANCH)
5354 nextbranch= regnext((regnode*)nextbranch);
5356 if (min1 > (SSize_t)(minnext + trie->minlen))
5357 min1 = minnext + trie->minlen;
5358 if (deltanext == SSize_t_MAX) {
5359 is_inf = is_inf_internal = 1;
5361 } else if (max1 < (SSize_t)(minnext + deltanext + trie->maxlen))
5362 max1 = minnext + deltanext + trie->maxlen;
5364 if (data_fake.flags & (SF_HAS_PAR|SF_IN_PAR))
5366 if (data_fake.flags & SCF_SEEN_ACCEPT) {
5367 if ( stopmin > min + min1)
5368 stopmin = min + min1;
5369 flags &= ~SCF_DO_SUBSTR;
5371 data->flags |= SCF_SEEN_ACCEPT;
5374 if (data_fake.flags & SF_HAS_EVAL)
5375 data->flags |= SF_HAS_EVAL;
5376 data->whilem_c = data_fake.whilem_c;
5378 if (flags & SCF_DO_STCLASS)
5379 ssc_or(pRExC_state, &accum, (regnode_charclass *) &this_class);
5382 if (flags & SCF_DO_SUBSTR) {
5383 data->pos_min += min1;
5384 data->pos_delta += max1 - min1;
5385 if (max1 != min1 || is_inf)
5386 data->longest = &(data->longest_float);
5389 delta += max1 - min1;
5390 if (flags & SCF_DO_STCLASS_OR) {
5391 ssc_or(pRExC_state, data->start_class, (regnode_charclass *) &accum);
5393 ssc_and(pRExC_state, data->start_class, (regnode_charclass *) and_withp);
5394 flags &= ~SCF_DO_STCLASS;
5397 else if (flags & SCF_DO_STCLASS_AND) {
5399 ssc_and(pRExC_state, data->start_class, (regnode_charclass *) &accum);
5400 flags &= ~SCF_DO_STCLASS;
5403 /* Switch to OR mode: cache the old value of
5404 * data->start_class */
5406 StructCopy(data->start_class, and_withp, regnode_ssc);
5407 flags &= ~SCF_DO_STCLASS_AND;
5408 StructCopy(&accum, data->start_class, regnode_ssc);
5409 flags |= SCF_DO_STCLASS_OR;
5416 else if (PL_regkind[OP(scan)] == TRIE) {
5417 reg_trie_data *trie = (reg_trie_data*)RExC_rxi->data->data[ ARG(scan) ];
5420 min += trie->minlen;
5421 delta += (trie->maxlen - trie->minlen);
5422 flags &= ~SCF_DO_STCLASS; /* xxx */
5423 if (flags & SCF_DO_SUBSTR) {
5424 /* Cannot expect anything... */
5425 scan_commit(pRExC_state, data, minlenp, is_inf);
5426 data->pos_min += trie->minlen;
5427 data->pos_delta += (trie->maxlen - trie->minlen);
5428 if (trie->maxlen != trie->minlen)
5429 data->longest = &(data->longest_float);
5431 if (trie->jump) /* no more substrings -- for now /grr*/
5432 flags &= ~SCF_DO_SUBSTR;
5434 #endif /* old or new */
5435 #endif /* TRIE_STUDY_OPT */
5437 /* Else: zero-length, ignore. */
5438 scan = regnext(scan);
5440 /* If we are exiting a recursion we can unset its recursed bit
5441 * and allow ourselves to enter it again - no danger of an
5442 * infinite loop there.
5443 if (stopparen > -1 && recursed) {
5444 DEBUG_STUDYDATA("unset:", data,depth);
5445 PAREN_UNSET( recursed, stopparen);
5449 DEBUG_STUDYDATA("frame-end:",data,depth);
5450 DEBUG_PEEP("fend", scan, depth);
5451 /* restore previous context */
5454 stopparen = frame->stop;
5455 recursed_depth = frame->prev_recursed_depth;
5458 frame = frame->prev;
5459 goto fake_study_recurse;
5464 DEBUG_STUDYDATA("pre-fin:",data,depth);
5467 *deltap = is_inf_internal ? SSize_t_MAX : delta;
5469 if (flags & SCF_DO_SUBSTR && is_inf)
5470 data->pos_delta = SSize_t_MAX - data->pos_min;
5471 if (is_par > (I32)U8_MAX)
5473 if (is_par && pars==1 && data) {
5474 data->flags |= SF_IN_PAR;
5475 data->flags &= ~SF_HAS_PAR;
5477 else if (pars && data) {
5478 data->flags |= SF_HAS_PAR;
5479 data->flags &= ~SF_IN_PAR;
5481 if (flags & SCF_DO_STCLASS_OR)
5482 ssc_and(pRExC_state, data->start_class, (regnode_charclass *) and_withp);
5483 if (flags & SCF_TRIE_RESTUDY)
5484 data->flags |= SCF_TRIE_RESTUDY;
5486 DEBUG_STUDYDATA("post-fin:",data,depth);
5489 SSize_t final_minlen= min < stopmin ? min : stopmin;
5491 if (!(RExC_seen & REG_UNBOUNDED_QUANTIFIER_SEEN) && (RExC_maxlen < final_minlen + delta)) {
5492 RExC_maxlen = final_minlen + delta;
5494 return final_minlen;
5500 S_add_data(RExC_state_t* const pRExC_state, const char* const s, const U32 n)
5502 U32 count = RExC_rxi->data ? RExC_rxi->data->count : 0;
5504 PERL_ARGS_ASSERT_ADD_DATA;
5506 Renewc(RExC_rxi->data,
5507 sizeof(*RExC_rxi->data) + sizeof(void*) * (count + n - 1),
5508 char, struct reg_data);
5510 Renew(RExC_rxi->data->what, count + n, U8);
5512 Newx(RExC_rxi->data->what, n, U8);
5513 RExC_rxi->data->count = count + n;
5514 Copy(s, RExC_rxi->data->what + count, n, U8);
5518 /*XXX: todo make this not included in a non debugging perl, but appears to be
5519 * used anyway there, in 'use re' */
5520 #ifndef PERL_IN_XSUB_RE
5522 Perl_reginitcolors(pTHX)
5524 const char * const s = PerlEnv_getenv("PERL_RE_COLORS");
5526 char *t = savepv(s);
5530 t = strchr(t, '\t');
5536 PL_colors[i] = t = (char *)"";
5541 PL_colors[i++] = (char *)"";
5548 #ifdef TRIE_STUDY_OPT
5549 #define CHECK_RESTUDY_GOTO_butfirst(dOsomething) \
5552 (data.flags & SCF_TRIE_RESTUDY) \
5560 #define CHECK_RESTUDY_GOTO_butfirst
5564 * pregcomp - compile a regular expression into internal code
5566 * Decides which engine's compiler to call based on the hint currently in
5570 #ifndef PERL_IN_XSUB_RE
5572 /* return the currently in-scope regex engine (or the default if none) */
5574 regexp_engine const *
5575 Perl_current_re_engine(pTHX)
5577 if (IN_PERL_COMPILETIME) {
5578 HV * const table = GvHV(PL_hintgv);
5581 if (!table || !(PL_hints & HINT_LOCALIZE_HH))
5582 return &reh_regexp_engine;
5583 ptr = hv_fetchs(table, "regcomp", FALSE);
5584 if ( !(ptr && SvIOK(*ptr) && SvIV(*ptr)))
5585 return &reh_regexp_engine;
5586 return INT2PTR(regexp_engine*,SvIV(*ptr));
5590 if (!PL_curcop->cop_hints_hash)
5591 return &reh_regexp_engine;
5592 ptr = cop_hints_fetch_pvs(PL_curcop, "regcomp", 0);
5593 if ( !(ptr && SvIOK(ptr) && SvIV(ptr)))
5594 return &reh_regexp_engine;
5595 return INT2PTR(regexp_engine*,SvIV(ptr));
5601 Perl_pregcomp(pTHX_ SV * const pattern, const U32 flags)
5603 regexp_engine const *eng = current_re_engine();
5604 GET_RE_DEBUG_FLAGS_DECL;
5606 PERL_ARGS_ASSERT_PREGCOMP;
5608 /* Dispatch a request to compile a regexp to correct regexp engine. */
5610 PerlIO_printf(Perl_debug_log, "Using engine %"UVxf"\n",
5613 return CALLREGCOMP_ENG(eng, pattern, flags);
5617 /* public(ish) entry point for the perl core's own regex compiling code.
5618 * It's actually a wrapper for Perl_re_op_compile that only takes an SV
5619 * pattern rather than a list of OPs, and uses the internal engine rather
5620 * than the current one */
5623 Perl_re_compile(pTHX_ SV * const pattern, U32 rx_flags)
5625 SV *pat = pattern; /* defeat constness! */
5626 PERL_ARGS_ASSERT_RE_COMPILE;
5627 return Perl_re_op_compile(aTHX_ &pat, 1, NULL,
5628 #ifdef PERL_IN_XSUB_RE
5633 NULL, NULL, rx_flags, 0);
5637 /* upgrade pattern pat_p of length plen_p to UTF8, and if there are code
5638 * blocks, recalculate the indices. Update pat_p and plen_p in-place to
5639 * point to the realloced string and length.
5641 * This is essentially a copy of Perl_bytes_to_utf8() with the code index
5645 S_pat_upgrade_to_utf8(pTHX_ RExC_state_t * const pRExC_state,
5646 char **pat_p, STRLEN *plen_p, int num_code_blocks)
5648 U8 *const src = (U8*)*pat_p;
5651 STRLEN s = 0, d = 0;
5653 GET_RE_DEBUG_FLAGS_DECL;
5655 DEBUG_PARSE_r(PerlIO_printf(Perl_debug_log,
5656 "UTF8 mismatch! Converting to utf8 for resizing and compile\n"));
5658 Newx(dst, *plen_p * 2 + 1, U8);
5660 while (s < *plen_p) {
5661 if (NATIVE_BYTE_IS_INVARIANT(src[s]))
5664 dst[d++] = UTF8_EIGHT_BIT_HI(src[s]);
5665 dst[d] = UTF8_EIGHT_BIT_LO(src[s]);
5667 if (n < num_code_blocks) {
5668 if (!do_end && pRExC_state->code_blocks[n].start == s) {
5669 pRExC_state->code_blocks[n].start = d;
5670 assert(dst[d] == '(');
5673 else if (do_end && pRExC_state->code_blocks[n].end == s) {
5674 pRExC_state->code_blocks[n].end = d;
5675 assert(dst[d] == ')');
5685 *pat_p = (char*) dst;
5687 RExC_orig_utf8 = RExC_utf8 = 1;
5692 /* S_concat_pat(): concatenate a list of args to the pattern string pat,
5693 * while recording any code block indices, and handling overloading,
5694 * nested qr// objects etc. If pat is null, it will allocate a new
5695 * string, or just return the first arg, if there's only one.
5697 * Returns the malloced/updated pat.
5698 * patternp and pat_count is the array of SVs to be concatted;
5699 * oplist is the optional list of ops that generated the SVs;
5700 * recompile_p is a pointer to a boolean that will be set if
5701 * the regex will need to be recompiled.
5702 * delim, if non-null is an SV that will be inserted between each element
5706 S_concat_pat(pTHX_ RExC_state_t * const pRExC_state,
5707 SV *pat, SV ** const patternp, int pat_count,
5708 OP *oplist, bool *recompile_p, SV *delim)
5712 bool use_delim = FALSE;
5713 bool alloced = FALSE;
5715 /* if we know we have at least two args, create an empty string,
5716 * then concatenate args to that. For no args, return an empty string */
5717 if (!pat && pat_count != 1) {
5723 for (svp = patternp; svp < patternp + pat_count; svp++) {
5726 STRLEN orig_patlen = 0;
5728 SV *msv = use_delim ? delim : *svp;
5729 if (!msv) msv = &PL_sv_undef;
5731 /* if we've got a delimiter, we go round the loop twice for each
5732 * svp slot (except the last), using the delimiter the second
5741 if (SvTYPE(msv) == SVt_PVAV) {
5742 /* we've encountered an interpolated array within
5743 * the pattern, e.g. /...@a..../. Expand the list of elements,
5744 * then recursively append elements.
5745 * The code in this block is based on S_pushav() */
5747 AV *const av = (AV*)msv;
5748 const SSize_t maxarg = AvFILL(av) + 1;
5752 assert(oplist->op_type == OP_PADAV
5753 || oplist->op_type == OP_RV2AV);
5754 oplist = OP_SIBLING(oplist);
5757 if (SvRMAGICAL(av)) {
5760 Newx(array, maxarg, SV*);
5762 for (i=0; i < maxarg; i++) {
5763 SV ** const svp = av_fetch(av, i, FALSE);
5764 array[i] = svp ? *svp : &PL_sv_undef;
5768 array = AvARRAY(av);
5770 pat = S_concat_pat(aTHX_ pRExC_state, pat,
5771 array, maxarg, NULL, recompile_p,
5773 GvSV((gv_fetchpvs("\"", GV_ADDMULTI, SVt_PV))));
5779 /* we make the assumption here that each op in the list of
5780 * op_siblings maps to one SV pushed onto the stack,
5781 * except for code blocks, with have both an OP_NULL and
5783 * This allows us to match up the list of SVs against the
5784 * list of OPs to find the next code block.
5786 * Note that PUSHMARK PADSV PADSV ..
5788 * PADRANGE PADSV PADSV ..
5789 * so the alignment still works. */
5792 if (oplist->op_type == OP_NULL
5793 && (oplist->op_flags & OPf_SPECIAL))
5795 assert(n < pRExC_state->num_code_blocks);
5796 pRExC_state->code_blocks[n].start = pat ? SvCUR(pat) : 0;
5797 pRExC_state->code_blocks[n].block = oplist;
5798 pRExC_state->code_blocks[n].src_regex = NULL;
5801 oplist = OP_SIBLING(oplist); /* skip CONST */
5804 oplist = OP_SIBLING(oplist);;
5807 /* apply magic and QR overloading to arg */
5810 if (SvROK(msv) && SvAMAGIC(msv)) {
5811 SV *sv = AMG_CALLunary(msv, regexp_amg);
5815 if (SvTYPE(sv) != SVt_REGEXP)
5816 Perl_croak(aTHX_ "Overloaded qr did not return a REGEXP");
5821 /* try concatenation overload ... */
5822 if (pat && (SvAMAGIC(pat) || SvAMAGIC(msv)) &&
5823 (sv = amagic_call(pat, msv, concat_amg, AMGf_assign)))
5826 /* overloading involved: all bets are off over literal
5827 * code. Pretend we haven't seen it */
5828 pRExC_state->num_code_blocks -= n;
5832 /* ... or failing that, try "" overload */
5833 while (SvAMAGIC(msv)
5834 && (sv = AMG_CALLunary(msv, string_amg))
5838 && SvRV(msv) == SvRV(sv))
5843 if (SvROK(msv) && SvTYPE(SvRV(msv)) == SVt_REGEXP)
5847 /* this is a partially unrolled
5848 * sv_catsv_nomg(pat, msv);
5849 * that allows us to adjust code block indices if
5852 char *dst = SvPV_force_nomg(pat, dlen);
5854 if (SvUTF8(msv) && !SvUTF8(pat)) {
5855 S_pat_upgrade_to_utf8(aTHX_ pRExC_state, &dst, &dlen, n);
5856 sv_setpvn(pat, dst, dlen);
5859 sv_catsv_nomg(pat, msv);
5866 pRExC_state->code_blocks[n-1].end = SvCUR(pat)-1;
5869 /* extract any code blocks within any embedded qr//'s */
5870 if (rx && SvTYPE(rx) == SVt_REGEXP
5871 && RX_ENGINE((REGEXP*)rx)->op_comp)
5874 RXi_GET_DECL(ReANY((REGEXP *)rx), ri);
5875 if (ri->num_code_blocks) {
5877 /* the presence of an embedded qr// with code means
5878 * we should always recompile: the text of the
5879 * qr// may not have changed, but it may be a
5880 * different closure than last time */
5882 Renew(pRExC_state->code_blocks,
5883 pRExC_state->num_code_blocks + ri->num_code_blocks,
5884 struct reg_code_block);
5885 pRExC_state->num_code_blocks += ri->num_code_blocks;
5887 for (i=0; i < ri->num_code_blocks; i++) {
5888 struct reg_code_block *src, *dst;
5889 STRLEN offset = orig_patlen
5890 + ReANY((REGEXP *)rx)->pre_prefix;
5891 assert(n < pRExC_state->num_code_blocks);
5892 src = &ri->code_blocks[i];
5893 dst = &pRExC_state->code_blocks[n];
5894 dst->start = src->start + offset;
5895 dst->end = src->end + offset;
5896 dst->block = src->block;
5897 dst->src_regex = (REGEXP*) SvREFCNT_inc( (SV*)
5906 /* avoid calling magic multiple times on a single element e.g. =~ $qr */
5915 /* see if there are any run-time code blocks in the pattern.
5916 * False positives are allowed */
5919 S_has_runtime_code(pTHX_ RExC_state_t * const pRExC_state,
5920 char *pat, STRLEN plen)
5925 PERL_UNUSED_CONTEXT;
5927 for (s = 0; s < plen; s++) {
5928 if (n < pRExC_state->num_code_blocks
5929 && s == pRExC_state->code_blocks[n].start)
5931 s = pRExC_state->code_blocks[n].end;
5935 /* TODO ideally should handle [..], (#..), /#.../x to reduce false
5937 if (pat[s] == '(' && s+2 <= plen && pat[s+1] == '?' &&
5939 || (s + 2 <= plen && pat[s+2] == '?' && pat[s+3] == '{'))
5946 /* Handle run-time code blocks. We will already have compiled any direct
5947 * or indirect literal code blocks. Now, take the pattern 'pat' and make a
5948 * copy of it, but with any literal code blocks blanked out and
5949 * appropriate chars escaped; then feed it into
5951 * eval "qr'modified_pattern'"
5955 * a\bc(?{"this was literal"})def'ghi\\jkl(?{"this is runtime"})mno
5959 * qr'a\\bc_______________________def\'ghi\\\\jkl(?{"this is runtime"})mno'
5961 * After eval_sv()-ing that, grab any new code blocks from the returned qr
5962 * and merge them with any code blocks of the original regexp.
5964 * If the pat is non-UTF8, while the evalled qr is UTF8, don't merge;
5965 * instead, just save the qr and return FALSE; this tells our caller that
5966 * the original pattern needs upgrading to utf8.
5970 S_compile_runtime_code(pTHX_ RExC_state_t * const pRExC_state,
5971 char *pat, STRLEN plen)
5975 GET_RE_DEBUG_FLAGS_DECL;
5977 if (pRExC_state->runtime_code_qr) {
5978 /* this is the second time we've been called; this should
5979 * only happen if the main pattern got upgraded to utf8
5980 * during compilation; re-use the qr we compiled first time
5981 * round (which should be utf8 too)
5983 qr = pRExC_state->runtime_code_qr;
5984 pRExC_state->runtime_code_qr = NULL;
5985 assert(RExC_utf8 && SvUTF8(qr));
5991 int newlen = plen + 6; /* allow for "qr''x\0" extra chars */
5995 /* determine how many extra chars we need for ' and \ escaping */
5996 for (s = 0; s < plen; s++) {
5997 if (pat[s] == '\'' || pat[s] == '\\')
6001 Newx(newpat, newlen, char);
6003 *p++ = 'q'; *p++ = 'r'; *p++ = '\'';
6005 for (s = 0; s < plen; s++) {
6006 if (n < pRExC_state->num_code_blocks
6007 && s == pRExC_state->code_blocks[n].start)
6009 /* blank out literal code block */
6010 assert(pat[s] == '(');
6011 while (s <= pRExC_state->code_blocks[n].end) {
6019 if (pat[s] == '\'' || pat[s] == '\\')
6024 if (pRExC_state->pm_flags & RXf_PMf_EXTENDED)
6028 PerlIO_printf(Perl_debug_log,
6029 "%sre-parsing pattern for runtime code:%s %s\n",
6030 PL_colors[4],PL_colors[5],newpat);
6033 sv = newSVpvn_flags(newpat, p-newpat-1, RExC_utf8 ? SVf_UTF8 : 0);
6039 PUSHSTACKi(PERLSI_REQUIRE);
6040 /* G_RE_REPARSING causes the toker to collapse \\ into \ when
6041 * parsing qr''; normally only q'' does this. It also alters
6043 eval_sv(sv, G_SCALAR|G_RE_REPARSING);
6044 SvREFCNT_dec_NN(sv);
6049 SV * const errsv = ERRSV;
6050 if (SvTRUE_NN(errsv))
6052 Safefree(pRExC_state->code_blocks);
6053 /* use croak_sv ? */
6054 Perl_croak_nocontext("%"SVf, SVfARG(errsv));
6057 assert(SvROK(qr_ref));
6059 assert(SvTYPE(qr) == SVt_REGEXP && RX_ENGINE((REGEXP*)qr)->op_comp);
6060 /* the leaving below frees the tmp qr_ref.
6061 * Give qr a life of its own */
6069 if (!RExC_utf8 && SvUTF8(qr)) {
6070 /* first time through; the pattern got upgraded; save the
6071 * qr for the next time through */
6072 assert(!pRExC_state->runtime_code_qr);
6073 pRExC_state->runtime_code_qr = qr;
6078 /* extract any code blocks within the returned qr// */
6081 /* merge the main (r1) and run-time (r2) code blocks into one */
6083 RXi_GET_DECL(ReANY((REGEXP *)qr), r2);
6084 struct reg_code_block *new_block, *dst;
6085 RExC_state_t * const r1 = pRExC_state; /* convenient alias */
6088 if (!r2->num_code_blocks) /* we guessed wrong */
6090 SvREFCNT_dec_NN(qr);
6095 r1->num_code_blocks + r2->num_code_blocks,
6096 struct reg_code_block);
6099 while ( i1 < r1->num_code_blocks
6100 || i2 < r2->num_code_blocks)
6102 struct reg_code_block *src;
6105 if (i1 == r1->num_code_blocks) {
6106 src = &r2->code_blocks[i2++];
6109 else if (i2 == r2->num_code_blocks)
6110 src = &r1->code_blocks[i1++];
6111 else if ( r1->code_blocks[i1].start
6112 < r2->code_blocks[i2].start)
6114 src = &r1->code_blocks[i1++];
6115 assert(src->end < r2->code_blocks[i2].start);
6118 assert( r1->code_blocks[i1].start
6119 > r2->code_blocks[i2].start);
6120 src = &r2->code_blocks[i2++];
6122 assert(src->end < r1->code_blocks[i1].start);
6125 assert(pat[src->start] == '(');
6126 assert(pat[src->end] == ')');
6127 dst->start = src->start;
6128 dst->end = src->end;
6129 dst->block = src->block;
6130 dst->src_regex = is_qr ? (REGEXP*) SvREFCNT_inc( (SV*) qr)
6134 r1->num_code_blocks += r2->num_code_blocks;
6135 Safefree(r1->code_blocks);
6136 r1->code_blocks = new_block;
6139 SvREFCNT_dec_NN(qr);
6145 S_setup_longest(pTHX_ RExC_state_t *pRExC_state, SV* sv_longest,
6146 SV** rx_utf8, SV** rx_substr, SSize_t* rx_end_shift,
6147 SSize_t lookbehind, SSize_t offset, SSize_t *minlen,
6148 STRLEN longest_length, bool eol, bool meol)
6150 /* This is the common code for setting up the floating and fixed length
6151 * string data extracted from Perl_re_op_compile() below. Returns a boolean
6152 * as to whether succeeded or not */
6157 if (! (longest_length
6158 || (eol /* Can't have SEOL and MULTI */
6159 && (! meol || (RExC_flags & RXf_PMf_MULTILINE)))
6161 /* See comments for join_exact for why REG_UNFOLDED_MULTI_SEEN */
6162 || (RExC_seen & REG_UNFOLDED_MULTI_SEEN))
6167 /* copy the information about the longest from the reg_scan_data
6168 over to the program. */
6169 if (SvUTF8(sv_longest)) {
6170 *rx_utf8 = sv_longest;
6173 *rx_substr = sv_longest;
6176 /* end_shift is how many chars that must be matched that
6177 follow this item. We calculate it ahead of time as once the
6178 lookbehind offset is added in we lose the ability to correctly
6180 ml = minlen ? *(minlen) : (SSize_t)longest_length;
6181 *rx_end_shift = ml - offset
6182 - longest_length + (SvTAIL(sv_longest) != 0)
6185 t = (eol/* Can't have SEOL and MULTI */
6186 && (! meol || (RExC_flags & RXf_PMf_MULTILINE)));
6187 fbm_compile(sv_longest, t ? FBMcf_TAIL : 0);
6193 * Perl_re_op_compile - the perl internal RE engine's function to compile a
6194 * regular expression into internal code.
6195 * The pattern may be passed either as:
6196 * a list of SVs (patternp plus pat_count)
6197 * a list of OPs (expr)
6198 * If both are passed, the SV list is used, but the OP list indicates
6199 * which SVs are actually pre-compiled code blocks
6201 * The SVs in the list have magic and qr overloading applied to them (and
6202 * the list may be modified in-place with replacement SVs in the latter
6205 * If the pattern hasn't changed from old_re, then old_re will be
6208 * eng is the current engine. If that engine has an op_comp method, then
6209 * handle directly (i.e. we assume that op_comp was us); otherwise, just
6210 * do the initial concatenation of arguments and pass on to the external
6213 * If is_bare_re is not null, set it to a boolean indicating whether the
6214 * arg list reduced (after overloading) to a single bare regex which has
6215 * been returned (i.e. /$qr/).
6217 * orig_rx_flags contains RXf_* flags. See perlreapi.pod for more details.
6219 * pm_flags contains the PMf_* flags, typically based on those from the
6220 * pm_flags field of the related PMOP. Currently we're only interested in
6221 * PMf_HAS_CV, PMf_IS_QR, PMf_USE_RE_EVAL.
6223 * We can't allocate space until we know how big the compiled form will be,
6224 * but we can't compile it (and thus know how big it is) until we've got a
6225 * place to put the code. So we cheat: we compile it twice, once with code
6226 * generation turned off and size counting turned on, and once "for real".
6227 * This also means that we don't allocate space until we are sure that the
6228 * thing really will compile successfully, and we never have to move the
6229 * code and thus invalidate pointers into it. (Note that it has to be in
6230 * one piece because free() must be able to free it all.) [NB: not true in perl]
6232 * Beware that the optimization-preparation code in here knows about some
6233 * of the structure of the compiled regexp. [I'll say.]
6237 Perl_re_op_compile(pTHX_ SV ** const patternp, int pat_count,
6238 OP *expr, const regexp_engine* eng, REGEXP *old_re,
6239 bool *is_bare_re, U32 orig_rx_flags, U32 pm_flags)
6243 regexp_internal *ri;
6251 SV *code_blocksv = NULL;
6252 SV** new_patternp = patternp;
6254 /* these are all flags - maybe they should be turned
6255 * into a single int with different bit masks */
6256 I32 sawlookahead = 0;
6261 regex_charset initial_charset = get_regex_charset(orig_rx_flags);
6263 bool runtime_code = 0;
6265 RExC_state_t RExC_state;
6266 RExC_state_t * const pRExC_state = &RExC_state;
6267 #ifdef TRIE_STUDY_OPT
6269 RExC_state_t copyRExC_state;
6271 GET_RE_DEBUG_FLAGS_DECL;
6273 PERL_ARGS_ASSERT_RE_OP_COMPILE;
6275 DEBUG_r(if (!PL_colorset) reginitcolors());
6277 #ifndef PERL_IN_XSUB_RE
6278 /* Initialize these here instead of as-needed, as is quick and avoids
6279 * having to test them each time otherwise */
6280 if (! PL_AboveLatin1) {
6281 PL_AboveLatin1 = _new_invlist_C_array(AboveLatin1_invlist);
6282 PL_Latin1 = _new_invlist_C_array(Latin1_invlist);
6283 PL_UpperLatin1 = _new_invlist_C_array(UpperLatin1_invlist);
6284 PL_utf8_foldable = _new_invlist_C_array(_Perl_Any_Folds_invlist);
6285 PL_HasMultiCharFold =
6286 _new_invlist_C_array(_Perl_Folds_To_Multi_Char_invlist);
6290 pRExC_state->code_blocks = NULL;
6291 pRExC_state->num_code_blocks = 0;
6294 *is_bare_re = FALSE;
6296 if (expr && (expr->op_type == OP_LIST ||
6297 (expr->op_type == OP_NULL && expr->op_targ == OP_LIST))) {
6298 /* allocate code_blocks if needed */
6302 for (o = cLISTOPx(expr)->op_first; o; o = OP_SIBLING(o))
6303 if (o->op_type == OP_NULL && (o->op_flags & OPf_SPECIAL))
6304 ncode++; /* count of DO blocks */
6306 pRExC_state->num_code_blocks = ncode;
6307 Newx(pRExC_state->code_blocks, ncode, struct reg_code_block);
6312 /* compile-time pattern with just OP_CONSTs and DO blocks */
6317 /* find how many CONSTs there are */
6320 if (expr->op_type == OP_CONST)
6323 for (o = cLISTOPx(expr)->op_first; o; o = OP_SIBLING(o)) {
6324 if (o->op_type == OP_CONST)
6328 /* fake up an SV array */
6330 assert(!new_patternp);
6331 Newx(new_patternp, n, SV*);
6332 SAVEFREEPV(new_patternp);
6336 if (expr->op_type == OP_CONST)
6337 new_patternp[n] = cSVOPx_sv(expr);
6339 for (o = cLISTOPx(expr)->op_first; o; o = OP_SIBLING(o)) {
6340 if (o->op_type == OP_CONST)
6341 new_patternp[n++] = cSVOPo_sv;
6346 DEBUG_PARSE_r(PerlIO_printf(Perl_debug_log,
6347 "Assembling pattern from %d elements%s\n", pat_count,
6348 orig_rx_flags & RXf_SPLIT ? " for split" : ""));
6350 /* set expr to the first arg op */
6352 if (pRExC_state->num_code_blocks
6353 && expr->op_type != OP_CONST)
6355 expr = cLISTOPx(expr)->op_first;
6356 assert( expr->op_type == OP_PUSHMARK
6357 || (expr->op_type == OP_NULL && expr->op_targ == OP_PUSHMARK)
6358 || expr->op_type == OP_PADRANGE);
6359 expr = OP_SIBLING(expr);
6362 pat = S_concat_pat(aTHX_ pRExC_state, NULL, new_patternp, pat_count,
6363 expr, &recompile, NULL);
6365 /* handle bare (possibly after overloading) regex: foo =~ $re */
6370 if (SvTYPE(re) == SVt_REGEXP) {
6374 Safefree(pRExC_state->code_blocks);
6375 DEBUG_PARSE_r(PerlIO_printf(Perl_debug_log,
6376 "Precompiled pattern%s\n",
6377 orig_rx_flags & RXf_SPLIT ? " for split" : ""));
6383 exp = SvPV_nomg(pat, plen);
6385 if (!eng->op_comp) {
6386 if ((SvUTF8(pat) && IN_BYTES)
6387 || SvGMAGICAL(pat) || SvAMAGIC(pat))
6389 /* make a temporary copy; either to convert to bytes,
6390 * or to avoid repeating get-magic / overloaded stringify */
6391 pat = newSVpvn_flags(exp, plen, SVs_TEMP |
6392 (IN_BYTES ? 0 : SvUTF8(pat)));
6394 Safefree(pRExC_state->code_blocks);
6395 return CALLREGCOMP_ENG(eng, pat, orig_rx_flags);
6398 /* ignore the utf8ness if the pattern is 0 length */
6399 RExC_utf8 = RExC_orig_utf8 = (plen == 0 || IN_BYTES) ? 0 : SvUTF8(pat);
6400 RExC_uni_semantics = 0;
6401 RExC_contains_locale = 0;
6402 RExC_contains_i = 0;
6403 pRExC_state->runtime_code_qr = NULL;
6406 SV *dsv= sv_newmortal();
6407 RE_PV_QUOTED_DECL(s, RExC_utf8, dsv, exp, plen, 60);
6408 PerlIO_printf(Perl_debug_log, "%sCompiling REx%s %s\n",
6409 PL_colors[4],PL_colors[5],s);
6413 /* we jump here if we upgrade the pattern to utf8 and have to
6416 if ((pm_flags & PMf_USE_RE_EVAL)
6417 /* this second condition covers the non-regex literal case,
6418 * i.e. $foo =~ '(?{})'. */
6419 || (IN_PERL_COMPILETIME && (PL_hints & HINT_RE_EVAL))
6421 runtime_code = S_has_runtime_code(aTHX_ pRExC_state, exp, plen);
6423 /* return old regex if pattern hasn't changed */
6424 /* XXX: note in the below we have to check the flags as well as the
6427 * Things get a touch tricky as we have to compare the utf8 flag
6428 * independently from the compile flags. */
6432 && !!RX_UTF8(old_re) == !!RExC_utf8
6433 && ( RX_COMPFLAGS(old_re) == ( orig_rx_flags & RXf_PMf_FLAGCOPYMASK ) )
6434 && RX_PRECOMP(old_re)
6435 && RX_PRELEN(old_re) == plen
6436 && memEQ(RX_PRECOMP(old_re), exp, plen)
6437 && !runtime_code /* with runtime code, always recompile */ )
6439 Safefree(pRExC_state->code_blocks);
6443 rx_flags = orig_rx_flags;
6445 if (rx_flags & PMf_FOLD) {
6446 RExC_contains_i = 1;
6448 if (RExC_utf8 && initial_charset == REGEX_DEPENDS_CHARSET) {
6450 /* Set to use unicode semantics if the pattern is in utf8 and has the
6451 * 'depends' charset specified, as it means unicode when utf8 */
6452 set_regex_charset(&rx_flags, REGEX_UNICODE_CHARSET);
6456 RExC_flags = rx_flags;
6457 RExC_pm_flags = pm_flags;
6460 if (TAINTING_get && TAINT_get)
6461 Perl_croak(aTHX_ "Eval-group in insecure regular expression");
6463 if (!S_compile_runtime_code(aTHX_ pRExC_state, exp, plen)) {
6464 /* whoops, we have a non-utf8 pattern, whilst run-time code
6465 * got compiled as utf8. Try again with a utf8 pattern */
6466 S_pat_upgrade_to_utf8(aTHX_ pRExC_state, &exp, &plen,
6467 pRExC_state->num_code_blocks);
6468 goto redo_first_pass;
6471 assert(!pRExC_state->runtime_code_qr);
6477 RExC_in_lookbehind = 0;
6478 RExC_seen_zerolen = *exp == '^' ? -1 : 0;
6480 RExC_override_recoding = 0;
6481 RExC_in_multi_char_class = 0;
6483 /* First pass: determine size, legality. */
6486 RExC_end = exp + plen;
6491 RExC_emit = (regnode *) &RExC_emit_dummy;
6492 RExC_whilem_seen = 0;
6493 RExC_open_parens = NULL;
6494 RExC_close_parens = NULL;
6496 RExC_paren_names = NULL;
6498 RExC_paren_name_list = NULL;
6500 RExC_recurse = NULL;
6501 RExC_study_chunk_recursed = NULL;
6502 RExC_study_chunk_recursed_bytes= 0;
6503 RExC_recurse_count = 0;
6504 pRExC_state->code_index = 0;
6506 #if 0 /* REGC() is (currently) a NOP at the first pass.
6507 * Clever compilers notice this and complain. --jhi */
6508 REGC((U8)REG_MAGIC, (char*)RExC_emit);
6511 PerlIO_printf(Perl_debug_log, "Starting first pass (sizing)\n");
6513 RExC_lastparse=NULL;
6515 /* reg may croak on us, not giving us a chance to free
6516 pRExC_state->code_blocks. We cannot SAVEFREEPV it now, as we may
6517 need it to survive as long as the regexp (qr/(?{})/).
6518 We must check that code_blocksv is not already set, because we may
6519 have jumped back to restart the sizing pass. */
6520 if (pRExC_state->code_blocks && !code_blocksv) {
6521 code_blocksv = newSV_type(SVt_PV);
6522 SAVEFREESV(code_blocksv);
6523 SvPV_set(code_blocksv, (char *)pRExC_state->code_blocks);
6524 SvLEN_set(code_blocksv, 1); /*sufficient to make sv_clear free it*/
6526 if (reg(pRExC_state, 0, &flags,1) == NULL) {
6527 /* It's possible to write a regexp in ascii that represents Unicode
6528 codepoints outside of the byte range, such as via \x{100}. If we
6529 detect such a sequence we have to convert the entire pattern to utf8
6530 and then recompile, as our sizing calculation will have been based
6531 on 1 byte == 1 character, but we will need to use utf8 to encode
6532 at least some part of the pattern, and therefore must convert the whole
6535 if (flags & RESTART_UTF8) {
6536 S_pat_upgrade_to_utf8(aTHX_ pRExC_state, &exp, &plen,
6537 pRExC_state->num_code_blocks);
6538 goto redo_first_pass;
6540 Perl_croak(aTHX_ "panic: reg returned NULL to re_op_compile for sizing pass, flags=%#"UVxf"", (UV) flags);
6543 SvLEN_set(code_blocksv,0); /* no you can't have it, sv_clear */
6546 PerlIO_printf(Perl_debug_log,
6547 "Required size %"IVdf" nodes\n"
6548 "Starting second pass (creation)\n",
6551 RExC_lastparse=NULL;
6554 /* The first pass could have found things that force Unicode semantics */
6555 if ((RExC_utf8 || RExC_uni_semantics)
6556 && get_regex_charset(rx_flags) == REGEX_DEPENDS_CHARSET)
6558 set_regex_charset(&rx_flags, REGEX_UNICODE_CHARSET);
6561 /* Small enough for pointer-storage convention?
6562 If extralen==0, this means that we will not need long jumps. */
6563 if (RExC_size >= 0x10000L && RExC_extralen)
6564 RExC_size += RExC_extralen;
6567 if (RExC_whilem_seen > 15)
6568 RExC_whilem_seen = 15;
6570 /* Allocate space and zero-initialize. Note, the two step process
6571 of zeroing when in debug mode, thus anything assigned has to
6572 happen after that */
6573 rx = (REGEXP*) newSV_type(SVt_REGEXP);
6575 Newxc(ri, sizeof(regexp_internal) + (unsigned)RExC_size * sizeof(regnode),
6576 char, regexp_internal);
6577 if ( r == NULL || ri == NULL )
6578 FAIL("Regexp out of space");
6580 /* avoid reading uninitialized memory in DEBUGGING code in study_chunk() */
6581 Zero(ri, sizeof(regexp_internal) + (unsigned)RExC_size * sizeof(regnode),
6584 /* bulk initialize base fields with 0. */
6585 Zero(ri, sizeof(regexp_internal), char);
6588 /* non-zero initialization begins here */
6591 r->extflags = rx_flags;
6592 RXp_COMPFLAGS(r) = orig_rx_flags & RXf_PMf_FLAGCOPYMASK;
6594 if (pm_flags & PMf_IS_QR) {
6595 ri->code_blocks = pRExC_state->code_blocks;
6596 ri->num_code_blocks = pRExC_state->num_code_blocks;
6601 for (n = 0; n < pRExC_state->num_code_blocks; n++)
6602 if (pRExC_state->code_blocks[n].src_regex)
6603 SAVEFREESV(pRExC_state->code_blocks[n].src_regex);
6604 SAVEFREEPV(pRExC_state->code_blocks);
6608 bool has_p = ((r->extflags & RXf_PMf_KEEPCOPY) == RXf_PMf_KEEPCOPY);
6609 bool has_charset = (get_regex_charset(r->extflags)
6610 != REGEX_DEPENDS_CHARSET);
6612 /* The caret is output if there are any defaults: if not all the STD
6613 * flags are set, or if no character set specifier is needed */
6615 (((r->extflags & RXf_PMf_STD_PMMOD) != RXf_PMf_STD_PMMOD)
6617 bool has_runon = ((RExC_seen & REG_RUN_ON_COMMENT_SEEN)
6618 == REG_RUN_ON_COMMENT_SEEN);
6619 U16 reganch = (U16)((r->extflags & RXf_PMf_STD_PMMOD)
6620 >> RXf_PMf_STD_PMMOD_SHIFT);
6621 const char *fptr = STD_PAT_MODS; /*"msix"*/
6623 /* Allocate for the worst case, which is all the std flags are turned
6624 * on. If more precision is desired, we could do a population count of
6625 * the flags set. This could be done with a small lookup table, or by
6626 * shifting, masking and adding, or even, when available, assembly
6627 * language for a machine-language population count.
6628 * We never output a minus, as all those are defaults, so are
6629 * covered by the caret */
6630 const STRLEN wraplen = plen + has_p + has_runon
6631 + has_default /* If needs a caret */
6633 /* If needs a character set specifier */
6634 + ((has_charset) ? MAX_CHARSET_NAME_LENGTH : 0)
6635 + (sizeof(STD_PAT_MODS) - 1)
6636 + (sizeof("(?:)") - 1);
6638 Newx(p, wraplen + 1, char); /* +1 for the ending NUL */
6639 r->xpv_len_u.xpvlenu_pv = p;
6641 SvFLAGS(rx) |= SVf_UTF8;
6644 /* If a default, cover it using the caret */
6646 *p++= DEFAULT_PAT_MOD;
6650 const char* const name = get_regex_charset_name(r->extflags, &len);
6651 Copy(name, p, len, char);
6655 *p++ = KEEPCOPY_PAT_MOD; /*'p'*/
6658 while((ch = *fptr++)) {
6666 Copy(RExC_precomp, p, plen, char);
6667 assert ((RX_WRAPPED(rx) - p) < 16);
6668 r->pre_prefix = p - RX_WRAPPED(rx);
6674 SvCUR_set(rx, p - RX_WRAPPED(rx));
6678 r->nparens = RExC_npar - 1; /* set early to validate backrefs */
6680 /* setup various meta data about recursion, this all requires
6681 * RExC_npar to be correctly set, and a bit later on we clear it */
6682 if (RExC_seen & REG_RECURSE_SEEN) {
6683 Newxz(RExC_open_parens, RExC_npar,regnode *);
6684 SAVEFREEPV(RExC_open_parens);
6685 Newxz(RExC_close_parens,RExC_npar,regnode *);
6686 SAVEFREEPV(RExC_close_parens);
6688 if (RExC_seen & (REG_RECURSE_SEEN | REG_GOSTART_SEEN)) {
6689 /* Note, RExC_npar is 1 + the number of parens in a pattern.
6690 * So its 1 if there are no parens. */
6691 RExC_study_chunk_recursed_bytes= (RExC_npar >> 3) +
6692 ((RExC_npar & 0x07) != 0);
6693 Newx(RExC_study_chunk_recursed,
6694 RExC_study_chunk_recursed_bytes * RExC_npar, U8);
6695 SAVEFREEPV(RExC_study_chunk_recursed);
6698 /* Useful during FAIL. */
6699 #ifdef RE_TRACK_PATTERN_OFFSETS
6700 Newxz(ri->u.offsets, 2*RExC_size+1, U32); /* MJD 20001228 */
6701 DEBUG_OFFSETS_r(PerlIO_printf(Perl_debug_log,
6702 "%s %"UVuf" bytes for offset annotations.\n",
6703 ri->u.offsets ? "Got" : "Couldn't get",
6704 (UV)((2*RExC_size+1) * sizeof(U32))));
6706 SetProgLen(ri,RExC_size);
6710 REH_CALL_COMP_BEGIN_HOOK(pRExC_state->rx);
6712 /* Second pass: emit code. */
6713 RExC_flags = rx_flags; /* don't let top level (?i) bleed */
6714 RExC_pm_flags = pm_flags;
6716 RExC_end = exp + plen;
6719 RExC_emit_start = ri->program;
6720 RExC_emit = ri->program;
6721 RExC_emit_bound = ri->program + RExC_size + 1;
6722 pRExC_state->code_index = 0;
6724 REGC((U8)REG_MAGIC, (char*) RExC_emit++);
6725 if (reg(pRExC_state, 0, &flags,1) == NULL) {
6727 Perl_croak(aTHX_ "panic: reg returned NULL to re_op_compile for generation pass, flags=%#"UVxf"", (UV) flags);
6729 /* XXXX To minimize changes to RE engine we always allocate
6730 3-units-long substrs field. */
6731 Newx(r->substrs, 1, struct reg_substr_data);
6732 if (RExC_recurse_count) {
6733 Newxz(RExC_recurse,RExC_recurse_count,regnode *);
6734 SAVEFREEPV(RExC_recurse);
6738 r->minlen = minlen = sawlookahead = sawplus = sawopen = sawminmod = 0;
6739 Zero(r->substrs, 1, struct reg_substr_data);
6740 if (RExC_study_chunk_recursed)
6741 Zero(RExC_study_chunk_recursed,
6742 RExC_study_chunk_recursed_bytes * RExC_npar, U8);
6744 #ifdef TRIE_STUDY_OPT
6746 StructCopy(&zero_scan_data, &data, scan_data_t);
6747 copyRExC_state = RExC_state;
6750 DEBUG_OPTIMISE_r(PerlIO_printf(Perl_debug_log,"Restudying\n"));
6752 RExC_state = copyRExC_state;
6753 if (seen & REG_TOP_LEVEL_BRANCHES_SEEN)
6754 RExC_seen |= REG_TOP_LEVEL_BRANCHES_SEEN;
6756 RExC_seen &= ~REG_TOP_LEVEL_BRANCHES_SEEN;
6757 StructCopy(&zero_scan_data, &data, scan_data_t);
6760 StructCopy(&zero_scan_data, &data, scan_data_t);
6763 /* Dig out information for optimizations. */
6764 r->extflags = RExC_flags; /* was pm_op */
6765 /*dmq: removed as part of de-PMOP: pm->op_pmflags = RExC_flags; */
6768 SvUTF8_on(rx); /* Unicode in it? */
6769 ri->regstclass = NULL;
6770 if (RExC_naughty >= 10) /* Probably an expensive pattern. */
6771 r->intflags |= PREGf_NAUGHTY;
6772 scan = ri->program + 1; /* First BRANCH. */
6774 /* testing for BRANCH here tells us whether there is "must appear"
6775 data in the pattern. If there is then we can use it for optimisations */
6776 if (!(RExC_seen & REG_TOP_LEVEL_BRANCHES_SEEN)) { /* Only one top-level choice.
6779 STRLEN longest_float_length, longest_fixed_length;
6780 regnode_ssc ch_class; /* pointed to by data */
6782 SSize_t last_close = 0; /* pointed to by data */
6783 regnode *first= scan;
6784 regnode *first_next= regnext(first);
6786 * Skip introductions and multiplicators >= 1
6787 * so that we can extract the 'meat' of the pattern that must
6788 * match in the large if() sequence following.
6789 * NOTE that EXACT is NOT covered here, as it is normally
6790 * picked up by the optimiser separately.
6792 * This is unfortunate as the optimiser isnt handling lookahead
6793 * properly currently.
6796 while ((OP(first) == OPEN && (sawopen = 1)) ||
6797 /* An OR of *one* alternative - should not happen now. */
6798 (OP(first) == BRANCH && OP(first_next) != BRANCH) ||
6799 /* for now we can't handle lookbehind IFMATCH*/
6800 (OP(first) == IFMATCH && !first->flags && (sawlookahead = 1)) ||
6801 (OP(first) == PLUS) ||
6802 (OP(first) == MINMOD) ||
6803 /* An {n,m} with n>0 */
6804 (PL_regkind[OP(first)] == CURLY && ARG1(first) > 0) ||
6805 (OP(first) == NOTHING && PL_regkind[OP(first_next)] != END ))
6808 * the only op that could be a regnode is PLUS, all the rest
6809 * will be regnode_1 or regnode_2.
6811 * (yves doesn't think this is true)
6813 if (OP(first) == PLUS)
6816 if (OP(first) == MINMOD)
6818 first += regarglen[OP(first)];
6820 first = NEXTOPER(first);
6821 first_next= regnext(first);
6824 /* Starting-point info. */
6826 DEBUG_PEEP("first:",first,0);
6827 /* Ignore EXACT as we deal with it later. */
6828 if (PL_regkind[OP(first)] == EXACT) {
6829 if (OP(first) == EXACT)
6830 NOOP; /* Empty, get anchored substr later. */
6832 ri->regstclass = first;
6835 else if (PL_regkind[OP(first)] == TRIE &&
6836 ((reg_trie_data *)ri->data->data[ ARG(first) ])->minlen>0)
6838 /* this can happen only on restudy */
6839 ri->regstclass = construct_ahocorasick_from_trie(pRExC_state, (regnode *)first, 0);
6842 else if (REGNODE_SIMPLE(OP(first)))
6843 ri->regstclass = first;
6844 else if (PL_regkind[OP(first)] == BOUND ||
6845 PL_regkind[OP(first)] == NBOUND)
6846 ri->regstclass = first;
6847 else if (PL_regkind[OP(first)] == BOL) {
6848 r->intflags |= (OP(first) == MBOL
6850 : (OP(first) == SBOL
6853 first = NEXTOPER(first);
6856 else if (OP(first) == GPOS) {
6857 r->intflags |= PREGf_ANCH_GPOS;
6858 first = NEXTOPER(first);
6861 else if ((!sawopen || !RExC_sawback) &&
6863 (OP(first) == STAR &&
6864 PL_regkind[OP(NEXTOPER(first))] == REG_ANY) &&
6865 !(r->intflags & PREGf_ANCH) && !pRExC_state->num_code_blocks)
6867 /* turn .* into ^.* with an implied $*=1 */
6869 (OP(NEXTOPER(first)) == REG_ANY)
6872 r->intflags |= (type | PREGf_IMPLICIT);
6873 first = NEXTOPER(first);
6876 if (sawplus && !sawminmod && !sawlookahead
6877 && (!sawopen || !RExC_sawback)
6878 && !pRExC_state->num_code_blocks) /* May examine pos and $& */
6879 /* x+ must match at the 1st pos of run of x's */
6880 r->intflags |= PREGf_SKIP;
6882 /* Scan is after the zeroth branch, first is atomic matcher. */
6883 #ifdef TRIE_STUDY_OPT
6886 PerlIO_printf(Perl_debug_log, "first at %"IVdf"\n",
6887 (IV)(first - scan + 1))
6891 PerlIO_printf(Perl_debug_log, "first at %"IVdf"\n",
6892 (IV)(first - scan + 1))
6898 * If there's something expensive in the r.e., find the
6899 * longest literal string that must appear and make it the
6900 * regmust. Resolve ties in favor of later strings, since
6901 * the regstart check works with the beginning of the r.e.
6902 * and avoiding duplication strengthens checking. Not a
6903 * strong reason, but sufficient in the absence of others.
6904 * [Now we resolve ties in favor of the earlier string if
6905 * it happens that c_offset_min has been invalidated, since the
6906 * earlier string may buy us something the later one won't.]
6909 data.longest_fixed = newSVpvs("");
6910 data.longest_float = newSVpvs("");
6911 data.last_found = newSVpvs("");
6912 data.longest = &(data.longest_fixed);
6913 ENTER_with_name("study_chunk");
6914 SAVEFREESV(data.longest_fixed);
6915 SAVEFREESV(data.longest_float);
6916 SAVEFREESV(data.last_found);
6918 if (!ri->regstclass) {
6919 ssc_init(pRExC_state, &ch_class);
6920 data.start_class = &ch_class;
6921 stclass_flag = SCF_DO_STCLASS_AND;
6922 } else /* XXXX Check for BOUND? */
6924 data.last_closep = &last_close;
6927 minlen = study_chunk(pRExC_state, &first, &minlen, &fake,
6928 scan + RExC_size, /* Up to end */
6930 SCF_DO_SUBSTR | SCF_WHILEM_VISITED_POS | stclass_flag
6931 | (restudied ? SCF_TRIE_DOING_RESTUDY : 0),
6935 CHECK_RESTUDY_GOTO_butfirst(LEAVE_with_name("study_chunk"));
6938 if ( RExC_npar == 1 && data.longest == &(data.longest_fixed)
6939 && data.last_start_min == 0 && data.last_end > 0
6940 && !RExC_seen_zerolen
6941 && !(RExC_seen & REG_VERBARG_SEEN)
6942 && !(RExC_seen & REG_GPOS_SEEN)
6944 r->extflags |= RXf_CHECK_ALL;
6946 scan_commit(pRExC_state, &data,&minlen,0);
6948 longest_float_length = CHR_SVLEN(data.longest_float);
6950 if (! ((SvCUR(data.longest_fixed) /* ok to leave SvCUR */
6951 && data.offset_fixed == data.offset_float_min
6952 && SvCUR(data.longest_fixed) == SvCUR(data.longest_float)))
6953 && S_setup_longest (aTHX_ pRExC_state,
6957 &(r->float_end_shift),
6958 data.lookbehind_float,
6959 data.offset_float_min,
6961 longest_float_length,
6962 cBOOL(data.flags & SF_FL_BEFORE_EOL),
6963 cBOOL(data.flags & SF_FL_BEFORE_MEOL)))
6965 r->float_min_offset = data.offset_float_min - data.lookbehind_float;
6966 r->float_max_offset = data.offset_float_max;
6967 if (data.offset_float_max < SSize_t_MAX) /* Don't offset infinity */
6968 r->float_max_offset -= data.lookbehind_float;
6969 SvREFCNT_inc_simple_void_NN(data.longest_float);
6972 r->float_substr = r->float_utf8 = NULL;
6973 longest_float_length = 0;
6976 longest_fixed_length = CHR_SVLEN(data.longest_fixed);
6978 if (S_setup_longest (aTHX_ pRExC_state,
6980 &(r->anchored_utf8),
6981 &(r->anchored_substr),
6982 &(r->anchored_end_shift),
6983 data.lookbehind_fixed,
6986 longest_fixed_length,
6987 cBOOL(data.flags & SF_FIX_BEFORE_EOL),
6988 cBOOL(data.flags & SF_FIX_BEFORE_MEOL)))
6990 r->anchored_offset = data.offset_fixed - data.lookbehind_fixed;
6991 SvREFCNT_inc_simple_void_NN(data.longest_fixed);
6994 r->anchored_substr = r->anchored_utf8 = NULL;
6995 longest_fixed_length = 0;
6997 LEAVE_with_name("study_chunk");
7000 && (OP(ri->regstclass) == REG_ANY || OP(ri->regstclass) == SANY))
7001 ri->regstclass = NULL;
7003 if ((!(r->anchored_substr || r->anchored_utf8) || r->anchored_offset)
7005 && ! (ANYOF_FLAGS(data.start_class) & ANYOF_EMPTY_STRING)
7006 && !ssc_is_anything(data.start_class))
7008 const U32 n = add_data(pRExC_state, STR_WITH_LEN("f"));
7010 ssc_finalize(pRExC_state, data.start_class);
7012 Newx(RExC_rxi->data->data[n], 1, regnode_ssc);
7013 StructCopy(data.start_class,
7014 (regnode_ssc*)RExC_rxi->data->data[n],
7016 ri->regstclass = (regnode*)RExC_rxi->data->data[n];
7017 r->intflags &= ~PREGf_SKIP; /* Used in find_byclass(). */
7018 DEBUG_COMPILE_r({ SV *sv = sv_newmortal();
7019 regprop(r, sv, (regnode*)data.start_class, NULL);
7020 PerlIO_printf(Perl_debug_log,
7021 "synthetic stclass \"%s\".\n",
7022 SvPVX_const(sv));});
7023 data.start_class = NULL;
7026 /* A temporary algorithm prefers floated substr to fixed one to dig
7028 if (longest_fixed_length > longest_float_length) {
7029 r->substrs->check_ix = 0;
7030 r->check_end_shift = r->anchored_end_shift;
7031 r->check_substr = r->anchored_substr;
7032 r->check_utf8 = r->anchored_utf8;
7033 r->check_offset_min = r->check_offset_max = r->anchored_offset;
7034 if (r->intflags & (PREGf_ANCH_SBOL|PREGf_ANCH_GPOS))
7035 r->intflags |= PREGf_NOSCAN;
7038 r->substrs->check_ix = 1;
7039 r->check_end_shift = r->float_end_shift;
7040 r->check_substr = r->float_substr;
7041 r->check_utf8 = r->float_utf8;
7042 r->check_offset_min = r->float_min_offset;
7043 r->check_offset_max = r->float_max_offset;
7045 if ((r->check_substr || r->check_utf8) ) {
7046 r->extflags |= RXf_USE_INTUIT;
7047 if (SvTAIL(r->check_substr ? r->check_substr : r->check_utf8))
7048 r->extflags |= RXf_INTUIT_TAIL;
7050 r->substrs->data[0].max_offset = r->substrs->data[0].min_offset;
7052 /* XXX Unneeded? dmq (shouldn't as this is handled elsewhere)
7053 if ( (STRLEN)minlen < longest_float_length )
7054 minlen= longest_float_length;
7055 if ( (STRLEN)minlen < longest_fixed_length )
7056 minlen= longest_fixed_length;
7060 /* Several toplevels. Best we can is to set minlen. */
7062 regnode_ssc ch_class;
7063 SSize_t last_close = 0;
7065 DEBUG_PARSE_r(PerlIO_printf(Perl_debug_log, "\nMulti Top Level\n"));
7067 scan = ri->program + 1;
7068 ssc_init(pRExC_state, &ch_class);
7069 data.start_class = &ch_class;
7070 data.last_closep = &last_close;
7073 minlen = study_chunk(pRExC_state,
7074 &scan, &minlen, &fake, scan + RExC_size, &data, -1, 0, NULL,
7075 SCF_DO_STCLASS_AND|SCF_WHILEM_VISITED_POS|(restudied
7076 ? SCF_TRIE_DOING_RESTUDY
7080 CHECK_RESTUDY_GOTO_butfirst(NOOP);
7082 r->check_substr = r->check_utf8 = r->anchored_substr = r->anchored_utf8
7083 = r->float_substr = r->float_utf8 = NULL;
7085 if (! (ANYOF_FLAGS(data.start_class) & ANYOF_EMPTY_STRING)
7086 && ! ssc_is_anything(data.start_class))
7088 const U32 n = add_data(pRExC_state, STR_WITH_LEN("f"));
7090 ssc_finalize(pRExC_state, data.start_class);
7092 Newx(RExC_rxi->data->data[n], 1, regnode_ssc);
7093 StructCopy(data.start_class,
7094 (regnode_ssc*)RExC_rxi->data->data[n],
7096 ri->regstclass = (regnode*)RExC_rxi->data->data[n];
7097 r->intflags &= ~PREGf_SKIP; /* Used in find_byclass(). */
7098 DEBUG_COMPILE_r({ SV* sv = sv_newmortal();
7099 regprop(r, sv, (regnode*)data.start_class, NULL);
7100 PerlIO_printf(Perl_debug_log,
7101 "synthetic stclass \"%s\".\n",
7102 SvPVX_const(sv));});
7103 data.start_class = NULL;
7107 if (RExC_seen & REG_UNBOUNDED_QUANTIFIER_SEEN) {
7108 r->extflags |= RXf_UNBOUNDED_QUANTIFIER_SEEN;
7109 r->maxlen = REG_INFTY;
7112 r->maxlen = RExC_maxlen;
7115 /* Guard against an embedded (?=) or (?<=) with a longer minlen than
7116 the "real" pattern. */
7118 PerlIO_printf(Perl_debug_log,"minlen: %"IVdf" r->minlen:%"IVdf" maxlen:%ld\n",
7119 (IV)minlen, (IV)r->minlen, RExC_maxlen);
7121 r->minlenret = minlen;
7122 if (r->minlen < minlen)
7125 if (RExC_seen & REG_GPOS_SEEN)
7126 r->intflags |= PREGf_GPOS_SEEN;
7127 if (RExC_seen & REG_LOOKBEHIND_SEEN)
7128 r->extflags |= RXf_NO_INPLACE_SUBST; /* inplace might break the
7130 if (pRExC_state->num_code_blocks)
7131 r->extflags |= RXf_EVAL_SEEN;
7132 if (RExC_seen & REG_CANY_SEEN)
7133 r->intflags |= PREGf_CANY_SEEN;
7134 if (RExC_seen & REG_VERBARG_SEEN)
7136 r->intflags |= PREGf_VERBARG_SEEN;
7137 r->extflags |= RXf_NO_INPLACE_SUBST; /* don't understand this! Yves */
7139 if (RExC_seen & REG_CUTGROUP_SEEN)
7140 r->intflags |= PREGf_CUTGROUP_SEEN;
7141 if (pm_flags & PMf_USE_RE_EVAL)
7142 r->intflags |= PREGf_USE_RE_EVAL;
7143 if (RExC_paren_names)
7144 RXp_PAREN_NAMES(r) = MUTABLE_HV(SvREFCNT_inc(RExC_paren_names));
7146 RXp_PAREN_NAMES(r) = NULL;
7148 /* If we have seen an anchor in our pattern then we set the extflag RXf_IS_ANCHORED
7149 * so it can be used in pp.c */
7150 if (r->intflags & PREGf_ANCH)
7151 r->extflags |= RXf_IS_ANCHORED;
7155 /* this is used to identify "special" patterns that might result
7156 * in Perl NOT calling the regex engine and instead doing the match "itself",
7157 * particularly special cases in split//. By having the regex compiler
7158 * do this pattern matching at a regop level (instead of by inspecting the pattern)
7159 * we avoid weird issues with equivalent patterns resulting in different behavior,
7160 * AND we allow non Perl engines to get the same optimizations by the setting the
7161 * flags appropriately - Yves */
7162 regnode *first = ri->program + 1;
7164 regnode *next = NEXTOPER(first);
7167 if (PL_regkind[fop] == NOTHING && nop == END)
7168 r->extflags |= RXf_NULL;
7169 else if (PL_regkind[fop] == BOL && nop == END)
7170 r->extflags |= RXf_START_ONLY;
7171 else if (fop == PLUS
7172 && PL_regkind[nop] == POSIXD && FLAGS(next) == _CC_SPACE
7173 && OP(regnext(first)) == END)
7174 r->extflags |= RXf_WHITE;
7175 else if ( r->extflags & RXf_SPLIT
7177 && STR_LEN(first) == 1
7178 && *(STRING(first)) == ' '
7179 && OP(regnext(first)) == END )
7180 r->extflags |= (RXf_SKIPWHITE|RXf_WHITE);
7184 if (RExC_contains_locale) {
7185 RXp_EXTFLAGS(r) |= RXf_TAINTED;
7189 if (RExC_paren_names) {
7190 ri->name_list_idx = add_data( pRExC_state, STR_WITH_LEN("a"));
7191 ri->data->data[ri->name_list_idx]
7192 = (void*)SvREFCNT_inc(RExC_paren_name_list);
7195 ri->name_list_idx = 0;
7197 if (RExC_recurse_count) {
7198 for ( ; RExC_recurse_count ; RExC_recurse_count-- ) {
7199 const regnode *scan = RExC_recurse[RExC_recurse_count-1];
7200 ARG2L_SET( scan, RExC_open_parens[ARG(scan)-1] - scan );
7203 Newxz(r->offs, RExC_npar, regexp_paren_pair);
7204 /* assume we don't need to swap parens around before we match */
7208 PerlIO_printf(Perl_debug_log,"Final program:\n");
7211 #ifdef RE_TRACK_PATTERN_OFFSETS
7212 DEBUG_OFFSETS_r(if (ri->u.offsets) {
7213 const STRLEN len = ri->u.offsets[0];
7215 GET_RE_DEBUG_FLAGS_DECL;
7216 PerlIO_printf(Perl_debug_log,
7217 "Offsets: [%"UVuf"]\n\t", (UV)ri->u.offsets[0]);
7218 for (i = 1; i <= len; i++) {
7219 if (ri->u.offsets[i*2-1] || ri->u.offsets[i*2])
7220 PerlIO_printf(Perl_debug_log, "%"UVuf":%"UVuf"[%"UVuf"] ",
7221 (UV)i, (UV)ri->u.offsets[i*2-1], (UV)ri->u.offsets[i*2]);
7223 PerlIO_printf(Perl_debug_log, "\n");
7228 /* under ithreads the ?pat? PMf_USED flag on the pmop is simulated
7229 * by setting the regexp SV to readonly-only instead. If the
7230 * pattern's been recompiled, the USEDness should remain. */
7231 if (old_re && SvREADONLY(old_re))
7239 Perl_reg_named_buff(pTHX_ REGEXP * const rx, SV * const key, SV * const value,
7242 PERL_ARGS_ASSERT_REG_NAMED_BUFF;
7244 PERL_UNUSED_ARG(value);
7246 if (flags & RXapif_FETCH) {
7247 return reg_named_buff_fetch(rx, key, flags);
7248 } else if (flags & (RXapif_STORE | RXapif_DELETE | RXapif_CLEAR)) {
7249 Perl_croak_no_modify();
7251 } else if (flags & RXapif_EXISTS) {
7252 return reg_named_buff_exists(rx, key, flags)
7255 } else if (flags & RXapif_REGNAMES) {
7256 return reg_named_buff_all(rx, flags);
7257 } else if (flags & (RXapif_SCALAR | RXapif_REGNAMES_COUNT)) {
7258 return reg_named_buff_scalar(rx, flags);
7260 Perl_croak(aTHX_ "panic: Unknown flags %d in named_buff", (int)flags);
7266 Perl_reg_named_buff_iter(pTHX_ REGEXP * const rx, const SV * const lastkey,
7269 PERL_ARGS_ASSERT_REG_NAMED_BUFF_ITER;
7270 PERL_UNUSED_ARG(lastkey);
7272 if (flags & RXapif_FIRSTKEY)
7273 return reg_named_buff_firstkey(rx, flags);
7274 else if (flags & RXapif_NEXTKEY)
7275 return reg_named_buff_nextkey(rx, flags);
7277 Perl_croak(aTHX_ "panic: Unknown flags %d in named_buff_iter",
7284 Perl_reg_named_buff_fetch(pTHX_ REGEXP * const r, SV * const namesv,
7287 AV *retarray = NULL;
7289 struct regexp *const rx = ReANY(r);
7291 PERL_ARGS_ASSERT_REG_NAMED_BUFF_FETCH;
7293 if (flags & RXapif_ALL)
7296 if (rx && RXp_PAREN_NAMES(rx)) {
7297 HE *he_str = hv_fetch_ent( RXp_PAREN_NAMES(rx), namesv, 0, 0 );
7300 SV* sv_dat=HeVAL(he_str);
7301 I32 *nums=(I32*)SvPVX(sv_dat);
7302 for ( i=0; i<SvIVX(sv_dat); i++ ) {
7303 if ((I32)(rx->nparens) >= nums[i]
7304 && rx->offs[nums[i]].start != -1
7305 && rx->offs[nums[i]].end != -1)
7308 CALLREG_NUMBUF_FETCH(r,nums[i],ret);
7313 ret = newSVsv(&PL_sv_undef);
7316 av_push(retarray, ret);
7319 return newRV_noinc(MUTABLE_SV(retarray));
7326 Perl_reg_named_buff_exists(pTHX_ REGEXP * const r, SV * const key,
7329 struct regexp *const rx = ReANY(r);
7331 PERL_ARGS_ASSERT_REG_NAMED_BUFF_EXISTS;
7333 if (rx && RXp_PAREN_NAMES(rx)) {
7334 if (flags & RXapif_ALL) {
7335 return hv_exists_ent(RXp_PAREN_NAMES(rx), key, 0);
7337 SV *sv = CALLREG_NAMED_BUFF_FETCH(r, key, flags);
7339 SvREFCNT_dec_NN(sv);
7351 Perl_reg_named_buff_firstkey(pTHX_ REGEXP * const r, const U32 flags)
7353 struct regexp *const rx = ReANY(r);
7355 PERL_ARGS_ASSERT_REG_NAMED_BUFF_FIRSTKEY;
7357 if ( rx && RXp_PAREN_NAMES(rx) ) {
7358 (void)hv_iterinit(RXp_PAREN_NAMES(rx));
7360 return CALLREG_NAMED_BUFF_NEXTKEY(r, NULL, flags & ~RXapif_FIRSTKEY);
7367 Perl_reg_named_buff_nextkey(pTHX_ REGEXP * const r, const U32 flags)
7369 struct regexp *const rx = ReANY(r);
7370 GET_RE_DEBUG_FLAGS_DECL;
7372 PERL_ARGS_ASSERT_REG_NAMED_BUFF_NEXTKEY;
7374 if (rx && RXp_PAREN_NAMES(rx)) {
7375 HV *hv = RXp_PAREN_NAMES(rx);
7377 while ( (temphe = hv_iternext_flags(hv,0)) ) {
7380 SV* sv_dat = HeVAL(temphe);
7381 I32 *nums = (I32*)SvPVX(sv_dat);
7382 for ( i = 0; i < SvIVX(sv_dat); i++ ) {
7383 if ((I32)(rx->lastparen) >= nums[i] &&
7384 rx->offs[nums[i]].start != -1 &&
7385 rx->offs[nums[i]].end != -1)
7391 if (parno || flags & RXapif_ALL) {
7392 return newSVhek(HeKEY_hek(temphe));
7400 Perl_reg_named_buff_scalar(pTHX_ REGEXP * const r, const U32 flags)
7405 struct regexp *const rx = ReANY(r);
7407 PERL_ARGS_ASSERT_REG_NAMED_BUFF_SCALAR;
7409 if (rx && RXp_PAREN_NAMES(rx)) {
7410 if (flags & (RXapif_ALL | RXapif_REGNAMES_COUNT)) {
7411 return newSViv(HvTOTALKEYS(RXp_PAREN_NAMES(rx)));
7412 } else if (flags & RXapif_ONE) {
7413 ret = CALLREG_NAMED_BUFF_ALL(r, (flags | RXapif_REGNAMES));
7414 av = MUTABLE_AV(SvRV(ret));
7415 length = av_tindex(av);
7416 SvREFCNT_dec_NN(ret);
7417 return newSViv(length + 1);
7419 Perl_croak(aTHX_ "panic: Unknown flags %d in named_buff_scalar",
7424 return &PL_sv_undef;
7428 Perl_reg_named_buff_all(pTHX_ REGEXP * const r, const U32 flags)
7430 struct regexp *const rx = ReANY(r);
7433 PERL_ARGS_ASSERT_REG_NAMED_BUFF_ALL;
7435 if (rx && RXp_PAREN_NAMES(rx)) {
7436 HV *hv= RXp_PAREN_NAMES(rx);
7438 (void)hv_iterinit(hv);
7439 while ( (temphe = hv_iternext_flags(hv,0)) ) {
7442 SV* sv_dat = HeVAL(temphe);
7443 I32 *nums = (I32*)SvPVX(sv_dat);
7444 for ( i = 0; i < SvIVX(sv_dat); i++ ) {
7445 if ((I32)(rx->lastparen) >= nums[i] &&
7446 rx->offs[nums[i]].start != -1 &&
7447 rx->offs[nums[i]].end != -1)
7453 if (parno || flags & RXapif_ALL) {
7454 av_push(av, newSVhek(HeKEY_hek(temphe)));
7459 return newRV_noinc(MUTABLE_SV(av));
7463 Perl_reg_numbered_buff_fetch(pTHX_ REGEXP * const r, const I32 paren,
7466 struct regexp *const rx = ReANY(r);
7472 PERL_ARGS_ASSERT_REG_NUMBERED_BUFF_FETCH;
7474 if ( n == RX_BUFF_IDX_CARET_PREMATCH
7475 || n == RX_BUFF_IDX_CARET_FULLMATCH
7476 || n == RX_BUFF_IDX_CARET_POSTMATCH
7479 bool keepcopy = cBOOL(rx->extflags & RXf_PMf_KEEPCOPY);
7481 /* on something like
7484 * the KEEPCOPY is set on the PMOP rather than the regex */
7485 if (PL_curpm && r == PM_GETRE(PL_curpm))
7486 keepcopy = cBOOL(PL_curpm->op_pmflags & PMf_KEEPCOPY);
7495 if (n == RX_BUFF_IDX_CARET_FULLMATCH)
7496 /* no need to distinguish between them any more */
7497 n = RX_BUFF_IDX_FULLMATCH;
7499 if ((n == RX_BUFF_IDX_PREMATCH || n == RX_BUFF_IDX_CARET_PREMATCH)
7500 && rx->offs[0].start != -1)
7502 /* $`, ${^PREMATCH} */
7503 i = rx->offs[0].start;
7507 if ((n == RX_BUFF_IDX_POSTMATCH || n == RX_BUFF_IDX_CARET_POSTMATCH)
7508 && rx->offs[0].end != -1)
7510 /* $', ${^POSTMATCH} */
7511 s = rx->subbeg - rx->suboffset + rx->offs[0].end;
7512 i = rx->sublen + rx->suboffset - rx->offs[0].end;
7515 if ( 0 <= n && n <= (I32)rx->nparens &&
7516 (s1 = rx->offs[n].start) != -1 &&
7517 (t1 = rx->offs[n].end) != -1)
7519 /* $&, ${^MATCH}, $1 ... */
7521 s = rx->subbeg + s1 - rx->suboffset;
7526 assert(s >= rx->subbeg);
7527 assert((STRLEN)rx->sublen >= (STRLEN)((s - rx->subbeg) + i) );
7529 #ifdef NO_TAINT_SUPPORT
7530 sv_setpvn(sv, s, i);
7532 const int oldtainted = TAINT_get;
7534 sv_setpvn(sv, s, i);
7535 TAINT_set(oldtainted);
7537 if ( (rx->intflags & PREGf_CANY_SEEN)
7538 ? (RXp_MATCH_UTF8(rx)
7539 && (!i || is_utf8_string((U8*)s, i)))
7540 : (RXp_MATCH_UTF8(rx)) )
7547 if (RXp_MATCH_TAINTED(rx)) {
7548 if (SvTYPE(sv) >= SVt_PVMG) {
7549 MAGIC* const mg = SvMAGIC(sv);
7552 SvMAGIC_set(sv, mg->mg_moremagic);
7554 if ((mgt = SvMAGIC(sv))) {
7555 mg->mg_moremagic = mgt;
7556 SvMAGIC_set(sv, mg);
7567 sv_setsv(sv,&PL_sv_undef);
7573 Perl_reg_numbered_buff_store(pTHX_ REGEXP * const rx, const I32 paren,
7574 SV const * const value)
7576 PERL_ARGS_ASSERT_REG_NUMBERED_BUFF_STORE;
7578 PERL_UNUSED_ARG(rx);
7579 PERL_UNUSED_ARG(paren);
7580 PERL_UNUSED_ARG(value);
7583 Perl_croak_no_modify();
7587 Perl_reg_numbered_buff_length(pTHX_ REGEXP * const r, const SV * const sv,
7590 struct regexp *const rx = ReANY(r);
7594 PERL_ARGS_ASSERT_REG_NUMBERED_BUFF_LENGTH;
7596 if ( paren == RX_BUFF_IDX_CARET_PREMATCH
7597 || paren == RX_BUFF_IDX_CARET_FULLMATCH
7598 || paren == RX_BUFF_IDX_CARET_POSTMATCH
7601 bool keepcopy = cBOOL(rx->extflags & RXf_PMf_KEEPCOPY);
7603 /* on something like
7606 * the KEEPCOPY is set on the PMOP rather than the regex */
7607 if (PL_curpm && r == PM_GETRE(PL_curpm))
7608 keepcopy = cBOOL(PL_curpm->op_pmflags & PMf_KEEPCOPY);
7614 /* Some of this code was originally in C<Perl_magic_len> in F<mg.c> */
7616 case RX_BUFF_IDX_CARET_PREMATCH: /* ${^PREMATCH} */
7617 case RX_BUFF_IDX_PREMATCH: /* $` */
7618 if (rx->offs[0].start != -1) {
7619 i = rx->offs[0].start;
7628 case RX_BUFF_IDX_CARET_POSTMATCH: /* ${^POSTMATCH} */
7629 case RX_BUFF_IDX_POSTMATCH: /* $' */
7630 if (rx->offs[0].end != -1) {
7631 i = rx->sublen - rx->offs[0].end;
7633 s1 = rx->offs[0].end;
7640 default: /* $& / ${^MATCH}, $1, $2, ... */
7641 if (paren <= (I32)rx->nparens &&
7642 (s1 = rx->offs[paren].start) != -1 &&
7643 (t1 = rx->offs[paren].end) != -1)
7649 if (ckWARN(WARN_UNINITIALIZED))
7650 report_uninit((const SV *)sv);
7655 if (i > 0 && RXp_MATCH_UTF8(rx)) {
7656 const char * const s = rx->subbeg - rx->suboffset + s1;
7661 if (is_utf8_string_loclen((U8*)s, i, &ep, &el))
7668 Perl_reg_qr_package(pTHX_ REGEXP * const rx)
7670 PERL_ARGS_ASSERT_REG_QR_PACKAGE;
7671 PERL_UNUSED_ARG(rx);
7675 return newSVpvs("Regexp");
7678 /* Scans the name of a named buffer from the pattern.
7679 * If flags is REG_RSN_RETURN_NULL returns null.
7680 * If flags is REG_RSN_RETURN_NAME returns an SV* containing the name
7681 * If flags is REG_RSN_RETURN_DATA returns the data SV* corresponding
7682 * to the parsed name as looked up in the RExC_paren_names hash.
7683 * If there is an error throws a vFAIL().. type exception.
7686 #define REG_RSN_RETURN_NULL 0
7687 #define REG_RSN_RETURN_NAME 1
7688 #define REG_RSN_RETURN_DATA 2
7691 S_reg_scan_name(pTHX_ RExC_state_t *pRExC_state, U32 flags)
7693 char *name_start = RExC_parse;
7695 PERL_ARGS_ASSERT_REG_SCAN_NAME;
7697 assert (RExC_parse <= RExC_end);
7698 if (RExC_parse == RExC_end) NOOP;
7699 else if (isIDFIRST_lazy_if(RExC_parse, UTF)) {
7700 /* skip IDFIRST by using do...while */
7703 RExC_parse += UTF8SKIP(RExC_parse);
7704 } while (isWORDCHAR_utf8((U8*)RExC_parse));
7708 } while (isWORDCHAR(*RExC_parse));
7710 RExC_parse++; /* so the <- from the vFAIL is after the offending
7712 vFAIL("Group name must start with a non-digit word character");
7716 = newSVpvn_flags(name_start, (int)(RExC_parse - name_start),
7717 SVs_TEMP | (UTF ? SVf_UTF8 : 0));
7718 if ( flags == REG_RSN_RETURN_NAME)
7720 else if (flags==REG_RSN_RETURN_DATA) {
7723 if ( ! sv_name ) /* should not happen*/
7724 Perl_croak(aTHX_ "panic: no svname in reg_scan_name");
7725 if (RExC_paren_names)
7726 he_str = hv_fetch_ent( RExC_paren_names, sv_name, 0, 0 );
7728 sv_dat = HeVAL(he_str);
7730 vFAIL("Reference to nonexistent named group");
7734 Perl_croak(aTHX_ "panic: bad flag %lx in reg_scan_name",
7735 (unsigned long) flags);
7737 assert(0); /* NOT REACHED */
7742 #define DEBUG_PARSE_MSG(funcname) DEBUG_PARSE_r({ \
7743 int rem=(int)(RExC_end - RExC_parse); \
7752 if (RExC_lastparse!=RExC_parse) \
7753 PerlIO_printf(Perl_debug_log," >%.*s%-*s", \
7756 iscut ? "..." : "<" \
7759 PerlIO_printf(Perl_debug_log,"%16s",""); \
7762 num = RExC_size + 1; \
7764 num=REG_NODE_NUM(RExC_emit); \
7765 if (RExC_lastnum!=num) \
7766 PerlIO_printf(Perl_debug_log,"|%4d",num); \
7768 PerlIO_printf(Perl_debug_log,"|%4s",""); \
7769 PerlIO_printf(Perl_debug_log,"|%*s%-4s", \
7770 (int)((depth*2)), "", \
7774 RExC_lastparse=RExC_parse; \
7779 #define DEBUG_PARSE(funcname) DEBUG_PARSE_r({ \
7780 DEBUG_PARSE_MSG((funcname)); \
7781 PerlIO_printf(Perl_debug_log,"%4s","\n"); \
7783 #define DEBUG_PARSE_FMT(funcname,fmt,args) DEBUG_PARSE_r({ \
7784 DEBUG_PARSE_MSG((funcname)); \
7785 PerlIO_printf(Perl_debug_log,fmt "\n",args); \
7788 /* This section of code defines the inversion list object and its methods. The
7789 * interfaces are highly subject to change, so as much as possible is static to
7790 * this file. An inversion list is here implemented as a malloc'd C UV array
7791 * as an SVt_INVLIST scalar.
7793 * An inversion list for Unicode is an array of code points, sorted by ordinal
7794 * number. The zeroth element is the first code point in the list. The 1th
7795 * element is the first element beyond that not in the list. In other words,
7796 * the first range is
7797 * invlist[0]..(invlist[1]-1)
7798 * The other ranges follow. Thus every element whose index is divisible by two
7799 * marks the beginning of a range that is in the list, and every element not
7800 * divisible by two marks the beginning of a range not in the list. A single
7801 * element inversion list that contains the single code point N generally
7802 * consists of two elements
7805 * (The exception is when N is the highest representable value on the
7806 * machine, in which case the list containing just it would be a single
7807 * element, itself. By extension, if the last range in the list extends to
7808 * infinity, then the first element of that range will be in the inversion list
7809 * at a position that is divisible by two, and is the final element in the
7811 * Taking the complement (inverting) an inversion list is quite simple, if the
7812 * first element is 0, remove it; otherwise add a 0 element at the beginning.
7813 * This implementation reserves an element at the beginning of each inversion
7814 * list to always contain 0; there is an additional flag in the header which
7815 * indicates if the list begins at the 0, or is offset to begin at the next
7818 * More about inversion lists can be found in "Unicode Demystified"
7819 * Chapter 13 by Richard Gillam, published by Addison-Wesley.
7820 * More will be coming when functionality is added later.
7822 * The inversion list data structure is currently implemented as an SV pointing
7823 * to an array of UVs that the SV thinks are bytes. This allows us to have an
7824 * array of UV whose memory management is automatically handled by the existing
7825 * facilities for SV's.
7827 * Some of the methods should always be private to the implementation, and some
7828 * should eventually be made public */
7830 /* The header definitions are in F<inline_invlist.c> */
7832 PERL_STATIC_INLINE UV*
7833 S__invlist_array_init(SV* const invlist, const bool will_have_0)
7835 /* Returns a pointer to the first element in the inversion list's array.
7836 * This is called upon initialization of an inversion list. Where the
7837 * array begins depends on whether the list has the code point U+0000 in it
7838 * or not. The other parameter tells it whether the code that follows this
7839 * call is about to put a 0 in the inversion list or not. The first
7840 * element is either the element reserved for 0, if TRUE, or the element
7841 * after it, if FALSE */
7843 bool* offset = get_invlist_offset_addr(invlist);
7844 UV* zero_addr = (UV *) SvPVX(invlist);
7846 PERL_ARGS_ASSERT__INVLIST_ARRAY_INIT;
7849 assert(! _invlist_len(invlist));
7853 /* 1^1 = 0; 1^0 = 1 */
7854 *offset = 1 ^ will_have_0;
7855 return zero_addr + *offset;
7858 PERL_STATIC_INLINE UV*
7859 S_invlist_array(SV* const invlist)
7861 /* Returns the pointer to the inversion list's array. Every time the
7862 * length changes, this needs to be called in case malloc or realloc moved
7865 PERL_ARGS_ASSERT_INVLIST_ARRAY;
7867 /* Must not be empty. If these fail, you probably didn't check for <len>
7868 * being non-zero before trying to get the array */
7869 assert(_invlist_len(invlist));
7871 /* The very first element always contains zero, The array begins either
7872 * there, or if the inversion list is offset, at the element after it.
7873 * The offset header field determines which; it contains 0 or 1 to indicate
7874 * how much additionally to add */
7875 assert(0 == *(SvPVX(invlist)));
7876 return ((UV *) SvPVX(invlist) + *get_invlist_offset_addr(invlist));
7879 PERL_STATIC_INLINE void
7880 S_invlist_set_len(pTHX_ SV* const invlist, const UV len, const bool offset)
7882 /* Sets the current number of elements stored in the inversion list.
7883 * Updates SvCUR correspondingly */
7884 PERL_UNUSED_CONTEXT;
7885 PERL_ARGS_ASSERT_INVLIST_SET_LEN;
7887 assert(SvTYPE(invlist) == SVt_INVLIST);
7892 : TO_INTERNAL_SIZE(len + offset));
7893 assert(SvLEN(invlist) == 0 || SvCUR(invlist) <= SvLEN(invlist));
7896 PERL_STATIC_INLINE IV*
7897 S_get_invlist_previous_index_addr(SV* invlist)
7899 /* Return the address of the IV that is reserved to hold the cached index
7901 PERL_ARGS_ASSERT_GET_INVLIST_PREVIOUS_INDEX_ADDR;
7903 assert(SvTYPE(invlist) == SVt_INVLIST);
7905 return &(((XINVLIST*) SvANY(invlist))->prev_index);
7908 PERL_STATIC_INLINE IV
7909 S_invlist_previous_index(SV* const invlist)
7911 /* Returns cached index of previous search */
7913 PERL_ARGS_ASSERT_INVLIST_PREVIOUS_INDEX;
7915 return *get_invlist_previous_index_addr(invlist);
7918 PERL_STATIC_INLINE void
7919 S_invlist_set_previous_index(SV* const invlist, const IV index)
7921 /* Caches <index> for later retrieval */
7923 PERL_ARGS_ASSERT_INVLIST_SET_PREVIOUS_INDEX;
7925 assert(index == 0 || index < (int) _invlist_len(invlist));
7927 *get_invlist_previous_index_addr(invlist) = index;
7930 PERL_STATIC_INLINE UV
7931 S_invlist_max(SV* const invlist)
7933 /* Returns the maximum number of elements storable in the inversion list's
7934 * array, without having to realloc() */
7936 PERL_ARGS_ASSERT_INVLIST_MAX;
7938 assert(SvTYPE(invlist) == SVt_INVLIST);
7940 /* Assumes worst case, in which the 0 element is not counted in the
7941 * inversion list, so subtracts 1 for that */
7942 return SvLEN(invlist) == 0 /* This happens under _new_invlist_C_array */
7943 ? FROM_INTERNAL_SIZE(SvCUR(invlist)) - 1
7944 : FROM_INTERNAL_SIZE(SvLEN(invlist)) - 1;
7947 #ifndef PERL_IN_XSUB_RE
7949 Perl__new_invlist(pTHX_ IV initial_size)
7952 /* Return a pointer to a newly constructed inversion list, with enough
7953 * space to store 'initial_size' elements. If that number is negative, a
7954 * system default is used instead */
7958 if (initial_size < 0) {
7962 /* Allocate the initial space */
7963 new_list = newSV_type(SVt_INVLIST);
7965 /* First 1 is in case the zero element isn't in the list; second 1 is for
7967 SvGROW(new_list, TO_INTERNAL_SIZE(initial_size + 1) + 1);
7968 invlist_set_len(new_list, 0, 0);
7970 /* Force iterinit() to be used to get iteration to work */
7971 *get_invlist_iter_addr(new_list) = (STRLEN) UV_MAX;
7973 *get_invlist_previous_index_addr(new_list) = 0;
7979 Perl__new_invlist_C_array(pTHX_ const UV* const list)
7981 /* Return a pointer to a newly constructed inversion list, initialized to
7982 * point to <list>, which has to be in the exact correct inversion list
7983 * form, including internal fields. Thus this is a dangerous routine that
7984 * should not be used in the wrong hands. The passed in 'list' contains
7985 * several header fields at the beginning that are not part of the
7986 * inversion list body proper */
7988 const STRLEN length = (STRLEN) list[0];
7989 const UV version_id = list[1];
7990 const bool offset = cBOOL(list[2]);
7991 #define HEADER_LENGTH 3
7992 /* If any of the above changes in any way, you must change HEADER_LENGTH
7993 * (if appropriate) and regenerate INVLIST_VERSION_ID by running
7994 * perl -E 'say int(rand 2**31-1)'
7996 #define INVLIST_VERSION_ID 148565664 /* This is a combination of a version and
7997 data structure type, so that one being
7998 passed in can be validated to be an
7999 inversion list of the correct vintage.
8002 SV* invlist = newSV_type(SVt_INVLIST);
8004 PERL_ARGS_ASSERT__NEW_INVLIST_C_ARRAY;
8006 if (version_id != INVLIST_VERSION_ID) {
8007 Perl_croak(aTHX_ "panic: Incorrect version for previously generated inversion list");
8010 /* The generated array passed in includes header elements that aren't part
8011 * of the list proper, so start it just after them */
8012 SvPV_set(invlist, (char *) (list + HEADER_LENGTH));
8014 SvLEN_set(invlist, 0); /* Means we own the contents, and the system
8015 shouldn't touch it */
8017 *(get_invlist_offset_addr(invlist)) = offset;
8019 /* The 'length' passed to us is the physical number of elements in the
8020 * inversion list. But if there is an offset the logical number is one
8022 invlist_set_len(invlist, length - offset, offset);
8024 invlist_set_previous_index(invlist, 0);
8026 /* Initialize the iteration pointer. */
8027 invlist_iterfinish(invlist);
8029 SvREADONLY_on(invlist);
8033 #endif /* ifndef PERL_IN_XSUB_RE */
8036 S_invlist_extend(pTHX_ SV* const invlist, const UV new_max)
8038 /* Grow the maximum size of an inversion list */
8040 PERL_ARGS_ASSERT_INVLIST_EXTEND;
8042 assert(SvTYPE(invlist) == SVt_INVLIST);
8044 /* Add one to account for the zero element at the beginning which may not
8045 * be counted by the calling parameters */
8046 SvGROW((SV *)invlist, TO_INTERNAL_SIZE(new_max + 1));
8049 PERL_STATIC_INLINE void
8050 S_invlist_trim(SV* const invlist)
8052 PERL_ARGS_ASSERT_INVLIST_TRIM;
8054 assert(SvTYPE(invlist) == SVt_INVLIST);
8056 /* Change the length of the inversion list to how many entries it currently
8058 SvPV_shrink_to_cur((SV *) invlist);
8062 S__append_range_to_invlist(pTHX_ SV* const invlist,
8063 const UV start, const UV end)
8065 /* Subject to change or removal. Append the range from 'start' to 'end' at
8066 * the end of the inversion list. The range must be above any existing
8070 UV max = invlist_max(invlist);
8071 UV len = _invlist_len(invlist);
8074 PERL_ARGS_ASSERT__APPEND_RANGE_TO_INVLIST;
8076 if (len == 0) { /* Empty lists must be initialized */
8077 offset = start != 0;
8078 array = _invlist_array_init(invlist, ! offset);
8081 /* Here, the existing list is non-empty. The current max entry in the
8082 * list is generally the first value not in the set, except when the
8083 * set extends to the end of permissible values, in which case it is
8084 * the first entry in that final set, and so this call is an attempt to
8085 * append out-of-order */
8087 UV final_element = len - 1;
8088 array = invlist_array(invlist);
8089 if (array[final_element] > start
8090 || ELEMENT_RANGE_MATCHES_INVLIST(final_element))
8092 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",
8093 array[final_element], start,
8094 ELEMENT_RANGE_MATCHES_INVLIST(final_element) ? 't' : 'f');
8097 /* Here, it is a legal append. If the new range begins with the first
8098 * value not in the set, it is extending the set, so the new first
8099 * value not in the set is one greater than the newly extended range.
8101 offset = *get_invlist_offset_addr(invlist);
8102 if (array[final_element] == start) {
8103 if (end != UV_MAX) {
8104 array[final_element] = end + 1;
8107 /* But if the end is the maximum representable on the machine,
8108 * just let the range that this would extend to have no end */
8109 invlist_set_len(invlist, len - 1, offset);
8115 /* Here the new range doesn't extend any existing set. Add it */
8117 len += 2; /* Includes an element each for the start and end of range */
8119 /* If wll overflow the existing space, extend, which may cause the array to
8122 invlist_extend(invlist, len);
8124 /* Have to set len here to avoid assert failure in invlist_array() */
8125 invlist_set_len(invlist, len, offset);
8127 array = invlist_array(invlist);
8130 invlist_set_len(invlist, len, offset);
8133 /* The next item on the list starts the range, the one after that is
8134 * one past the new range. */
8135 array[len - 2] = start;
8136 if (end != UV_MAX) {
8137 array[len - 1] = end + 1;
8140 /* But if the end is the maximum representable on the machine, just let
8141 * the range have no end */
8142 invlist_set_len(invlist, len - 1, offset);
8146 #ifndef PERL_IN_XSUB_RE
8149 Perl__invlist_search(SV* const invlist, const UV cp)
8151 /* Searches the inversion list for the entry that contains the input code
8152 * point <cp>. If <cp> is not in the list, -1 is returned. Otherwise, the
8153 * return value is the index into the list's array of the range that
8158 IV high = _invlist_len(invlist);
8159 const IV highest_element = high - 1;
8162 PERL_ARGS_ASSERT__INVLIST_SEARCH;
8164 /* If list is empty, return failure. */
8169 /* (We can't get the array unless we know the list is non-empty) */
8170 array = invlist_array(invlist);
8172 mid = invlist_previous_index(invlist);
8173 assert(mid >=0 && mid <= highest_element);
8175 /* <mid> contains the cache of the result of the previous call to this
8176 * function (0 the first time). See if this call is for the same result,
8177 * or if it is for mid-1. This is under the theory that calls to this
8178 * function will often be for related code points that are near each other.
8179 * And benchmarks show that caching gives better results. We also test
8180 * here if the code point is within the bounds of the list. These tests
8181 * replace others that would have had to be made anyway to make sure that
8182 * the array bounds were not exceeded, and these give us extra information
8183 * at the same time */
8184 if (cp >= array[mid]) {
8185 if (cp >= array[highest_element]) {
8186 return highest_element;
8189 /* Here, array[mid] <= cp < array[highest_element]. This means that
8190 * the final element is not the answer, so can exclude it; it also
8191 * means that <mid> is not the final element, so can refer to 'mid + 1'
8193 if (cp < array[mid + 1]) {
8199 else { /* cp < aray[mid] */
8200 if (cp < array[0]) { /* Fail if outside the array */
8204 if (cp >= array[mid - 1]) {
8209 /* Binary search. What we are looking for is <i> such that
8210 * array[i] <= cp < array[i+1]
8211 * The loop below converges on the i+1. Note that there may not be an
8212 * (i+1)th element in the array, and things work nonetheless */
8213 while (low < high) {
8214 mid = (low + high) / 2;
8215 assert(mid <= highest_element);
8216 if (array[mid] <= cp) { /* cp >= array[mid] */
8219 /* We could do this extra test to exit the loop early.
8220 if (cp < array[low]) {
8225 else { /* cp < array[mid] */
8232 invlist_set_previous_index(invlist, high);
8237 Perl__invlist_populate_swatch(SV* const invlist,
8238 const UV start, const UV end, U8* swatch)
8240 /* populates a swatch of a swash the same way swatch_get() does in utf8.c,
8241 * but is used when the swash has an inversion list. This makes this much
8242 * faster, as it uses a binary search instead of a linear one. This is
8243 * intimately tied to that function, and perhaps should be in utf8.c,
8244 * except it is intimately tied to inversion lists as well. It assumes
8245 * that <swatch> is all 0's on input */
8248 const IV len = _invlist_len(invlist);
8252 PERL_ARGS_ASSERT__INVLIST_POPULATE_SWATCH;
8254 if (len == 0) { /* Empty inversion list */
8258 array = invlist_array(invlist);
8260 /* Find which element it is */
8261 i = _invlist_search(invlist, start);
8263 /* We populate from <start> to <end> */
8264 while (current < end) {
8267 /* The inversion list gives the results for every possible code point
8268 * after the first one in the list. Only those ranges whose index is
8269 * even are ones that the inversion list matches. For the odd ones,
8270 * and if the initial code point is not in the list, we have to skip
8271 * forward to the next element */
8272 if (i == -1 || ! ELEMENT_RANGE_MATCHES_INVLIST(i)) {
8274 if (i >= len) { /* Finished if beyond the end of the array */
8278 if (current >= end) { /* Finished if beyond the end of what we
8280 if (LIKELY(end < UV_MAX)) {
8284 /* We get here when the upper bound is the maximum
8285 * representable on the machine, and we are looking for just
8286 * that code point. Have to special case it */
8288 goto join_end_of_list;
8291 assert(current >= start);
8293 /* The current range ends one below the next one, except don't go past
8296 upper = (i < len && array[i] < end) ? array[i] : end;
8298 /* Here we are in a range that matches. Populate a bit in the 3-bit U8
8299 * for each code point in it */
8300 for (; current < upper; current++) {
8301 const STRLEN offset = (STRLEN)(current - start);
8302 swatch[offset >> 3] |= 1 << (offset & 7);
8307 /* Quit if at the end of the list */
8310 /* But first, have to deal with the highest possible code point on
8311 * the platform. The previous code assumes that <end> is one
8312 * beyond where we want to populate, but that is impossible at the
8313 * platform's infinity, so have to handle it specially */
8314 if (UNLIKELY(end == UV_MAX && ELEMENT_RANGE_MATCHES_INVLIST(len-1)))
8316 const STRLEN offset = (STRLEN)(end - start);
8317 swatch[offset >> 3] |= 1 << (offset & 7);
8322 /* Advance to the next range, which will be for code points not in the
8331 Perl__invlist_union_maybe_complement_2nd(pTHX_ SV* const a, SV* const b,
8332 const bool complement_b, SV** output)
8334 /* Take the union of two inversion lists and point <output> to it. *output
8335 * SHOULD BE DEFINED upon input, and if it points to one of the two lists,
8336 * the reference count to that list will be decremented if not already a
8337 * temporary (mortal); otherwise *output will be made correspondingly
8338 * mortal. The first list, <a>, may be NULL, in which case a copy of the
8339 * second list is returned. If <complement_b> is TRUE, the union is taken
8340 * of the complement (inversion) of <b> instead of b itself.
8342 * The basis for this comes from "Unicode Demystified" Chapter 13 by
8343 * Richard Gillam, published by Addison-Wesley, and explained at some
8344 * length there. The preface says to incorporate its examples into your
8345 * code at your own risk.
8347 * The algorithm is like a merge sort.
8349 * XXX A potential performance improvement is to keep track as we go along
8350 * if only one of the inputs contributes to the result, meaning the other
8351 * is a subset of that one. In that case, we can skip the final copy and
8352 * return the larger of the input lists, but then outside code might need
8353 * to keep track of whether to free the input list or not */
8355 const UV* array_a; /* a's array */
8357 UV len_a; /* length of a's array */
8360 SV* u; /* the resulting union */
8364 UV i_a = 0; /* current index into a's array */
8368 /* running count, as explained in the algorithm source book; items are
8369 * stopped accumulating and are output when the count changes to/from 0.
8370 * The count is incremented when we start a range that's in the set, and
8371 * decremented when we start a range that's not in the set. So its range
8372 * is 0 to 2. Only when the count is zero is something not in the set.
8376 PERL_ARGS_ASSERT__INVLIST_UNION_MAYBE_COMPLEMENT_2ND;
8379 /* If either one is empty, the union is the other one */
8380 if (a == NULL || ((len_a = _invlist_len(a)) == 0)) {
8381 bool make_temp = FALSE; /* Should we mortalize the result? */
8385 if (! (make_temp = cBOOL(SvTEMP(a)))) {
8391 *output = invlist_clone(b);
8393 _invlist_invert(*output);
8395 } /* else *output already = b; */
8398 sv_2mortal(*output);
8402 else if ((len_b = _invlist_len(b)) == 0) {
8403 bool make_temp = FALSE;
8405 if (! (make_temp = cBOOL(SvTEMP(b)))) {
8410 /* The complement of an empty list is a list that has everything in it,
8411 * so the union with <a> includes everything too */
8414 if (! (make_temp = cBOOL(SvTEMP(a)))) {
8418 *output = _new_invlist(1);
8419 _append_range_to_invlist(*output, 0, UV_MAX);
8421 else if (*output != a) {
8422 *output = invlist_clone(a);
8424 /* else *output already = a; */
8427 sv_2mortal(*output);
8432 /* Here both lists exist and are non-empty */
8433 array_a = invlist_array(a);
8434 array_b = invlist_array(b);
8436 /* If are to take the union of 'a' with the complement of b, set it
8437 * up so are looking at b's complement. */
8440 /* To complement, we invert: if the first element is 0, remove it. To
8441 * do this, we just pretend the array starts one later */
8442 if (array_b[0] == 0) {
8448 /* But if the first element is not zero, we pretend the list starts
8449 * at the 0 that is always stored immediately before the array. */
8455 /* Size the union for the worst case: that the sets are completely
8457 u = _new_invlist(len_a + len_b);
8459 /* Will contain U+0000 if either component does */
8460 array_u = _invlist_array_init(u, (len_a > 0 && array_a[0] == 0)
8461 || (len_b > 0 && array_b[0] == 0));
8463 /* Go through each list item by item, stopping when exhausted one of
8465 while (i_a < len_a && i_b < len_b) {
8466 UV cp; /* The element to potentially add to the union's array */
8467 bool cp_in_set; /* is it in the the input list's set or not */
8469 /* We need to take one or the other of the two inputs for the union.
8470 * Since we are merging two sorted lists, we take the smaller of the
8471 * next items. In case of a tie, we take the one that is in its set
8472 * first. If we took one not in the set first, it would decrement the
8473 * count, possibly to 0 which would cause it to be output as ending the
8474 * range, and the next time through we would take the same number, and
8475 * output it again as beginning the next range. By doing it the
8476 * opposite way, there is no possibility that the count will be
8477 * momentarily decremented to 0, and thus the two adjoining ranges will
8478 * be seamlessly merged. (In a tie and both are in the set or both not
8479 * in the set, it doesn't matter which we take first.) */
8480 if (array_a[i_a] < array_b[i_b]
8481 || (array_a[i_a] == array_b[i_b]
8482 && ELEMENT_RANGE_MATCHES_INVLIST(i_a)))
8484 cp_in_set = ELEMENT_RANGE_MATCHES_INVLIST(i_a);
8488 cp_in_set = ELEMENT_RANGE_MATCHES_INVLIST(i_b);
8489 cp = array_b[i_b++];
8492 /* Here, have chosen which of the two inputs to look at. Only output
8493 * if the running count changes to/from 0, which marks the
8494 * beginning/end of a range in that's in the set */
8497 array_u[i_u++] = cp;
8504 array_u[i_u++] = cp;
8509 /* Here, we are finished going through at least one of the lists, which
8510 * means there is something remaining in at most one. We check if the list
8511 * that hasn't been exhausted is positioned such that we are in the middle
8512 * of a range in its set or not. (i_a and i_b point to the element beyond
8513 * the one we care about.) If in the set, we decrement 'count'; if 0, there
8514 * is potentially more to output.
8515 * There are four cases:
8516 * 1) Both weren't in their sets, count is 0, and remains 0. What's left
8517 * in the union is entirely from the non-exhausted set.
8518 * 2) Both were in their sets, count is 2. Nothing further should
8519 * be output, as everything that remains will be in the exhausted
8520 * list's set, hence in the union; decrementing to 1 but not 0 insures
8522 * 3) the exhausted was in its set, non-exhausted isn't, count is 1.
8523 * Nothing further should be output because the union includes
8524 * everything from the exhausted set. Not decrementing ensures that.
8525 * 4) the exhausted wasn't in its set, non-exhausted is, count is 1;
8526 * decrementing to 0 insures that we look at the remainder of the
8527 * non-exhausted set */
8528 if ((i_a != len_a && PREV_RANGE_MATCHES_INVLIST(i_a))
8529 || (i_b != len_b && PREV_RANGE_MATCHES_INVLIST(i_b)))
8534 /* The final length is what we've output so far, plus what else is about to
8535 * be output. (If 'count' is non-zero, then the input list we exhausted
8536 * has everything remaining up to the machine's limit in its set, and hence
8537 * in the union, so there will be no further output. */
8540 /* At most one of the subexpressions will be non-zero */
8541 len_u += (len_a - i_a) + (len_b - i_b);
8544 /* Set result to final length, which can change the pointer to array_u, so
8546 if (len_u != _invlist_len(u)) {
8547 invlist_set_len(u, len_u, *get_invlist_offset_addr(u));
8549 array_u = invlist_array(u);
8552 /* When 'count' is 0, the list that was exhausted (if one was shorter than
8553 * the other) ended with everything above it not in its set. That means
8554 * that the remaining part of the union is precisely the same as the
8555 * non-exhausted list, so can just copy it unchanged. (If both list were
8556 * exhausted at the same time, then the operations below will be both 0.)
8559 IV copy_count; /* At most one will have a non-zero copy count */
8560 if ((copy_count = len_a - i_a) > 0) {
8561 Copy(array_a + i_a, array_u + i_u, copy_count, UV);
8563 else if ((copy_count = len_b - i_b) > 0) {
8564 Copy(array_b + i_b, array_u + i_u, copy_count, UV);
8568 /* We may be removing a reference to one of the inputs. If so, the output
8569 * is made mortal if the input was. (Mortal SVs shouldn't have their ref
8570 * count decremented) */
8571 if (a == *output || b == *output) {
8572 assert(! invlist_is_iterating(*output));
8573 if ((SvTEMP(*output))) {
8577 SvREFCNT_dec_NN(*output);
8587 Perl__invlist_intersection_maybe_complement_2nd(pTHX_ SV* const a, SV* const b,
8588 const bool complement_b, SV** i)
8590 /* Take the intersection of two inversion lists and point <i> to it. *i
8591 * SHOULD BE DEFINED upon input, and if it points to one of the two lists,
8592 * the reference count to that list will be decremented if not already a
8593 * temporary (mortal); otherwise *i will be made correspondingly mortal.
8594 * The first list, <a>, may be NULL, in which case an empty list is
8595 * returned. If <complement_b> is TRUE, the result will be the
8596 * intersection of <a> and the complement (or inversion) of <b> instead of
8599 * The basis for this comes from "Unicode Demystified" Chapter 13 by
8600 * Richard Gillam, published by Addison-Wesley, and explained at some
8601 * length there. The preface says to incorporate its examples into your
8602 * code at your own risk. In fact, it had bugs
8604 * The algorithm is like a merge sort, and is essentially the same as the
8608 const UV* array_a; /* a's array */
8610 UV len_a; /* length of a's array */
8613 SV* r; /* the resulting intersection */
8617 UV i_a = 0; /* current index into a's array */
8621 /* running count, as explained in the algorithm source book; items are
8622 * stopped accumulating and are output when the count changes to/from 2.
8623 * The count is incremented when we start a range that's in the set, and
8624 * decremented when we start a range that's not in the set. So its range
8625 * is 0 to 2. Only when the count is 2 is something in the intersection.
8629 PERL_ARGS_ASSERT__INVLIST_INTERSECTION_MAYBE_COMPLEMENT_2ND;
8632 /* Special case if either one is empty */
8633 len_a = (a == NULL) ? 0 : _invlist_len(a);
8634 if ((len_a == 0) || ((len_b = _invlist_len(b)) == 0)) {
8635 bool make_temp = FALSE;
8637 if (len_a != 0 && complement_b) {
8639 /* Here, 'a' is not empty, therefore from the above 'if', 'b' must
8640 * be empty. Here, also we are using 'b's complement, which hence
8641 * must be every possible code point. Thus the intersection is
8645 if (! (make_temp = cBOOL(SvTEMP(b)))) {
8650 *i = invlist_clone(a);
8652 /* else *i is already 'a' */
8660 /* Here, 'a' or 'b' is empty and not using the complement of 'b'. The
8661 * intersection must be empty */
8663 if (! (make_temp = cBOOL(SvTEMP(a)))) {
8668 if (! (make_temp = cBOOL(SvTEMP(b)))) {
8672 *i = _new_invlist(0);
8680 /* Here both lists exist and are non-empty */
8681 array_a = invlist_array(a);
8682 array_b = invlist_array(b);
8684 /* If are to take the intersection of 'a' with the complement of b, set it
8685 * up so are looking at b's complement. */
8688 /* To complement, we invert: if the first element is 0, remove it. To
8689 * do this, we just pretend the array starts one later */
8690 if (array_b[0] == 0) {
8696 /* But if the first element is not zero, we pretend the list starts
8697 * at the 0 that is always stored immediately before the array. */
8703 /* Size the intersection for the worst case: that the intersection ends up
8704 * fragmenting everything to be completely disjoint */
8705 r= _new_invlist(len_a + len_b);
8707 /* Will contain U+0000 iff both components do */
8708 array_r = _invlist_array_init(r, len_a > 0 && array_a[0] == 0
8709 && len_b > 0 && array_b[0] == 0);
8711 /* Go through each list item by item, stopping when exhausted one of
8713 while (i_a < len_a && i_b < len_b) {
8714 UV cp; /* The element to potentially add to the intersection's
8716 bool cp_in_set; /* Is it in the input list's set or not */
8718 /* We need to take one or the other of the two inputs for the
8719 * intersection. Since we are merging two sorted lists, we take the
8720 * smaller of the next items. In case of a tie, we take the one that
8721 * is not in its set first (a difference from the union algorithm). If
8722 * we took one in the set first, it would increment the count, possibly
8723 * to 2 which would cause it to be output as starting a range in the
8724 * intersection, and the next time through we would take that same
8725 * number, and output it again as ending the set. By doing it the
8726 * opposite of this, there is no possibility that the count will be
8727 * momentarily incremented to 2. (In a tie and both are in the set or
8728 * both not in the set, it doesn't matter which we take first.) */
8729 if (array_a[i_a] < array_b[i_b]
8730 || (array_a[i_a] == array_b[i_b]
8731 && ! ELEMENT_RANGE_MATCHES_INVLIST(i_a)))
8733 cp_in_set = ELEMENT_RANGE_MATCHES_INVLIST(i_a);
8737 cp_in_set = ELEMENT_RANGE_MATCHES_INVLIST(i_b);
8741 /* Here, have chosen which of the two inputs to look at. Only output
8742 * if the running count changes to/from 2, which marks the
8743 * beginning/end of a range that's in the intersection */
8747 array_r[i_r++] = cp;
8752 array_r[i_r++] = cp;
8758 /* Here, we are finished going through at least one of the lists, which
8759 * means there is something remaining in at most one. We check if the list
8760 * that has been exhausted is positioned such that we are in the middle
8761 * of a range in its set or not. (i_a and i_b point to elements 1 beyond
8762 * the ones we care about.) There are four cases:
8763 * 1) Both weren't in their sets, count is 0, and remains 0. There's
8764 * nothing left in the intersection.
8765 * 2) Both were in their sets, count is 2 and perhaps is incremented to
8766 * above 2. What should be output is exactly that which is in the
8767 * non-exhausted set, as everything it has is also in the intersection
8768 * set, and everything it doesn't have can't be in the intersection
8769 * 3) The exhausted was in its set, non-exhausted isn't, count is 1, and
8770 * gets incremented to 2. Like the previous case, the intersection is
8771 * everything that remains in the non-exhausted set.
8772 * 4) the exhausted wasn't in its set, non-exhausted is, count is 1, and
8773 * remains 1. And the intersection has nothing more. */
8774 if ((i_a == len_a && PREV_RANGE_MATCHES_INVLIST(i_a))
8775 || (i_b == len_b && PREV_RANGE_MATCHES_INVLIST(i_b)))
8780 /* The final length is what we've output so far plus what else is in the
8781 * intersection. At most one of the subexpressions below will be non-zero
8785 len_r += (len_a - i_a) + (len_b - i_b);
8788 /* Set result to final length, which can change the pointer to array_r, so
8790 if (len_r != _invlist_len(r)) {
8791 invlist_set_len(r, len_r, *get_invlist_offset_addr(r));
8793 array_r = invlist_array(r);
8796 /* Finish outputting any remaining */
8797 if (count >= 2) { /* At most one will have a non-zero copy count */
8799 if ((copy_count = len_a - i_a) > 0) {
8800 Copy(array_a + i_a, array_r + i_r, copy_count, UV);
8802 else if ((copy_count = len_b - i_b) > 0) {
8803 Copy(array_b + i_b, array_r + i_r, copy_count, UV);
8807 /* We may be removing a reference to one of the inputs. If so, the output
8808 * is made mortal if the input was. (Mortal SVs shouldn't have their ref
8809 * count decremented) */
8810 if (a == *i || b == *i) {
8811 assert(! invlist_is_iterating(*i));
8816 SvREFCNT_dec_NN(*i);
8826 Perl__add_range_to_invlist(pTHX_ SV* invlist, const UV start, const UV end)
8828 /* Add the range from 'start' to 'end' inclusive to the inversion list's
8829 * set. A pointer to the inversion list is returned. This may actually be
8830 * a new list, in which case the passed in one has been destroyed. The
8831 * passed in inversion list can be NULL, in which case a new one is created
8832 * with just the one range in it */
8837 if (invlist == NULL) {
8838 invlist = _new_invlist(2);
8842 len = _invlist_len(invlist);
8845 /* If comes after the final entry actually in the list, can just append it
8848 || (! ELEMENT_RANGE_MATCHES_INVLIST(len - 1)
8849 && start >= invlist_array(invlist)[len - 1]))
8851 _append_range_to_invlist(invlist, start, end);
8855 /* Here, can't just append things, create and return a new inversion list
8856 * which is the union of this range and the existing inversion list */
8857 range_invlist = _new_invlist(2);
8858 _append_range_to_invlist(range_invlist, start, end);
8860 _invlist_union(invlist, range_invlist, &invlist);
8862 /* The temporary can be freed */
8863 SvREFCNT_dec_NN(range_invlist);
8869 Perl__setup_canned_invlist(pTHX_ const STRLEN size, const UV element0,
8870 UV** other_elements_ptr)
8872 /* Create and return an inversion list whose contents are to be populated
8873 * by the caller. The caller gives the number of elements (in 'size') and
8874 * the very first element ('element0'). This function will set
8875 * '*other_elements_ptr' to an array of UVs, where the remaining elements
8878 * Obviously there is some trust involved that the caller will properly
8879 * fill in the other elements of the array.
8881 * (The first element needs to be passed in, as the underlying code does
8882 * things differently depending on whether it is zero or non-zero) */
8884 SV* invlist = _new_invlist(size);
8887 PERL_ARGS_ASSERT__SETUP_CANNED_INVLIST;
8889 _append_range_to_invlist(invlist, element0, element0);
8890 offset = *get_invlist_offset_addr(invlist);
8892 invlist_set_len(invlist, size, offset);
8893 *other_elements_ptr = invlist_array(invlist) + 1;
8899 PERL_STATIC_INLINE SV*
8900 S_add_cp_to_invlist(pTHX_ SV* invlist, const UV cp) {
8901 return _add_range_to_invlist(invlist, cp, cp);
8904 #ifndef PERL_IN_XSUB_RE
8906 Perl__invlist_invert(pTHX_ SV* const invlist)
8908 /* Complement the input inversion list. This adds a 0 if the list didn't
8909 * have a zero; removes it otherwise. As described above, the data
8910 * structure is set up so that this is very efficient */
8912 PERL_ARGS_ASSERT__INVLIST_INVERT;
8914 assert(! invlist_is_iterating(invlist));
8916 /* The inverse of matching nothing is matching everything */
8917 if (_invlist_len(invlist) == 0) {
8918 _append_range_to_invlist(invlist, 0, UV_MAX);
8922 *get_invlist_offset_addr(invlist) = ! *get_invlist_offset_addr(invlist);
8927 PERL_STATIC_INLINE SV*
8928 S_invlist_clone(pTHX_ SV* const invlist)
8931 /* Return a new inversion list that is a copy of the input one, which is
8932 * unchanged. The new list will not be mortal even if the old one was. */
8934 /* Need to allocate extra space to accommodate Perl's addition of a
8935 * trailing NUL to SvPV's, since it thinks they are always strings */
8936 SV* new_invlist = _new_invlist(_invlist_len(invlist) + 1);
8937 STRLEN physical_length = SvCUR(invlist);
8938 bool offset = *(get_invlist_offset_addr(invlist));
8940 PERL_ARGS_ASSERT_INVLIST_CLONE;
8942 *(get_invlist_offset_addr(new_invlist)) = offset;
8943 invlist_set_len(new_invlist, _invlist_len(invlist), offset);
8944 Copy(SvPVX(invlist), SvPVX(new_invlist), physical_length, char);
8949 PERL_STATIC_INLINE STRLEN*
8950 S_get_invlist_iter_addr(SV* invlist)
8952 /* Return the address of the UV that contains the current iteration
8955 PERL_ARGS_ASSERT_GET_INVLIST_ITER_ADDR;
8957 assert(SvTYPE(invlist) == SVt_INVLIST);
8959 return &(((XINVLIST*) SvANY(invlist))->iterator);
8962 PERL_STATIC_INLINE void
8963 S_invlist_iterinit(SV* invlist) /* Initialize iterator for invlist */
8965 PERL_ARGS_ASSERT_INVLIST_ITERINIT;
8967 *get_invlist_iter_addr(invlist) = 0;
8970 PERL_STATIC_INLINE void
8971 S_invlist_iterfinish(SV* invlist)
8973 /* Terminate iterator for invlist. This is to catch development errors.
8974 * Any iteration that is interrupted before completed should call this
8975 * function. Functions that add code points anywhere else but to the end
8976 * of an inversion list assert that they are not in the middle of an
8977 * iteration. If they were, the addition would make the iteration
8978 * problematical: if the iteration hadn't reached the place where things
8979 * were being added, it would be ok */
8981 PERL_ARGS_ASSERT_INVLIST_ITERFINISH;
8983 *get_invlist_iter_addr(invlist) = (STRLEN) UV_MAX;
8987 S_invlist_iternext(SV* invlist, UV* start, UV* end)
8989 /* An C<invlist_iterinit> call on <invlist> must be used to set this up.
8990 * This call sets in <*start> and <*end>, the next range in <invlist>.
8991 * Returns <TRUE> if successful and the next call will return the next
8992 * range; <FALSE> if was already at the end of the list. If the latter,
8993 * <*start> and <*end> are unchanged, and the next call to this function
8994 * will start over at the beginning of the list */
8996 STRLEN* pos = get_invlist_iter_addr(invlist);
8997 UV len = _invlist_len(invlist);
9000 PERL_ARGS_ASSERT_INVLIST_ITERNEXT;
9003 *pos = (STRLEN) UV_MAX; /* Force iterinit() to be required next time */
9007 array = invlist_array(invlist);
9009 *start = array[(*pos)++];
9015 *end = array[(*pos)++] - 1;
9021 PERL_STATIC_INLINE bool
9022 S_invlist_is_iterating(SV* const invlist)
9024 PERL_ARGS_ASSERT_INVLIST_IS_ITERATING;
9026 return *(get_invlist_iter_addr(invlist)) < (STRLEN) UV_MAX;
9029 PERL_STATIC_INLINE UV
9030 S_invlist_highest(SV* const invlist)
9032 /* Returns the highest code point that matches an inversion list. This API
9033 * has an ambiguity, as it returns 0 under either the highest is actually
9034 * 0, or if the list is empty. If this distinction matters to you, check
9035 * for emptiness before calling this function */
9037 UV len = _invlist_len(invlist);
9040 PERL_ARGS_ASSERT_INVLIST_HIGHEST;
9046 array = invlist_array(invlist);
9048 /* The last element in the array in the inversion list always starts a
9049 * range that goes to infinity. That range may be for code points that are
9050 * matched in the inversion list, or it may be for ones that aren't
9051 * matched. In the latter case, the highest code point in the set is one
9052 * less than the beginning of this range; otherwise it is the final element
9053 * of this range: infinity */
9054 return (ELEMENT_RANGE_MATCHES_INVLIST(len - 1))
9056 : array[len - 1] - 1;
9059 #ifndef PERL_IN_XSUB_RE
9061 Perl__invlist_contents(pTHX_ SV* const invlist)
9063 /* Get the contents of an inversion list into a string SV so that they can
9064 * be printed out. It uses the format traditionally done for debug tracing
9068 SV* output = newSVpvs("\n");
9070 PERL_ARGS_ASSERT__INVLIST_CONTENTS;
9072 assert(! invlist_is_iterating(invlist));
9074 invlist_iterinit(invlist);
9075 while (invlist_iternext(invlist, &start, &end)) {
9076 if (end == UV_MAX) {
9077 Perl_sv_catpvf(aTHX_ output, "%04"UVXf"\tINFINITY\n", start);
9079 else if (end != start) {
9080 Perl_sv_catpvf(aTHX_ output, "%04"UVXf"\t%04"UVXf"\n",
9084 Perl_sv_catpvf(aTHX_ output, "%04"UVXf"\n", start);
9092 #ifndef PERL_IN_XSUB_RE
9094 Perl__invlist_dump(pTHX_ PerlIO *file, I32 level,
9095 const char * const indent, SV* const invlist)
9097 /* Designed to be called only by do_sv_dump(). Dumps out the ranges of the
9098 * inversion list 'invlist' to 'file' at 'level' Each line is prefixed by
9099 * the string 'indent'. The output looks like this:
9100 [0] 0x000A .. 0x000D
9102 [4] 0x2028 .. 0x2029
9103 [6] 0x3104 .. INFINITY
9104 * This means that the first range of code points matched by the list are
9105 * 0xA through 0xD; the second range contains only the single code point
9106 * 0x85, etc. An inversion list is an array of UVs. Two array elements
9107 * are used to define each range (except if the final range extends to
9108 * infinity, only a single element is needed). The array index of the
9109 * first element for the corresponding range is given in brackets. */
9114 PERL_ARGS_ASSERT__INVLIST_DUMP;
9116 if (invlist_is_iterating(invlist)) {
9117 Perl_dump_indent(aTHX_ level, file,
9118 "%sCan't dump inversion list because is in middle of iterating\n",
9123 invlist_iterinit(invlist);
9124 while (invlist_iternext(invlist, &start, &end)) {
9125 if (end == UV_MAX) {
9126 Perl_dump_indent(aTHX_ level, file,
9127 "%s[%"UVuf"] 0x%04"UVXf" .. INFINITY\n",
9128 indent, (UV)count, start);
9130 else if (end != start) {
9131 Perl_dump_indent(aTHX_ level, file,
9132 "%s[%"UVuf"] 0x%04"UVXf" .. 0x%04"UVXf"\n",
9133 indent, (UV)count, start, end);
9136 Perl_dump_indent(aTHX_ level, file, "%s[%"UVuf"] 0x%04"UVXf"\n",
9137 indent, (UV)count, start);
9144 Perl__load_PL_utf8_foldclosures (pTHX)
9146 assert(! PL_utf8_foldclosures);
9148 /* If the folds haven't been read in, call a fold function
9150 if (! PL_utf8_tofold) {
9151 U8 dummy[UTF8_MAXBYTES_CASE+1];
9153 /* This string is just a short named one above \xff */
9154 to_utf8_fold((U8*) HYPHEN_UTF8, dummy, NULL);
9155 assert(PL_utf8_tofold); /* Verify that worked */
9157 PL_utf8_foldclosures = _swash_inversion_hash(PL_utf8_tofold);
9161 #ifdef PERL_ARGS_ASSERT__INVLISTEQ
9163 S__invlistEQ(pTHX_ SV* const a, SV* const b, const bool complement_b)
9165 /* Return a boolean as to if the two passed in inversion lists are
9166 * identical. The final argument, if TRUE, says to take the complement of
9167 * the second inversion list before doing the comparison */
9169 const UV* array_a = invlist_array(a);
9170 const UV* array_b = invlist_array(b);
9171 UV len_a = _invlist_len(a);
9172 UV len_b = _invlist_len(b);
9174 UV i = 0; /* current index into the arrays */
9175 bool retval = TRUE; /* Assume are identical until proven otherwise */
9177 PERL_ARGS_ASSERT__INVLISTEQ;
9179 /* If are to compare 'a' with the complement of b, set it
9180 * up so are looking at b's complement. */
9183 /* The complement of nothing is everything, so <a> would have to have
9184 * just one element, starting at zero (ending at infinity) */
9186 return (len_a == 1 && array_a[0] == 0);
9188 else if (array_b[0] == 0) {
9190 /* Otherwise, to complement, we invert. Here, the first element is
9191 * 0, just remove it. To do this, we just pretend the array starts
9199 /* But if the first element is not zero, we pretend the list starts
9200 * at the 0 that is always stored immediately before the array. */
9206 /* Make sure that the lengths are the same, as well as the final element
9207 * before looping through the remainder. (Thus we test the length, final,
9208 * and first elements right off the bat) */
9209 if (len_a != len_b || array_a[len_a-1] != array_b[len_a-1]) {
9212 else for (i = 0; i < len_a - 1; i++) {
9213 if (array_a[i] != array_b[i]) {
9223 #undef HEADER_LENGTH
9224 #undef TO_INTERNAL_SIZE
9225 #undef FROM_INTERNAL_SIZE
9226 #undef INVLIST_VERSION_ID
9228 /* End of inversion list object */
9231 S_parse_lparen_question_flags(pTHX_ RExC_state_t *pRExC_state)
9233 /* This parses the flags that are in either the '(?foo)' or '(?foo:bar)'
9234 * constructs, and updates RExC_flags with them. On input, RExC_parse
9235 * should point to the first flag; it is updated on output to point to the
9236 * final ')' or ':'. There needs to be at least one flag, or this will
9239 /* for (?g), (?gc), and (?o) warnings; warning
9240 about (?c) will warn about (?g) -- japhy */
9242 #define WASTED_O 0x01
9243 #define WASTED_G 0x02
9244 #define WASTED_C 0x04
9245 #define WASTED_GC (WASTED_G|WASTED_C)
9246 I32 wastedflags = 0x00;
9247 U32 posflags = 0, negflags = 0;
9248 U32 *flagsp = &posflags;
9249 char has_charset_modifier = '\0';
9251 bool has_use_defaults = FALSE;
9252 const char* const seqstart = RExC_parse - 1; /* Point to the '?' */
9254 PERL_ARGS_ASSERT_PARSE_LPAREN_QUESTION_FLAGS;
9256 /* '^' as an initial flag sets certain defaults */
9257 if (UCHARAT(RExC_parse) == '^') {
9259 has_use_defaults = TRUE;
9260 STD_PMMOD_FLAGS_CLEAR(&RExC_flags);
9261 set_regex_charset(&RExC_flags, (RExC_utf8 || RExC_uni_semantics)
9262 ? REGEX_UNICODE_CHARSET
9263 : REGEX_DEPENDS_CHARSET);
9266 cs = get_regex_charset(RExC_flags);
9267 if (cs == REGEX_DEPENDS_CHARSET
9268 && (RExC_utf8 || RExC_uni_semantics))
9270 cs = REGEX_UNICODE_CHARSET;
9273 while (*RExC_parse) {
9274 /* && strchr("iogcmsx", *RExC_parse) */
9275 /* (?g), (?gc) and (?o) are useless here
9276 and must be globally applied -- japhy */
9277 switch (*RExC_parse) {
9279 /* Code for the imsx flags */
9280 CASE_STD_PMMOD_FLAGS_PARSE_SET(flagsp);
9282 case LOCALE_PAT_MOD:
9283 if (has_charset_modifier) {
9284 goto excess_modifier;
9286 else if (flagsp == &negflags) {
9289 cs = REGEX_LOCALE_CHARSET;
9290 has_charset_modifier = LOCALE_PAT_MOD;
9292 case UNICODE_PAT_MOD:
9293 if (has_charset_modifier) {
9294 goto excess_modifier;
9296 else if (flagsp == &negflags) {
9299 cs = REGEX_UNICODE_CHARSET;
9300 has_charset_modifier = UNICODE_PAT_MOD;
9302 case ASCII_RESTRICT_PAT_MOD:
9303 if (flagsp == &negflags) {
9306 if (has_charset_modifier) {
9307 if (cs != REGEX_ASCII_RESTRICTED_CHARSET) {
9308 goto excess_modifier;
9310 /* Doubled modifier implies more restricted */
9311 cs = REGEX_ASCII_MORE_RESTRICTED_CHARSET;
9314 cs = REGEX_ASCII_RESTRICTED_CHARSET;
9316 has_charset_modifier = ASCII_RESTRICT_PAT_MOD;
9318 case DEPENDS_PAT_MOD:
9319 if (has_use_defaults) {
9320 goto fail_modifiers;
9322 else if (flagsp == &negflags) {
9325 else if (has_charset_modifier) {
9326 goto excess_modifier;
9329 /* The dual charset means unicode semantics if the
9330 * pattern (or target, not known until runtime) are
9331 * utf8, or something in the pattern indicates unicode
9333 cs = (RExC_utf8 || RExC_uni_semantics)
9334 ? REGEX_UNICODE_CHARSET
9335 : REGEX_DEPENDS_CHARSET;
9336 has_charset_modifier = DEPENDS_PAT_MOD;
9340 if (has_charset_modifier == ASCII_RESTRICT_PAT_MOD) {
9341 vFAIL2("Regexp modifier \"%c\" may appear a maximum of twice", ASCII_RESTRICT_PAT_MOD);
9343 else if (has_charset_modifier == *(RExC_parse - 1)) {
9344 vFAIL2("Regexp modifier \"%c\" may not appear twice",
9348 vFAIL3("Regexp modifiers \"%c\" and \"%c\" are mutually exclusive", has_charset_modifier, *(RExC_parse - 1));
9353 vFAIL2("Regexp modifier \"%c\" may not appear after the \"-\"",
9356 case ONCE_PAT_MOD: /* 'o' */
9357 case GLOBAL_PAT_MOD: /* 'g' */
9358 if (SIZE_ONLY && ckWARN(WARN_REGEXP)) {
9359 const I32 wflagbit = *RExC_parse == 'o'
9362 if (! (wastedflags & wflagbit) ) {
9363 wastedflags |= wflagbit;
9364 /* diag_listed_as: Useless (?-%s) - don't use /%s modifier in regex; marked by <-- HERE in m/%s/ */
9367 "Useless (%s%c) - %suse /%c modifier",
9368 flagsp == &negflags ? "?-" : "?",
9370 flagsp == &negflags ? "don't " : "",
9377 case CONTINUE_PAT_MOD: /* 'c' */
9378 if (SIZE_ONLY && ckWARN(WARN_REGEXP)) {
9379 if (! (wastedflags & WASTED_C) ) {
9380 wastedflags |= WASTED_GC;
9381 /* diag_listed_as: Useless (?-%s) - don't use /%s modifier in regex; marked by <-- HERE in m/%s/ */
9384 "Useless (%sc) - %suse /gc modifier",
9385 flagsp == &negflags ? "?-" : "?",
9386 flagsp == &negflags ? "don't " : ""
9391 case KEEPCOPY_PAT_MOD: /* 'p' */
9392 if (flagsp == &negflags) {
9394 ckWARNreg(RExC_parse + 1,"Useless use of (?-p)");
9396 *flagsp |= RXf_PMf_KEEPCOPY;
9400 /* A flag is a default iff it is following a minus, so
9401 * if there is a minus, it means will be trying to
9402 * re-specify a default which is an error */
9403 if (has_use_defaults || flagsp == &negflags) {
9404 goto fail_modifiers;
9407 wastedflags = 0; /* reset so (?g-c) warns twice */
9411 RExC_flags |= posflags;
9412 RExC_flags &= ~negflags;
9413 set_regex_charset(&RExC_flags, cs);
9414 if (RExC_flags & RXf_PMf_FOLD) {
9415 RExC_contains_i = 1;
9421 RExC_parse += UTF ? UTF8SKIP(RExC_parse) : 1;
9422 /* diag_listed_as: Sequence (?%s...) not recognized in regex; marked by <-- HERE in m/%s/ */
9423 vFAIL2utf8f("Sequence (%"UTF8f"...) not recognized",
9424 UTF8fARG(UTF, RExC_parse-seqstart, seqstart));
9433 - reg - regular expression, i.e. main body or parenthesized thing
9435 * Caller must absorb opening parenthesis.
9437 * Combining parenthesis handling with the base level of regular expression
9438 * is a trifle forced, but the need to tie the tails of the branches to what
9439 * follows makes it hard to avoid.
9441 #define REGTAIL(x,y,z) regtail((x),(y),(z),depth+1)
9443 #define REGTAIL_STUDY(x,y,z) regtail_study((x),(y),(z),depth+1)
9445 #define REGTAIL_STUDY(x,y,z) regtail((x),(y),(z),depth+1)
9448 /* Returns NULL, setting *flagp to TRYAGAIN at the end of (?) that only sets
9449 flags. Returns NULL, setting *flagp to RESTART_UTF8 if the sizing scan
9450 needs to be restarted.
9451 Otherwise would only return NULL if regbranch() returns NULL, which
9454 S_reg(pTHX_ RExC_state_t *pRExC_state, I32 paren, I32 *flagp,U32 depth)
9455 /* paren: Parenthesized? 0=top; 1,2=inside '(': changed to letter.
9456 * 2 is like 1, but indicates that nextchar() has been called to advance
9457 * RExC_parse beyond the '('. Things like '(?' are indivisible tokens, and
9458 * this flag alerts us to the need to check for that */
9460 regnode *ret; /* Will be the head of the group. */
9463 regnode *ender = NULL;
9466 U32 oregflags = RExC_flags;
9467 bool have_branch = 0;
9469 I32 freeze_paren = 0;
9470 I32 after_freeze = 0;
9471 I32 num; /* numeric backreferences */
9473 char * parse_start = RExC_parse; /* MJD */
9474 char * const oregcomp_parse = RExC_parse;
9476 GET_RE_DEBUG_FLAGS_DECL;
9478 PERL_ARGS_ASSERT_REG;
9479 DEBUG_PARSE("reg ");
9481 *flagp = 0; /* Tentatively. */
9484 /* Make an OPEN node, if parenthesized. */
9487 /* Under /x, space and comments can be gobbled up between the '(' and
9488 * here (if paren ==2). The forms '(*VERB' and '(?...' disallow such
9489 * intervening space, as the sequence is a token, and a token should be
9491 bool has_intervening_patws = paren == 2 && *(RExC_parse - 1) != '(';
9493 if ( *RExC_parse == '*') { /* (*VERB:ARG) */
9494 char *start_verb = RExC_parse;
9495 STRLEN verb_len = 0;
9496 char *start_arg = NULL;
9497 unsigned char op = 0;
9499 int internal_argval = 0; /* internal_argval is only useful if
9502 if (has_intervening_patws) {
9504 vFAIL("In '(*VERB...)', the '(' and '*' must be adjacent");
9506 while ( *RExC_parse && *RExC_parse != ')' ) {
9507 if ( *RExC_parse == ':' ) {
9508 start_arg = RExC_parse + 1;
9514 verb_len = RExC_parse - start_verb;
9517 while ( *RExC_parse && *RExC_parse != ')' )
9519 if ( *RExC_parse != ')' )
9520 vFAIL("Unterminated verb pattern argument");
9521 if ( RExC_parse == start_arg )
9524 if ( *RExC_parse != ')' )
9525 vFAIL("Unterminated verb pattern");
9528 switch ( *start_verb ) {
9529 case 'A': /* (*ACCEPT) */
9530 if ( memEQs(start_verb,verb_len,"ACCEPT") ) {
9532 internal_argval = RExC_nestroot;
9535 case 'C': /* (*COMMIT) */
9536 if ( memEQs(start_verb,verb_len,"COMMIT") )
9539 case 'F': /* (*FAIL) */
9540 if ( verb_len==1 || memEQs(start_verb,verb_len,"FAIL") ) {
9545 case ':': /* (*:NAME) */
9546 case 'M': /* (*MARK:NAME) */
9547 if ( verb_len==0 || memEQs(start_verb,verb_len,"MARK") ) {
9552 case 'P': /* (*PRUNE) */
9553 if ( memEQs(start_verb,verb_len,"PRUNE") )
9556 case 'S': /* (*SKIP) */
9557 if ( memEQs(start_verb,verb_len,"SKIP") )
9560 case 'T': /* (*THEN) */
9561 /* [19:06] <TimToady> :: is then */
9562 if ( memEQs(start_verb,verb_len,"THEN") ) {
9564 RExC_seen |= REG_CUTGROUP_SEEN;
9569 RExC_parse += UTF ? UTF8SKIP(RExC_parse) : 1;
9571 "Unknown verb pattern '%"UTF8f"'",
9572 UTF8fARG(UTF, verb_len, start_verb));
9575 if ( start_arg && internal_argval ) {
9576 vFAIL3("Verb pattern '%.*s' may not have an argument",
9577 verb_len, start_verb);
9578 } else if ( argok < 0 && !start_arg ) {
9579 vFAIL3("Verb pattern '%.*s' has a mandatory argument",
9580 verb_len, start_verb);
9582 ret = reganode(pRExC_state, op, internal_argval);
9583 if ( ! internal_argval && ! SIZE_ONLY ) {
9585 SV *sv = newSVpvn( start_arg,
9586 RExC_parse - start_arg);
9587 ARG(ret) = add_data( pRExC_state,
9589 RExC_rxi->data->data[ARG(ret)]=(void*)sv;
9596 if (!internal_argval)
9597 RExC_seen |= REG_VERBARG_SEEN;
9598 } else if ( start_arg ) {
9599 vFAIL3("Verb pattern '%.*s' may not have an argument",
9600 verb_len, start_verb);
9602 ret = reg_node(pRExC_state, op);
9604 nextchar(pRExC_state);
9607 else if (*RExC_parse == '?') { /* (?...) */
9608 bool is_logical = 0;
9609 const char * const seqstart = RExC_parse;
9610 if (has_intervening_patws) {
9612 vFAIL("In '(?...)', the '(' and '?' must be adjacent");
9616 paren = *RExC_parse++;
9617 ret = NULL; /* For look-ahead/behind. */
9620 case 'P': /* (?P...) variants for those used to PCRE/Python */
9621 paren = *RExC_parse++;
9622 if ( paren == '<') /* (?P<...>) named capture */
9624 else if (paren == '>') { /* (?P>name) named recursion */
9625 goto named_recursion;
9627 else if (paren == '=') { /* (?P=...) named backref */
9628 /* this pretty much dupes the code for \k<NAME> in
9629 * regatom(), if you change this make sure you change that
9631 char* name_start = RExC_parse;
9633 SV *sv_dat = reg_scan_name(pRExC_state,
9634 SIZE_ONLY ? REG_RSN_RETURN_NULL : REG_RSN_RETURN_DATA);
9635 if (RExC_parse == name_start || *RExC_parse != ')')
9636 /* diag_listed_as: Sequence ?P=... not terminated in regex; marked by <-- HERE in m/%s/ */
9637 vFAIL2("Sequence %.3s... not terminated",parse_start);
9640 num = add_data( pRExC_state, STR_WITH_LEN("S"));
9641 RExC_rxi->data->data[num]=(void*)sv_dat;
9642 SvREFCNT_inc_simple_void(sv_dat);
9645 ret = reganode(pRExC_state,
9648 : (ASCII_FOLD_RESTRICTED)
9650 : (AT_LEAST_UNI_SEMANTICS)
9658 Set_Node_Offset(ret, parse_start+1);
9659 Set_Node_Cur_Length(ret, parse_start);
9661 nextchar(pRExC_state);
9665 /* diag_listed_as: Sequence (?%s...) not recognized in regex; marked by <-- HERE in m/%s/ */
9666 vFAIL3("Sequence (%.*s...) not recognized",
9667 RExC_parse-seqstart, seqstart);
9669 case '<': /* (?<...) */
9670 if (*RExC_parse == '!')
9672 else if (*RExC_parse != '=')
9678 case '\'': /* (?'...') */
9679 name_start= RExC_parse;
9680 svname = reg_scan_name(pRExC_state,
9681 SIZE_ONLY /* reverse test from the others */
9682 ? REG_RSN_RETURN_NAME
9683 : REG_RSN_RETURN_NULL);
9684 if (RExC_parse == name_start || *RExC_parse != paren)
9685 vFAIL2("Sequence (?%c... not terminated",
9686 paren=='>' ? '<' : paren);
9690 if (!svname) /* shouldn't happen */
9692 "panic: reg_scan_name returned NULL");
9693 if (!RExC_paren_names) {
9694 RExC_paren_names= newHV();
9695 sv_2mortal(MUTABLE_SV(RExC_paren_names));
9697 RExC_paren_name_list= newAV();
9698 sv_2mortal(MUTABLE_SV(RExC_paren_name_list));
9701 he_str = hv_fetch_ent( RExC_paren_names, svname, 1, 0 );
9703 sv_dat = HeVAL(he_str);
9705 /* croak baby croak */
9707 "panic: paren_name hash element allocation failed");
9708 } else if ( SvPOK(sv_dat) ) {
9709 /* (?|...) can mean we have dupes so scan to check
9710 its already been stored. Maybe a flag indicating
9711 we are inside such a construct would be useful,
9712 but the arrays are likely to be quite small, so
9713 for now we punt -- dmq */
9714 IV count = SvIV(sv_dat);
9715 I32 *pv = (I32*)SvPVX(sv_dat);
9717 for ( i = 0 ; i < count ; i++ ) {
9718 if ( pv[i] == RExC_npar ) {
9724 pv = (I32*)SvGROW(sv_dat,
9725 SvCUR(sv_dat) + sizeof(I32)+1);
9726 SvCUR_set(sv_dat, SvCUR(sv_dat) + sizeof(I32));
9727 pv[count] = RExC_npar;
9728 SvIV_set(sv_dat, SvIVX(sv_dat) + 1);
9731 (void)SvUPGRADE(sv_dat,SVt_PVNV);
9732 sv_setpvn(sv_dat, (char *)&(RExC_npar),
9735 SvIV_set(sv_dat, 1);
9738 /* Yes this does cause a memory leak in debugging Perls
9740 if (!av_store(RExC_paren_name_list,
9741 RExC_npar, SvREFCNT_inc(svname)))
9742 SvREFCNT_dec_NN(svname);
9745 /*sv_dump(sv_dat);*/
9747 nextchar(pRExC_state);
9749 goto capturing_parens;
9751 RExC_seen |= REG_LOOKBEHIND_SEEN;
9752 RExC_in_lookbehind++;
9755 case '=': /* (?=...) */
9756 RExC_seen_zerolen++;
9758 case '!': /* (?!...) */
9759 RExC_seen_zerolen++;
9760 if (*RExC_parse == ')') {
9761 ret=reg_node(pRExC_state, OPFAIL);
9762 nextchar(pRExC_state);
9766 case '|': /* (?|...) */
9767 /* branch reset, behave like a (?:...) except that
9768 buffers in alternations share the same numbers */
9770 after_freeze = freeze_paren = RExC_npar;
9772 case ':': /* (?:...) */
9773 case '>': /* (?>...) */
9775 case '$': /* (?$...) */
9776 case '@': /* (?@...) */
9777 vFAIL2("Sequence (?%c...) not implemented", (int)paren);
9779 case '0' : /* (?0) */
9780 case 'R' : /* (?R) */
9781 if (*RExC_parse != ')')
9782 FAIL("Sequence (?R) not terminated");
9783 ret = reg_node(pRExC_state, GOSTART);
9784 RExC_seen |= REG_GOSTART_SEEN;
9785 *flagp |= POSTPONED;
9786 nextchar(pRExC_state);
9789 /* named and numeric backreferences */
9790 case '&': /* (?&NAME) */
9791 parse_start = RExC_parse - 1;
9794 SV *sv_dat = reg_scan_name(pRExC_state,
9795 SIZE_ONLY ? REG_RSN_RETURN_NULL : REG_RSN_RETURN_DATA);
9796 num = sv_dat ? *((I32 *)SvPVX(sv_dat)) : 0;
9798 if (RExC_parse == RExC_end || *RExC_parse != ')')
9799 vFAIL("Sequence (?&... not terminated");
9800 goto gen_recurse_regop;
9801 assert(0); /* NOT REACHED */
9803 if (!(RExC_parse[0] >= '1' && RExC_parse[0] <= '9')) {
9805 vFAIL("Illegal pattern");
9807 goto parse_recursion;
9809 case '-': /* (?-1) */
9810 if (!(RExC_parse[0] >= '1' && RExC_parse[0] <= '9')) {
9811 RExC_parse--; /* rewind to let it be handled later */
9815 case '1': case '2': case '3': case '4': /* (?1) */
9816 case '5': case '6': case '7': case '8': case '9':
9819 num = atoi(RExC_parse);
9820 parse_start = RExC_parse - 1; /* MJD */
9821 if (*RExC_parse == '-')
9823 while (isDIGIT(*RExC_parse))
9825 if (*RExC_parse!=')')
9826 vFAIL("Expecting close bracket");
9829 if ( paren == '-' ) {
9831 Diagram of capture buffer numbering.
9832 Top line is the normal capture buffer numbers
9833 Bottom line is the negative indexing as from
9837 /(a(x)y)(a(b(c(?-2)d)e)f)(g(h))/
9841 num = RExC_npar + num;
9844 vFAIL("Reference to nonexistent group");
9846 } else if ( paren == '+' ) {
9847 num = RExC_npar + num - 1;
9850 ret = reganode(pRExC_state, GOSUB, num);
9852 if (num > (I32)RExC_rx->nparens) {
9854 vFAIL("Reference to nonexistent group");
9856 ARG2L_SET( ret, RExC_recurse_count++);
9858 DEBUG_OPTIMISE_MORE_r(PerlIO_printf(Perl_debug_log,
9859 "Recurse #%"UVuf" to %"IVdf"\n",
9860 (UV)ARG(ret), (IV)ARG2L(ret)));
9864 RExC_seen |= REG_RECURSE_SEEN;
9865 Set_Node_Length(ret, 1 + regarglen[OP(ret)]); /* MJD */
9866 Set_Node_Offset(ret, parse_start); /* MJD */
9868 *flagp |= POSTPONED;
9869 nextchar(pRExC_state);
9872 assert(0); /* NOT REACHED */
9874 case '?': /* (??...) */
9876 if (*RExC_parse != '{') {
9878 /* diag_listed_as: Sequence (?%s...) not recognized in regex; marked by <-- HERE in m/%s/ */
9880 "Sequence (%"UTF8f"...) not recognized",
9881 UTF8fARG(UTF, RExC_parse-seqstart, seqstart));
9884 *flagp |= POSTPONED;
9885 paren = *RExC_parse++;
9887 case '{': /* (?{...}) */
9890 struct reg_code_block *cb;
9892 RExC_seen_zerolen++;
9894 if ( !pRExC_state->num_code_blocks
9895 || pRExC_state->code_index >= pRExC_state->num_code_blocks
9896 || pRExC_state->code_blocks[pRExC_state->code_index].start
9897 != (STRLEN)((RExC_parse -3 - (is_logical ? 1 : 0))
9900 if (RExC_pm_flags & PMf_USE_RE_EVAL)
9901 FAIL("panic: Sequence (?{...}): no code block found\n");
9902 FAIL("Eval-group not allowed at runtime, use re 'eval'");
9904 /* this is a pre-compiled code block (?{...}) */
9905 cb = &pRExC_state->code_blocks[pRExC_state->code_index];
9906 RExC_parse = RExC_start + cb->end;
9909 if (cb->src_regex) {
9910 n = add_data(pRExC_state, STR_WITH_LEN("rl"));
9911 RExC_rxi->data->data[n] =
9912 (void*)SvREFCNT_inc((SV*)cb->src_regex);
9913 RExC_rxi->data->data[n+1] = (void*)o;
9916 n = add_data(pRExC_state,
9917 (RExC_pm_flags & PMf_HAS_CV) ? "L" : "l", 1);
9918 RExC_rxi->data->data[n] = (void*)o;
9921 pRExC_state->code_index++;
9922 nextchar(pRExC_state);
9926 ret = reg_node(pRExC_state, LOGICAL);
9927 eval = reganode(pRExC_state, EVAL, n);
9930 /* for later propagation into (??{}) return value */
9931 eval->flags = (U8) (RExC_flags & RXf_PMf_COMPILETIME);
9933 REGTAIL(pRExC_state, ret, eval);
9934 /* deal with the length of this later - MJD */
9937 ret = reganode(pRExC_state, EVAL, n);
9938 Set_Node_Length(ret, RExC_parse - parse_start + 1);
9939 Set_Node_Offset(ret, parse_start);
9942 case '(': /* (?(?{...})...) and (?(?=...)...) */
9945 if (RExC_parse[0] == '?') { /* (?(?...)) */
9946 if (RExC_parse[1] == '=' || RExC_parse[1] == '!'
9947 || RExC_parse[1] == '<'
9948 || RExC_parse[1] == '{') { /* Lookahead or eval. */
9952 ret = reg_node(pRExC_state, LOGICAL);
9956 tail = reg(pRExC_state, 1, &flag, depth+1);
9957 if (flag & RESTART_UTF8) {
9958 *flagp = RESTART_UTF8;
9961 REGTAIL(pRExC_state, ret, tail);
9965 else if ( RExC_parse[0] == '<' /* (?(<NAME>)...) */
9966 || RExC_parse[0] == '\'' ) /* (?('NAME')...) */
9968 char ch = RExC_parse[0] == '<' ? '>' : '\'';
9969 char *name_start= RExC_parse++;
9971 SV *sv_dat=reg_scan_name(pRExC_state,
9972 SIZE_ONLY ? REG_RSN_RETURN_NULL : REG_RSN_RETURN_DATA);
9973 if (RExC_parse == name_start || *RExC_parse != ch)
9974 vFAIL2("Sequence (?(%c... not terminated",
9975 (ch == '>' ? '<' : ch));
9978 num = add_data( pRExC_state, STR_WITH_LEN("S"));
9979 RExC_rxi->data->data[num]=(void*)sv_dat;
9980 SvREFCNT_inc_simple_void(sv_dat);
9982 ret = reganode(pRExC_state,NGROUPP,num);
9983 goto insert_if_check_paren;
9985 else if (RExC_parse[0] == 'D' &&
9986 RExC_parse[1] == 'E' &&
9987 RExC_parse[2] == 'F' &&
9988 RExC_parse[3] == 'I' &&
9989 RExC_parse[4] == 'N' &&
9990 RExC_parse[5] == 'E')
9992 ret = reganode(pRExC_state,DEFINEP,0);
9995 goto insert_if_check_paren;
9997 else if (RExC_parse[0] == 'R') {
10000 if (RExC_parse[0] >= '1' && RExC_parse[0] <= '9' ) {
10001 parno = atoi(RExC_parse++);
10002 while (isDIGIT(*RExC_parse))
10004 } else if (RExC_parse[0] == '&') {
10007 sv_dat = reg_scan_name(pRExC_state,
10009 ? REG_RSN_RETURN_NULL
10010 : REG_RSN_RETURN_DATA);
10011 parno = sv_dat ? *((I32 *)SvPVX(sv_dat)) : 0;
10013 ret = reganode(pRExC_state,INSUBP,parno);
10014 goto insert_if_check_paren;
10016 else if (RExC_parse[0] >= '1' && RExC_parse[0] <= '9' ) {
10020 parno = atoi(RExC_parse++);
10022 while (isDIGIT(*RExC_parse))
10024 ret = reganode(pRExC_state, GROUPP, parno);
10026 insert_if_check_paren:
10027 if (*(tmp = nextchar(pRExC_state)) != ')') {
10028 /* nextchar also skips comments, so undo its work
10029 * and skip over the the next character.
10032 RExC_parse += UTF ? UTF8SKIP(RExC_parse) : 1;
10033 vFAIL("Switch condition not recognized");
10036 REGTAIL(pRExC_state, ret, reganode(pRExC_state, IFTHEN, 0));
10037 br = regbranch(pRExC_state, &flags, 1,depth+1);
10039 if (flags & RESTART_UTF8) {
10040 *flagp = RESTART_UTF8;
10043 FAIL2("panic: regbranch returned NULL, flags=%#"UVxf"",
10046 REGTAIL(pRExC_state, br, reganode(pRExC_state,
10048 c = *nextchar(pRExC_state);
10049 if (flags&HASWIDTH)
10050 *flagp |= HASWIDTH;
10053 vFAIL("(?(DEFINE)....) does not allow branches");
10055 /* Fake one for optimizer. */
10056 lastbr = reganode(pRExC_state, IFTHEN, 0);
10058 if (!regbranch(pRExC_state, &flags, 1,depth+1)) {
10059 if (flags & RESTART_UTF8) {
10060 *flagp = RESTART_UTF8;
10063 FAIL2("panic: regbranch returned NULL, flags=%#"UVxf"",
10066 REGTAIL(pRExC_state, ret, lastbr);
10067 if (flags&HASWIDTH)
10068 *flagp |= HASWIDTH;
10069 c = *nextchar(pRExC_state);
10074 vFAIL("Switch (?(condition)... contains too many branches");
10075 ender = reg_node(pRExC_state, TAIL);
10076 REGTAIL(pRExC_state, br, ender);
10078 REGTAIL(pRExC_state, lastbr, ender);
10079 REGTAIL(pRExC_state, NEXTOPER(NEXTOPER(lastbr)), ender);
10082 REGTAIL(pRExC_state, ret, ender);
10083 RExC_size++; /* XXX WHY do we need this?!!
10084 For large programs it seems to be required
10085 but I can't figure out why. -- dmq*/
10089 RExC_parse += UTF ? UTF8SKIP(RExC_parse) : 1;
10090 vFAIL("Unknown switch condition (?(...))");
10093 case '[': /* (?[ ... ]) */
10094 return handle_regex_sets(pRExC_state, NULL, flagp, depth,
10097 RExC_parse--; /* for vFAIL to print correctly */
10098 vFAIL("Sequence (? incomplete");
10100 default: /* e.g., (?i) */
10103 parse_lparen_question_flags(pRExC_state);
10104 if (UCHARAT(RExC_parse) != ':') {
10105 nextchar(pRExC_state);
10110 nextchar(pRExC_state);
10120 ret = reganode(pRExC_state, OPEN, parno);
10122 if (!RExC_nestroot)
10123 RExC_nestroot = parno;
10124 if (RExC_seen & REG_RECURSE_SEEN
10125 && !RExC_open_parens[parno-1])
10127 DEBUG_OPTIMISE_MORE_r(PerlIO_printf(Perl_debug_log,
10128 "Setting open paren #%"IVdf" to %d\n",
10129 (IV)parno, REG_NODE_NUM(ret)));
10130 RExC_open_parens[parno-1]= ret;
10133 Set_Node_Length(ret, 1); /* MJD */
10134 Set_Node_Offset(ret, RExC_parse); /* MJD */
10142 /* Pick up the branches, linking them together. */
10143 parse_start = RExC_parse; /* MJD */
10144 br = regbranch(pRExC_state, &flags, 1,depth+1);
10146 /* branch_len = (paren != 0); */
10149 if (flags & RESTART_UTF8) {
10150 *flagp = RESTART_UTF8;
10153 FAIL2("panic: regbranch returned NULL, flags=%#"UVxf"", (UV) flags);
10155 if (*RExC_parse == '|') {
10156 if (!SIZE_ONLY && RExC_extralen) {
10157 reginsert(pRExC_state, BRANCHJ, br, depth+1);
10160 reginsert(pRExC_state, BRANCH, br, depth+1);
10161 Set_Node_Length(br, paren != 0);
10162 Set_Node_Offset_To_R(br-RExC_emit_start, parse_start-RExC_start);
10166 RExC_extralen += 1; /* For BRANCHJ-BRANCH. */
10168 else if (paren == ':') {
10169 *flagp |= flags&SIMPLE;
10171 if (is_open) { /* Starts with OPEN. */
10172 REGTAIL(pRExC_state, ret, br); /* OPEN -> first. */
10174 else if (paren != '?') /* Not Conditional */
10176 *flagp |= flags & (SPSTART | HASWIDTH | POSTPONED);
10178 while (*RExC_parse == '|') {
10179 if (!SIZE_ONLY && RExC_extralen) {
10180 ender = reganode(pRExC_state, LONGJMP,0);
10182 /* Append to the previous. */
10183 REGTAIL(pRExC_state, NEXTOPER(NEXTOPER(lastbr)), ender);
10186 RExC_extralen += 2; /* Account for LONGJMP. */
10187 nextchar(pRExC_state);
10188 if (freeze_paren) {
10189 if (RExC_npar > after_freeze)
10190 after_freeze = RExC_npar;
10191 RExC_npar = freeze_paren;
10193 br = regbranch(pRExC_state, &flags, 0, depth+1);
10196 if (flags & RESTART_UTF8) {
10197 *flagp = RESTART_UTF8;
10200 FAIL2("panic: regbranch returned NULL, flags=%#"UVxf"", (UV) flags);
10202 REGTAIL(pRExC_state, lastbr, br); /* BRANCH -> BRANCH. */
10204 *flagp |= flags & (SPSTART | HASWIDTH | POSTPONED);
10207 if (have_branch || paren != ':') {
10208 /* Make a closing node, and hook it on the end. */
10211 ender = reg_node(pRExC_state, TAIL);
10214 ender = reganode(pRExC_state, CLOSE, parno);
10215 if (!SIZE_ONLY && RExC_seen & REG_RECURSE_SEEN) {
10216 DEBUG_OPTIMISE_MORE_r(PerlIO_printf(Perl_debug_log,
10217 "Setting close paren #%"IVdf" to %d\n",
10218 (IV)parno, REG_NODE_NUM(ender)));
10219 RExC_close_parens[parno-1]= ender;
10220 if (RExC_nestroot == parno)
10223 Set_Node_Offset(ender,RExC_parse+1); /* MJD */
10224 Set_Node_Length(ender,1); /* MJD */
10230 *flagp &= ~HASWIDTH;
10233 ender = reg_node(pRExC_state, SUCCEED);
10236 ender = reg_node(pRExC_state, END);
10238 assert(!RExC_opend); /* there can only be one! */
10239 RExC_opend = ender;
10243 DEBUG_PARSE_r(if (!SIZE_ONLY) {
10244 SV * const mysv_val1=sv_newmortal();
10245 SV * const mysv_val2=sv_newmortal();
10246 DEBUG_PARSE_MSG("lsbr");
10247 regprop(RExC_rx, mysv_val1, lastbr, NULL);
10248 regprop(RExC_rx, mysv_val2, ender, NULL);
10249 PerlIO_printf(Perl_debug_log, "~ tying lastbr %s (%"IVdf") to ender %s (%"IVdf") offset %"IVdf"\n",
10250 SvPV_nolen_const(mysv_val1),
10251 (IV)REG_NODE_NUM(lastbr),
10252 SvPV_nolen_const(mysv_val2),
10253 (IV)REG_NODE_NUM(ender),
10254 (IV)(ender - lastbr)
10257 REGTAIL(pRExC_state, lastbr, ender);
10259 if (have_branch && !SIZE_ONLY) {
10260 char is_nothing= 1;
10262 RExC_seen |= REG_TOP_LEVEL_BRANCHES_SEEN;
10264 /* Hook the tails of the branches to the closing node. */
10265 for (br = ret; br; br = regnext(br)) {
10266 const U8 op = PL_regkind[OP(br)];
10267 if (op == BRANCH) {
10268 REGTAIL_STUDY(pRExC_state, NEXTOPER(br), ender);
10269 if ( OP(NEXTOPER(br)) != NOTHING
10270 || regnext(NEXTOPER(br)) != ender)
10273 else if (op == BRANCHJ) {
10274 REGTAIL_STUDY(pRExC_state, NEXTOPER(NEXTOPER(br)), ender);
10275 /* for now we always disable this optimisation * /
10276 if ( OP(NEXTOPER(NEXTOPER(br))) != NOTHING
10277 || regnext(NEXTOPER(NEXTOPER(br))) != ender)
10283 br= PL_regkind[OP(ret)] != BRANCH ? regnext(ret) : ret;
10284 DEBUG_PARSE_r(if (!SIZE_ONLY) {
10285 SV * const mysv_val1=sv_newmortal();
10286 SV * const mysv_val2=sv_newmortal();
10287 DEBUG_PARSE_MSG("NADA");
10288 regprop(RExC_rx, mysv_val1, ret, NULL);
10289 regprop(RExC_rx, mysv_val2, ender, NULL);
10290 PerlIO_printf(Perl_debug_log, "~ converting ret %s (%"IVdf") to ender %s (%"IVdf") offset %"IVdf"\n",
10291 SvPV_nolen_const(mysv_val1),
10292 (IV)REG_NODE_NUM(ret),
10293 SvPV_nolen_const(mysv_val2),
10294 (IV)REG_NODE_NUM(ender),
10299 if (OP(ender) == TAIL) {
10304 for ( opt= br + 1; opt < ender ; opt++ )
10305 OP(opt)= OPTIMIZED;
10306 NEXT_OFF(br)= ender - br;
10314 static const char parens[] = "=!<,>";
10316 if (paren && (p = strchr(parens, paren))) {
10317 U8 node = ((p - parens) % 2) ? UNLESSM : IFMATCH;
10318 int flag = (p - parens) > 1;
10321 node = SUSPEND, flag = 0;
10322 reginsert(pRExC_state, node,ret, depth+1);
10323 Set_Node_Cur_Length(ret, parse_start);
10324 Set_Node_Offset(ret, parse_start + 1);
10326 REGTAIL_STUDY(pRExC_state, ret, reg_node(pRExC_state, TAIL));
10330 /* Check for proper termination. */
10332 /* restore original flags, but keep (?p) */
10333 RExC_flags = oregflags | (RExC_flags & RXf_PMf_KEEPCOPY);
10334 if (RExC_parse >= RExC_end || *nextchar(pRExC_state) != ')') {
10335 RExC_parse = oregcomp_parse;
10336 vFAIL("Unmatched (");
10339 else if (!paren && RExC_parse < RExC_end) {
10340 if (*RExC_parse == ')') {
10342 vFAIL("Unmatched )");
10345 FAIL("Junk on end of regexp"); /* "Can't happen". */
10346 assert(0); /* NOTREACHED */
10349 if (RExC_in_lookbehind) {
10350 RExC_in_lookbehind--;
10352 if (after_freeze > RExC_npar)
10353 RExC_npar = after_freeze;
10358 - regbranch - one alternative of an | operator
10360 * Implements the concatenation operator.
10362 * Returns NULL, setting *flagp to RESTART_UTF8 if the sizing scan needs to be
10366 S_regbranch(pTHX_ RExC_state_t *pRExC_state, I32 *flagp, I32 first, U32 depth)
10369 regnode *chain = NULL;
10371 I32 flags = 0, c = 0;
10372 GET_RE_DEBUG_FLAGS_DECL;
10374 PERL_ARGS_ASSERT_REGBRANCH;
10376 DEBUG_PARSE("brnc");
10381 if (!SIZE_ONLY && RExC_extralen)
10382 ret = reganode(pRExC_state, BRANCHJ,0);
10384 ret = reg_node(pRExC_state, BRANCH);
10385 Set_Node_Length(ret, 1);
10389 if (!first && SIZE_ONLY)
10390 RExC_extralen += 1; /* BRANCHJ */
10392 *flagp = WORST; /* Tentatively. */
10395 nextchar(pRExC_state);
10396 while (RExC_parse < RExC_end && *RExC_parse != '|' && *RExC_parse != ')') {
10397 flags &= ~TRYAGAIN;
10398 latest = regpiece(pRExC_state, &flags,depth+1);
10399 if (latest == NULL) {
10400 if (flags & TRYAGAIN)
10402 if (flags & RESTART_UTF8) {
10403 *flagp = RESTART_UTF8;
10406 FAIL2("panic: regpiece returned NULL, flags=%#"UVxf"", (UV) flags);
10408 else if (ret == NULL)
10410 *flagp |= flags&(HASWIDTH|POSTPONED);
10411 if (chain == NULL) /* First piece. */
10412 *flagp |= flags&SPSTART;
10415 REGTAIL(pRExC_state, chain, latest);
10420 if (chain == NULL) { /* Loop ran zero times. */
10421 chain = reg_node(pRExC_state, NOTHING);
10426 *flagp |= flags&SIMPLE;
10433 - regpiece - something followed by possible [*+?]
10435 * Note that the branching code sequences used for ? and the general cases
10436 * of * and + are somewhat optimized: they use the same NOTHING node as
10437 * both the endmarker for their branch list and the body of the last branch.
10438 * It might seem that this node could be dispensed with entirely, but the
10439 * endmarker role is not redundant.
10441 * Returns NULL, setting *flagp to TRYAGAIN if regatom() returns NULL with
10443 * Returns NULL, setting *flagp to RESTART_UTF8 if the sizing scan needs to be
10447 S_regpiece(pTHX_ RExC_state_t *pRExC_state, I32 *flagp, U32 depth)
10453 const char * const origparse = RExC_parse;
10455 I32 max = REG_INFTY;
10456 #ifdef RE_TRACK_PATTERN_OFFSETS
10459 const char *maxpos = NULL;
10461 /* Save the original in case we change the emitted regop to a FAIL. */
10462 regnode * const orig_emit = RExC_emit;
10464 GET_RE_DEBUG_FLAGS_DECL;
10466 PERL_ARGS_ASSERT_REGPIECE;
10468 DEBUG_PARSE("piec");
10470 ret = regatom(pRExC_state, &flags,depth+1);
10472 if (flags & (TRYAGAIN|RESTART_UTF8))
10473 *flagp |= flags & (TRYAGAIN|RESTART_UTF8);
10475 FAIL2("panic: regatom returned NULL, flags=%#"UVxf"", (UV) flags);
10481 if (op == '{' && regcurly(RExC_parse)) {
10483 #ifdef RE_TRACK_PATTERN_OFFSETS
10484 parse_start = RExC_parse; /* MJD */
10486 next = RExC_parse + 1;
10487 while (isDIGIT(*next) || *next == ',') {
10488 if (*next == ',') {
10496 if (*next == '}') { /* got one */
10500 min = atoi(RExC_parse);
10501 if (*maxpos == ',')
10504 maxpos = RExC_parse;
10505 max = atoi(maxpos);
10506 if (!max && *maxpos != '0')
10507 max = REG_INFTY; /* meaning "infinity" */
10508 else if (max >= REG_INFTY)
10509 vFAIL2("Quantifier in {,} bigger than %d", REG_INFTY - 1);
10511 nextchar(pRExC_state);
10512 if (max < min) { /* If can't match, warn and optimize to fail
10515 ckWARNreg(RExC_parse, "Quantifier {n,m} with n > m can't match");
10517 /* We can't back off the size because we have to reserve
10518 * enough space for all the things we are about to throw
10519 * away, but we can shrink it by the ammount we are about
10520 * to re-use here */
10521 RExC_size = PREVOPER(RExC_size) - regarglen[(U8)OPFAIL];
10524 RExC_emit = orig_emit;
10526 ret = reg_node(pRExC_state, OPFAIL);
10529 else if (min == max
10530 && RExC_parse < RExC_end
10531 && (*RExC_parse == '?' || *RExC_parse == '+'))
10534 ckWARN2reg(RExC_parse + 1,
10535 "Useless use of greediness modifier '%c'",
10538 /* Absorb the modifier, so later code doesn't see nor use
10540 nextchar(pRExC_state);
10544 if ((flags&SIMPLE)) {
10545 RExC_naughty += 2 + RExC_naughty / 2;
10546 reginsert(pRExC_state, CURLY, ret, depth+1);
10547 Set_Node_Offset(ret, parse_start+1); /* MJD */
10548 Set_Node_Cur_Length(ret, parse_start);
10551 regnode * const w = reg_node(pRExC_state, WHILEM);
10554 REGTAIL(pRExC_state, ret, w);
10555 if (!SIZE_ONLY && RExC_extralen) {
10556 reginsert(pRExC_state, LONGJMP,ret, depth+1);
10557 reginsert(pRExC_state, NOTHING,ret, depth+1);
10558 NEXT_OFF(ret) = 3; /* Go over LONGJMP. */
10560 reginsert(pRExC_state, CURLYX,ret, depth+1);
10562 Set_Node_Offset(ret, parse_start+1);
10563 Set_Node_Length(ret,
10564 op == '{' ? (RExC_parse - parse_start) : 1);
10566 if (!SIZE_ONLY && RExC_extralen)
10567 NEXT_OFF(ret) = 3; /* Go over NOTHING to LONGJMP. */
10568 REGTAIL(pRExC_state, ret, reg_node(pRExC_state, NOTHING));
10570 RExC_whilem_seen++, RExC_extralen += 3;
10571 RExC_naughty += 4 + RExC_naughty; /* compound interest */
10578 *flagp |= HASWIDTH;
10580 ARG1_SET(ret, (U16)min);
10581 ARG2_SET(ret, (U16)max);
10583 if (max == REG_INFTY)
10584 RExC_seen |= REG_UNBOUNDED_QUANTIFIER_SEEN;
10590 if (!ISMULT1(op)) {
10595 #if 0 /* Now runtime fix should be reliable. */
10597 /* if this is reinstated, don't forget to put this back into perldiag:
10599 =item Regexp *+ operand could be empty at {#} in regex m/%s/
10601 (F) The part of the regexp subject to either the * or + quantifier
10602 could match an empty string. The {#} shows in the regular
10603 expression about where the problem was discovered.
10607 if (!(flags&HASWIDTH) && op != '?')
10608 vFAIL("Regexp *+ operand could be empty");
10611 #ifdef RE_TRACK_PATTERN_OFFSETS
10612 parse_start = RExC_parse;
10614 nextchar(pRExC_state);
10616 *flagp = (op != '+') ? (WORST|SPSTART|HASWIDTH) : (WORST|HASWIDTH);
10618 if (op == '*' && (flags&SIMPLE)) {
10619 reginsert(pRExC_state, STAR, ret, depth+1);
10622 RExC_seen |= REG_UNBOUNDED_QUANTIFIER_SEEN;
10624 else if (op == '*') {
10628 else if (op == '+' && (flags&SIMPLE)) {
10629 reginsert(pRExC_state, PLUS, ret, depth+1);
10632 RExC_seen |= REG_UNBOUNDED_QUANTIFIER_SEEN;
10634 else if (op == '+') {
10638 else if (op == '?') {
10643 if (!SIZE_ONLY && !(flags&(HASWIDTH|POSTPONED)) && max > REG_INFTY/3) {
10644 SAVEFREESV(RExC_rx_sv); /* in case of fatal warnings */
10645 ckWARN2reg(RExC_parse,
10646 "%"UTF8f" matches null string many times",
10647 UTF8fARG(UTF, (RExC_parse >= origparse
10648 ? RExC_parse - origparse
10651 (void)ReREFCNT_inc(RExC_rx_sv);
10654 if (RExC_parse < RExC_end && *RExC_parse == '?') {
10655 nextchar(pRExC_state);
10656 reginsert(pRExC_state, MINMOD, ret, depth+1);
10657 REGTAIL(pRExC_state, ret, ret + NODE_STEP_REGNODE);
10660 if (RExC_parse < RExC_end && *RExC_parse == '+') {
10662 nextchar(pRExC_state);
10663 ender = reg_node(pRExC_state, SUCCEED);
10664 REGTAIL(pRExC_state, ret, ender);
10665 reginsert(pRExC_state, SUSPEND, ret, depth+1);
10667 ender = reg_node(pRExC_state, TAIL);
10668 REGTAIL(pRExC_state, ret, ender);
10671 if (RExC_parse < RExC_end && ISMULT2(RExC_parse)) {
10673 vFAIL("Nested quantifiers");
10680 S_grok_bslash_N(pTHX_ RExC_state_t *pRExC_state, regnode** node_p,
10681 UV *valuep, I32 *flagp, U32 depth, bool in_char_class,
10682 const bool strict /* Apply stricter parsing rules? */
10686 /* This is expected to be called by a parser routine that has recognized '\N'
10687 and needs to handle the rest. RExC_parse is expected to point at the first
10688 char following the N at the time of the call. On successful return,
10689 RExC_parse has been updated to point to just after the sequence identified
10690 by this routine, and <*flagp> has been updated.
10692 The \N may be inside (indicated by the boolean <in_char_class>) or outside a
10695 \N may begin either a named sequence, or if outside a character class, mean
10696 to match a non-newline. For non single-quoted regexes, the tokenizer has
10697 attempted to decide which, and in the case of a named sequence, converted it
10698 into one of the forms: \N{} (if the sequence is null), or \N{U+c1.c2...},
10699 where c1... are the characters in the sequence. For single-quoted regexes,
10700 the tokenizer passes the \N sequence through unchanged; this code will not
10701 attempt to determine this nor expand those, instead raising a syntax error.
10702 The net effect is that if the beginning of the passed-in pattern isn't '{U+'
10703 or there is no '}', it signals that this \N occurrence means to match a
10706 Only the \N{U+...} form should occur in a character class, for the same
10707 reason that '.' inside a character class means to just match a period: it
10708 just doesn't make sense.
10710 The function raises an error (via vFAIL), and doesn't return for various
10711 syntax errors. Otherwise it returns TRUE and sets <node_p> or <valuep> on
10712 success; it returns FALSE otherwise. Returns FALSE, setting *flagp to
10713 RESTART_UTF8 if the sizing scan needs to be restarted. Such a restart is
10714 only possible if node_p is non-NULL.
10717 If <valuep> is non-null, it means the caller can accept an input sequence
10718 consisting of a just a single code point; <*valuep> is set to that value
10719 if the input is such.
10721 If <node_p> is non-null it signifies that the caller can accept any other
10722 legal sequence (i.e., one that isn't just a single code point). <*node_p>
10724 1) \N means not-a-NL: points to a newly created REG_ANY node;
10725 2) \N{}: points to a new NOTHING node;
10726 3) otherwise: points to a new EXACT node containing the resolved
10728 Note that FALSE is returned for single code point sequences if <valuep> is
10732 char * endbrace; /* '}' following the name */
10734 char *endchar; /* Points to '.' or '}' ending cur char in the input
10736 bool has_multiple_chars; /* true if the input stream contains a sequence of
10737 more than one character */
10739 GET_RE_DEBUG_FLAGS_DECL;
10741 PERL_ARGS_ASSERT_GROK_BSLASH_N;
10743 GET_RE_DEBUG_FLAGS;
10745 assert(cBOOL(node_p) ^ cBOOL(valuep)); /* Exactly one should be set */
10747 /* The [^\n] meaning of \N ignores spaces and comments under the /x
10748 * modifier. The other meaning does not, so use a temporary until we find
10749 * out which we are being called with */
10750 p = (RExC_flags & RXf_PMf_EXTENDED)
10751 ? regpatws(pRExC_state, RExC_parse,
10752 TRUE) /* means recognize comments */
10755 /* Disambiguate between \N meaning a named character versus \N meaning
10756 * [^\n]. The former is assumed when it can't be the latter. */
10757 if (*p != '{' || regcurly(p)) {
10760 /* no bare \N allowed in a charclass */
10761 if (in_char_class) {
10762 vFAIL("\\N in a character class must be a named character: \\N{...}");
10766 RExC_parse--; /* Need to back off so nextchar() doesn't skip the
10768 nextchar(pRExC_state);
10769 *node_p = reg_node(pRExC_state, REG_ANY);
10770 *flagp |= HASWIDTH|SIMPLE;
10772 Set_Node_Length(*node_p, 1); /* MJD */
10776 /* Here, we have decided it should be a named character or sequence */
10778 /* The test above made sure that the next real character is a '{', but
10779 * under the /x modifier, it could be separated by space (or a comment and
10780 * \n) and this is not allowed (for consistency with \x{...} and the
10781 * tokenizer handling of \N{NAME}). */
10782 if (*RExC_parse != '{') {
10783 vFAIL("Missing braces on \\N{}");
10786 RExC_parse++; /* Skip past the '{' */
10788 if (! (endbrace = strchr(RExC_parse, '}')) /* no trailing brace */
10789 || ! (endbrace == RExC_parse /* nothing between the {} */
10790 || (endbrace - RExC_parse >= 2 /* U+ (bad hex is checked below
10792 && strnEQ(RExC_parse, "U+", 2)))) /* for a better error msg)
10795 if (endbrace) RExC_parse = endbrace; /* position msg's '<--HERE' */
10796 vFAIL("\\N{NAME} must be resolved by the lexer");
10799 if (endbrace == RExC_parse) { /* empty: \N{} */
10802 *node_p = reg_node(pRExC_state,NOTHING);
10804 else if (in_char_class) {
10805 if (SIZE_ONLY && in_char_class) {
10807 RExC_parse++; /* Position after the "}" */
10808 vFAIL("Zero length \\N{}");
10811 ckWARNreg(RExC_parse,
10812 "Ignoring zero length \\N{} in character class");
10820 nextchar(pRExC_state);
10824 RExC_uni_semantics = 1; /* Unicode named chars imply Unicode semantics */
10825 RExC_parse += 2; /* Skip past the 'U+' */
10827 endchar = RExC_parse + strcspn(RExC_parse, ".}");
10829 /* Code points are separated by dots. If none, there is only one code
10830 * point, and is terminated by the brace */
10831 has_multiple_chars = (endchar < endbrace);
10833 if (valuep && (! has_multiple_chars || in_char_class)) {
10834 /* We only pay attention to the first char of
10835 multichar strings being returned in char classes. I kinda wonder
10836 if this makes sense as it does change the behaviour
10837 from earlier versions, OTOH that behaviour was broken
10838 as well. XXX Solution is to recharacterize as
10839 [rest-of-class]|multi1|multi2... */
10841 STRLEN length_of_hex = (STRLEN)(endchar - RExC_parse);
10842 I32 grok_hex_flags = PERL_SCAN_ALLOW_UNDERSCORES
10843 | PERL_SCAN_DISALLOW_PREFIX
10844 | (SIZE_ONLY ? PERL_SCAN_SILENT_ILLDIGIT : 0);
10846 *valuep = grok_hex(RExC_parse, &length_of_hex, &grok_hex_flags, NULL);
10848 /* The tokenizer should have guaranteed validity, but it's possible to
10849 * bypass it by using single quoting, so check */
10850 if (length_of_hex == 0
10851 || length_of_hex != (STRLEN)(endchar - RExC_parse) )
10853 RExC_parse += length_of_hex; /* Includes all the valid */
10854 RExC_parse += (RExC_orig_utf8) /* point to after 1st invalid */
10855 ? UTF8SKIP(RExC_parse)
10857 /* Guard against malformed utf8 */
10858 if (RExC_parse >= endchar) {
10859 RExC_parse = endchar;
10861 vFAIL("Invalid hexadecimal number in \\N{U+...}");
10864 if (in_char_class && has_multiple_chars) {
10866 RExC_parse = endbrace;
10867 vFAIL("\\N{} in character class restricted to one character");
10870 ckWARNreg(endchar, "Using just the first character returned by \\N{} in character class");
10874 RExC_parse = endbrace + 1;
10876 else if (! node_p || ! has_multiple_chars) {
10878 /* Here, the input is legal, but not according to the caller's
10879 * options. We fail without advancing the parse, so that the
10880 * caller can try again */
10886 /* What is done here is to convert this to a sub-pattern of the form
10887 * (?:\x{char1}\x{char2}...)
10888 * and then call reg recursively. That way, it retains its atomicness,
10889 * while not having to worry about special handling that some code
10890 * points may have. toke.c has converted the original Unicode values
10891 * to native, so that we can just pass on the hex values unchanged. We
10892 * do have to set a flag to keep recoding from happening in the
10895 SV * substitute_parse = newSVpvn_flags("?:", 2, SVf_UTF8|SVs_TEMP);
10897 char *orig_end = RExC_end;
10900 while (RExC_parse < endbrace) {
10902 /* Convert to notation the rest of the code understands */
10903 sv_catpv(substitute_parse, "\\x{");
10904 sv_catpvn(substitute_parse, RExC_parse, endchar - RExC_parse);
10905 sv_catpv(substitute_parse, "}");
10907 /* Point to the beginning of the next character in the sequence. */
10908 RExC_parse = endchar + 1;
10909 endchar = RExC_parse + strcspn(RExC_parse, ".}");
10911 sv_catpv(substitute_parse, ")");
10913 RExC_parse = SvPV(substitute_parse, len);
10915 /* Don't allow empty number */
10917 vFAIL("Invalid hexadecimal number in \\N{U+...}");
10919 RExC_end = RExC_parse + len;
10921 /* The values are Unicode, and therefore not subject to recoding */
10922 RExC_override_recoding = 1;
10924 if (!(*node_p = reg(pRExC_state, 1, &flags, depth+1))) {
10925 if (flags & RESTART_UTF8) {
10926 *flagp = RESTART_UTF8;
10929 FAIL2("panic: reg returned NULL to grok_bslash_N, flags=%#"UVxf"",
10932 *flagp |= flags&(HASWIDTH|SPSTART|SIMPLE|POSTPONED);
10934 RExC_parse = endbrace;
10935 RExC_end = orig_end;
10936 RExC_override_recoding = 0;
10938 nextchar(pRExC_state);
10948 * It returns the code point in utf8 for the value in *encp.
10949 * value: a code value in the source encoding
10950 * encp: a pointer to an Encode object
10952 * If the result from Encode is not a single character,
10953 * it returns U+FFFD (Replacement character) and sets *encp to NULL.
10956 S_reg_recode(pTHX_ const char value, SV **encp)
10959 SV * const sv = newSVpvn_flags(&value, numlen, SVs_TEMP);
10960 const char * const s = *encp ? sv_recode_to_utf8(sv, *encp) : SvPVX(sv);
10961 const STRLEN newlen = SvCUR(sv);
10962 UV uv = UNICODE_REPLACEMENT;
10964 PERL_ARGS_ASSERT_REG_RECODE;
10968 ? utf8n_to_uvchr((U8*)s, newlen, &numlen, UTF8_ALLOW_DEFAULT)
10971 if (!newlen || numlen != newlen) {
10972 uv = UNICODE_REPLACEMENT;
10978 PERL_STATIC_INLINE U8
10979 S_compute_EXACTish(RExC_state_t *pRExC_state)
10983 PERL_ARGS_ASSERT_COMPUTE_EXACTISH;
10989 op = get_regex_charset(RExC_flags);
10990 if (op >= REGEX_ASCII_RESTRICTED_CHARSET) {
10991 op--; /* /a is same as /u, and map /aa's offset to what /a's would have
10992 been, so there is no hole */
10995 return op + EXACTF;
10998 PERL_STATIC_INLINE void
10999 S_alloc_maybe_populate_EXACT(pTHX_ RExC_state_t *pRExC_state,
11000 regnode *node, I32* flagp, STRLEN len, UV code_point,
11003 /* This knows the details about sizing an EXACTish node, setting flags for
11004 * it (by setting <*flagp>, and potentially populating it with a single
11007 * If <len> (the length in bytes) is non-zero, this function assumes that
11008 * the node has already been populated, and just does the sizing. In this
11009 * case <code_point> should be the final code point that has already been
11010 * placed into the node. This value will be ignored except that under some
11011 * circumstances <*flagp> is set based on it.
11013 * If <len> is zero, the function assumes that the node is to contain only
11014 * the single character given by <code_point> and calculates what <len>
11015 * should be. In pass 1, it sizes the node appropriately. In pass 2, it
11016 * additionally will populate the node's STRING with <code_point> or its
11019 * In both cases <*flagp> is appropriately set
11021 * It knows that under FOLD, the Latin Sharp S and UTF characters above
11022 * 255, must be folded (the former only when the rules indicate it can
11025 * When it does the populating, it looks at the flag 'downgradable'. If
11026 * true with a node that folds, it checks if the single code point
11027 * participates in a fold, and if not downgrades the node to an EXACT.
11028 * This helps the optimizer */
11030 bool len_passed_in = cBOOL(len != 0);
11031 U8 character[UTF8_MAXBYTES_CASE+1];
11033 PERL_ARGS_ASSERT_ALLOC_MAYBE_POPULATE_EXACT;
11035 /* Don't bother to check for downgrading in PASS1, as it doesn't make any
11036 * sizing difference, and is extra work that is thrown away */
11037 if (downgradable && ! PASS2) {
11038 downgradable = FALSE;
11041 if (! len_passed_in) {
11043 if (UNI_IS_INVARIANT(code_point)) {
11044 if (LOC || ! FOLD) { /* /l defers folding until runtime */
11045 *character = (U8) code_point;
11047 else { /* Here is /i and not /l (toFOLD() is defined on just
11048 ASCII, which isn't the same thing as INVARIANT on
11049 EBCDIC, but it works there, as the extra invariants
11050 fold to themselves) */
11051 *character = toFOLD((U8) code_point);
11053 && *character == code_point
11054 && ! HAS_NONLATIN1_FOLD_CLOSURE(code_point))
11061 else if (FOLD && (! LOC
11062 || ! is_PROBLEMATIC_LOCALE_FOLD_cp(code_point)))
11063 { /* Folding, and ok to do so now */
11064 UV folded = _to_uni_fold_flags(
11068 FOLD_FLAGS_FULL | ((ASCII_FOLD_RESTRICTED)
11069 ? FOLD_FLAGS_NOMIX_ASCII
11072 && folded == code_point
11073 && ! _invlist_contains_cp(PL_utf8_foldable, code_point))
11078 else if (code_point <= MAX_UTF8_TWO_BYTE) {
11080 /* Not folding this cp, and can output it directly */
11081 *character = UTF8_TWO_BYTE_HI(code_point);
11082 *(character + 1) = UTF8_TWO_BYTE_LO(code_point);
11086 uvchr_to_utf8( character, code_point);
11087 len = UTF8SKIP(character);
11089 } /* Else pattern isn't UTF8. */
11091 *character = (U8) code_point;
11093 } /* Else is folded non-UTF8 */
11094 else if (LIKELY(code_point != LATIN_SMALL_LETTER_SHARP_S)) {
11096 /* We don't fold any non-UTF8 except possibly the Sharp s (see
11097 * comments at join_exact()); */
11098 *character = (U8) code_point;
11101 /* Can turn into an EXACT node if we know the fold at compile time,
11102 * and it folds to itself and doesn't particpate in other folds */
11105 && PL_fold_latin1[code_point] == code_point
11106 && (! HAS_NONLATIN1_FOLD_CLOSURE(code_point)
11107 || (isASCII(code_point) && ASCII_FOLD_RESTRICTED)))
11111 } /* else is Sharp s. May need to fold it */
11112 else if (AT_LEAST_UNI_SEMANTICS && ! ASCII_FOLD_RESTRICTED) {
11114 *(character + 1) = 's';
11118 *character = LATIN_SMALL_LETTER_SHARP_S;
11124 RExC_size += STR_SZ(len);
11127 RExC_emit += STR_SZ(len);
11128 STR_LEN(node) = len;
11129 if (! len_passed_in) {
11130 Copy((char *) character, STRING(node), len, char);
11134 *flagp |= HASWIDTH;
11136 /* A single character node is SIMPLE, except for the special-cased SHARP S
11138 if ((len == 1 || (UTF && len == UNISKIP(code_point)))
11139 && (code_point != LATIN_SMALL_LETTER_SHARP_S
11140 || ! FOLD || ! DEPENDS_SEMANTICS))
11145 /* The OP may not be well defined in PASS1 */
11146 if (PASS2 && OP(node) == EXACTFL) {
11147 RExC_contains_locale = 1;
11152 /* return atoi(p), unless it's too big to sensibly be a backref,
11153 * in which case return I32_MAX (rather than possibly 32-bit wrapping) */
11156 S_backref_value(char *p)
11160 for (;isDIGIT(*q); q++) {} /* calculate length of num */
11161 if (q - p == 0 || q - p > 9)
11168 - regatom - the lowest level
11170 Try to identify anything special at the start of the pattern. If there
11171 is, then handle it as required. This may involve generating a single regop,
11172 such as for an assertion; or it may involve recursing, such as to
11173 handle a () structure.
11175 If the string doesn't start with something special then we gobble up
11176 as much literal text as we can.
11178 Once we have been able to handle whatever type of thing started the
11179 sequence, we return.
11181 Note: we have to be careful with escapes, as they can be both literal
11182 and special, and in the case of \10 and friends, context determines which.
11184 A summary of the code structure is:
11186 switch (first_byte) {
11187 cases for each special:
11188 handle this special;
11191 switch (2nd byte) {
11192 cases for each unambiguous special:
11193 handle this special;
11195 cases for each ambigous special/literal:
11197 if (special) handle here
11199 default: // unambiguously literal:
11202 default: // is a literal char
11205 create EXACTish node for literal;
11206 while (more input and node isn't full) {
11207 switch (input_byte) {
11208 cases for each special;
11209 make sure parse pointer is set so that the next call to
11210 regatom will see this special first
11211 goto loopdone; // EXACTish node terminated by prev. char
11213 append char to EXACTISH node;
11215 get next input byte;
11219 return the generated node;
11221 Specifically there are two separate switches for handling
11222 escape sequences, with the one for handling literal escapes requiring
11223 a dummy entry for all of the special escapes that are actually handled
11226 Returns NULL, setting *flagp to TRYAGAIN if reg() returns NULL with
11228 Returns NULL, setting *flagp to RESTART_UTF8 if the sizing scan needs to be
11230 Otherwise does not return NULL.
11234 S_regatom(pTHX_ RExC_state_t *pRExC_state, I32 *flagp, U32 depth)
11236 regnode *ret = NULL;
11238 char *parse_start = RExC_parse;
11243 GET_RE_DEBUG_FLAGS_DECL;
11245 *flagp = WORST; /* Tentatively. */
11247 DEBUG_PARSE("atom");
11249 PERL_ARGS_ASSERT_REGATOM;
11252 switch ((U8)*RExC_parse) {
11254 RExC_seen_zerolen++;
11255 nextchar(pRExC_state);
11256 if (RExC_flags & RXf_PMf_MULTILINE)
11257 ret = reg_node(pRExC_state, MBOL);
11258 else if (RExC_flags & RXf_PMf_SINGLELINE)
11259 ret = reg_node(pRExC_state, SBOL);
11261 ret = reg_node(pRExC_state, BOL);
11262 Set_Node_Length(ret, 1); /* MJD */
11265 nextchar(pRExC_state);
11267 RExC_seen_zerolen++;
11268 if (RExC_flags & RXf_PMf_MULTILINE)
11269 ret = reg_node(pRExC_state, MEOL);
11270 else if (RExC_flags & RXf_PMf_SINGLELINE)
11271 ret = reg_node(pRExC_state, SEOL);
11273 ret = reg_node(pRExC_state, EOL);
11274 Set_Node_Length(ret, 1); /* MJD */
11277 nextchar(pRExC_state);
11278 if (RExC_flags & RXf_PMf_SINGLELINE)
11279 ret = reg_node(pRExC_state, SANY);
11281 ret = reg_node(pRExC_state, REG_ANY);
11282 *flagp |= HASWIDTH|SIMPLE;
11284 Set_Node_Length(ret, 1); /* MJD */
11288 char * const oregcomp_parse = ++RExC_parse;
11289 ret = regclass(pRExC_state, flagp,depth+1,
11290 FALSE, /* means parse the whole char class */
11291 TRUE, /* allow multi-char folds */
11292 FALSE, /* don't silence non-portable warnings. */
11294 if (*RExC_parse != ']') {
11295 RExC_parse = oregcomp_parse;
11296 vFAIL("Unmatched [");
11299 if (*flagp & RESTART_UTF8)
11301 FAIL2("panic: regclass returned NULL to regatom, flags=%#"UVxf"",
11304 nextchar(pRExC_state);
11305 Set_Node_Length(ret, RExC_parse - oregcomp_parse + 1); /* MJD */
11309 nextchar(pRExC_state);
11310 ret = reg(pRExC_state, 2, &flags,depth+1);
11312 if (flags & TRYAGAIN) {
11313 if (RExC_parse == RExC_end) {
11314 /* Make parent create an empty node if needed. */
11315 *flagp |= TRYAGAIN;
11320 if (flags & RESTART_UTF8) {
11321 *flagp = RESTART_UTF8;
11324 FAIL2("panic: reg returned NULL to regatom, flags=%#"UVxf"",
11327 *flagp |= flags&(HASWIDTH|SPSTART|SIMPLE|POSTPONED);
11331 if (flags & TRYAGAIN) {
11332 *flagp |= TRYAGAIN;
11335 vFAIL("Internal urp");
11336 /* Supposed to be caught earlier. */
11342 vFAIL("Quantifier follows nothing");
11347 This switch handles escape sequences that resolve to some kind
11348 of special regop and not to literal text. Escape sequnces that
11349 resolve to literal text are handled below in the switch marked
11352 Every entry in this switch *must* have a corresponding entry
11353 in the literal escape switch. However, the opposite is not
11354 required, as the default for this switch is to jump to the
11355 literal text handling code.
11357 switch ((U8)*++RExC_parse) {
11358 /* Special Escapes */
11360 RExC_seen_zerolen++;
11361 ret = reg_node(pRExC_state, SBOL);
11363 goto finish_meta_pat;
11365 ret = reg_node(pRExC_state, GPOS);
11366 RExC_seen |= REG_GPOS_SEEN;
11368 goto finish_meta_pat;
11370 RExC_seen_zerolen++;
11371 ret = reg_node(pRExC_state, KEEPS);
11373 /* XXX:dmq : disabling in-place substitution seems to
11374 * be necessary here to avoid cases of memory corruption, as
11375 * with: C<$_="x" x 80; s/x\K/y/> -- rgs
11377 RExC_seen |= REG_LOOKBEHIND_SEEN;
11378 goto finish_meta_pat;
11380 ret = reg_node(pRExC_state, SEOL);
11382 RExC_seen_zerolen++; /* Do not optimize RE away */
11383 goto finish_meta_pat;
11385 ret = reg_node(pRExC_state, EOS);
11387 RExC_seen_zerolen++; /* Do not optimize RE away */
11388 goto finish_meta_pat;
11390 ret = reg_node(pRExC_state, CANY);
11391 RExC_seen |= REG_CANY_SEEN;
11392 *flagp |= HASWIDTH|SIMPLE;
11394 ckWARNdep(RExC_parse+1, "\\C is deprecated");
11396 goto finish_meta_pat;
11398 ret = reg_node(pRExC_state, CLUMP);
11399 *flagp |= HASWIDTH;
11400 goto finish_meta_pat;
11406 arg = ANYOF_WORDCHAR;
11410 RExC_seen_zerolen++;
11411 RExC_seen |= REG_LOOKBEHIND_SEEN;
11412 op = BOUND + get_regex_charset(RExC_flags);
11413 if (op > BOUNDA) { /* /aa is same as /a */
11416 else if (op == BOUNDL) {
11417 RExC_contains_locale = 1;
11419 ret = reg_node(pRExC_state, op);
11420 FLAGS(ret) = get_regex_charset(RExC_flags);
11422 if (! SIZE_ONLY && (U8) *(RExC_parse + 1) == '{') {
11423 /* diag_listed_as: Use "%s" instead of "%s" */
11424 vFAIL("Use \"\\b\\{\" instead of \"\\b{\"");
11426 goto finish_meta_pat;
11428 RExC_seen_zerolen++;
11429 RExC_seen |= REG_LOOKBEHIND_SEEN;
11430 op = NBOUND + get_regex_charset(RExC_flags);
11431 if (op > NBOUNDA) { /* /aa is same as /a */
11434 else if (op == NBOUNDL) {
11435 RExC_contains_locale = 1;
11437 ret = reg_node(pRExC_state, op);
11438 FLAGS(ret) = get_regex_charset(RExC_flags);
11440 if (! SIZE_ONLY && (U8) *(RExC_parse + 1) == '{') {
11441 /* diag_listed_as: Use "%s" instead of "%s" */
11442 vFAIL("Use \"\\B\\{\" instead of \"\\B{\"");
11444 goto finish_meta_pat;
11454 ret = reg_node(pRExC_state, LNBREAK);
11455 *flagp |= HASWIDTH|SIMPLE;
11456 goto finish_meta_pat;
11464 goto join_posix_op_known;
11470 arg = ANYOF_VERTWS;
11472 goto join_posix_op_known;
11482 op = POSIXD + get_regex_charset(RExC_flags);
11483 if (op > POSIXA) { /* /aa is same as /a */
11486 else if (op == POSIXL) {
11487 RExC_contains_locale = 1;
11490 join_posix_op_known:
11493 op += NPOSIXD - POSIXD;
11496 ret = reg_node(pRExC_state, op);
11498 FLAGS(ret) = namedclass_to_classnum(arg);
11501 *flagp |= HASWIDTH|SIMPLE;
11505 nextchar(pRExC_state);
11506 Set_Node_Length(ret, 2); /* MJD */
11512 char* parse_start = RExC_parse - 2;
11517 ret = regclass(pRExC_state, flagp,depth+1,
11518 TRUE, /* means just parse this element */
11519 FALSE, /* don't allow multi-char folds */
11520 FALSE, /* don't silence non-portable warnings.
11521 It would be a bug if these returned
11524 /* regclass() can only return RESTART_UTF8 if multi-char folds
11527 FAIL2("panic: regclass returned NULL to regatom, flags=%#"UVxf"",
11532 Set_Node_Offset(ret, parse_start + 2);
11533 Set_Node_Cur_Length(ret, parse_start);
11534 nextchar(pRExC_state);
11538 /* Handle \N and \N{NAME} with multiple code points here and not
11539 * below because it can be multicharacter. join_exact() will join
11540 * them up later on. Also this makes sure that things like
11541 * /\N{BLAH}+/ and \N{BLAH} being multi char Just Happen. dmq.
11542 * The options to the grok function call causes it to fail if the
11543 * sequence is just a single code point. We then go treat it as
11544 * just another character in the current EXACT node, and hence it
11545 * gets uniform treatment with all the other characters. The
11546 * special treatment for quantifiers is not needed for such single
11547 * character sequences */
11549 if (! grok_bslash_N(pRExC_state, &ret, NULL, flagp, depth, FALSE,
11550 FALSE /* not strict */ )) {
11551 if (*flagp & RESTART_UTF8)
11557 case 'k': /* Handle \k<NAME> and \k'NAME' */
11560 char ch= RExC_parse[1];
11561 if (ch != '<' && ch != '\'' && ch != '{') {
11563 /* diag_listed_as: Sequence \%s... not terminated in regex; marked by <-- HERE in m/%s/ */
11564 vFAIL2("Sequence %.2s... not terminated",parse_start);
11566 /* this pretty much dupes the code for (?P=...) in reg(), if
11567 you change this make sure you change that */
11568 char* name_start = (RExC_parse += 2);
11570 SV *sv_dat = reg_scan_name(pRExC_state,
11571 SIZE_ONLY ? REG_RSN_RETURN_NULL : REG_RSN_RETURN_DATA);
11572 ch= (ch == '<') ? '>' : (ch == '{') ? '}' : '\'';
11573 if (RExC_parse == name_start || *RExC_parse != ch)
11574 /* diag_listed_as: Sequence \%s... not terminated in regex; marked by <-- HERE in m/%s/ */
11575 vFAIL2("Sequence %.3s... not terminated",parse_start);
11578 num = add_data( pRExC_state, STR_WITH_LEN("S"));
11579 RExC_rxi->data->data[num]=(void*)sv_dat;
11580 SvREFCNT_inc_simple_void(sv_dat);
11584 ret = reganode(pRExC_state,
11587 : (ASCII_FOLD_RESTRICTED)
11589 : (AT_LEAST_UNI_SEMANTICS)
11595 *flagp |= HASWIDTH;
11597 /* override incorrect value set in reganode MJD */
11598 Set_Node_Offset(ret, parse_start+1);
11599 Set_Node_Cur_Length(ret, parse_start);
11600 nextchar(pRExC_state);
11606 case '1': case '2': case '3': case '4':
11607 case '5': case '6': case '7': case '8': case '9':
11612 if (*RExC_parse == 'g') {
11616 if (*RExC_parse == '{') {
11620 if (*RExC_parse == '-') {
11624 if (hasbrace && !isDIGIT(*RExC_parse)) {
11625 if (isrel) RExC_parse--;
11627 goto parse_named_seq;
11630 num = S_backref_value(RExC_parse);
11632 vFAIL("Reference to invalid group 0");
11633 else if (num == I32_MAX) {
11634 if (isDIGIT(*RExC_parse))
11635 vFAIL("Reference to nonexistent group");
11637 vFAIL("Unterminated \\g... pattern");
11641 num = RExC_npar - num;
11643 vFAIL("Reference to nonexistent or unclosed group");
11647 num = S_backref_value(RExC_parse);
11648 /* bare \NNN might be backref or octal - if it is larger than or equal
11649 * RExC_npar then it is assumed to be and octal escape.
11650 * Note RExC_npar is +1 from the actual number of parens*/
11651 if (num == I32_MAX || (num > 9 && num >= RExC_npar
11652 && *RExC_parse != '8' && *RExC_parse != '9'))
11654 /* Probably a character specified in octal, e.g. \35 */
11659 /* at this point RExC_parse definitely points to a backref
11662 #ifdef RE_TRACK_PATTERN_OFFSETS
11663 char * const parse_start = RExC_parse - 1; /* MJD */
11665 while (isDIGIT(*RExC_parse))
11668 if (*RExC_parse != '}')
11669 vFAIL("Unterminated \\g{...} pattern");
11673 if (num > (I32)RExC_rx->nparens)
11674 vFAIL("Reference to nonexistent group");
11677 ret = reganode(pRExC_state,
11680 : (ASCII_FOLD_RESTRICTED)
11682 : (AT_LEAST_UNI_SEMANTICS)
11688 *flagp |= HASWIDTH;
11690 /* override incorrect value set in reganode MJD */
11691 Set_Node_Offset(ret, parse_start+1);
11692 Set_Node_Cur_Length(ret, parse_start);
11694 nextchar(pRExC_state);
11699 if (RExC_parse >= RExC_end)
11700 FAIL("Trailing \\");
11703 /* Do not generate "unrecognized" warnings here, we fall
11704 back into the quick-grab loop below */
11711 if (RExC_flags & RXf_PMf_EXTENDED) {
11712 RExC_parse = reg_skipcomment( pRExC_state, RExC_parse );
11713 if (RExC_parse < RExC_end)
11720 parse_start = RExC_parse - 1;
11729 #define MAX_NODE_STRING_SIZE 127
11730 char foldbuf[MAX_NODE_STRING_SIZE+UTF8_MAXBYTES_CASE];
11732 U8 upper_parse = MAX_NODE_STRING_SIZE;
11733 U8 node_type = compute_EXACTish(pRExC_state);
11734 bool next_is_quantifier;
11735 char * oldp = NULL;
11737 /* We can convert EXACTF nodes to EXACTFU if they contain only
11738 * characters that match identically regardless of the target
11739 * string's UTF8ness. The reason to do this is that EXACTF is not
11740 * trie-able, EXACTFU is.
11742 * Similarly, we can convert EXACTFL nodes to EXACTFU if they
11743 * contain only above-Latin1 characters (hence must be in UTF8),
11744 * which don't participate in folds with Latin1-range characters,
11745 * as the latter's folds aren't known until runtime. (We don't
11746 * need to figure this out until pass 2) */
11747 bool maybe_exactfu = PASS2
11748 && (node_type == EXACTF || node_type == EXACTFL);
11750 /* If a folding node contains only code points that don't
11751 * participate in folds, it can be changed into an EXACT node,
11752 * which allows the optimizer more things to look for */
11755 ret = reg_node(pRExC_state, node_type);
11757 /* In pass1, folded, we use a temporary buffer instead of the
11758 * actual node, as the node doesn't exist yet */
11759 s = (SIZE_ONLY && FOLD) ? foldbuf : STRING(ret);
11765 /* We do the EXACTFish to EXACT node only if folding. (And we
11766 * don't need to figure this out until pass 2) */
11767 maybe_exact = FOLD && PASS2;
11769 /* XXX The node can hold up to 255 bytes, yet this only goes to
11770 * 127. I (khw) do not know why. Keeping it somewhat less than
11771 * 255 allows us to not have to worry about overflow due to
11772 * converting to utf8 and fold expansion, but that value is
11773 * 255-UTF8_MAXBYTES_CASE. join_exact() may join adjacent nodes
11774 * split up by this limit into a single one using the real max of
11775 * 255. Even at 127, this breaks under rare circumstances. If
11776 * folding, we do not want to split a node at a character that is a
11777 * non-final in a multi-char fold, as an input string could just
11778 * happen to want to match across the node boundary. The join
11779 * would solve that problem if the join actually happens. But a
11780 * series of more than two nodes in a row each of 127 would cause
11781 * the first join to succeed to get to 254, but then there wouldn't
11782 * be room for the next one, which could at be one of those split
11783 * multi-char folds. I don't know of any fool-proof solution. One
11784 * could back off to end with only a code point that isn't such a
11785 * non-final, but it is possible for there not to be any in the
11787 for (p = RExC_parse - 1;
11788 len < upper_parse && p < RExC_end;
11793 if (RExC_flags & RXf_PMf_EXTENDED)
11794 p = regpatws(pRExC_state, p,
11795 TRUE); /* means recognize comments */
11806 /* Literal Escapes Switch
11808 This switch is meant to handle escape sequences that
11809 resolve to a literal character.
11811 Every escape sequence that represents something
11812 else, like an assertion or a char class, is handled
11813 in the switch marked 'Special Escapes' above in this
11814 routine, but also has an entry here as anything that
11815 isn't explicitly mentioned here will be treated as
11816 an unescaped equivalent literal.
11819 switch ((U8)*++p) {
11820 /* These are all the special escapes. */
11821 case 'A': /* Start assertion */
11822 case 'b': case 'B': /* Word-boundary assertion*/
11823 case 'C': /* Single char !DANGEROUS! */
11824 case 'd': case 'D': /* digit class */
11825 case 'g': case 'G': /* generic-backref, pos assertion */
11826 case 'h': case 'H': /* HORIZWS */
11827 case 'k': case 'K': /* named backref, keep marker */
11828 case 'p': case 'P': /* Unicode property */
11829 case 'R': /* LNBREAK */
11830 case 's': case 'S': /* space class */
11831 case 'v': case 'V': /* VERTWS */
11832 case 'w': case 'W': /* word class */
11833 case 'X': /* eXtended Unicode "combining
11834 character sequence" */
11835 case 'z': case 'Z': /* End of line/string assertion */
11839 /* Anything after here is an escape that resolves to a
11840 literal. (Except digits, which may or may not)
11846 case 'N': /* Handle a single-code point named character. */
11847 /* The options cause it to fail if a multiple code
11848 * point sequence. Handle those in the switch() above
11850 RExC_parse = p + 1;
11851 if (! grok_bslash_N(pRExC_state, NULL, &ender,
11852 flagp, depth, FALSE,
11853 FALSE /* not strict */ ))
11855 if (*flagp & RESTART_UTF8)
11856 FAIL("panic: grok_bslash_N set RESTART_UTF8");
11857 RExC_parse = p = oldp;
11861 if (ender > 0xff) {
11878 ender = ASCII_TO_NATIVE('\033');
11888 const char* error_msg;
11890 bool valid = grok_bslash_o(&p,
11893 TRUE, /* out warnings */
11894 FALSE, /* not strict */
11895 TRUE, /* Output warnings
11900 RExC_parse = p; /* going to die anyway; point
11901 to exact spot of failure */
11905 if (PL_encoding && ender < 0x100) {
11906 goto recode_encoding;
11908 if (ender > 0xff) {
11915 UV result = UV_MAX; /* initialize to erroneous
11917 const char* error_msg;
11919 bool valid = grok_bslash_x(&p,
11922 TRUE, /* out warnings */
11923 FALSE, /* not strict */
11924 TRUE, /* Output warnings
11929 RExC_parse = p; /* going to die anyway; point
11930 to exact spot of failure */
11935 if (PL_encoding && ender < 0x100) {
11936 goto recode_encoding;
11938 if (ender > 0xff) {
11945 ender = grok_bslash_c(*p++, SIZE_ONLY);
11947 case '8': case '9': /* must be a backreference */
11950 case '1': case '2': case '3':case '4':
11951 case '5': case '6': case '7':
11952 /* When we parse backslash escapes there is ambiguity
11953 * between backreferences and octal escapes. Any escape
11954 * from \1 - \9 is a backreference, any multi-digit
11955 * escape which does not start with 0 and which when
11956 * evaluated as decimal could refer to an already
11957 * parsed capture buffer is a backslash. Anything else
11960 * Note this implies that \118 could be interpreted as
11961 * 118 OR as "\11" . "8" depending on whether there
11962 * were 118 capture buffers defined already in the
11965 /* NOTE, RExC_npar is 1 more than the actual number of
11966 * parens we have seen so far, hence the < RExC_npar below. */
11968 if ( !isDIGIT(p[1]) || S_backref_value(p) < RExC_npar)
11969 { /* Not to be treated as an octal constant, go
11977 I32 flags = PERL_SCAN_SILENT_ILLDIGIT;
11979 ender = grok_oct(p, &numlen, &flags, NULL);
11980 if (ender > 0xff) {
11984 if (SIZE_ONLY /* like \08, \178 */
11987 && isDIGIT(*p) && ckWARN(WARN_REGEXP))
11989 reg_warn_non_literal_string(
11991 form_short_octal_warning(p, numlen));
11994 if (PL_encoding && ender < 0x100)
11995 goto recode_encoding;
11998 if (! RExC_override_recoding) {
11999 SV* enc = PL_encoding;
12000 ender = reg_recode((const char)(U8)ender, &enc);
12001 if (!enc && SIZE_ONLY)
12002 ckWARNreg(p, "Invalid escape in the specified encoding");
12008 FAIL("Trailing \\");
12011 if (!SIZE_ONLY&& isALPHANUMERIC(*p)) {
12012 /* Include any { following the alpha to emphasize
12013 * that it could be part of an escape at some point
12015 int len = (isALPHA(*p) && *(p + 1) == '{') ? 2 : 1;
12016 ckWARN3reg(p + len, "Unrecognized escape \\%.*s passed through", len, p);
12018 goto normal_default;
12019 } /* End of switch on '\' */
12022 /* Currently we don't warn when the lbrace is at the start
12023 * of a construct. This catches it in the middle of a
12024 * literal string, or when its the first thing after
12025 * something like "\b" */
12027 && (len || (p > RExC_start && isALPHA_A(*(p -1)))))
12029 ckWARNregdep(p + 1, "Unescaped left brace in regex is deprecated, passed through");
12032 default: /* A literal character */
12034 if (UTF8_IS_START(*p) && UTF) {
12036 ender = utf8n_to_uvchr((U8*)p, RExC_end - p,
12037 &numlen, UTF8_ALLOW_DEFAULT);
12043 } /* End of switch on the literal */
12045 /* Here, have looked at the literal character and <ender>
12046 * contains its ordinal, <p> points to the character after it
12049 if ( RExC_flags & RXf_PMf_EXTENDED)
12050 p = regpatws(pRExC_state, p,
12051 TRUE); /* means recognize comments */
12053 /* If the next thing is a quantifier, it applies to this
12054 * character only, which means that this character has to be in
12055 * its own node and can't just be appended to the string in an
12056 * existing node, so if there are already other characters in
12057 * the node, close the node with just them, and set up to do
12058 * this character again next time through, when it will be the
12059 * only thing in its new node */
12060 if ((next_is_quantifier = (p < RExC_end && ISMULT2(p))) && len)
12066 if (! FOLD /* The simple case, just append the literal */
12067 || (LOC /* Also don't fold for tricky chars under /l */
12068 && is_PROBLEMATIC_LOCALE_FOLD_cp(ender)))
12071 const STRLEN unilen = reguni(pRExC_state, ender, s);
12077 /* The loop increments <len> each time, as all but this
12078 * path (and one other) through it add a single byte to
12079 * the EXACTish node. But this one has changed len to
12080 * be the correct final value, so subtract one to
12081 * cancel out the increment that follows */
12085 REGC((char)ender, s++);
12088 /* Can get here if folding only if is one of the /l
12089 * characters whose fold depends on the locale. The
12090 * occurrence of any of these indicate that we can't
12091 * simplify things */
12093 maybe_exact = FALSE;
12094 maybe_exactfu = FALSE;
12099 /* See comments for join_exact() as to why we fold this
12100 * non-UTF at compile time */
12101 || (node_type == EXACTFU
12102 && ender == LATIN_SMALL_LETTER_SHARP_S)))
12104 /* Here, are folding and are not UTF-8 encoded; therefore
12105 * the character must be in the range 0-255, and is not /l
12106 * (Not /l because we already handled these under /l in
12107 * is_PROBLEMATIC_LOCALE_FOLD_cp */
12108 if (IS_IN_SOME_FOLD_L1(ender)) {
12109 maybe_exact = FALSE;
12111 /* See if the character's fold differs between /d and
12112 * /u. This includes the multi-char fold SHARP S to
12115 && (PL_fold[ender] != PL_fold_latin1[ender]
12116 || ender == LATIN_SMALL_LETTER_SHARP_S
12118 && isARG2_lower_or_UPPER_ARG1('s', ender)
12119 && isARG2_lower_or_UPPER_ARG1('s',
12122 maybe_exactfu = FALSE;
12126 /* Even when folding, we store just the input character, as
12127 * we have an array that finds its fold quickly */
12128 *(s++) = (char) ender;
12130 else { /* FOLD and UTF */
12131 /* Unlike the non-fold case, we do actually have to
12132 * calculate the results here in pass 1. This is for two
12133 * reasons, the folded length may be longer than the
12134 * unfolded, and we have to calculate how many EXACTish
12135 * nodes it will take; and we may run out of room in a node
12136 * in the middle of a potential multi-char fold, and have
12137 * to back off accordingly. (Hence we can't use REGC for
12138 * the simple case just below.) */
12141 if (isASCII(ender)) {
12142 folded = toFOLD(ender);
12143 *(s)++ = (U8) folded;
12148 folded = _to_uni_fold_flags(
12152 FOLD_FLAGS_FULL | ((ASCII_FOLD_RESTRICTED)
12153 ? FOLD_FLAGS_NOMIX_ASCII
12157 /* The loop increments <len> each time, as all but this
12158 * path (and one other) through it add a single byte to
12159 * the EXACTish node. But this one has changed len to
12160 * be the correct final value, so subtract one to
12161 * cancel out the increment that follows */
12162 len += foldlen - 1;
12164 /* If this node only contains non-folding code points so
12165 * far, see if this new one is also non-folding */
12167 if (folded != ender) {
12168 maybe_exact = FALSE;
12171 /* Here the fold is the original; we have to check
12172 * further to see if anything folds to it */
12173 if (_invlist_contains_cp(PL_utf8_foldable,
12176 maybe_exact = FALSE;
12183 if (next_is_quantifier) {
12185 /* Here, the next input is a quantifier, and to get here,
12186 * the current character is the only one in the node.
12187 * Also, here <len> doesn't include the final byte for this
12193 } /* End of loop through literal characters */
12195 /* Here we have either exhausted the input or ran out of room in
12196 * the node. (If we encountered a character that can't be in the
12197 * node, transfer is made directly to <loopdone>, and so we
12198 * wouldn't have fallen off the end of the loop.) In the latter
12199 * case, we artificially have to split the node into two, because
12200 * we just don't have enough space to hold everything. This
12201 * creates a problem if the final character participates in a
12202 * multi-character fold in the non-final position, as a match that
12203 * should have occurred won't, due to the way nodes are matched,
12204 * and our artificial boundary. So back off until we find a non-
12205 * problematic character -- one that isn't at the beginning or
12206 * middle of such a fold. (Either it doesn't participate in any
12207 * folds, or appears only in the final position of all the folds it
12208 * does participate in.) A better solution with far fewer false
12209 * positives, and that would fill the nodes more completely, would
12210 * be to actually have available all the multi-character folds to
12211 * test against, and to back-off only far enough to be sure that
12212 * this node isn't ending with a partial one. <upper_parse> is set
12213 * further below (if we need to reparse the node) to include just
12214 * up through that final non-problematic character that this code
12215 * identifies, so when it is set to less than the full node, we can
12216 * skip the rest of this */
12217 if (FOLD && p < RExC_end && upper_parse == MAX_NODE_STRING_SIZE) {
12219 const STRLEN full_len = len;
12221 assert(len >= MAX_NODE_STRING_SIZE);
12223 /* Here, <s> points to the final byte of the final character.
12224 * Look backwards through the string until find a non-
12225 * problematic character */
12229 /* This has no multi-char folds to non-UTF characters */
12230 if (ASCII_FOLD_RESTRICTED) {
12234 while (--s >= s0 && IS_NON_FINAL_FOLD(*s)) { }
12238 if (! PL_NonL1NonFinalFold) {
12239 PL_NonL1NonFinalFold = _new_invlist_C_array(
12240 NonL1_Perl_Non_Final_Folds_invlist);
12243 /* Point to the first byte of the final character */
12244 s = (char *) utf8_hop((U8 *) s, -1);
12246 while (s >= s0) { /* Search backwards until find
12247 non-problematic char */
12248 if (UTF8_IS_INVARIANT(*s)) {
12250 /* There are no ascii characters that participate
12251 * in multi-char folds under /aa. In EBCDIC, the
12252 * non-ascii invariants are all control characters,
12253 * so don't ever participate in any folds. */
12254 if (ASCII_FOLD_RESTRICTED
12255 || ! IS_NON_FINAL_FOLD(*s))
12260 else if (UTF8_IS_DOWNGRADEABLE_START(*s)) {
12261 if (! IS_NON_FINAL_FOLD(TWO_BYTE_UTF8_TO_NATIVE(
12267 else if (! _invlist_contains_cp(
12268 PL_NonL1NonFinalFold,
12269 valid_utf8_to_uvchr((U8 *) s, NULL)))
12274 /* Here, the current character is problematic in that
12275 * it does occur in the non-final position of some
12276 * fold, so try the character before it, but have to
12277 * special case the very first byte in the string, so
12278 * we don't read outside the string */
12279 s = (s == s0) ? s -1 : (char *) utf8_hop((U8 *) s, -1);
12280 } /* End of loop backwards through the string */
12282 /* If there were only problematic characters in the string,
12283 * <s> will point to before s0, in which case the length
12284 * should be 0, otherwise include the length of the
12285 * non-problematic character just found */
12286 len = (s < s0) ? 0 : s - s0 + UTF8SKIP(s);
12289 /* Here, have found the final character, if any, that is
12290 * non-problematic as far as ending the node without splitting
12291 * it across a potential multi-char fold. <len> contains the
12292 * number of bytes in the node up-to and including that
12293 * character, or is 0 if there is no such character, meaning
12294 * the whole node contains only problematic characters. In
12295 * this case, give up and just take the node as-is. We can't
12300 /* If the node ends in an 's' we make sure it stays EXACTF,
12301 * as if it turns into an EXACTFU, it could later get
12302 * joined with another 's' that would then wrongly match
12304 if (maybe_exactfu && isARG2_lower_or_UPPER_ARG1('s', ender))
12306 maybe_exactfu = FALSE;
12310 /* Here, the node does contain some characters that aren't
12311 * problematic. If one such is the final character in the
12312 * node, we are done */
12313 if (len == full_len) {
12316 else if (len + ((UTF) ? UTF8SKIP(s) : 1) == full_len) {
12318 /* If the final character is problematic, but the
12319 * penultimate is not, back-off that last character to
12320 * later start a new node with it */
12325 /* Here, the final non-problematic character is earlier
12326 * in the input than the penultimate character. What we do
12327 * is reparse from the beginning, going up only as far as
12328 * this final ok one, thus guaranteeing that the node ends
12329 * in an acceptable character. The reason we reparse is
12330 * that we know how far in the character is, but we don't
12331 * know how to correlate its position with the input parse.
12332 * An alternate implementation would be to build that
12333 * correlation as we go along during the original parse,
12334 * but that would entail extra work for every node, whereas
12335 * this code gets executed only when the string is too
12336 * large for the node, and the final two characters are
12337 * problematic, an infrequent occurrence. Yet another
12338 * possible strategy would be to save the tail of the
12339 * string, and the next time regatom is called, initialize
12340 * with that. The problem with this is that unless you
12341 * back off one more character, you won't be guaranteed
12342 * regatom will get called again, unless regbranch,
12343 * regpiece ... are also changed. If you do back off that
12344 * extra character, so that there is input guaranteed to
12345 * force calling regatom, you can't handle the case where
12346 * just the first character in the node is acceptable. I
12347 * (khw) decided to try this method which doesn't have that
12348 * pitfall; if performance issues are found, we can do a
12349 * combination of the current approach plus that one */
12355 } /* End of verifying node ends with an appropriate char */
12357 loopdone: /* Jumped to when encounters something that shouldn't be in
12360 /* I (khw) don't know if you can get here with zero length, but the
12361 * old code handled this situation by creating a zero-length EXACT
12362 * node. Might as well be NOTHING instead */
12368 /* If 'maybe_exact' is still set here, means there are no
12369 * code points in the node that participate in folds;
12370 * similarly for 'maybe_exactfu' and code points that match
12371 * differently depending on UTF8ness of the target string
12372 * (for /u), or depending on locale for /l */
12376 else if (maybe_exactfu) {
12380 alloc_maybe_populate_EXACT(pRExC_state, ret, flagp, len, ender,
12381 FALSE /* Don't look to see if could
12382 be turned into an EXACT
12383 node, as we have already
12388 RExC_parse = p - 1;
12389 Set_Node_Cur_Length(ret, parse_start);
12390 nextchar(pRExC_state);
12392 /* len is STRLEN which is unsigned, need to copy to signed */
12395 vFAIL("Internal disaster");
12398 } /* End of label 'defchar:' */
12400 } /* End of giant switch on input character */
12406 S_regpatws(RExC_state_t *pRExC_state, char *p , const bool recognize_comment )
12408 /* Returns the next non-pattern-white space, non-comment character (the
12409 * latter only if 'recognize_comment is true) in the string p, which is
12410 * ended by RExC_end. See also reg_skipcomment */
12411 const char *e = RExC_end;
12413 PERL_ARGS_ASSERT_REGPATWS;
12417 if ((len = is_PATWS_safe(p, e, UTF))) {
12420 else if (recognize_comment && *p == '#') {
12421 p = reg_skipcomment(pRExC_state, p);
12430 S_populate_ANYOF_from_invlist(pTHX_ regnode *node, SV** invlist_ptr)
12432 /* Uses the inversion list '*invlist_ptr' to populate the ANYOF 'node'. It
12433 * sets up the bitmap and any flags, removing those code points from the
12434 * inversion list, setting it to NULL should it become completely empty */
12436 PERL_ARGS_ASSERT_POPULATE_ANYOF_FROM_INVLIST;
12437 assert(PL_regkind[OP(node)] == ANYOF);
12439 ANYOF_BITMAP_ZERO(node);
12440 if (*invlist_ptr) {
12442 /* This gets set if we actually need to modify things */
12443 bool change_invlist = FALSE;
12447 /* Start looking through *invlist_ptr */
12448 invlist_iterinit(*invlist_ptr);
12449 while (invlist_iternext(*invlist_ptr, &start, &end)) {
12453 if (end == UV_MAX && start <= 256) {
12454 ANYOF_FLAGS(node) |= ANYOF_ABOVE_LATIN1_ALL;
12456 else if (end >= 256) {
12457 ANYOF_FLAGS(node) |= ANYOF_UTF8;
12460 /* Quit if are above what we should change */
12465 change_invlist = TRUE;
12467 /* Set all the bits in the range, up to the max that we are doing */
12468 high = (end < 255) ? end : 255;
12469 for (i = start; i <= (int) high; i++) {
12470 if (! ANYOF_BITMAP_TEST(node, i)) {
12471 ANYOF_BITMAP_SET(node, i);
12475 invlist_iterfinish(*invlist_ptr);
12477 /* Done with loop; remove any code points that are in the bitmap from
12478 * *invlist_ptr; similarly for code points above latin1 if we have a
12479 * flag to match all of them anyways */
12480 if (change_invlist) {
12481 _invlist_subtract(*invlist_ptr, PL_Latin1, invlist_ptr);
12483 if (ANYOF_FLAGS(node) & ANYOF_ABOVE_LATIN1_ALL) {
12484 _invlist_intersection(*invlist_ptr, PL_Latin1, invlist_ptr);
12487 /* If have completely emptied it, remove it completely */
12488 if (_invlist_len(*invlist_ptr) == 0) {
12489 SvREFCNT_dec_NN(*invlist_ptr);
12490 *invlist_ptr = NULL;
12495 /* Parse POSIX character classes: [[:foo:]], [[=foo=]], [[.foo.]].
12496 Character classes ([:foo:]) can also be negated ([:^foo:]).
12497 Returns a named class id (ANYOF_XXX) if successful, -1 otherwise.
12498 Equivalence classes ([=foo=]) and composites ([.foo.]) are parsed,
12499 but trigger failures because they are currently unimplemented. */
12501 #define POSIXCC_DONE(c) ((c) == ':')
12502 #define POSIXCC_NOTYET(c) ((c) == '=' || (c) == '.')
12503 #define POSIXCC(c) (POSIXCC_DONE(c) || POSIXCC_NOTYET(c))
12505 PERL_STATIC_INLINE I32
12506 S_regpposixcc(pTHX_ RExC_state_t *pRExC_state, I32 value, const bool strict)
12508 I32 namedclass = OOB_NAMEDCLASS;
12510 PERL_ARGS_ASSERT_REGPPOSIXCC;
12512 if (value == '[' && RExC_parse + 1 < RExC_end &&
12513 /* I smell either [: or [= or [. -- POSIX has been here, right? */
12514 POSIXCC(UCHARAT(RExC_parse)))
12516 const char c = UCHARAT(RExC_parse);
12517 char* const s = RExC_parse++;
12519 while (RExC_parse < RExC_end && UCHARAT(RExC_parse) != c)
12521 if (RExC_parse == RExC_end) {
12524 /* Try to give a better location for the error (than the end of
12525 * the string) by looking for the matching ']' */
12527 while (RExC_parse < RExC_end && UCHARAT(RExC_parse) != ']') {
12530 vFAIL2("Unmatched '%c' in POSIX class", c);
12532 /* Grandfather lone [:, [=, [. */
12536 const char* const t = RExC_parse++; /* skip over the c */
12539 if (UCHARAT(RExC_parse) == ']') {
12540 const char *posixcc = s + 1;
12541 RExC_parse++; /* skip over the ending ] */
12544 const I32 complement = *posixcc == '^' ? *posixcc++ : 0;
12545 const I32 skip = t - posixcc;
12547 /* Initially switch on the length of the name. */
12550 if (memEQ(posixcc, "word", 4)) /* this is not POSIX,
12551 this is the Perl \w
12553 namedclass = ANYOF_WORDCHAR;
12556 /* Names all of length 5. */
12557 /* alnum alpha ascii blank cntrl digit graph lower
12558 print punct space upper */
12559 /* Offset 4 gives the best switch position. */
12560 switch (posixcc[4]) {
12562 if (memEQ(posixcc, "alph", 4)) /* alpha */
12563 namedclass = ANYOF_ALPHA;
12566 if (memEQ(posixcc, "spac", 4)) /* space */
12567 namedclass = ANYOF_PSXSPC;
12570 if (memEQ(posixcc, "grap", 4)) /* graph */
12571 namedclass = ANYOF_GRAPH;
12574 if (memEQ(posixcc, "asci", 4)) /* ascii */
12575 namedclass = ANYOF_ASCII;
12578 if (memEQ(posixcc, "blan", 4)) /* blank */
12579 namedclass = ANYOF_BLANK;
12582 if (memEQ(posixcc, "cntr", 4)) /* cntrl */
12583 namedclass = ANYOF_CNTRL;
12586 if (memEQ(posixcc, "alnu", 4)) /* alnum */
12587 namedclass = ANYOF_ALPHANUMERIC;
12590 if (memEQ(posixcc, "lowe", 4)) /* lower */
12591 namedclass = (FOLD) ? ANYOF_CASED : ANYOF_LOWER;
12592 else if (memEQ(posixcc, "uppe", 4)) /* upper */
12593 namedclass = (FOLD) ? ANYOF_CASED : ANYOF_UPPER;
12596 if (memEQ(posixcc, "digi", 4)) /* digit */
12597 namedclass = ANYOF_DIGIT;
12598 else if (memEQ(posixcc, "prin", 4)) /* print */
12599 namedclass = ANYOF_PRINT;
12600 else if (memEQ(posixcc, "punc", 4)) /* punct */
12601 namedclass = ANYOF_PUNCT;
12606 if (memEQ(posixcc, "xdigit", 6))
12607 namedclass = ANYOF_XDIGIT;
12611 if (namedclass == OOB_NAMEDCLASS)
12613 "POSIX class [:%"UTF8f":] unknown",
12614 UTF8fARG(UTF, t - s - 1, s + 1));
12616 /* The #defines are structured so each complement is +1 to
12617 * the normal one */
12621 assert (posixcc[skip] == ':');
12622 assert (posixcc[skip+1] == ']');
12623 } else if (!SIZE_ONLY) {
12624 /* [[=foo=]] and [[.foo.]] are still future. */
12626 /* adjust RExC_parse so the warning shows after
12627 the class closes */
12628 while (UCHARAT(RExC_parse) && UCHARAT(RExC_parse) != ']')
12630 vFAIL3("POSIX syntax [%c %c] is reserved for future extensions", c, c);
12633 /* Maternal grandfather:
12634 * "[:" ending in ":" but not in ":]" */
12636 vFAIL("Unmatched '[' in POSIX class");
12639 /* Grandfather lone [:, [=, [. */
12649 S_could_it_be_a_POSIX_class(RExC_state_t *pRExC_state)
12651 /* This applies some heuristics at the current parse position (which should
12652 * be at a '[') to see if what follows might be intended to be a [:posix:]
12653 * class. It returns true if it really is a posix class, of course, but it
12654 * also can return true if it thinks that what was intended was a posix
12655 * class that didn't quite make it.
12657 * It will return true for
12659 * [:alphanumerics] (as long as the ] isn't followed immediately by a
12660 * ')' indicating the end of the (?[
12661 * [:any garbage including %^&$ punctuation:]
12663 * This is designed to be called only from S_handle_regex_sets; it could be
12664 * easily adapted to be called from the spot at the beginning of regclass()
12665 * that checks to see in a normal bracketed class if the surrounding []
12666 * have been omitted ([:word:] instead of [[:word:]]). But doing so would
12667 * change long-standing behavior, so I (khw) didn't do that */
12668 char* p = RExC_parse + 1;
12669 char first_char = *p;
12671 PERL_ARGS_ASSERT_COULD_IT_BE_A_POSIX_CLASS;
12673 assert(*(p - 1) == '[');
12675 if (! POSIXCC(first_char)) {
12680 while (p < RExC_end && isWORDCHAR(*p)) p++;
12682 if (p >= RExC_end) {
12686 if (p - RExC_parse > 2 /* Got at least 1 word character */
12687 && (*p == first_char
12688 || (*p == ']' && p + 1 < RExC_end && *(p + 1) != ')')))
12693 p = (char *) memchr(RExC_parse, ']', RExC_end - RExC_parse);
12696 && p - RExC_parse > 2 /* [:] evaluates to colon;
12697 [::] is a bad posix class. */
12698 && first_char == *(p - 1));
12702 S_handle_regex_sets(pTHX_ RExC_state_t *pRExC_state, SV** return_invlist,
12703 I32 *flagp, U32 depth,
12704 char * const oregcomp_parse)
12706 /* Handle the (?[...]) construct to do set operations */
12709 UV start, end; /* End points of code point ranges */
12711 char *save_end, *save_parse;
12716 const bool save_fold = FOLD;
12718 GET_RE_DEBUG_FLAGS_DECL;
12720 PERL_ARGS_ASSERT_HANDLE_REGEX_SETS;
12723 vFAIL("(?[...]) not valid in locale");
12725 RExC_uni_semantics = 1;
12727 /* This will return only an ANYOF regnode, or (unlikely) something smaller
12728 * (such as EXACT). Thus we can skip most everything if just sizing. We
12729 * call regclass to handle '[]' so as to not have to reinvent its parsing
12730 * rules here (throwing away the size it computes each time). And, we exit
12731 * upon an unescaped ']' that isn't one ending a regclass. To do both
12732 * these things, we need to realize that something preceded by a backslash
12733 * is escaped, so we have to keep track of backslashes */
12735 UV depth = 0; /* how many nested (?[...]) constructs */
12737 Perl_ck_warner_d(aTHX_
12738 packWARN(WARN_EXPERIMENTAL__REGEX_SETS),
12739 "The regex_sets feature is experimental" REPORT_LOCATION,
12740 UTF8fARG(UTF, (RExC_parse - RExC_precomp), RExC_precomp),
12742 RExC_end - RExC_start - (RExC_parse - RExC_precomp),
12743 RExC_precomp + (RExC_parse - RExC_precomp)));
12745 while (RExC_parse < RExC_end) {
12746 SV* current = NULL;
12747 RExC_parse = regpatws(pRExC_state, RExC_parse,
12748 TRUE); /* means recognize comments */
12749 switch (*RExC_parse) {
12751 if (RExC_parse[1] == '[') depth++, RExC_parse++;
12756 /* Skip the next byte (which could cause us to end up in
12757 * the middle of a UTF-8 character, but since none of those
12758 * are confusable with anything we currently handle in this
12759 * switch (invariants all), it's safe. We'll just hit the
12760 * default: case next time and keep on incrementing until
12761 * we find one of the invariants we do handle. */
12766 /* If this looks like it is a [:posix:] class, leave the
12767 * parse pointer at the '[' to fool regclass() into
12768 * thinking it is part of a '[[:posix:]]'. That function
12769 * will use strict checking to force a syntax error if it
12770 * doesn't work out to a legitimate class */
12771 bool is_posix_class
12772 = could_it_be_a_POSIX_class(pRExC_state);
12773 if (! is_posix_class) {
12777 /* regclass() can only return RESTART_UTF8 if multi-char
12778 folds are allowed. */
12779 if (!regclass(pRExC_state, flagp,depth+1,
12780 is_posix_class, /* parse the whole char
12781 class only if not a
12783 FALSE, /* don't allow multi-char folds */
12784 TRUE, /* silence non-portable warnings. */
12786 FAIL2("panic: regclass returned NULL to handle_sets, flags=%#"UVxf"",
12789 /* function call leaves parse pointing to the ']', except
12790 * if we faked it */
12791 if (is_posix_class) {
12795 SvREFCNT_dec(current); /* In case it returned something */
12800 if (depth--) break;
12802 if (RExC_parse < RExC_end
12803 && *RExC_parse == ')')
12805 node = reganode(pRExC_state, ANYOF, 0);
12806 RExC_size += ANYOF_SKIP;
12807 nextchar(pRExC_state);
12808 Set_Node_Length(node,
12809 RExC_parse - oregcomp_parse + 1); /* MJD */
12818 FAIL("Syntax error in (?[...])");
12821 /* Pass 2 only after this. Everything in this construct is a
12822 * metacharacter. Operands begin with either a '\' (for an escape
12823 * sequence), or a '[' for a bracketed character class. Any other
12824 * character should be an operator, or parenthesis for grouping. Both
12825 * types of operands are handled by calling regclass() to parse them. It
12826 * is called with a parameter to indicate to return the computed inversion
12827 * list. The parsing here is implemented via a stack. Each entry on the
12828 * stack is a single character representing one of the operators, or the
12829 * '('; or else a pointer to an operand inversion list. */
12831 #define IS_OPERAND(a) (! SvIOK(a))
12833 /* The stack starts empty. It is a syntax error if the first thing parsed
12834 * is a binary operator; everything else is pushed on the stack. When an
12835 * operand is parsed, the top of the stack is examined. If it is a binary
12836 * operator, the item before it should be an operand, and both are replaced
12837 * by the result of doing that operation on the new operand and the one on
12838 * the stack. Thus a sequence of binary operands is reduced to a single
12839 * one before the next one is parsed.
12841 * A unary operator may immediately follow a binary in the input, for
12844 * When an operand is parsed and the top of the stack is a unary operator,
12845 * the operation is performed, and then the stack is rechecked to see if
12846 * this new operand is part of a binary operation; if so, it is handled as
12849 * A '(' is simply pushed on the stack; it is valid only if the stack is
12850 * empty, or the top element of the stack is an operator or another '('
12851 * (for which the parenthesized expression will become an operand). By the
12852 * time the corresponding ')' is parsed everything in between should have
12853 * been parsed and evaluated to a single operand (or else is a syntax
12854 * error), and is handled as a regular operand */
12856 sv_2mortal((SV *)(stack = newAV()));
12858 while (RExC_parse < RExC_end) {
12859 I32 top_index = av_tindex(stack);
12861 SV* current = NULL;
12863 /* Skip white space */
12864 RExC_parse = regpatws(pRExC_state, RExC_parse,
12865 TRUE /* means recognize comments */ );
12866 if (RExC_parse >= RExC_end) {
12867 Perl_croak(aTHX_ "panic: Read past end of '(?[ ])'");
12869 if ((curchar = UCHARAT(RExC_parse)) == ']') {
12876 if (av_tindex(stack) >= 0 /* This makes sure that we can
12877 safely subtract 1 from
12878 RExC_parse in the next clause.
12879 If we have something on the
12880 stack, we have parsed something
12882 && UCHARAT(RExC_parse - 1) == '('
12883 && RExC_parse < RExC_end)
12885 /* If is a '(?', could be an embedded '(?flags:(?[...])'.
12886 * This happens when we have some thing like
12888 * my $thai_or_lao = qr/(?[ \p{Thai} + \p{Lao} ])/;
12890 * qr/(?[ \p{Digit} & $thai_or_lao ])/;
12892 * Here we would be handling the interpolated
12893 * '$thai_or_lao'. We handle this by a recursive call to
12894 * ourselves which returns the inversion list the
12895 * interpolated expression evaluates to. We use the flags
12896 * from the interpolated pattern. */
12897 U32 save_flags = RExC_flags;
12898 const char * const save_parse = ++RExC_parse;
12900 parse_lparen_question_flags(pRExC_state);
12902 if (RExC_parse == save_parse /* Makes sure there was at
12903 least one flag (or this
12904 embedding wasn't compiled)
12906 || RExC_parse >= RExC_end - 4
12907 || UCHARAT(RExC_parse) != ':'
12908 || UCHARAT(++RExC_parse) != '('
12909 || UCHARAT(++RExC_parse) != '?'
12910 || UCHARAT(++RExC_parse) != '[')
12913 /* In combination with the above, this moves the
12914 * pointer to the point just after the first erroneous
12915 * character (or if there are no flags, to where they
12916 * should have been) */
12917 if (RExC_parse >= RExC_end - 4) {
12918 RExC_parse = RExC_end;
12920 else if (RExC_parse != save_parse) {
12921 RExC_parse += (UTF) ? UTF8SKIP(RExC_parse) : 1;
12923 vFAIL("Expecting '(?flags:(?[...'");
12926 (void) handle_regex_sets(pRExC_state, ¤t, flagp,
12927 depth+1, oregcomp_parse);
12929 /* Here, 'current' contains the embedded expression's
12930 * inversion list, and RExC_parse points to the trailing
12931 * ']'; the next character should be the ')' which will be
12932 * paired with the '(' that has been put on the stack, so
12933 * the whole embedded expression reduces to '(operand)' */
12936 RExC_flags = save_flags;
12937 goto handle_operand;
12942 RExC_parse += (UTF) ? UTF8SKIP(RExC_parse) : 1;
12943 vFAIL("Unexpected character");
12946 /* regclass() can only return RESTART_UTF8 if multi-char
12947 folds are allowed. */
12948 if (!regclass(pRExC_state, flagp,depth+1,
12949 TRUE, /* means parse just the next thing */
12950 FALSE, /* don't allow multi-char folds */
12951 FALSE, /* don't silence non-portable warnings. */
12953 FAIL2("panic: regclass returned NULL to handle_sets, flags=%#"UVxf"",
12955 /* regclass() will return with parsing just the \ sequence,
12956 * leaving the parse pointer at the next thing to parse */
12958 goto handle_operand;
12960 case '[': /* Is a bracketed character class */
12962 bool is_posix_class = could_it_be_a_POSIX_class(pRExC_state);
12964 if (! is_posix_class) {
12968 /* regclass() can only return RESTART_UTF8 if multi-char
12969 folds are allowed. */
12970 if(!regclass(pRExC_state, flagp,depth+1,
12971 is_posix_class, /* parse the whole char class
12972 only if not a posix class */
12973 FALSE, /* don't allow multi-char folds */
12974 FALSE, /* don't silence non-portable warnings. */
12976 FAIL2("panic: regclass returned NULL to handle_sets, flags=%#"UVxf"",
12978 /* function call leaves parse pointing to the ']', except if we
12980 if (is_posix_class) {
12984 goto handle_operand;
12993 || ( ! (top_ptr = av_fetch(stack, top_index, FALSE)))
12994 || ! IS_OPERAND(*top_ptr))
12997 vFAIL2("Unexpected binary operator '%c' with no preceding operand", curchar);
12999 av_push(stack, newSVuv(curchar));
13003 av_push(stack, newSVuv(curchar));
13007 if (top_index >= 0) {
13008 top_ptr = av_fetch(stack, top_index, FALSE);
13010 if (IS_OPERAND(*top_ptr)) {
13012 vFAIL("Unexpected '(' with no preceding operator");
13015 av_push(stack, newSVuv(curchar));
13022 || ! (current = av_pop(stack))
13023 || ! IS_OPERAND(current)
13024 || ! (lparen = av_pop(stack))
13025 || IS_OPERAND(lparen)
13026 || SvUV(lparen) != '(')
13028 SvREFCNT_dec(current);
13030 vFAIL("Unexpected ')'");
13033 SvREFCNT_dec_NN(lparen);
13040 /* Here, we have an operand to process, in 'current' */
13042 if (top_index < 0) { /* Just push if stack is empty */
13043 av_push(stack, current);
13046 SV* top = av_pop(stack);
13048 char current_operator;
13050 if (IS_OPERAND(top)) {
13051 SvREFCNT_dec_NN(top);
13052 SvREFCNT_dec_NN(current);
13053 vFAIL("Operand with no preceding operator");
13055 current_operator = (char) SvUV(top);
13056 switch (current_operator) {
13057 case '(': /* Push the '(' back on followed by the new
13059 av_push(stack, top);
13060 av_push(stack, current);
13061 SvREFCNT_inc(top); /* Counters the '_dec' done
13062 just after the 'break', so
13063 it doesn't get wrongly freed
13068 _invlist_invert(current);
13070 /* Unlike binary operators, the top of the stack,
13071 * now that this unary one has been popped off, may
13072 * legally be an operator, and we now have operand
13075 SvREFCNT_dec_NN(top);
13076 goto handle_operand;
13079 prev = av_pop(stack);
13080 _invlist_intersection(prev,
13083 av_push(stack, current);
13088 prev = av_pop(stack);
13089 _invlist_union(prev, current, ¤t);
13090 av_push(stack, current);
13094 prev = av_pop(stack);;
13095 _invlist_subtract(prev, current, ¤t);
13096 av_push(stack, current);
13099 case '^': /* The union minus the intersection */
13105 prev = av_pop(stack);
13106 _invlist_union(prev, current, &u);
13107 _invlist_intersection(prev, current, &i);
13108 /* _invlist_subtract will overwrite current
13109 without freeing what it already contains */
13111 _invlist_subtract(u, i, ¤t);
13112 av_push(stack, current);
13113 SvREFCNT_dec_NN(i);
13114 SvREFCNT_dec_NN(u);
13115 SvREFCNT_dec_NN(element);
13120 Perl_croak(aTHX_ "panic: Unexpected item on '(?[ ])' stack");
13122 SvREFCNT_dec_NN(top);
13123 SvREFCNT_dec(prev);
13127 RExC_parse += (UTF) ? UTF8SKIP(RExC_parse) : 1;
13130 if (av_tindex(stack) < 0 /* Was empty */
13131 || ((final = av_pop(stack)) == NULL)
13132 || ! IS_OPERAND(final)
13133 || av_tindex(stack) >= 0) /* More left on stack */
13135 vFAIL("Incomplete expression within '(?[ ])'");
13138 /* Here, 'final' is the resultant inversion list from evaluating the
13139 * expression. Return it if so requested */
13140 if (return_invlist) {
13141 *return_invlist = final;
13145 /* Otherwise generate a resultant node, based on 'final'. regclass() is
13146 * expecting a string of ranges and individual code points */
13147 invlist_iterinit(final);
13148 result_string = newSVpvs("");
13149 while (invlist_iternext(final, &start, &end)) {
13150 if (start == end) {
13151 Perl_sv_catpvf(aTHX_ result_string, "\\x{%"UVXf"}", start);
13154 Perl_sv_catpvf(aTHX_ result_string, "\\x{%"UVXf"}-\\x{%"UVXf"}",
13159 save_parse = RExC_parse;
13160 RExC_parse = SvPV(result_string, len);
13161 save_end = RExC_end;
13162 RExC_end = RExC_parse + len;
13164 /* We turn off folding around the call, as the class we have constructed
13165 * already has all folding taken into consideration, and we don't want
13166 * regclass() to add to that */
13167 RExC_flags &= ~RXf_PMf_FOLD;
13168 /* regclass() can only return RESTART_UTF8 if multi-char folds are allowed.
13170 node = regclass(pRExC_state, flagp,depth+1,
13171 FALSE, /* means parse the whole char class */
13172 FALSE, /* don't allow multi-char folds */
13173 TRUE, /* silence non-portable warnings. The above may very
13174 well have generated non-portable code points, but
13175 they're valid on this machine */
13178 FAIL2("panic: regclass returned NULL to handle_sets, flags=%#"UVxf,
13181 RExC_flags |= RXf_PMf_FOLD;
13183 RExC_parse = save_parse + 1;
13184 RExC_end = save_end;
13185 SvREFCNT_dec_NN(final);
13186 SvREFCNT_dec_NN(result_string);
13188 nextchar(pRExC_state);
13189 Set_Node_Length(node, RExC_parse - oregcomp_parse + 1); /* MJD */
13195 S_add_above_Latin1_folds(pTHX_ RExC_state_t *pRExC_state, const U8 cp, SV** invlist)
13197 /* This hard-codes the Latin1/above-Latin1 folding rules, so that an
13198 * innocent-looking character class, like /[ks]/i won't have to go out to
13199 * disk to find the possible matches.
13201 * This should be called only for a Latin1-range code points, cp, which is
13202 * known to be involved in a simple fold with other code points above
13203 * Latin1. It would give false results if /aa has been specified.
13204 * Multi-char folds are outside the scope of this, and must be handled
13207 * XXX It would be better to generate these via regen, in case a new
13208 * version of the Unicode standard adds new mappings, though that is not
13209 * really likely, and may be caught by the default: case of the switch
13212 PERL_ARGS_ASSERT_ADD_ABOVE_LATIN1_FOLDS;
13214 assert(HAS_NONLATIN1_SIMPLE_FOLD_CLOSURE(cp));
13220 add_cp_to_invlist(*invlist, KELVIN_SIGN);
13224 *invlist = add_cp_to_invlist(*invlist, LATIN_SMALL_LETTER_LONG_S);
13227 *invlist = add_cp_to_invlist(*invlist, GREEK_CAPITAL_LETTER_MU);
13228 *invlist = add_cp_to_invlist(*invlist, GREEK_SMALL_LETTER_MU);
13230 case LATIN_CAPITAL_LETTER_A_WITH_RING_ABOVE:
13231 case LATIN_SMALL_LETTER_A_WITH_RING_ABOVE:
13232 *invlist = add_cp_to_invlist(*invlist, ANGSTROM_SIGN);
13234 case LATIN_SMALL_LETTER_Y_WITH_DIAERESIS:
13235 *invlist = add_cp_to_invlist(*invlist,
13236 LATIN_CAPITAL_LETTER_Y_WITH_DIAERESIS);
13238 case LATIN_SMALL_LETTER_SHARP_S:
13239 *invlist = add_cp_to_invlist(*invlist, LATIN_CAPITAL_LETTER_SHARP_S);
13242 /* Use deprecated warning to increase the chances of this being
13244 ckWARN2reg_d(RExC_parse, "Perl folding rules are not up-to-date for 0x%02X; please use the perlbug utility to report;", cp);
13249 /* The names of properties whose definitions are not known at compile time are
13250 * stored in this SV, after a constant heading. So if the length has been
13251 * changed since initialization, then there is a run-time definition. */
13252 #define HAS_NONLOCALE_RUNTIME_PROPERTY_DEFINITION \
13253 (SvCUR(listsv) != initial_listsv_len)
13256 S_regclass(pTHX_ RExC_state_t *pRExC_state, I32 *flagp, U32 depth,
13257 const bool stop_at_1, /* Just parse the next thing, don't
13258 look for a full character class */
13259 bool allow_multi_folds,
13260 const bool silence_non_portable, /* Don't output warnings
13263 SV** ret_invlist) /* Return an inversion list, not a node */
13265 /* parse a bracketed class specification. Most of these will produce an
13266 * ANYOF node; but something like [a] will produce an EXACT node; [aA], an
13267 * EXACTFish node; [[:ascii:]], a POSIXA node; etc. It is more complex
13268 * under /i with multi-character folds: it will be rewritten following the
13269 * paradigm of this example, where the <multi-fold>s are characters which
13270 * fold to multiple character sequences:
13271 * /[abc\x{multi-fold1}def\x{multi-fold2}ghi]/i
13272 * gets effectively rewritten as:
13273 * /(?:\x{multi-fold1}|\x{multi-fold2}|[abcdefghi]/i
13274 * reg() gets called (recursively) on the rewritten version, and this
13275 * function will return what it constructs. (Actually the <multi-fold>s
13276 * aren't physically removed from the [abcdefghi], it's just that they are
13277 * ignored in the recursion by means of a flag:
13278 * <RExC_in_multi_char_class>.)
13280 * ANYOF nodes contain a bit map for the first 256 characters, with the
13281 * corresponding bit set if that character is in the list. For characters
13282 * above 255, a range list or swash is used. There are extra bits for \w,
13283 * etc. in locale ANYOFs, as what these match is not determinable at
13286 * Returns NULL, setting *flagp to RESTART_UTF8 if the sizing scan needs
13287 * to be restarted. This can only happen if ret_invlist is non-NULL.
13290 UV prevvalue = OOB_UNICODE, save_prevvalue = OOB_UNICODE;
13292 UV value = OOB_UNICODE, save_value = OOB_UNICODE;
13295 IV namedclass = OOB_NAMEDCLASS;
13296 char *rangebegin = NULL;
13297 bool need_class = 0;
13299 STRLEN initial_listsv_len = 0; /* Kind of a kludge to see if it is more
13300 than just initialized. */
13301 SV* properties = NULL; /* Code points that match \p{} \P{} */
13302 SV* posixes = NULL; /* Code points that match classes like [:word:],
13303 extended beyond the Latin1 range. These have to
13304 be kept separate from other code points for much
13305 of this function because their handling is
13306 different under /i, and for most classes under
13308 SV* nposixes = NULL; /* Similarly for [:^word:]. These are kept
13309 separate for a while from the non-complemented
13310 versions because of complications with /d
13312 UV element_count = 0; /* Number of distinct elements in the class.
13313 Optimizations may be possible if this is tiny */
13314 AV * multi_char_matches = NULL; /* Code points that fold to more than one
13315 character; used under /i */
13317 char * stop_ptr = RExC_end; /* where to stop parsing */
13318 const bool skip_white = cBOOL(ret_invlist); /* ignore unescaped white
13320 const bool strict = cBOOL(ret_invlist); /* Apply strict parsing rules? */
13322 /* Unicode properties are stored in a swash; this holds the current one
13323 * being parsed. If this swash is the only above-latin1 component of the
13324 * character class, an optimization is to pass it directly on to the
13325 * execution engine. Otherwise, it is set to NULL to indicate that there
13326 * are other things in the class that have to be dealt with at execution
13328 SV* swash = NULL; /* Code points that match \p{} \P{} */
13330 /* Set if a component of this character class is user-defined; just passed
13331 * on to the engine */
13332 bool has_user_defined_property = FALSE;
13334 /* inversion list of code points this node matches only when the target
13335 * string is in UTF-8. (Because is under /d) */
13336 SV* depends_list = NULL;
13338 /* Inversion list of code points this node matches regardless of things
13339 * like locale, folding, utf8ness of the target string */
13340 SV* cp_list = NULL;
13342 /* Like cp_list, but code points on this list need to be checked for things
13343 * that fold to/from them under /i */
13344 SV* cp_foldable_list = NULL;
13346 /* Like cp_list, but code points on this list are valid only when the
13347 * runtime locale is UTF-8 */
13348 SV* only_utf8_locale_list = NULL;
13351 /* In a range, counts how many 0-2 of the ends of it came from literals,
13352 * not escapes. Thus we can tell if 'A' was input vs \x{C1} */
13353 UV literal_endpoint = 0;
13355 bool invert = FALSE; /* Is this class to be complemented */
13357 bool warn_super = ALWAYS_WARN_SUPER;
13359 regnode * const orig_emit = RExC_emit; /* Save the original RExC_emit in
13360 case we need to change the emitted regop to an EXACT. */
13361 const char * orig_parse = RExC_parse;
13362 const SSize_t orig_size = RExC_size;
13363 bool posixl_matches_all = FALSE; /* Does /l class have both e.g. \W,\w ? */
13364 GET_RE_DEBUG_FLAGS_DECL;
13366 PERL_ARGS_ASSERT_REGCLASS;
13368 PERL_UNUSED_ARG(depth);
13371 DEBUG_PARSE("clas");
13373 /* Assume we are going to generate an ANYOF node. */
13374 ret = reganode(pRExC_state, ANYOF, 0);
13377 RExC_size += ANYOF_SKIP;
13378 listsv = &PL_sv_undef; /* For code scanners: listsv always non-NULL. */
13381 ANYOF_FLAGS(ret) = 0;
13383 RExC_emit += ANYOF_SKIP;
13384 listsv = newSVpvs_flags("# comment\n", SVs_TEMP);
13385 initial_listsv_len = SvCUR(listsv);
13386 SvTEMP_off(listsv); /* Grr, TEMPs and mortals are conflated. */
13390 RExC_parse = regpatws(pRExC_state, RExC_parse,
13391 FALSE /* means don't recognize comments */ );
13394 if (UCHARAT(RExC_parse) == '^') { /* Complement of range. */
13397 allow_multi_folds = FALSE;
13400 RExC_parse = regpatws(pRExC_state, RExC_parse,
13401 FALSE /* means don't recognize comments */ );
13405 /* Check that they didn't say [:posix:] instead of [[:posix:]] */
13406 if (!SIZE_ONLY && RExC_parse < RExC_end && POSIXCC(UCHARAT(RExC_parse))) {
13407 const char *s = RExC_parse;
13408 const char c = *s++;
13410 while (isWORDCHAR(*s))
13412 if (*s && c == *s && s[1] == ']') {
13413 SAVEFREESV(RExC_rx_sv);
13415 "POSIX syntax [%c %c] belongs inside character classes",
13417 (void)ReREFCNT_inc(RExC_rx_sv);
13421 /* If the caller wants us to just parse a single element, accomplish this
13422 * by faking the loop ending condition */
13423 if (stop_at_1 && RExC_end > RExC_parse) {
13424 stop_ptr = RExC_parse + 1;
13427 /* allow 1st char to be ']' (allowing it to be '-' is dealt with later) */
13428 if (UCHARAT(RExC_parse) == ']')
13429 goto charclassloop;
13433 if (RExC_parse >= stop_ptr) {
13438 RExC_parse = regpatws(pRExC_state, RExC_parse,
13439 FALSE /* means don't recognize comments */ );
13442 if (UCHARAT(RExC_parse) == ']') {
13448 namedclass = OOB_NAMEDCLASS; /* initialize as illegal */
13449 save_value = value;
13450 save_prevvalue = prevvalue;
13453 rangebegin = RExC_parse;
13457 value = utf8n_to_uvchr((U8*)RExC_parse,
13458 RExC_end - RExC_parse,
13459 &numlen, UTF8_ALLOW_DEFAULT);
13460 RExC_parse += numlen;
13463 value = UCHARAT(RExC_parse++);
13466 && RExC_parse < RExC_end
13467 && POSIXCC(UCHARAT(RExC_parse)))
13469 namedclass = regpposixcc(pRExC_state, value, strict);
13471 else if (value == '\\') {
13473 value = utf8n_to_uvchr((U8*)RExC_parse,
13474 RExC_end - RExC_parse,
13475 &numlen, UTF8_ALLOW_DEFAULT);
13476 RExC_parse += numlen;
13479 value = UCHARAT(RExC_parse++);
13481 /* Some compilers cannot handle switching on 64-bit integer
13482 * values, therefore value cannot be an UV. Yes, this will
13483 * be a problem later if we want switch on Unicode.
13484 * A similar issue a little bit later when switching on
13485 * namedclass. --jhi */
13487 /* If the \ is escaping white space when white space is being
13488 * skipped, it means that that white space is wanted literally, and
13489 * is already in 'value'. Otherwise, need to translate the escape
13490 * into what it signifies. */
13491 if (! skip_white || ! is_PATWS_cp(value)) switch ((I32)value) {
13493 case 'w': namedclass = ANYOF_WORDCHAR; break;
13494 case 'W': namedclass = ANYOF_NWORDCHAR; break;
13495 case 's': namedclass = ANYOF_SPACE; break;
13496 case 'S': namedclass = ANYOF_NSPACE; break;
13497 case 'd': namedclass = ANYOF_DIGIT; break;
13498 case 'D': namedclass = ANYOF_NDIGIT; break;
13499 case 'v': namedclass = ANYOF_VERTWS; break;
13500 case 'V': namedclass = ANYOF_NVERTWS; break;
13501 case 'h': namedclass = ANYOF_HORIZWS; break;
13502 case 'H': namedclass = ANYOF_NHORIZWS; break;
13503 case 'N': /* Handle \N{NAME} in class */
13505 /* We only pay attention to the first char of
13506 multichar strings being returned. I kinda wonder
13507 if this makes sense as it does change the behaviour
13508 from earlier versions, OTOH that behaviour was broken
13510 if (! grok_bslash_N(pRExC_state, NULL, &value, flagp, depth,
13511 TRUE, /* => charclass */
13514 if (*flagp & RESTART_UTF8)
13515 FAIL("panic: grok_bslash_N set RESTART_UTF8");
13525 /* We will handle any undefined properties ourselves */
13526 U8 swash_init_flags = _CORE_SWASH_INIT_RETURN_IF_UNDEF
13527 /* And we actually would prefer to get
13528 * the straight inversion list of the
13529 * swash, since we will be accessing it
13530 * anyway, to save a little time */
13531 |_CORE_SWASH_INIT_ACCEPT_INVLIST;
13533 if (RExC_parse >= RExC_end)
13534 vFAIL2("Empty \\%c{}", (U8)value);
13535 if (*RExC_parse == '{') {
13536 const U8 c = (U8)value;
13537 e = strchr(RExC_parse++, '}');
13539 vFAIL2("Missing right brace on \\%c{}", c);
13540 while (isSPACE(*RExC_parse))
13542 if (e == RExC_parse)
13543 vFAIL2("Empty \\%c{}", c);
13544 n = e - RExC_parse;
13545 while (isSPACE(*(RExC_parse + n - 1)))
13556 if (UCHARAT(RExC_parse) == '^') {
13559 /* toggle. (The rhs xor gets the single bit that
13560 * differs between P and p; the other xor inverts just
13562 value ^= 'P' ^ 'p';
13564 while (isSPACE(*RExC_parse)) {
13569 /* Try to get the definition of the property into
13570 * <invlist>. If /i is in effect, the effective property
13571 * will have its name be <__NAME_i>. The design is
13572 * discussed in commit
13573 * 2f833f5208e26b208886e51e09e2c072b5eabb46 */
13574 name = savepv(Perl_form(aTHX_
13576 (FOLD) ? "__" : "",
13582 /* Look up the property name, and get its swash and
13583 * inversion list, if the property is found */
13585 SvREFCNT_dec_NN(swash);
13587 swash = _core_swash_init("utf8", name, &PL_sv_undef,
13590 NULL, /* No inversion list */
13593 if (! swash || ! (invlist = _get_swash_invlist(swash))) {
13594 HV* curpkg = (IN_PERL_COMPILETIME)
13596 : CopSTASH(PL_curcop);
13598 SvREFCNT_dec_NN(swash);
13602 /* Here didn't find it. It could be a user-defined
13603 * property that will be available at run-time. If we
13604 * accept only compile-time properties, is an error;
13605 * otherwise add it to the list for run-time look up */
13607 RExC_parse = e + 1;
13609 "Property '%"UTF8f"' is unknown",
13610 UTF8fARG(UTF, n, name));
13613 /* If the property name doesn't already have a package
13614 * name, add the current one to it so that it can be
13615 * referred to outside it. [perl #121777] */
13616 if (curpkg && ! instr(name, "::")) {
13617 char* pkgname = HvNAME(curpkg);
13618 if (strNE(pkgname, "main")) {
13619 char* full_name = Perl_form(aTHX_
13623 n = strlen(full_name);
13625 name = savepvn(full_name, n);
13628 Perl_sv_catpvf(aTHX_ listsv, "%cutf8::%"UTF8f"\n",
13629 (value == 'p' ? '+' : '!'),
13630 UTF8fARG(UTF, n, name));
13631 has_user_defined_property = TRUE;
13633 /* We don't know yet, so have to assume that the
13634 * property could match something in the Latin1 range,
13635 * hence something that isn't utf8. Note that this
13636 * would cause things in <depends_list> to match
13637 * inappropriately, except that any \p{}, including
13638 * this one forces Unicode semantics, which means there
13639 * is no <depends_list> */
13640 ANYOF_FLAGS(ret) |= ANYOF_NONBITMAP_NON_UTF8;
13644 /* Here, did get the swash and its inversion list. If
13645 * the swash is from a user-defined property, then this
13646 * whole character class should be regarded as such */
13647 if (swash_init_flags
13648 & _CORE_SWASH_INIT_USER_DEFINED_PROPERTY)
13650 has_user_defined_property = TRUE;
13653 /* We warn on matching an above-Unicode code point
13654 * if the match would return true, except don't
13655 * warn for \p{All}, which has exactly one element
13657 (_invlist_contains_cp(invlist, 0x110000)
13658 && (! (_invlist_len(invlist) == 1
13659 && *invlist_array(invlist) == 0)))
13665 /* Invert if asking for the complement */
13666 if (value == 'P') {
13667 _invlist_union_complement_2nd(properties,
13671 /* The swash can't be used as-is, because we've
13672 * inverted things; delay removing it to here after
13673 * have copied its invlist above */
13674 SvREFCNT_dec_NN(swash);
13678 _invlist_union(properties, invlist, &properties);
13683 RExC_parse = e + 1;
13684 namedclass = ANYOF_UNIPROP; /* no official name, but it's
13687 /* \p means they want Unicode semantics */
13688 RExC_uni_semantics = 1;
13691 case 'n': value = '\n'; break;
13692 case 'r': value = '\r'; break;
13693 case 't': value = '\t'; break;
13694 case 'f': value = '\f'; break;
13695 case 'b': value = '\b'; break;
13696 case 'e': value = ASCII_TO_NATIVE('\033');break;
13697 case 'a': value = '\a'; break;
13699 RExC_parse--; /* function expects to be pointed at the 'o' */
13701 const char* error_msg;
13702 bool valid = grok_bslash_o(&RExC_parse,
13705 SIZE_ONLY, /* warnings in pass
13708 silence_non_portable,
13714 if (PL_encoding && value < 0x100) {
13715 goto recode_encoding;
13719 RExC_parse--; /* function expects to be pointed at the 'x' */
13721 const char* error_msg;
13722 bool valid = grok_bslash_x(&RExC_parse,
13725 TRUE, /* Output warnings */
13727 silence_non_portable,
13733 if (PL_encoding && value < 0x100)
13734 goto recode_encoding;
13737 value = grok_bslash_c(*RExC_parse++, SIZE_ONLY);
13739 case '0': case '1': case '2': case '3': case '4':
13740 case '5': case '6': case '7':
13742 /* Take 1-3 octal digits */
13743 I32 flags = PERL_SCAN_SILENT_ILLDIGIT;
13744 numlen = (strict) ? 4 : 3;
13745 value = grok_oct(--RExC_parse, &numlen, &flags, NULL);
13746 RExC_parse += numlen;
13749 RExC_parse += (UTF) ? UTF8SKIP(RExC_parse) : 1;
13750 vFAIL("Need exactly 3 octal digits");
13752 else if (! SIZE_ONLY /* like \08, \178 */
13754 && RExC_parse < RExC_end
13755 && isDIGIT(*RExC_parse)
13756 && ckWARN(WARN_REGEXP))
13758 SAVEFREESV(RExC_rx_sv);
13759 reg_warn_non_literal_string(
13761 form_short_octal_warning(RExC_parse, numlen));
13762 (void)ReREFCNT_inc(RExC_rx_sv);
13765 if (PL_encoding && value < 0x100)
13766 goto recode_encoding;
13770 if (! RExC_override_recoding) {
13771 SV* enc = PL_encoding;
13772 value = reg_recode((const char)(U8)value, &enc);
13775 vFAIL("Invalid escape in the specified encoding");
13777 else if (SIZE_ONLY) {
13778 ckWARNreg(RExC_parse,
13779 "Invalid escape in the specified encoding");
13785 /* Allow \_ to not give an error */
13786 if (!SIZE_ONLY && isWORDCHAR(value) && value != '_') {
13788 vFAIL2("Unrecognized escape \\%c in character class",
13792 SAVEFREESV(RExC_rx_sv);
13793 ckWARN2reg(RExC_parse,
13794 "Unrecognized escape \\%c in character class passed through",
13796 (void)ReREFCNT_inc(RExC_rx_sv);
13800 } /* End of switch on char following backslash */
13801 } /* end of handling backslash escape sequences */
13804 literal_endpoint++;
13807 /* Here, we have the current token in 'value' */
13809 if (namedclass > OOB_NAMEDCLASS) { /* this is a named class \blah */
13812 /* a bad range like a-\d, a-[:digit:]. The '-' is taken as a
13813 * literal, as is the character that began the false range, i.e.
13814 * the 'a' in the examples */
13817 const int w = (RExC_parse >= rangebegin)
13818 ? RExC_parse - rangebegin
13822 "False [] range \"%"UTF8f"\"",
13823 UTF8fARG(UTF, w, rangebegin));
13826 SAVEFREESV(RExC_rx_sv); /* in case of fatal warnings */
13827 ckWARN2reg(RExC_parse,
13828 "False [] range \"%"UTF8f"\"",
13829 UTF8fARG(UTF, w, rangebegin));
13830 (void)ReREFCNT_inc(RExC_rx_sv);
13831 cp_list = add_cp_to_invlist(cp_list, '-');
13832 cp_foldable_list = add_cp_to_invlist(cp_foldable_list,
13837 range = 0; /* this was not a true range */
13838 element_count += 2; /* So counts for three values */
13841 classnum = namedclass_to_classnum(namedclass);
13843 if (LOC && namedclass < ANYOF_POSIXL_MAX
13844 #ifndef HAS_ISASCII
13845 && classnum != _CC_ASCII
13848 /* What the Posix classes (like \w, [:space:]) match in locale
13849 * isn't knowable under locale until actual match time. Room
13850 * must be reserved (one time per outer bracketed class) to
13851 * store such classes. The space will contain a bit for each
13852 * named class that is to be matched against. This isn't
13853 * needed for \p{} and pseudo-classes, as they are not affected
13854 * by locale, and hence are dealt with separately */
13855 if (! need_class) {
13858 RExC_size += ANYOF_POSIXL_SKIP - ANYOF_SKIP;
13861 RExC_emit += ANYOF_POSIXL_SKIP - ANYOF_SKIP;
13863 ANYOF_FLAGS(ret) |= ANYOF_POSIXL;
13864 ANYOF_POSIXL_ZERO(ret);
13867 /* Coverity thinks it is possible for this to be negative; both
13868 * jhi and khw think it's not, but be safer */
13869 assert(! (ANYOF_FLAGS(ret) & ANYOF_POSIXL)
13870 || (namedclass + ((namedclass % 2) ? -1 : 1)) >= 0);
13872 /* See if it already matches the complement of this POSIX
13874 if ((ANYOF_FLAGS(ret) & ANYOF_POSIXL)
13875 && ANYOF_POSIXL_TEST(ret, namedclass + ((namedclass % 2)
13879 posixl_matches_all = TRUE;
13880 break; /* No need to continue. Since it matches both
13881 e.g., \w and \W, it matches everything, and the
13882 bracketed class can be optimized into qr/./s */
13885 /* Add this class to those that should be checked at runtime */
13886 ANYOF_POSIXL_SET(ret, namedclass);
13888 /* The above-Latin1 characters are not subject to locale rules.
13889 * Just add them, in the second pass, to the
13890 * unconditionally-matched list */
13892 SV* scratch_list = NULL;
13894 /* Get the list of the above-Latin1 code points this
13896 _invlist_intersection_maybe_complement_2nd(PL_AboveLatin1,
13897 PL_XPosix_ptrs[classnum],
13899 /* Odd numbers are complements, like
13900 * NDIGIT, NASCII, ... */
13901 namedclass % 2 != 0,
13903 /* Checking if 'cp_list' is NULL first saves an extra
13904 * clone. Its reference count will be decremented at the
13905 * next union, etc, or if this is the only instance, at the
13906 * end of the routine */
13908 cp_list = scratch_list;
13911 _invlist_union(cp_list, scratch_list, &cp_list);
13912 SvREFCNT_dec_NN(scratch_list);
13914 continue; /* Go get next character */
13917 else if (! SIZE_ONLY) {
13919 /* Here, not in pass1 (in that pass we skip calculating the
13920 * contents of this class), and is /l, or is a POSIX class for
13921 * which /l doesn't matter (or is a Unicode property, which is
13922 * skipped here). */
13923 if (namedclass >= ANYOF_POSIXL_MAX) { /* If a special class */
13924 if (namedclass != ANYOF_UNIPROP) { /* UNIPROP = \p and \P */
13926 /* Here, should be \h, \H, \v, or \V. None of /d, /i
13927 * nor /l make a difference in what these match,
13928 * therefore we just add what they match to cp_list. */
13929 if (classnum != _CC_VERTSPACE) {
13930 assert( namedclass == ANYOF_HORIZWS
13931 || namedclass == ANYOF_NHORIZWS);
13933 /* It turns out that \h is just a synonym for
13935 classnum = _CC_BLANK;
13938 _invlist_union_maybe_complement_2nd(
13940 PL_XPosix_ptrs[classnum],
13941 namedclass % 2 != 0, /* Complement if odd
13942 (NHORIZWS, NVERTWS)
13947 else { /* Garden variety class. If is NASCII, NDIGIT, ...
13948 complement and use nposixes */
13949 SV** posixes_ptr = namedclass % 2 == 0
13952 SV** source_ptr = &PL_XPosix_ptrs[classnum];
13953 _invlist_union_maybe_complement_2nd(
13956 namedclass % 2 != 0,
13959 continue; /* Go get next character */
13961 } /* end of namedclass \blah */
13963 /* Here, we have a single value. If 'range' is set, it is the ending
13964 * of a range--check its validity. Later, we will handle each
13965 * individual code point in the range. If 'range' isn't set, this
13966 * could be the beginning of a range, so check for that by looking
13967 * ahead to see if the next real character to be processed is the range
13968 * indicator--the minus sign */
13971 RExC_parse = regpatws(pRExC_state, RExC_parse,
13972 FALSE /* means don't recognize comments */ );
13976 if (prevvalue > value) /* b-a */ {
13977 const int w = RExC_parse - rangebegin;
13979 "Invalid [] range \"%"UTF8f"\"",
13980 UTF8fARG(UTF, w, rangebegin));
13981 range = 0; /* not a valid range */
13985 prevvalue = value; /* save the beginning of the potential range */
13986 if (! stop_at_1 /* Can't be a range if parsing just one thing */
13987 && *RExC_parse == '-')
13989 char* next_char_ptr = RExC_parse + 1;
13990 if (skip_white) { /* Get the next real char after the '-' */
13991 next_char_ptr = regpatws(pRExC_state,
13993 FALSE); /* means don't recognize
13997 /* If the '-' is at the end of the class (just before the ']',
13998 * it is a literal minus; otherwise it is a range */
13999 if (next_char_ptr < RExC_end && *next_char_ptr != ']') {
14000 RExC_parse = next_char_ptr;
14002 /* a bad range like \w-, [:word:]- ? */
14003 if (namedclass > OOB_NAMEDCLASS) {
14004 if (strict || ckWARN(WARN_REGEXP)) {
14006 RExC_parse >= rangebegin ?
14007 RExC_parse - rangebegin : 0;
14009 vFAIL4("False [] range \"%*.*s\"",
14014 "False [] range \"%*.*s\"",
14019 cp_list = add_cp_to_invlist(cp_list, '-');
14023 range = 1; /* yeah, it's a range! */
14024 continue; /* but do it the next time */
14029 /* Here, <prevvalue> is the beginning of the range, if any; or <value>
14032 /* non-Latin1 code point implies unicode semantics. Must be set in
14033 * pass1 so is there for the whole of pass 2 */
14035 RExC_uni_semantics = 1;
14038 /* Ready to process either the single value, or the completed range.
14039 * For single-valued non-inverted ranges, we consider the possibility
14040 * of multi-char folds. (We made a conscious decision to not do this
14041 * for the other cases because it can often lead to non-intuitive
14042 * results. For example, you have the peculiar case that:
14043 * "s s" =~ /^[^\xDF]+$/i => Y
14044 * "ss" =~ /^[^\xDF]+$/i => N
14046 * See [perl #89750] */
14047 if (FOLD && allow_multi_folds && value == prevvalue) {
14048 if (value == LATIN_SMALL_LETTER_SHARP_S
14049 || (value > 255 && _invlist_contains_cp(PL_HasMultiCharFold,
14052 /* Here <value> is indeed a multi-char fold. Get what it is */
14054 U8 foldbuf[UTF8_MAXBYTES_CASE];
14057 UV folded = _to_uni_fold_flags(
14061 FOLD_FLAGS_FULL | (ASCII_FOLD_RESTRICTED
14062 ? FOLD_FLAGS_NOMIX_ASCII
14066 /* Here, <folded> should be the first character of the
14067 * multi-char fold of <value>, with <foldbuf> containing the
14068 * whole thing. But, if this fold is not allowed (because of
14069 * the flags), <fold> will be the same as <value>, and should
14070 * be processed like any other character, so skip the special
14072 if (folded != value) {
14074 /* Skip if we are recursed, currently parsing the class
14075 * again. Otherwise add this character to the list of
14076 * multi-char folds. */
14077 if (! RExC_in_multi_char_class) {
14078 AV** this_array_ptr;
14080 STRLEN cp_count = utf8_length(foldbuf,
14081 foldbuf + foldlen);
14082 SV* multi_fold = sv_2mortal(newSVpvs(""));
14084 Perl_sv_catpvf(aTHX_ multi_fold, "\\x{%"UVXf"}", value);
14087 if (! multi_char_matches) {
14088 multi_char_matches = newAV();
14091 /* <multi_char_matches> is actually an array of arrays.
14092 * There will be one or two top-level elements: [2],
14093 * and/or [3]. The [2] element is an array, each
14094 * element thereof is a character which folds to TWO
14095 * characters; [3] is for folds to THREE characters.
14096 * (Unicode guarantees a maximum of 3 characters in any
14097 * fold.) When we rewrite the character class below,
14098 * we will do so such that the longest folds are
14099 * written first, so that it prefers the longest
14100 * matching strings first. This is done even if it
14101 * turns out that any quantifier is non-greedy, out of
14102 * programmer laziness. Tom Christiansen has agreed
14103 * that this is ok. This makes the test for the
14104 * ligature 'ffi' come before the test for 'ff' */
14105 if (av_exists(multi_char_matches, cp_count)) {
14106 this_array_ptr = (AV**) av_fetch(multi_char_matches,
14108 this_array = *this_array_ptr;
14111 this_array = newAV();
14112 av_store(multi_char_matches, cp_count,
14115 av_push(this_array, multi_fold);
14118 /* This element should not be processed further in this
14121 value = save_value;
14122 prevvalue = save_prevvalue;
14128 /* Deal with this element of the class */
14131 cp_foldable_list = _add_range_to_invlist(cp_foldable_list,
14134 SV* this_range = _new_invlist(1);
14135 _append_range_to_invlist(this_range, prevvalue, value);
14137 /* In EBCDIC, the ranges 'A-Z' and 'a-z' are each not contiguous.
14138 * If this range was specified using something like 'i-j', we want
14139 * to include only the 'i' and the 'j', and not anything in
14140 * between, so exclude non-ASCII, non-alphabetics from it.
14141 * However, if the range was specified with something like
14142 * [\x89-\x91] or [\x89-j], all code points within it should be
14143 * included. literal_endpoint==2 means both ends of the range used
14144 * a literal character, not \x{foo} */
14145 if (literal_endpoint == 2
14146 && ((prevvalue >= 'a' && value <= 'z')
14147 || (prevvalue >= 'A' && value <= 'Z')))
14149 _invlist_intersection(this_range, PL_XPosix_ptrs[_CC_ASCII],
14152 /* Since this above only contains ascii, the intersection of it
14153 * with anything will still yield only ascii */
14154 _invlist_intersection(this_range, PL_XPosix_ptrs[_CC_ALPHA],
14157 _invlist_union(cp_foldable_list, this_range, &cp_foldable_list);
14158 literal_endpoint = 0;
14162 range = 0; /* this range (if it was one) is done now */
14163 } /* End of loop through all the text within the brackets */
14165 /* If anything in the class expands to more than one character, we have to
14166 * deal with them by building up a substitute parse string, and recursively
14167 * calling reg() on it, instead of proceeding */
14168 if (multi_char_matches) {
14169 SV * substitute_parse = newSVpvn_flags("?:", 2, SVs_TEMP);
14172 char *save_end = RExC_end;
14173 char *save_parse = RExC_parse;
14174 bool first_time = TRUE; /* First multi-char occurrence doesn't get
14179 #if 0 /* Have decided not to deal with multi-char folds in inverted classes,
14180 because too confusing */
14182 sv_catpv(substitute_parse, "(?:");
14186 /* Look at the longest folds first */
14187 for (cp_count = av_tindex(multi_char_matches); cp_count > 0; cp_count--) {
14189 if (av_exists(multi_char_matches, cp_count)) {
14190 AV** this_array_ptr;
14193 this_array_ptr = (AV**) av_fetch(multi_char_matches,
14195 while ((this_sequence = av_pop(*this_array_ptr)) !=
14198 if (! first_time) {
14199 sv_catpv(substitute_parse, "|");
14201 first_time = FALSE;
14203 sv_catpv(substitute_parse, SvPVX(this_sequence));
14208 /* If the character class contains anything else besides these
14209 * multi-character folds, have to include it in recursive parsing */
14210 if (element_count) {
14211 sv_catpv(substitute_parse, "|[");
14212 sv_catpvn(substitute_parse, orig_parse, RExC_parse - orig_parse);
14213 sv_catpv(substitute_parse, "]");
14216 sv_catpv(substitute_parse, ")");
14219 /* This is a way to get the parse to skip forward a whole named
14220 * sequence instead of matching the 2nd character when it fails the
14222 sv_catpv(substitute_parse, "(*THEN)(*SKIP)(*FAIL)|.)");
14226 RExC_parse = SvPV(substitute_parse, len);
14227 RExC_end = RExC_parse + len;
14228 RExC_in_multi_char_class = 1;
14229 RExC_emit = (regnode *)orig_emit;
14231 ret = reg(pRExC_state, 1, ®_flags, depth+1);
14233 *flagp |= reg_flags&(HASWIDTH|SIMPLE|SPSTART|POSTPONED|RESTART_UTF8);
14235 RExC_parse = save_parse;
14236 RExC_end = save_end;
14237 RExC_in_multi_char_class = 0;
14238 SvREFCNT_dec_NN(multi_char_matches);
14242 /* Here, we've gone through the entire class and dealt with multi-char
14243 * folds. We are now in a position that we can do some checks to see if we
14244 * can optimize this ANYOF node into a simpler one, even in Pass 1.
14245 * Currently we only do two checks:
14246 * 1) is in the unlikely event that the user has specified both, eg. \w and
14247 * \W under /l, then the class matches everything. (This optimization
14248 * is done only to make the optimizer code run later work.)
14249 * 2) if the character class contains only a single element (including a
14250 * single range), we see if there is an equivalent node for it.
14251 * Other checks are possible */
14252 if (! ret_invlist /* Can't optimize if returning the constructed
14254 && (UNLIKELY(posixl_matches_all) || element_count == 1))
14259 if (UNLIKELY(posixl_matches_all)) {
14262 else if (namedclass > OOB_NAMEDCLASS) { /* this is a named class, like
14263 \w or [:digit:] or \p{foo}
14266 /* All named classes are mapped into POSIXish nodes, with its FLAG
14267 * argument giving which class it is */
14268 switch ((I32)namedclass) {
14269 case ANYOF_UNIPROP:
14272 /* These don't depend on the charset modifiers. They always
14273 * match under /u rules */
14274 case ANYOF_NHORIZWS:
14275 case ANYOF_HORIZWS:
14276 namedclass = ANYOF_BLANK + namedclass - ANYOF_HORIZWS;
14279 case ANYOF_NVERTWS:
14284 /* The actual POSIXish node for all the rest depends on the
14285 * charset modifier. The ones in the first set depend only on
14286 * ASCII or, if available on this platform, locale */
14290 op = (LOC) ? POSIXL : POSIXA;
14301 /* under /a could be alpha */
14303 if (ASCII_RESTRICTED) {
14304 namedclass = ANYOF_ALPHA + (namedclass % 2);
14312 /* The rest have more possibilities depending on the charset.
14313 * We take advantage of the enum ordering of the charset
14314 * modifiers to get the exact node type, */
14316 op = POSIXD + get_regex_charset(RExC_flags);
14317 if (op > POSIXA) { /* /aa is same as /a */
14322 /* The odd numbered ones are the complements of the
14323 * next-lower even number one */
14324 if (namedclass % 2 == 1) {
14328 arg = namedclass_to_classnum(namedclass);
14332 else if (value == prevvalue) {
14334 /* Here, the class consists of just a single code point */
14337 if (! LOC && value == '\n') {
14338 op = REG_ANY; /* Optimize [^\n] */
14339 *flagp |= HASWIDTH|SIMPLE;
14343 else if (value < 256 || UTF) {
14345 /* Optimize a single value into an EXACTish node, but not if it
14346 * would require converting the pattern to UTF-8. */
14347 op = compute_EXACTish(pRExC_state);
14349 } /* Otherwise is a range */
14350 else if (! LOC) { /* locale could vary these */
14351 if (prevvalue == '0') {
14352 if (value == '9') {
14357 else if (prevvalue == 'A') {
14360 && literal_endpoint == 2
14363 arg = (FOLD) ? _CC_ALPHA : _CC_UPPER;
14367 else if (prevvalue == 'a') {
14370 && literal_endpoint == 2
14373 arg = (FOLD) ? _CC_ALPHA : _CC_LOWER;
14379 /* Here, we have changed <op> away from its initial value iff we found
14380 * an optimization */
14383 /* Throw away this ANYOF regnode, and emit the calculated one,
14384 * which should correspond to the beginning, not current, state of
14386 const char * cur_parse = RExC_parse;
14387 RExC_parse = (char *)orig_parse;
14391 /* To get locale nodes to not use the full ANYOF size would
14392 * require moving the code above that writes the portions
14393 * of it that aren't in other nodes to after this point.
14394 * e.g. ANYOF_POSIXL_SET */
14395 RExC_size = orig_size;
14399 RExC_emit = (regnode *)orig_emit;
14400 if (PL_regkind[op] == POSIXD) {
14401 if (op == POSIXL) {
14402 RExC_contains_locale = 1;
14405 op += NPOSIXD - POSIXD;
14410 ret = reg_node(pRExC_state, op);
14412 if (PL_regkind[op] == POSIXD || PL_regkind[op] == NPOSIXD) {
14416 *flagp |= HASWIDTH|SIMPLE;
14418 else if (PL_regkind[op] == EXACT) {
14419 alloc_maybe_populate_EXACT(pRExC_state, ret, flagp, 0, value,
14420 TRUE /* downgradable to EXACT */
14424 RExC_parse = (char *) cur_parse;
14426 SvREFCNT_dec(posixes);
14427 SvREFCNT_dec(nposixes);
14428 SvREFCNT_dec(cp_list);
14429 SvREFCNT_dec(cp_foldable_list);
14436 /****** !SIZE_ONLY (Pass 2) AFTER HERE *********/
14438 /* If folding, we calculate all characters that could fold to or from the
14439 * ones already on the list */
14440 if (cp_foldable_list) {
14442 UV start, end; /* End points of code point ranges */
14444 SV* fold_intersection = NULL;
14447 /* Our calculated list will be for Unicode rules. For locale
14448 * matching, we have to keep a separate list that is consulted at
14449 * runtime only when the locale indicates Unicode rules. For
14450 * non-locale, we just use to the general list */
14452 use_list = &only_utf8_locale_list;
14455 use_list = &cp_list;
14458 /* Only the characters in this class that participate in folds need
14459 * be checked. Get the intersection of this class and all the
14460 * possible characters that are foldable. This can quickly narrow
14461 * down a large class */
14462 _invlist_intersection(PL_utf8_foldable, cp_foldable_list,
14463 &fold_intersection);
14465 /* The folds for all the Latin1 characters are hard-coded into this
14466 * program, but we have to go out to disk to get the others. */
14467 if (invlist_highest(cp_foldable_list) >= 256) {
14469 /* This is a hash that for a particular fold gives all
14470 * characters that are involved in it */
14471 if (! PL_utf8_foldclosures) {
14472 _load_PL_utf8_foldclosures();
14476 /* Now look at the foldable characters in this class individually */
14477 invlist_iterinit(fold_intersection);
14478 while (invlist_iternext(fold_intersection, &start, &end)) {
14481 /* Look at every character in the range */
14482 for (j = start; j <= end; j++) {
14483 U8 foldbuf[UTF8_MAXBYTES_CASE+1];
14489 if (IS_IN_SOME_FOLD_L1(j)) {
14491 /* ASCII is always matched; non-ASCII is matched
14492 * only under Unicode rules (which could happen
14493 * under /l if the locale is a UTF-8 one */
14494 if (isASCII(j) || ! DEPENDS_SEMANTICS) {
14495 *use_list = add_cp_to_invlist(*use_list,
14496 PL_fold_latin1[j]);
14500 add_cp_to_invlist(depends_list,
14501 PL_fold_latin1[j]);
14505 if (HAS_NONLATIN1_SIMPLE_FOLD_CLOSURE(j)
14506 && (! isASCII(j) || ! ASCII_FOLD_RESTRICTED))
14508 add_above_Latin1_folds(pRExC_state,
14515 /* Here is an above Latin1 character. We don't have the
14516 * rules hard-coded for it. First, get its fold. This is
14517 * the simple fold, as the multi-character folds have been
14518 * handled earlier and separated out */
14519 _to_uni_fold_flags(j, foldbuf, &foldlen,
14520 (ASCII_FOLD_RESTRICTED)
14521 ? FOLD_FLAGS_NOMIX_ASCII
14524 /* Single character fold of above Latin1. Add everything in
14525 * its fold closure to the list that this node should match.
14526 * The fold closures data structure is a hash with the keys
14527 * being the UTF-8 of every character that is folded to, like
14528 * 'k', and the values each an array of all code points that
14529 * fold to its key. e.g. [ 'k', 'K', KELVIN_SIGN ].
14530 * Multi-character folds are not included */
14531 if ((listp = hv_fetch(PL_utf8_foldclosures,
14532 (char *) foldbuf, foldlen, FALSE)))
14534 AV* list = (AV*) *listp;
14536 for (k = 0; k <= av_tindex(list); k++) {
14537 SV** c_p = av_fetch(list, k, FALSE);
14543 /* /aa doesn't allow folds between ASCII and non- */
14544 if ((ASCII_FOLD_RESTRICTED
14545 && (isASCII(c) != isASCII(j))))
14550 /* Folds under /l which cross the 255/256 boundary
14551 * are added to a separate list. (These are valid
14552 * only when the locale is UTF-8.) */
14553 if (c < 256 && LOC) {
14554 *use_list = add_cp_to_invlist(*use_list, c);
14558 if (isASCII(c) || c > 255 || AT_LEAST_UNI_SEMANTICS)
14560 cp_list = add_cp_to_invlist(cp_list, c);
14563 /* Similarly folds involving non-ascii Latin1
14564 * characters under /d are added to their list */
14565 depends_list = add_cp_to_invlist(depends_list,
14572 SvREFCNT_dec_NN(fold_intersection);
14575 /* Now that we have finished adding all the folds, there is no reason
14576 * to keep the foldable list separate */
14577 _invlist_union(cp_list, cp_foldable_list, &cp_list);
14578 SvREFCNT_dec_NN(cp_foldable_list);
14581 /* And combine the result (if any) with any inversion list from posix
14582 * classes. The lists are kept separate up to now because we don't want to
14583 * fold the classes (folding of those is automatically handled by the swash
14584 * fetching code) */
14585 if (posixes || nposixes) {
14586 if (posixes && AT_LEAST_ASCII_RESTRICTED) {
14587 /* Under /a and /aa, nothing above ASCII matches these */
14588 _invlist_intersection(posixes,
14589 PL_XPosix_ptrs[_CC_ASCII],
14593 if (DEPENDS_SEMANTICS) {
14594 /* Under /d, everything in the upper half of the Latin1 range
14595 * matches these complements */
14596 ANYOF_FLAGS(ret) |= ANYOF_NON_UTF8_NON_ASCII_ALL;
14598 else if (AT_LEAST_ASCII_RESTRICTED) {
14599 /* Under /a and /aa, everything above ASCII matches these
14601 _invlist_union_complement_2nd(nposixes,
14602 PL_XPosix_ptrs[_CC_ASCII],
14606 _invlist_union(posixes, nposixes, &posixes);
14607 SvREFCNT_dec_NN(nposixes);
14610 posixes = nposixes;
14613 if (! DEPENDS_SEMANTICS) {
14615 _invlist_union(cp_list, posixes, &cp_list);
14616 SvREFCNT_dec_NN(posixes);
14623 /* Under /d, we put into a separate list the Latin1 things that
14624 * match only when the target string is utf8 */
14625 SV* nonascii_but_latin1_properties = NULL;
14626 _invlist_intersection(posixes, PL_UpperLatin1,
14627 &nonascii_but_latin1_properties);
14628 _invlist_subtract(posixes, nonascii_but_latin1_properties,
14631 _invlist_union(cp_list, posixes, &cp_list);
14632 SvREFCNT_dec_NN(posixes);
14638 if (depends_list) {
14639 _invlist_union(depends_list, nonascii_but_latin1_properties,
14641 SvREFCNT_dec_NN(nonascii_but_latin1_properties);
14644 depends_list = nonascii_but_latin1_properties;
14649 /* And combine the result (if any) with any inversion list from properties.
14650 * The lists are kept separate up to now so that we can distinguish the two
14651 * in regards to matching above-Unicode. A run-time warning is generated
14652 * if a Unicode property is matched against a non-Unicode code point. But,
14653 * we allow user-defined properties to match anything, without any warning,
14654 * and we also suppress the warning if there is a portion of the character
14655 * class that isn't a Unicode property, and which matches above Unicode, \W
14656 * or [\x{110000}] for example.
14657 * (Note that in this case, unlike the Posix one above, there is no
14658 * <depends_list>, because having a Unicode property forces Unicode
14663 /* If it matters to the final outcome, see if a non-property
14664 * component of the class matches above Unicode. If so, the
14665 * warning gets suppressed. This is true even if just a single
14666 * such code point is specified, as though not strictly correct if
14667 * another such code point is matched against, the fact that they
14668 * are using above-Unicode code points indicates they should know
14669 * the issues involved */
14671 warn_super = ! (invert
14672 ^ (invlist_highest(cp_list) > PERL_UNICODE_MAX));
14675 _invlist_union(properties, cp_list, &cp_list);
14676 SvREFCNT_dec_NN(properties);
14679 cp_list = properties;
14683 ANYOF_FLAGS(ret) |= ANYOF_WARN_SUPER;
14687 /* Here, we have calculated what code points should be in the character
14690 * Now we can see about various optimizations. Fold calculation (which we
14691 * did above) needs to take place before inversion. Otherwise /[^k]/i
14692 * would invert to include K, which under /i would match k, which it
14693 * shouldn't. Therefore we can't invert folded locale now, as it won't be
14694 * folded until runtime */
14696 /* If we didn't do folding, it's because some information isn't available
14697 * until runtime; set the run-time fold flag for these. (We don't have to
14698 * worry about properties folding, as that is taken care of by the swash
14699 * fetching). We know to set the flag if we have a non-NULL list for UTF-8
14700 * locales, or the class matches at least one 0-255 range code point */
14702 if (only_utf8_locale_list) {
14703 ANYOF_FLAGS(ret) |= ANYOF_LOC_FOLD;
14705 else if (cp_list) { /* Look to see if there a 0-255 code point is in
14708 invlist_iterinit(cp_list);
14709 if (invlist_iternext(cp_list, &start, &end) && start < 256) {
14710 ANYOF_FLAGS(ret) |= ANYOF_LOC_FOLD;
14712 invlist_iterfinish(cp_list);
14716 /* Optimize inverted simple patterns (e.g. [^a-z]) when everything is known
14717 * at compile time. Besides not inverting folded locale now, we can't
14718 * invert if there are things such as \w, which aren't known until runtime
14722 && ! (ANYOF_FLAGS(ret) & (ANYOF_LOCALE_FLAGS))
14724 && ! HAS_NONLOCALE_RUNTIME_PROPERTY_DEFINITION)
14726 _invlist_invert(cp_list);
14728 /* Any swash can't be used as-is, because we've inverted things */
14730 SvREFCNT_dec_NN(swash);
14734 /* Clear the invert flag since have just done it here */
14739 *ret_invlist = cp_list;
14740 SvREFCNT_dec(swash);
14742 /* Discard the generated node */
14744 RExC_size = orig_size;
14747 RExC_emit = orig_emit;
14752 /* Some character classes are equivalent to other nodes. Such nodes take
14753 * up less room and generally fewer operations to execute than ANYOF nodes.
14754 * Above, we checked for and optimized into some such equivalents for
14755 * certain common classes that are easy to test. Getting to this point in
14756 * the code means that the class didn't get optimized there. Since this
14757 * code is only executed in Pass 2, it is too late to save space--it has
14758 * been allocated in Pass 1, and currently isn't given back. But turning
14759 * things into an EXACTish node can allow the optimizer to join it to any
14760 * adjacent such nodes. And if the class is equivalent to things like /./,
14761 * expensive run-time swashes can be avoided. Now that we have more
14762 * complete information, we can find things necessarily missed by the
14763 * earlier code. I (khw) am not sure how much to look for here. It would
14764 * be easy, but perhaps too slow, to check any candidates against all the
14765 * node types they could possibly match using _invlistEQ(). */
14770 && ! (ANYOF_FLAGS(ret) & (ANYOF_LOCALE_FLAGS))
14771 && ! HAS_NONLOCALE_RUNTIME_PROPERTY_DEFINITION
14773 /* We don't optimize if we are supposed to make sure all non-Unicode
14774 * code points raise a warning, as only ANYOF nodes have this check.
14776 && ! ((ANYOF_FLAGS(ret) & ANYOF_WARN_SUPER) && ALWAYS_WARN_SUPER))
14779 U8 op = END; /* The optimzation node-type */
14780 const char * cur_parse= RExC_parse;
14782 invlist_iterinit(cp_list);
14783 if (! invlist_iternext(cp_list, &start, &end)) {
14785 /* Here, the list is empty. This happens, for example, when a
14786 * Unicode property is the only thing in the character class, and
14787 * it doesn't match anything. (perluniprops.pod notes such
14790 *flagp |= HASWIDTH|SIMPLE;
14792 else if (start == end) { /* The range is a single code point */
14793 if (! invlist_iternext(cp_list, &start, &end)
14795 /* Don't do this optimization if it would require changing
14796 * the pattern to UTF-8 */
14797 && (start < 256 || UTF))
14799 /* Here, the list contains a single code point. Can optimize
14800 * into an EXACTish node */
14809 /* A locale node under folding with one code point can be
14810 * an EXACTFL, as its fold won't be calculated until
14816 /* Here, we are generally folding, but there is only one
14817 * code point to match. If we have to, we use an EXACT
14818 * node, but it would be better for joining with adjacent
14819 * nodes in the optimization pass if we used the same
14820 * EXACTFish node that any such are likely to be. We can
14821 * do this iff the code point doesn't participate in any
14822 * folds. For example, an EXACTF of a colon is the same as
14823 * an EXACT one, since nothing folds to or from a colon. */
14825 if (IS_IN_SOME_FOLD_L1(value)) {
14830 if (_invlist_contains_cp(PL_utf8_foldable, value)) {
14835 /* If we haven't found the node type, above, it means we
14836 * can use the prevailing one */
14838 op = compute_EXACTish(pRExC_state);
14843 else if (start == 0) {
14844 if (end == UV_MAX) {
14846 *flagp |= HASWIDTH|SIMPLE;
14849 else if (end == '\n' - 1
14850 && invlist_iternext(cp_list, &start, &end)
14851 && start == '\n' + 1 && end == UV_MAX)
14854 *flagp |= HASWIDTH|SIMPLE;
14858 invlist_iterfinish(cp_list);
14861 RExC_parse = (char *)orig_parse;
14862 RExC_emit = (regnode *)orig_emit;
14864 ret = reg_node(pRExC_state, op);
14866 RExC_parse = (char *)cur_parse;
14868 if (PL_regkind[op] == EXACT) {
14869 alloc_maybe_populate_EXACT(pRExC_state, ret, flagp, 0, value,
14870 TRUE /* downgradable to EXACT */
14874 SvREFCNT_dec_NN(cp_list);
14879 /* Here, <cp_list> contains all the code points we can determine at
14880 * compile time that match under all conditions. Go through it, and
14881 * for things that belong in the bitmap, put them there, and delete from
14882 * <cp_list>. While we are at it, see if everything above 255 is in the
14883 * list, and if so, set a flag to speed up execution */
14885 populate_ANYOF_from_invlist(ret, &cp_list);
14888 ANYOF_FLAGS(ret) |= ANYOF_INVERT;
14891 /* Here, the bitmap has been populated with all the Latin1 code points that
14892 * always match. Can now add to the overall list those that match only
14893 * when the target string is UTF-8 (<depends_list>). */
14894 if (depends_list) {
14896 _invlist_union(cp_list, depends_list, &cp_list);
14897 SvREFCNT_dec_NN(depends_list);
14900 cp_list = depends_list;
14902 ANYOF_FLAGS(ret) |= ANYOF_UTF8;
14905 /* If there is a swash and more than one element, we can't use the swash in
14906 * the optimization below. */
14907 if (swash && element_count > 1) {
14908 SvREFCNT_dec_NN(swash);
14912 set_ANYOF_arg(pRExC_state, ret, cp_list,
14913 (HAS_NONLOCALE_RUNTIME_PROPERTY_DEFINITION)
14915 only_utf8_locale_list,
14916 swash, has_user_defined_property);
14918 *flagp |= HASWIDTH|SIMPLE;
14920 if (ANYOF_FLAGS(ret) & ANYOF_LOCALE_FLAGS) {
14921 RExC_contains_locale = 1;
14927 #undef HAS_NONLOCALE_RUNTIME_PROPERTY_DEFINITION
14930 S_set_ANYOF_arg(pTHX_ RExC_state_t* const pRExC_state,
14931 regnode* const node,
14933 SV* const runtime_defns,
14934 SV* const only_utf8_locale_list,
14936 const bool has_user_defined_property)
14938 /* Sets the arg field of an ANYOF-type node 'node', using information about
14939 * the node passed-in. If there is nothing outside the node's bitmap, the
14940 * arg is set to ANYOF_NONBITMAP_EMPTY. Otherwise, it sets the argument to
14941 * the count returned by add_data(), having allocated and stored an array,
14942 * av, that that count references, as follows:
14943 * av[0] stores the character class description in its textual form.
14944 * This is used later (regexec.c:Perl_regclass_swash()) to
14945 * initialize the appropriate swash, and is also useful for dumping
14946 * the regnode. This is set to &PL_sv_undef if the textual
14947 * description is not needed at run-time (as happens if the other
14948 * elements completely define the class)
14949 * av[1] if &PL_sv_undef, is a placeholder to later contain the swash
14950 * computed from av[0]. But if no further computation need be done,
14951 * the swash is stored here now (and av[0] is &PL_sv_undef).
14952 * av[2] stores the inversion list of code points that match only if the
14953 * current locale is UTF-8
14954 * av[3] stores the cp_list inversion list for use in addition or instead
14955 * of av[0]; used only if cp_list exists and av[1] is &PL_sv_undef.
14956 * (Otherwise everything needed is already in av[0] and av[1])
14957 * av[4] is set if any component of the class is from a user-defined
14958 * property; used only if av[3] exists */
14962 PERL_ARGS_ASSERT_SET_ANYOF_ARG;
14964 if (! cp_list && ! runtime_defns && ! only_utf8_locale_list) {
14965 assert(! (ANYOF_FLAGS(node)
14966 & (ANYOF_UTF8|ANYOF_NONBITMAP_NON_UTF8)));
14967 ARG_SET(node, ANYOF_NONBITMAP_EMPTY);
14970 AV * const av = newAV();
14973 assert(ANYOF_FLAGS(node)
14974 & (ANYOF_UTF8|ANYOF_NONBITMAP_NON_UTF8|ANYOF_LOC_FOLD));
14976 av_store(av, 0, (runtime_defns)
14977 ? SvREFCNT_inc(runtime_defns) : &PL_sv_undef);
14980 av_store(av, 1, swash);
14981 SvREFCNT_dec_NN(cp_list);
14984 av_store(av, 1, &PL_sv_undef);
14986 av_store(av, 3, cp_list);
14987 av_store(av, 4, newSVuv(has_user_defined_property));
14991 if (only_utf8_locale_list) {
14992 av_store(av, 2, only_utf8_locale_list);
14995 av_store(av, 2, &PL_sv_undef);
14998 rv = newRV_noinc(MUTABLE_SV(av));
14999 n = add_data(pRExC_state, STR_WITH_LEN("s"));
15000 RExC_rxi->data->data[n] = (void*)rv;
15006 /* reg_skipcomment()
15008 Absorbs an /x style # comment from the input stream,
15009 returning a pointer to the first character beyond the comment, or if the
15010 comment terminates the pattern without anything following it, this returns
15011 one past the final character of the pattern (in other words, RExC_end) and
15012 sets the REG_RUN_ON_COMMENT_SEEN flag.
15014 Note it's the callers responsibility to ensure that we are
15015 actually in /x mode
15019 PERL_STATIC_INLINE char*
15020 S_reg_skipcomment(RExC_state_t *pRExC_state, char* p)
15022 PERL_ARGS_ASSERT_REG_SKIPCOMMENT;
15026 while (p < RExC_end) {
15027 if (*(++p) == '\n') {
15032 /* we ran off the end of the pattern without ending the comment, so we have
15033 * to add an \n when wrapping */
15034 RExC_seen |= REG_RUN_ON_COMMENT_SEEN;
15040 Advances the parse position, and optionally absorbs
15041 "whitespace" from the inputstream.
15043 Without /x "whitespace" means (?#...) style comments only,
15044 with /x this means (?#...) and # comments and whitespace proper.
15046 Returns the RExC_parse point from BEFORE the scan occurs.
15048 This is the /x friendly way of saying RExC_parse++.
15052 S_nextchar(pTHX_ RExC_state_t *pRExC_state)
15054 char* const retval = RExC_parse++;
15056 PERL_ARGS_ASSERT_NEXTCHAR;
15059 if (RExC_end - RExC_parse >= 3
15060 && *RExC_parse == '('
15061 && RExC_parse[1] == '?'
15062 && RExC_parse[2] == '#')
15064 while (*RExC_parse != ')') {
15065 if (RExC_parse == RExC_end)
15066 FAIL("Sequence (?#... not terminated");
15072 if (RExC_flags & RXf_PMf_EXTENDED) {
15073 char * p = regpatws(pRExC_state, RExC_parse,
15074 TRUE); /* means recognize comments */
15075 if (p != RExC_parse) {
15085 - reg_node - emit a node
15087 STATIC regnode * /* Location. */
15088 S_reg_node(pTHX_ RExC_state_t *pRExC_state, U8 op)
15091 regnode * const ret = RExC_emit;
15092 GET_RE_DEBUG_FLAGS_DECL;
15094 PERL_ARGS_ASSERT_REG_NODE;
15097 SIZE_ALIGN(RExC_size);
15101 if (RExC_emit >= RExC_emit_bound)
15102 Perl_croak(aTHX_ "panic: reg_node overrun trying to emit %d, %p>=%p",
15103 op, (void*)RExC_emit, (void*)RExC_emit_bound);
15105 NODE_ALIGN_FILL(ret);
15107 FILL_ADVANCE_NODE(ptr, op);
15108 REH_CALL_COMP_NODE_HOOK(pRExC_state->rx, (ptr) - 1);
15109 #ifdef RE_TRACK_PATTERN_OFFSETS
15110 if (RExC_offsets) { /* MJD */
15112 ("%s:%d: (op %s) %s %"UVuf" (len %"UVuf") (max %"UVuf").\n",
15113 "reg_node", __LINE__,
15115 (UV)(RExC_emit - RExC_emit_start) > RExC_offsets[0]
15116 ? "Overwriting end of array!\n" : "OK",
15117 (UV)(RExC_emit - RExC_emit_start),
15118 (UV)(RExC_parse - RExC_start),
15119 (UV)RExC_offsets[0]));
15120 Set_Node_Offset(RExC_emit, RExC_parse + (op == END));
15128 - reganode - emit a node with an argument
15130 STATIC regnode * /* Location. */
15131 S_reganode(pTHX_ RExC_state_t *pRExC_state, U8 op, U32 arg)
15134 regnode * const ret = RExC_emit;
15135 GET_RE_DEBUG_FLAGS_DECL;
15137 PERL_ARGS_ASSERT_REGANODE;
15140 SIZE_ALIGN(RExC_size);
15145 assert(2==regarglen[op]+1);
15147 Anything larger than this has to allocate the extra amount.
15148 If we changed this to be:
15150 RExC_size += (1 + regarglen[op]);
15152 then it wouldn't matter. Its not clear what side effect
15153 might come from that so its not done so far.
15158 if (RExC_emit >= RExC_emit_bound)
15159 Perl_croak(aTHX_ "panic: reg_node overrun trying to emit %d, %p>=%p",
15160 op, (void*)RExC_emit, (void*)RExC_emit_bound);
15162 NODE_ALIGN_FILL(ret);
15164 FILL_ADVANCE_NODE_ARG(ptr, op, arg);
15165 REH_CALL_COMP_NODE_HOOK(pRExC_state->rx, (ptr) - 2);
15166 #ifdef RE_TRACK_PATTERN_OFFSETS
15167 if (RExC_offsets) { /* MJD */
15169 ("%s(%d): (op %s) %s %"UVuf" <- %"UVuf" (max %"UVuf").\n",
15173 (UV)(RExC_emit - RExC_emit_start) > RExC_offsets[0] ?
15174 "Overwriting end of array!\n" : "OK",
15175 (UV)(RExC_emit - RExC_emit_start),
15176 (UV)(RExC_parse - RExC_start),
15177 (UV)RExC_offsets[0]));
15178 Set_Cur_Node_Offset;
15186 - reguni - emit (if appropriate) a Unicode character
15188 PERL_STATIC_INLINE STRLEN
15189 S_reguni(pTHX_ const RExC_state_t *pRExC_state, UV uv, char* s)
15191 PERL_ARGS_ASSERT_REGUNI;
15193 return SIZE_ONLY ? UNISKIP(uv) : (uvchr_to_utf8((U8*)s, uv) - (U8*)s);
15197 - reginsert - insert an operator in front of already-emitted operand
15199 * Means relocating the operand.
15202 S_reginsert(pTHX_ RExC_state_t *pRExC_state, U8 op, regnode *opnd, U32 depth)
15207 const int offset = regarglen[(U8)op];
15208 const int size = NODE_STEP_REGNODE + offset;
15209 GET_RE_DEBUG_FLAGS_DECL;
15211 PERL_ARGS_ASSERT_REGINSERT;
15212 PERL_UNUSED_CONTEXT;
15213 PERL_UNUSED_ARG(depth);
15214 /* (PL_regkind[(U8)op] == CURLY ? EXTRA_STEP_2ARGS : 0); */
15215 DEBUG_PARSE_FMT("inst"," - %s",PL_reg_name[op]);
15224 if (RExC_open_parens) {
15226 /*DEBUG_PARSE_FMT("inst"," - %"IVdf, (IV)RExC_npar);*/
15227 for ( paren=0 ; paren < RExC_npar ; paren++ ) {
15228 if ( RExC_open_parens[paren] >= opnd ) {
15229 /*DEBUG_PARSE_FMT("open"," - %d",size);*/
15230 RExC_open_parens[paren] += size;
15232 /*DEBUG_PARSE_FMT("open"," - %s","ok");*/
15234 if ( RExC_close_parens[paren] >= opnd ) {
15235 /*DEBUG_PARSE_FMT("close"," - %d",size);*/
15236 RExC_close_parens[paren] += size;
15238 /*DEBUG_PARSE_FMT("close"," - %s","ok");*/
15243 while (src > opnd) {
15244 StructCopy(--src, --dst, regnode);
15245 #ifdef RE_TRACK_PATTERN_OFFSETS
15246 if (RExC_offsets) { /* MJD 20010112 */
15248 ("%s(%d): (op %s) %s copy %"UVuf" -> %"UVuf" (max %"UVuf").\n",
15252 (UV)(dst - RExC_emit_start) > RExC_offsets[0]
15253 ? "Overwriting end of array!\n" : "OK",
15254 (UV)(src - RExC_emit_start),
15255 (UV)(dst - RExC_emit_start),
15256 (UV)RExC_offsets[0]));
15257 Set_Node_Offset_To_R(dst-RExC_emit_start, Node_Offset(src));
15258 Set_Node_Length_To_R(dst-RExC_emit_start, Node_Length(src));
15264 place = opnd; /* Op node, where operand used to be. */
15265 #ifdef RE_TRACK_PATTERN_OFFSETS
15266 if (RExC_offsets) { /* MJD */
15268 ("%s(%d): (op %s) %s %"UVuf" <- %"UVuf" (max %"UVuf").\n",
15272 (UV)(place - RExC_emit_start) > RExC_offsets[0]
15273 ? "Overwriting end of array!\n" : "OK",
15274 (UV)(place - RExC_emit_start),
15275 (UV)(RExC_parse - RExC_start),
15276 (UV)RExC_offsets[0]));
15277 Set_Node_Offset(place, RExC_parse);
15278 Set_Node_Length(place, 1);
15281 src = NEXTOPER(place);
15282 FILL_ADVANCE_NODE(place, op);
15283 REH_CALL_COMP_NODE_HOOK(pRExC_state->rx, (place) - 1);
15284 Zero(src, offset, regnode);
15288 - regtail - set the next-pointer at the end of a node chain of p to val.
15289 - SEE ALSO: regtail_study
15291 /* TODO: All three parms should be const */
15293 S_regtail(pTHX_ RExC_state_t *pRExC_state, regnode *p,
15294 const regnode *val,U32 depth)
15297 GET_RE_DEBUG_FLAGS_DECL;
15299 PERL_ARGS_ASSERT_REGTAIL;
15301 PERL_UNUSED_ARG(depth);
15307 /* Find last node. */
15310 regnode * const temp = regnext(scan);
15312 SV * const mysv=sv_newmortal();
15313 DEBUG_PARSE_MSG((scan==p ? "tail" : ""));
15314 regprop(RExC_rx, mysv, scan, NULL);
15315 PerlIO_printf(Perl_debug_log, "~ %s (%d) %s %s\n",
15316 SvPV_nolen_const(mysv), REG_NODE_NUM(scan),
15317 (temp == NULL ? "->" : ""),
15318 (temp == NULL ? PL_reg_name[OP(val)] : "")
15326 if (reg_off_by_arg[OP(scan)]) {
15327 ARG_SET(scan, val - scan);
15330 NEXT_OFF(scan) = val - scan;
15336 - regtail_study - set the next-pointer at the end of a node chain of p to val.
15337 - Look for optimizable sequences at the same time.
15338 - currently only looks for EXACT chains.
15340 This is experimental code. The idea is to use this routine to perform
15341 in place optimizations on branches and groups as they are constructed,
15342 with the long term intention of removing optimization from study_chunk so
15343 that it is purely analytical.
15345 Currently only used when in DEBUG mode. The macro REGTAIL_STUDY() is used
15346 to control which is which.
15349 /* TODO: All four parms should be const */
15352 S_regtail_study(pTHX_ RExC_state_t *pRExC_state, regnode *p,
15353 const regnode *val,U32 depth)
15358 #ifdef EXPERIMENTAL_INPLACESCAN
15361 GET_RE_DEBUG_FLAGS_DECL;
15363 PERL_ARGS_ASSERT_REGTAIL_STUDY;
15369 /* Find last node. */
15373 regnode * const temp = regnext(scan);
15374 #ifdef EXPERIMENTAL_INPLACESCAN
15375 if (PL_regkind[OP(scan)] == EXACT) {
15376 bool unfolded_multi_char; /* Unexamined in this routine */
15377 if (join_exact(pRExC_state, scan, &min,
15378 &unfolded_multi_char, 1, val, depth+1))
15383 switch (OP(scan)) {
15386 case EXACTFA_NO_TRIE:
15391 if( exact == PSEUDO )
15393 else if ( exact != OP(scan) )
15402 SV * const mysv=sv_newmortal();
15403 DEBUG_PARSE_MSG((scan==p ? "tsdy" : ""));
15404 regprop(RExC_rx, mysv, scan, NULL);
15405 PerlIO_printf(Perl_debug_log, "~ %s (%d) -> %s\n",
15406 SvPV_nolen_const(mysv),
15407 REG_NODE_NUM(scan),
15408 PL_reg_name[exact]);
15415 SV * const mysv_val=sv_newmortal();
15416 DEBUG_PARSE_MSG("");
15417 regprop(RExC_rx, mysv_val, val, NULL);
15418 PerlIO_printf(Perl_debug_log,
15419 "~ attach to %s (%"IVdf") offset to %"IVdf"\n",
15420 SvPV_nolen_const(mysv_val),
15421 (IV)REG_NODE_NUM(val),
15425 if (reg_off_by_arg[OP(scan)]) {
15426 ARG_SET(scan, val - scan);
15429 NEXT_OFF(scan) = val - scan;
15437 - regdump - dump a regexp onto Perl_debug_log in vaguely comprehensible form
15442 S_regdump_intflags(pTHX_ const char *lead, const U32 flags)
15447 ASSUME(REG_INTFLAGS_NAME_SIZE <= sizeof(flags)*8);
15449 for (bit=0; bit<REG_INTFLAGS_NAME_SIZE; bit++) {
15450 if (flags & (1<<bit)) {
15451 if (!set++ && lead)
15452 PerlIO_printf(Perl_debug_log, "%s",lead);
15453 PerlIO_printf(Perl_debug_log, "%s ",PL_reg_intflags_name[bit]);
15458 PerlIO_printf(Perl_debug_log, "\n");
15460 PerlIO_printf(Perl_debug_log, "%s[none-set]\n",lead);
15465 S_regdump_extflags(pTHX_ const char *lead, const U32 flags)
15471 ASSUME(REG_EXTFLAGS_NAME_SIZE <= sizeof(flags)*8);
15473 for (bit=0; bit<REG_EXTFLAGS_NAME_SIZE; bit++) {
15474 if (flags & (1<<bit)) {
15475 if ((1<<bit) & RXf_PMf_CHARSET) { /* Output separately, below */
15478 if (!set++ && lead)
15479 PerlIO_printf(Perl_debug_log, "%s",lead);
15480 PerlIO_printf(Perl_debug_log, "%s ",PL_reg_extflags_name[bit]);
15483 if ((cs = get_regex_charset(flags)) != REGEX_DEPENDS_CHARSET) {
15484 if (!set++ && lead) {
15485 PerlIO_printf(Perl_debug_log, "%s",lead);
15488 case REGEX_UNICODE_CHARSET:
15489 PerlIO_printf(Perl_debug_log, "UNICODE");
15491 case REGEX_LOCALE_CHARSET:
15492 PerlIO_printf(Perl_debug_log, "LOCALE");
15494 case REGEX_ASCII_RESTRICTED_CHARSET:
15495 PerlIO_printf(Perl_debug_log, "ASCII-RESTRICTED");
15497 case REGEX_ASCII_MORE_RESTRICTED_CHARSET:
15498 PerlIO_printf(Perl_debug_log, "ASCII-MORE_RESTRICTED");
15501 PerlIO_printf(Perl_debug_log, "UNKNOWN CHARACTER SET");
15507 PerlIO_printf(Perl_debug_log, "\n");
15509 PerlIO_printf(Perl_debug_log, "%s[none-set]\n",lead);
15515 Perl_regdump(pTHX_ const regexp *r)
15519 SV * const sv = sv_newmortal();
15520 SV *dsv= sv_newmortal();
15521 RXi_GET_DECL(r,ri);
15522 GET_RE_DEBUG_FLAGS_DECL;
15524 PERL_ARGS_ASSERT_REGDUMP;
15526 (void)dumpuntil(r, ri->program, ri->program + 1, NULL, NULL, sv, 0, 0);
15528 /* Header fields of interest. */
15529 if (r->anchored_substr) {
15530 RE_PV_QUOTED_DECL(s, 0, dsv, SvPVX_const(r->anchored_substr),
15531 RE_SV_DUMPLEN(r->anchored_substr), 30);
15532 PerlIO_printf(Perl_debug_log,
15533 "anchored %s%s at %"IVdf" ",
15534 s, RE_SV_TAIL(r->anchored_substr),
15535 (IV)r->anchored_offset);
15536 } else if (r->anchored_utf8) {
15537 RE_PV_QUOTED_DECL(s, 1, dsv, SvPVX_const(r->anchored_utf8),
15538 RE_SV_DUMPLEN(r->anchored_utf8), 30);
15539 PerlIO_printf(Perl_debug_log,
15540 "anchored utf8 %s%s at %"IVdf" ",
15541 s, RE_SV_TAIL(r->anchored_utf8),
15542 (IV)r->anchored_offset);
15544 if (r->float_substr) {
15545 RE_PV_QUOTED_DECL(s, 0, dsv, SvPVX_const(r->float_substr),
15546 RE_SV_DUMPLEN(r->float_substr), 30);
15547 PerlIO_printf(Perl_debug_log,
15548 "floating %s%s at %"IVdf"..%"UVuf" ",
15549 s, RE_SV_TAIL(r->float_substr),
15550 (IV)r->float_min_offset, (UV)r->float_max_offset);
15551 } else if (r->float_utf8) {
15552 RE_PV_QUOTED_DECL(s, 1, dsv, SvPVX_const(r->float_utf8),
15553 RE_SV_DUMPLEN(r->float_utf8), 30);
15554 PerlIO_printf(Perl_debug_log,
15555 "floating utf8 %s%s at %"IVdf"..%"UVuf" ",
15556 s, RE_SV_TAIL(r->float_utf8),
15557 (IV)r->float_min_offset, (UV)r->float_max_offset);
15559 if (r->check_substr || r->check_utf8)
15560 PerlIO_printf(Perl_debug_log,
15562 (r->check_substr == r->float_substr
15563 && r->check_utf8 == r->float_utf8
15564 ? "(checking floating" : "(checking anchored"));
15565 if (r->intflags & PREGf_NOSCAN)
15566 PerlIO_printf(Perl_debug_log, " noscan");
15567 if (r->extflags & RXf_CHECK_ALL)
15568 PerlIO_printf(Perl_debug_log, " isall");
15569 if (r->check_substr || r->check_utf8)
15570 PerlIO_printf(Perl_debug_log, ") ");
15572 if (ri->regstclass) {
15573 regprop(r, sv, ri->regstclass, NULL);
15574 PerlIO_printf(Perl_debug_log, "stclass %s ", SvPVX_const(sv));
15576 if (r->intflags & PREGf_ANCH) {
15577 PerlIO_printf(Perl_debug_log, "anchored");
15578 if (r->intflags & PREGf_ANCH_BOL)
15579 PerlIO_printf(Perl_debug_log, "(BOL)");
15580 if (r->intflags & PREGf_ANCH_MBOL)
15581 PerlIO_printf(Perl_debug_log, "(MBOL)");
15582 if (r->intflags & PREGf_ANCH_SBOL)
15583 PerlIO_printf(Perl_debug_log, "(SBOL)");
15584 if (r->intflags & PREGf_ANCH_GPOS)
15585 PerlIO_printf(Perl_debug_log, "(GPOS)");
15586 PerlIO_putc(Perl_debug_log, ' ');
15588 if (r->intflags & PREGf_GPOS_SEEN)
15589 PerlIO_printf(Perl_debug_log, "GPOS:%"UVuf" ", (UV)r->gofs);
15590 if (r->intflags & PREGf_SKIP)
15591 PerlIO_printf(Perl_debug_log, "plus ");
15592 if (r->intflags & PREGf_IMPLICIT)
15593 PerlIO_printf(Perl_debug_log, "implicit ");
15594 PerlIO_printf(Perl_debug_log, "minlen %"IVdf" ", (IV)r->minlen);
15595 if (r->extflags & RXf_EVAL_SEEN)
15596 PerlIO_printf(Perl_debug_log, "with eval ");
15597 PerlIO_printf(Perl_debug_log, "\n");
15599 regdump_extflags("r->extflags: ",r->extflags);
15600 regdump_intflags("r->intflags: ",r->intflags);
15603 PERL_ARGS_ASSERT_REGDUMP;
15604 PERL_UNUSED_CONTEXT;
15605 PERL_UNUSED_ARG(r);
15606 #endif /* DEBUGGING */
15610 - regprop - printable representation of opcode, with run time support
15614 Perl_regprop(pTHX_ const regexp *prog, SV *sv, const regnode *o, const regmatch_info *reginfo)
15620 /* Should be synchronized with * ANYOF_ #xdefines in regcomp.h */
15621 static const char * const anyofs[] = {
15622 #if _CC_WORDCHAR != 0 || _CC_DIGIT != 1 || _CC_ALPHA != 2 || _CC_LOWER != 3 \
15623 || _CC_UPPER != 4 || _CC_PUNCT != 5 || _CC_PRINT != 6 \
15624 || _CC_ALPHANUMERIC != 7 || _CC_GRAPH != 8 || _CC_CASED != 9 \
15625 || _CC_SPACE != 10 || _CC_BLANK != 11 || _CC_XDIGIT != 12 \
15626 || _CC_PSXSPC != 13 || _CC_CNTRL != 14 || _CC_ASCII != 15 \
15627 || _CC_VERTSPACE != 16
15628 #error Need to adjust order of anyofs[]
15665 RXi_GET_DECL(prog,progi);
15666 GET_RE_DEBUG_FLAGS_DECL;
15668 PERL_ARGS_ASSERT_REGPROP;
15672 if (OP(o) > REGNODE_MAX) /* regnode.type is unsigned */
15673 /* It would be nice to FAIL() here, but this may be called from
15674 regexec.c, and it would be hard to supply pRExC_state. */
15675 Perl_croak(aTHX_ "Corrupted regexp opcode %d > %d",
15676 (int)OP(o), (int)REGNODE_MAX);
15677 sv_catpv(sv, PL_reg_name[OP(o)]); /* Take off const! */
15679 k = PL_regkind[OP(o)];
15682 sv_catpvs(sv, " ");
15683 /* Using is_utf8_string() (via PERL_PV_UNI_DETECT)
15684 * is a crude hack but it may be the best for now since
15685 * we have no flag "this EXACTish node was UTF-8"
15687 pv_pretty(sv, STRING(o), STR_LEN(o), 60, PL_colors[0], PL_colors[1],
15688 PERL_PV_ESCAPE_UNI_DETECT |
15689 PERL_PV_ESCAPE_NONASCII |
15690 PERL_PV_PRETTY_ELLIPSES |
15691 PERL_PV_PRETTY_LTGT |
15692 PERL_PV_PRETTY_NOCLEAR
15694 } else if (k == TRIE) {
15695 /* print the details of the trie in dumpuntil instead, as
15696 * progi->data isn't available here */
15697 const char op = OP(o);
15698 const U32 n = ARG(o);
15699 const reg_ac_data * const ac = IS_TRIE_AC(op) ?
15700 (reg_ac_data *)progi->data->data[n] :
15702 const reg_trie_data * const trie
15703 = (reg_trie_data*)progi->data->data[!IS_TRIE_AC(op) ? n : ac->trie];
15705 Perl_sv_catpvf(aTHX_ sv, "-%s",PL_reg_name[o->flags]);
15706 DEBUG_TRIE_COMPILE_r(
15707 Perl_sv_catpvf(aTHX_ sv,
15708 "<S:%"UVuf"/%"IVdf" W:%"UVuf" L:%"UVuf"/%"UVuf" C:%"UVuf"/%"UVuf">",
15709 (UV)trie->startstate,
15710 (IV)trie->statecount-1, /* -1 because of the unused 0 element */
15711 (UV)trie->wordcount,
15714 (UV)TRIE_CHARCOUNT(trie),
15715 (UV)trie->uniquecharcount
15718 if ( IS_ANYOF_TRIE(op) || trie->bitmap ) {
15719 sv_catpvs(sv, "[");
15720 (void) put_latin1_charclass_innards(sv, IS_ANYOF_TRIE(op)
15722 : TRIE_BITMAP(trie));
15723 sv_catpvs(sv, "]");
15726 } else if (k == CURLY) {
15727 if (OP(o) == CURLYM || OP(o) == CURLYN || OP(o) == CURLYX)
15728 Perl_sv_catpvf(aTHX_ sv, "[%d]", o->flags); /* Parenth number */
15729 Perl_sv_catpvf(aTHX_ sv, " {%d,%d}", ARG1(o), ARG2(o));
15731 else if (k == WHILEM && o->flags) /* Ordinal/of */
15732 Perl_sv_catpvf(aTHX_ sv, "[%d/%d]", o->flags & 0xf, o->flags>>4);
15733 else if (k == REF || k == OPEN || k == CLOSE
15734 || k == GROUPP || OP(o)==ACCEPT)
15736 Perl_sv_catpvf(aTHX_ sv, "%d", (int)ARG(o)); /* Parenth number */
15737 if ( RXp_PAREN_NAMES(prog) ) {
15738 if ( k != REF || (OP(o) < NREF)) {
15739 AV *list= MUTABLE_AV(progi->data->data[progi->name_list_idx]);
15740 SV **name= av_fetch(list, ARG(o), 0 );
15742 Perl_sv_catpvf(aTHX_ sv, " '%"SVf"'", SVfARG(*name));
15745 AV *list= MUTABLE_AV(progi->data->data[ progi->name_list_idx ]);
15746 SV *sv_dat= MUTABLE_SV(progi->data->data[ ARG( o ) ]);
15747 I32 *nums=(I32*)SvPVX(sv_dat);
15748 SV **name= av_fetch(list, nums[0], 0 );
15751 for ( n=0; n<SvIVX(sv_dat); n++ ) {
15752 Perl_sv_catpvf(aTHX_ sv, "%s%"IVdf,
15753 (n ? "," : ""), (IV)nums[n]);
15755 Perl_sv_catpvf(aTHX_ sv, " '%"SVf"'", SVfARG(*name));
15759 if ( k == REF && reginfo) {
15760 U32 n = ARG(o); /* which paren pair */
15761 I32 ln = prog->offs[n].start;
15762 if (prog->lastparen < n || ln == -1)
15763 Perl_sv_catpvf(aTHX_ sv, ": FAIL");
15764 else if (ln == prog->offs[n].end)
15765 Perl_sv_catpvf(aTHX_ sv, ": ACCEPT - EMPTY STRING");
15767 const char *s = reginfo->strbeg + ln;
15768 Perl_sv_catpvf(aTHX_ sv, ": ");
15769 Perl_pv_pretty( aTHX_ sv, s, prog->offs[n].end - prog->offs[n].start, 32, 0, 0,
15770 PERL_PV_ESCAPE_UNI_DETECT|PERL_PV_PRETTY_NOCLEAR|PERL_PV_PRETTY_ELLIPSES|PERL_PV_PRETTY_QUOTE );
15773 } else if (k == GOSUB)
15774 /* Paren and offset */
15775 Perl_sv_catpvf(aTHX_ sv, "%d[%+d]", (int)ARG(o),(int)ARG2L(o));
15776 else if (k == VERB) {
15778 Perl_sv_catpvf(aTHX_ sv, ":%"SVf,
15779 SVfARG((MUTABLE_SV(progi->data->data[ ARG( o ) ]))));
15780 } else if (k == LOGICAL)
15781 /* 2: embedded, otherwise 1 */
15782 Perl_sv_catpvf(aTHX_ sv, "[%d]", o->flags);
15783 else if (k == ANYOF) {
15784 const U8 flags = ANYOF_FLAGS(o);
15788 if (flags & ANYOF_LOCALE_FLAGS)
15789 sv_catpvs(sv, "{loc}");
15790 if (flags & ANYOF_LOC_FOLD)
15791 sv_catpvs(sv, "{i}");
15792 Perl_sv_catpvf(aTHX_ sv, "[%s", PL_colors[0]);
15793 if (flags & ANYOF_INVERT)
15794 sv_catpvs(sv, "^");
15796 /* output what the standard cp 0-255 bitmap matches */
15797 do_sep = put_latin1_charclass_innards(sv, ANYOF_BITMAP(o));
15799 /* output any special charclass tests (used entirely under use
15801 if (ANYOF_POSIXL_TEST_ANY_SET(o)) {
15803 for (i = 0; i < ANYOF_POSIXL_MAX; i++) {
15804 if (ANYOF_POSIXL_TEST(o,i)) {
15805 sv_catpv(sv, anyofs[i]);
15811 if ((flags & (ANYOF_ABOVE_LATIN1_ALL
15813 |ANYOF_NONBITMAP_NON_UTF8
15817 Perl_sv_catpvf(aTHX_ sv,"%s][%s",PL_colors[1],PL_colors[0]);
15818 if (flags & ANYOF_INVERT)
15819 /*make sure the invert info is in each */
15820 sv_catpvs(sv, "^");
15823 if (flags & ANYOF_NON_UTF8_NON_ASCII_ALL) {
15824 sv_catpvs(sv, "{non-utf8-latin1-all}");
15827 /* output information about the unicode matching */
15828 if (flags & ANYOF_ABOVE_LATIN1_ALL)
15829 sv_catpvs(sv, "{unicode_all}");
15830 else if (ARG(o) != ANYOF_NONBITMAP_EMPTY) {
15831 SV *lv; /* Set if there is something outside the bit map. */
15832 bool byte_output = FALSE; /* If something in the bitmap has
15834 SV *only_utf8_locale;
15836 /* Get the stuff that wasn't in the bitmap */
15837 (void) _get_regclass_nonbitmap_data(prog, o, FALSE,
15838 &lv, &only_utf8_locale);
15839 if (lv && lv != &PL_sv_undef) {
15840 char *s = savesvpv(lv);
15841 char * const origs = s;
15843 while (*s && *s != '\n')
15847 const char * const t = ++s;
15849 if (flags & ANYOF_NONBITMAP_NON_UTF8) {
15850 sv_catpvs(sv, "{outside bitmap}");
15853 sv_catpvs(sv, "{utf8}");
15857 sv_catpvs(sv, " ");
15863 /* Truncate very long output */
15864 if (s - origs > 256) {
15865 Perl_sv_catpvf(aTHX_ sv,
15867 (int) (s - origs - 1),
15873 else if (*s == '\t') {
15887 SvREFCNT_dec_NN(lv);
15890 if ((flags & ANYOF_LOC_FOLD)
15891 && only_utf8_locale
15892 && only_utf8_locale != &PL_sv_undef)
15895 int max_entries = 256;
15897 sv_catpvs(sv, "{utf8 locale}");
15898 invlist_iterinit(only_utf8_locale);
15899 while (invlist_iternext(only_utf8_locale,
15901 put_range(sv, start, end);
15903 if (max_entries < 0) {
15904 sv_catpvs(sv, "...");
15908 invlist_iterfinish(only_utf8_locale);
15913 Perl_sv_catpvf(aTHX_ sv, "%s]", PL_colors[1]);
15915 else if (k == POSIXD || k == NPOSIXD) {
15916 U8 index = FLAGS(o) * 2;
15917 if (index < C_ARRAY_LENGTH(anyofs)) {
15918 if (*anyofs[index] != '[') {
15921 sv_catpv(sv, anyofs[index]);
15922 if (*anyofs[index] != '[') {
15927 Perl_sv_catpvf(aTHX_ sv, "[illegal type=%d])", index);
15930 else if (k == BRANCHJ && (OP(o) == UNLESSM || OP(o) == IFMATCH))
15931 Perl_sv_catpvf(aTHX_ sv, "[%d]", -(o->flags));
15933 PERL_UNUSED_CONTEXT;
15934 PERL_UNUSED_ARG(sv);
15935 PERL_UNUSED_ARG(o);
15936 PERL_UNUSED_ARG(prog);
15937 PERL_UNUSED_ARG(reginfo);
15938 #endif /* DEBUGGING */
15944 Perl_re_intuit_string(pTHX_ REGEXP * const r)
15945 { /* Assume that RE_INTUIT is set */
15946 struct regexp *const prog = ReANY(r);
15947 GET_RE_DEBUG_FLAGS_DECL;
15949 PERL_ARGS_ASSERT_RE_INTUIT_STRING;
15950 PERL_UNUSED_CONTEXT;
15954 const char * const s = SvPV_nolen_const(prog->check_substr
15955 ? prog->check_substr : prog->check_utf8);
15957 if (!PL_colorset) reginitcolors();
15958 PerlIO_printf(Perl_debug_log,
15959 "%sUsing REx %ssubstr:%s \"%s%.60s%s%s\"\n",
15961 prog->check_substr ? "" : "utf8 ",
15962 PL_colors[5],PL_colors[0],
15965 (strlen(s) > 60 ? "..." : ""));
15968 return prog->check_substr ? prog->check_substr : prog->check_utf8;
15974 handles refcounting and freeing the perl core regexp structure. When
15975 it is necessary to actually free the structure the first thing it
15976 does is call the 'free' method of the regexp_engine associated to
15977 the regexp, allowing the handling of the void *pprivate; member
15978 first. (This routine is not overridable by extensions, which is why
15979 the extensions free is called first.)
15981 See regdupe and regdupe_internal if you change anything here.
15983 #ifndef PERL_IN_XSUB_RE
15985 Perl_pregfree(pTHX_ REGEXP *r)
15991 Perl_pregfree2(pTHX_ REGEXP *rx)
15993 struct regexp *const r = ReANY(rx);
15994 GET_RE_DEBUG_FLAGS_DECL;
15996 PERL_ARGS_ASSERT_PREGFREE2;
15998 if (r->mother_re) {
15999 ReREFCNT_dec(r->mother_re);
16001 CALLREGFREE_PVT(rx); /* free the private data */
16002 SvREFCNT_dec(RXp_PAREN_NAMES(r));
16003 Safefree(r->xpv_len_u.xpvlenu_pv);
16006 SvREFCNT_dec(r->anchored_substr);
16007 SvREFCNT_dec(r->anchored_utf8);
16008 SvREFCNT_dec(r->float_substr);
16009 SvREFCNT_dec(r->float_utf8);
16010 Safefree(r->substrs);
16012 RX_MATCH_COPY_FREE(rx);
16013 #ifdef PERL_ANY_COW
16014 SvREFCNT_dec(r->saved_copy);
16017 SvREFCNT_dec(r->qr_anoncv);
16018 rx->sv_u.svu_rx = 0;
16023 This is a hacky workaround to the structural issue of match results
16024 being stored in the regexp structure which is in turn stored in
16025 PL_curpm/PL_reg_curpm. The problem is that due to qr// the pattern
16026 could be PL_curpm in multiple contexts, and could require multiple
16027 result sets being associated with the pattern simultaneously, such
16028 as when doing a recursive match with (??{$qr})
16030 The solution is to make a lightweight copy of the regexp structure
16031 when a qr// is returned from the code executed by (??{$qr}) this
16032 lightweight copy doesn't actually own any of its data except for
16033 the starp/end and the actual regexp structure itself.
16039 Perl_reg_temp_copy (pTHX_ REGEXP *ret_x, REGEXP *rx)
16041 struct regexp *ret;
16042 struct regexp *const r = ReANY(rx);
16043 const bool islv = ret_x && SvTYPE(ret_x) == SVt_PVLV;
16045 PERL_ARGS_ASSERT_REG_TEMP_COPY;
16048 ret_x = (REGEXP*) newSV_type(SVt_REGEXP);
16050 SvOK_off((SV *)ret_x);
16052 /* For PVLVs, SvANY points to the xpvlv body while sv_u points
16053 to the regexp. (For SVt_REGEXPs, sv_upgrade has already
16054 made both spots point to the same regexp body.) */
16055 REGEXP *temp = (REGEXP *)newSV_type(SVt_REGEXP);
16056 assert(!SvPVX(ret_x));
16057 ret_x->sv_u.svu_rx = temp->sv_any;
16058 temp->sv_any = NULL;
16059 SvFLAGS(temp) = (SvFLAGS(temp) & ~SVTYPEMASK) | SVt_NULL;
16060 SvREFCNT_dec_NN(temp);
16061 /* SvCUR still resides in the xpvlv struct, so the regexp copy-
16062 ing below will not set it. */
16063 SvCUR_set(ret_x, SvCUR(rx));
16066 /* This ensures that SvTHINKFIRST(sv) is true, and hence that
16067 sv_force_normal(sv) is called. */
16069 ret = ReANY(ret_x);
16071 SvFLAGS(ret_x) |= SvUTF8(rx);
16072 /* We share the same string buffer as the original regexp, on which we
16073 hold a reference count, incremented when mother_re is set below.
16074 The string pointer is copied here, being part of the regexp struct.
16076 memcpy(&(ret->xpv_cur), &(r->xpv_cur),
16077 sizeof(regexp) - STRUCT_OFFSET(regexp, xpv_cur));
16079 const I32 npar = r->nparens+1;
16080 Newx(ret->offs, npar, regexp_paren_pair);
16081 Copy(r->offs, ret->offs, npar, regexp_paren_pair);
16084 Newx(ret->substrs, 1, struct reg_substr_data);
16085 StructCopy(r->substrs, ret->substrs, struct reg_substr_data);
16087 SvREFCNT_inc_void(ret->anchored_substr);
16088 SvREFCNT_inc_void(ret->anchored_utf8);
16089 SvREFCNT_inc_void(ret->float_substr);
16090 SvREFCNT_inc_void(ret->float_utf8);
16092 /* check_substr and check_utf8, if non-NULL, point to either their
16093 anchored or float namesakes, and don't hold a second reference. */
16095 RX_MATCH_COPIED_off(ret_x);
16096 #ifdef PERL_ANY_COW
16097 ret->saved_copy = NULL;
16099 ret->mother_re = ReREFCNT_inc(r->mother_re ? r->mother_re : rx);
16100 SvREFCNT_inc_void(ret->qr_anoncv);
16106 /* regfree_internal()
16108 Free the private data in a regexp. This is overloadable by
16109 extensions. Perl takes care of the regexp structure in pregfree(),
16110 this covers the *pprivate pointer which technically perl doesn't
16111 know about, however of course we have to handle the
16112 regexp_internal structure when no extension is in use.
16114 Note this is called before freeing anything in the regexp
16119 Perl_regfree_internal(pTHX_ REGEXP * const rx)
16121 struct regexp *const r = ReANY(rx);
16122 RXi_GET_DECL(r,ri);
16123 GET_RE_DEBUG_FLAGS_DECL;
16125 PERL_ARGS_ASSERT_REGFREE_INTERNAL;
16131 SV *dsv= sv_newmortal();
16132 RE_PV_QUOTED_DECL(s, RX_UTF8(rx),
16133 dsv, RX_PRECOMP(rx), RX_PRELEN(rx), 60);
16134 PerlIO_printf(Perl_debug_log,"%sFreeing REx:%s %s\n",
16135 PL_colors[4],PL_colors[5],s);
16138 #ifdef RE_TRACK_PATTERN_OFFSETS
16140 Safefree(ri->u.offsets); /* 20010421 MJD */
16142 if (ri->code_blocks) {
16144 for (n = 0; n < ri->num_code_blocks; n++)
16145 SvREFCNT_dec(ri->code_blocks[n].src_regex);
16146 Safefree(ri->code_blocks);
16150 int n = ri->data->count;
16153 /* If you add a ->what type here, update the comment in regcomp.h */
16154 switch (ri->data->what[n]) {
16160 SvREFCNT_dec(MUTABLE_SV(ri->data->data[n]));
16163 Safefree(ri->data->data[n]);
16169 { /* Aho Corasick add-on structure for a trie node.
16170 Used in stclass optimization only */
16172 reg_ac_data *aho=(reg_ac_data*)ri->data->data[n];
16173 #ifdef USE_ITHREADS
16177 refcount = --aho->refcount;
16180 PerlMemShared_free(aho->states);
16181 PerlMemShared_free(aho->fail);
16182 /* do this last!!!! */
16183 PerlMemShared_free(ri->data->data[n]);
16184 /* we should only ever get called once, so
16185 * assert as much, and also guard the free
16186 * which /might/ happen twice. At the least
16187 * it will make code anlyzers happy and it
16188 * doesn't cost much. - Yves */
16189 assert(ri->regstclass);
16190 if (ri->regstclass) {
16191 PerlMemShared_free(ri->regstclass);
16192 ri->regstclass = 0;
16199 /* trie structure. */
16201 reg_trie_data *trie=(reg_trie_data*)ri->data->data[n];
16202 #ifdef USE_ITHREADS
16206 refcount = --trie->refcount;
16209 PerlMemShared_free(trie->charmap);
16210 PerlMemShared_free(trie->states);
16211 PerlMemShared_free(trie->trans);
16213 PerlMemShared_free(trie->bitmap);
16215 PerlMemShared_free(trie->jump);
16216 PerlMemShared_free(trie->wordinfo);
16217 /* do this last!!!! */
16218 PerlMemShared_free(ri->data->data[n]);
16223 Perl_croak(aTHX_ "panic: regfree data code '%c'",
16224 ri->data->what[n]);
16227 Safefree(ri->data->what);
16228 Safefree(ri->data);
16234 #define av_dup_inc(s,t) MUTABLE_AV(sv_dup_inc((const SV *)s,t))
16235 #define hv_dup_inc(s,t) MUTABLE_HV(sv_dup_inc((const SV *)s,t))
16236 #define SAVEPVN(p,n) ((p) ? savepvn(p,n) : NULL)
16239 re_dup - duplicate a regexp.
16241 This routine is expected to clone a given regexp structure. It is only
16242 compiled under USE_ITHREADS.
16244 After all of the core data stored in struct regexp is duplicated
16245 the regexp_engine.dupe method is used to copy any private data
16246 stored in the *pprivate pointer. This allows extensions to handle
16247 any duplication it needs to do.
16249 See pregfree() and regfree_internal() if you change anything here.
16251 #if defined(USE_ITHREADS)
16252 #ifndef PERL_IN_XSUB_RE
16254 Perl_re_dup_guts(pTHX_ const REGEXP *sstr, REGEXP *dstr, CLONE_PARAMS *param)
16258 const struct regexp *r = ReANY(sstr);
16259 struct regexp *ret = ReANY(dstr);
16261 PERL_ARGS_ASSERT_RE_DUP_GUTS;
16263 npar = r->nparens+1;
16264 Newx(ret->offs, npar, regexp_paren_pair);
16265 Copy(r->offs, ret->offs, npar, regexp_paren_pair);
16267 if (ret->substrs) {
16268 /* Do it this way to avoid reading from *r after the StructCopy().
16269 That way, if any of the sv_dup_inc()s dislodge *r from the L1
16270 cache, it doesn't matter. */
16271 const bool anchored = r->check_substr
16272 ? r->check_substr == r->anchored_substr
16273 : r->check_utf8 == r->anchored_utf8;
16274 Newx(ret->substrs, 1, struct reg_substr_data);
16275 StructCopy(r->substrs, ret->substrs, struct reg_substr_data);
16277 ret->anchored_substr = sv_dup_inc(ret->anchored_substr, param);
16278 ret->anchored_utf8 = sv_dup_inc(ret->anchored_utf8, param);
16279 ret->float_substr = sv_dup_inc(ret->float_substr, param);
16280 ret->float_utf8 = sv_dup_inc(ret->float_utf8, param);
16282 /* check_substr and check_utf8, if non-NULL, point to either their
16283 anchored or float namesakes, and don't hold a second reference. */
16285 if (ret->check_substr) {
16287 assert(r->check_utf8 == r->anchored_utf8);
16288 ret->check_substr = ret->anchored_substr;
16289 ret->check_utf8 = ret->anchored_utf8;
16291 assert(r->check_substr == r->float_substr);
16292 assert(r->check_utf8 == r->float_utf8);
16293 ret->check_substr = ret->float_substr;
16294 ret->check_utf8 = ret->float_utf8;
16296 } else if (ret->check_utf8) {
16298 ret->check_utf8 = ret->anchored_utf8;
16300 ret->check_utf8 = ret->float_utf8;
16305 RXp_PAREN_NAMES(ret) = hv_dup_inc(RXp_PAREN_NAMES(ret), param);
16306 ret->qr_anoncv = MUTABLE_CV(sv_dup_inc((const SV *)ret->qr_anoncv, param));
16309 RXi_SET(ret,CALLREGDUPE_PVT(dstr,param));
16311 if (RX_MATCH_COPIED(dstr))
16312 ret->subbeg = SAVEPVN(ret->subbeg, ret->sublen);
16314 ret->subbeg = NULL;
16315 #ifdef PERL_ANY_COW
16316 ret->saved_copy = NULL;
16319 /* Whether mother_re be set or no, we need to copy the string. We
16320 cannot refrain from copying it when the storage points directly to
16321 our mother regexp, because that's
16322 1: a buffer in a different thread
16323 2: something we no longer hold a reference on
16324 so we need to copy it locally. */
16325 RX_WRAPPED(dstr) = SAVEPVN(RX_WRAPPED(sstr), SvCUR(sstr)+1);
16326 ret->mother_re = NULL;
16328 #endif /* PERL_IN_XSUB_RE */
16333 This is the internal complement to regdupe() which is used to copy
16334 the structure pointed to by the *pprivate pointer in the regexp.
16335 This is the core version of the extension overridable cloning hook.
16336 The regexp structure being duplicated will be copied by perl prior
16337 to this and will be provided as the regexp *r argument, however
16338 with the /old/ structures pprivate pointer value. Thus this routine
16339 may override any copying normally done by perl.
16341 It returns a pointer to the new regexp_internal structure.
16345 Perl_regdupe_internal(pTHX_ REGEXP * const rx, CLONE_PARAMS *param)
16348 struct regexp *const r = ReANY(rx);
16349 regexp_internal *reti;
16351 RXi_GET_DECL(r,ri);
16353 PERL_ARGS_ASSERT_REGDUPE_INTERNAL;
16357 Newxc(reti, sizeof(regexp_internal) + len*sizeof(regnode),
16358 char, regexp_internal);
16359 Copy(ri->program, reti->program, len+1, regnode);
16361 reti->num_code_blocks = ri->num_code_blocks;
16362 if (ri->code_blocks) {
16364 Newxc(reti->code_blocks, ri->num_code_blocks, struct reg_code_block,
16365 struct reg_code_block);
16366 Copy(ri->code_blocks, reti->code_blocks, ri->num_code_blocks,
16367 struct reg_code_block);
16368 for (n = 0; n < ri->num_code_blocks; n++)
16369 reti->code_blocks[n].src_regex = (REGEXP*)
16370 sv_dup_inc((SV*)(ri->code_blocks[n].src_regex), param);
16373 reti->code_blocks = NULL;
16375 reti->regstclass = NULL;
16378 struct reg_data *d;
16379 const int count = ri->data->count;
16382 Newxc(d, sizeof(struct reg_data) + count*sizeof(void *),
16383 char, struct reg_data);
16384 Newx(d->what, count, U8);
16387 for (i = 0; i < count; i++) {
16388 d->what[i] = ri->data->what[i];
16389 switch (d->what[i]) {
16390 /* see also regcomp.h and regfree_internal() */
16391 case 'a': /* actually an AV, but the dup function is identical. */
16395 case 'u': /* actually an HV, but the dup function is identical. */
16396 d->data[i] = sv_dup_inc((const SV *)ri->data->data[i], param);
16399 /* This is cheating. */
16400 Newx(d->data[i], 1, regnode_ssc);
16401 StructCopy(ri->data->data[i], d->data[i], regnode_ssc);
16402 reti->regstclass = (regnode*)d->data[i];
16405 /* Trie stclasses are readonly and can thus be shared
16406 * without duplication. We free the stclass in pregfree
16407 * when the corresponding reg_ac_data struct is freed.
16409 reti->regstclass= ri->regstclass;
16413 ((reg_trie_data*)ri->data->data[i])->refcount++;
16418 d->data[i] = ri->data->data[i];
16421 Perl_croak(aTHX_ "panic: re_dup unknown data code '%c'",
16422 ri->data->what[i]);
16431 reti->name_list_idx = ri->name_list_idx;
16433 #ifdef RE_TRACK_PATTERN_OFFSETS
16434 if (ri->u.offsets) {
16435 Newx(reti->u.offsets, 2*len+1, U32);
16436 Copy(ri->u.offsets, reti->u.offsets, 2*len+1, U32);
16439 SetProgLen(reti,len);
16442 return (void*)reti;
16445 #endif /* USE_ITHREADS */
16447 #ifndef PERL_IN_XSUB_RE
16450 - regnext - dig the "next" pointer out of a node
16453 Perl_regnext(pTHX_ regnode *p)
16460 if (OP(p) > REGNODE_MAX) { /* regnode.type is unsigned */
16461 Perl_croak(aTHX_ "Corrupted regexp opcode %d > %d",
16462 (int)OP(p), (int)REGNODE_MAX);
16465 offset = (reg_off_by_arg[OP(p)] ? ARG(p) : NEXT_OFF(p));
16474 S_re_croak2(pTHX_ bool utf8, const char* pat1,const char* pat2,...)
16477 STRLEN l1 = strlen(pat1);
16478 STRLEN l2 = strlen(pat2);
16481 const char *message;
16483 PERL_ARGS_ASSERT_RE_CROAK2;
16489 Copy(pat1, buf, l1 , char);
16490 Copy(pat2, buf + l1, l2 , char);
16491 buf[l1 + l2] = '\n';
16492 buf[l1 + l2 + 1] = '\0';
16493 va_start(args, pat2);
16494 msv = vmess(buf, &args);
16496 message = SvPV_const(msv,l1);
16499 Copy(message, buf, l1 , char);
16500 /* l1-1 to avoid \n */
16501 Perl_croak(aTHX_ "%"UTF8f, UTF8fARG(utf8, l1-1, buf));
16504 /* XXX Here's a total kludge. But we need to re-enter for swash routines. */
16506 #ifndef PERL_IN_XSUB_RE
16508 Perl_save_re_context(pTHX)
16510 /* Save $1..$n (#18107: UTF-8 s/(\w+)/uc($1)/e); AMS 20021106. */
16512 const REGEXP * const rx = PM_GETRE(PL_curpm);
16515 for (i = 1; i <= RX_NPARENS(rx); i++) {
16516 char digits[TYPE_CHARS(long)];
16517 const STRLEN len = my_snprintf(digits, sizeof(digits),
16519 GV *const *const gvp
16520 = (GV**)hv_fetch(PL_defstash, digits, len, 0);
16523 GV * const gv = *gvp;
16524 if (SvTYPE(gv) == SVt_PVGV && GvSV(gv))
16536 S_put_byte(pTHX_ SV *sv, int c)
16538 PERL_ARGS_ASSERT_PUT_BYTE;
16542 case '\r': Perl_sv_catpvf(aTHX_ sv, "\\r"); break;
16543 case '\n': Perl_sv_catpvf(aTHX_ sv, "\\n"); break;
16544 case '\t': Perl_sv_catpvf(aTHX_ sv, "\\t"); break;
16545 case '\f': Perl_sv_catpvf(aTHX_ sv, "\\f"); break;
16546 case '\a': Perl_sv_catpvf(aTHX_ sv, "\\a"); break;
16549 Perl_sv_catpvf(aTHX_ sv, "\\x{%x}", c);
16554 const char string = c;
16555 if (c == '-' || c == ']' || c == '\\' || c == '^')
16556 sv_catpvs(sv, "\\");
16557 sv_catpvn(sv, &string, 1);
16562 S_put_range(pTHX_ SV *sv, UV start, UV end)
16565 /* Appends to 'sv' a displayable version of the range of code points from
16566 * 'start' to 'end'. It assumes that only ASCII printables are displayable
16567 * as-is (though some of these will be escaped by put_byte()). For the
16568 * time being, this subroutine only works for latin1 (< 256) code points */
16570 assert(start <= end);
16572 PERL_ARGS_ASSERT_PUT_RANGE;
16574 while (start <= end) {
16575 if (end - start < 3) { /* Individual chars in short ranges */
16576 for (; start <= end; start++) {
16577 put_byte(sv, start);
16582 /* For small ranges that include printable ASCII characters, it's more
16583 * legible to print those characters rather than hex values. For
16584 * larger ranges that include more than printables, it's probably
16585 * clearer to just give the start and end points of the range in hex,
16586 * and that's all we can do if there aren't any printables within the
16589 * On ASCII platforms the range of printables is contiguous. If the
16590 * entire range is printable, we print each character as such. If the
16591 * range is partially printable and partially not, it's less likely
16592 * that the individual printables are meaningful, especially if all or
16593 * almost all of them are in the range. But we err on the side of the
16594 * individual printables being meaningful by using the hex only if the
16595 * range contains all but 2 of the printables.
16597 * On EBCDIC platforms, the printables are scattered around so that the
16598 * maximum range length containing only them is about 10. Anything
16599 * longer we treat as hex; otherwise we examine the range character by
16600 * character to see */
16602 if (start < 256 && (((end < 255) ? end : 255) - start <= 10))
16604 if ((isPRINT_A(start) && isPRINT_A(end))
16605 || (end >= 0x7F && (isPRINT_A(start) && start > 0x21))
16606 || ((end < 0x7D && isPRINT_A(end)) && start < 0x20))
16609 /* If the range beginning isn't an ASCII printable, we find the
16610 * last such in the range, then split the output, so all the
16611 * non-printables are in one subrange; then process the remaining
16612 * portion as usual. If the entire range isn't printables, we
16613 * don't split, but drop down to print as hex */
16614 if (! isPRINT_A(start)) {
16615 UV temp_end = start + 1;
16616 while (temp_end <= end && ! isPRINT_A(temp_end)) {
16619 if (temp_end <= end) {
16620 put_range(sv, start, temp_end - 1);
16626 /* If the range beginning is a digit, output a subrange of just the
16627 * digits, then process the remaining portion as usual */
16628 if (isDIGIT_A(start)) {
16629 put_byte(sv, start);
16630 sv_catpvs(sv, "-");
16631 while (start <= end && isDIGIT_A(start)) start++;
16632 put_byte(sv, start - 1);
16636 /* Similarly for alphabetics. Because in both ASCII and EBCDIC,
16637 * the code points for upper and lower A-Z and a-z aren't
16638 * intermixed, the resulting subrange will consist solely of either
16639 * upper- or lower- alphabetics */
16640 if (isALPHA_A(start)) {
16641 put_byte(sv, start);
16642 sv_catpvs(sv, "-");
16643 while (start <= end && isALPHA_A(start)) start++;
16644 put_byte(sv, start - 1);
16648 /* We output any remaining printables as individual characters */
16649 if (isPUNCT_A(start) || isSPACE_A(start)) {
16650 while (start <= end && (isPUNCT_A(start) || isSPACE_A(start))) {
16651 put_byte(sv, start);
16658 /* Here is a control or non-ascii. Output the range or subrange as
16660 Perl_sv_catpvf(aTHX_ sv, "\\x{%02" UVXf "}-\\x{%02" UVXf "}",
16662 (end < 256) ? end : 255);
16668 S_put_latin1_charclass_innards(pTHX_ SV *sv, char *bitmap)
16670 /* Appends to 'sv' a displayable version of the innards of the bracketed
16671 * character class whose bitmap is 'bitmap'; Returns 'TRUE' if it actually
16672 * output anything */
16675 bool has_output_anything = FALSE;
16677 PERL_ARGS_ASSERT_PUT_LATIN1_CHARCLASS_INNARDS;
16679 for (i = 0; i < 256; i++) {
16680 if (i < 256 && BITMAP_TEST((U8 *) bitmap,i)) {
16682 /* The character at index i should be output. Find the next
16683 * character that should NOT be output */
16685 for (j = i + 1; j <= 256; j++) {
16686 if (! BITMAP_TEST((U8 *) bitmap, j)) {
16691 /* Everything between them is a single range that should be output
16693 put_range(sv, i, j - 1);
16694 has_output_anything = TRUE;
16699 return has_output_anything;
16702 #define CLEAR_OPTSTART \
16703 if (optstart) STMT_START { \
16704 DEBUG_OPTIMISE_r(PerlIO_printf(Perl_debug_log, \
16705 " (%"IVdf" nodes)\n", (IV)(node - optstart))); \
16709 #define DUMPUNTIL(b,e) \
16711 node=dumpuntil(r,start,(b),(e),last,sv,indent+1,depth+1);
16713 STATIC const regnode *
16714 S_dumpuntil(pTHX_ const regexp *r, const regnode *start, const regnode *node,
16715 const regnode *last, const regnode *plast,
16716 SV* sv, I32 indent, U32 depth)
16719 U8 op = PSEUDO; /* Arbitrary non-END op. */
16720 const regnode *next;
16721 const regnode *optstart= NULL;
16723 RXi_GET_DECL(r,ri);
16724 GET_RE_DEBUG_FLAGS_DECL;
16726 PERL_ARGS_ASSERT_DUMPUNTIL;
16728 #ifdef DEBUG_DUMPUNTIL
16729 PerlIO_printf(Perl_debug_log, "--- %d : %d - %d - %d\n",indent,node-start,
16730 last ? last-start : 0,plast ? plast-start : 0);
16733 if (plast && plast < last)
16736 while (PL_regkind[op] != END && (!last || node < last)) {
16738 /* While that wasn't END last time... */
16741 if (op == CLOSE || op == WHILEM)
16743 next = regnext((regnode *)node);
16746 if (OP(node) == OPTIMIZED) {
16747 if (!optstart && RE_DEBUG_FLAG(RE_DEBUG_COMPILE_OPTIMISE))
16754 regprop(r, sv, node, NULL);
16755 PerlIO_printf(Perl_debug_log, "%4"IVdf":%*s%s", (IV)(node - start),
16756 (int)(2*indent + 1), "", SvPVX_const(sv));
16758 if (OP(node) != OPTIMIZED) {
16759 if (next == NULL) /* Next ptr. */
16760 PerlIO_printf(Perl_debug_log, " (0)");
16761 else if (PL_regkind[(U8)op] == BRANCH
16762 && PL_regkind[OP(next)] != BRANCH )
16763 PerlIO_printf(Perl_debug_log, " (FAIL)");
16765 PerlIO_printf(Perl_debug_log, " (%"IVdf")", (IV)(next - start));
16766 (void)PerlIO_putc(Perl_debug_log, '\n');
16770 if (PL_regkind[(U8)op] == BRANCHJ) {
16773 const regnode *nnode = (OP(next) == LONGJMP
16774 ? regnext((regnode *)next)
16776 if (last && nnode > last)
16778 DUMPUNTIL(NEXTOPER(NEXTOPER(node)), nnode);
16781 else if (PL_regkind[(U8)op] == BRANCH) {
16783 DUMPUNTIL(NEXTOPER(node), next);
16785 else if ( PL_regkind[(U8)op] == TRIE ) {
16786 const regnode *this_trie = node;
16787 const char op = OP(node);
16788 const U32 n = ARG(node);
16789 const reg_ac_data * const ac = op>=AHOCORASICK ?
16790 (reg_ac_data *)ri->data->data[n] :
16792 const reg_trie_data * const trie =
16793 (reg_trie_data*)ri->data->data[op<AHOCORASICK ? n : ac->trie];
16795 AV *const trie_words
16796 = MUTABLE_AV(ri->data->data[n + TRIE_WORDS_OFFSET]);
16798 const regnode *nextbranch= NULL;
16801 for (word_idx= 0; word_idx < (I32)trie->wordcount; word_idx++) {
16802 SV ** const elem_ptr = av_fetch(trie_words,word_idx,0);
16804 PerlIO_printf(Perl_debug_log, "%*s%s ",
16805 (int)(2*(indent+3)), "",
16807 ? pv_pretty(sv, SvPV_nolen_const(*elem_ptr),
16808 SvCUR(*elem_ptr), 60,
16809 PL_colors[0], PL_colors[1],
16811 ? PERL_PV_ESCAPE_UNI
16813 | PERL_PV_PRETTY_ELLIPSES
16814 | PERL_PV_PRETTY_LTGT
16819 U16 dist= trie->jump[word_idx+1];
16820 PerlIO_printf(Perl_debug_log, "(%"UVuf")\n",
16821 (UV)((dist ? this_trie + dist : next) - start));
16824 nextbranch= this_trie + trie->jump[0];
16825 DUMPUNTIL(this_trie + dist, nextbranch);
16827 if (nextbranch && PL_regkind[OP(nextbranch)]==BRANCH)
16828 nextbranch= regnext((regnode *)nextbranch);
16830 PerlIO_printf(Perl_debug_log, "\n");
16833 if (last && next > last)
16838 else if ( op == CURLY ) { /* "next" might be very big: optimizer */
16839 DUMPUNTIL(NEXTOPER(node) + EXTRA_STEP_2ARGS,
16840 NEXTOPER(node) + EXTRA_STEP_2ARGS + 1);
16842 else if (PL_regkind[(U8)op] == CURLY && op != CURLYX) {
16844 DUMPUNTIL(NEXTOPER(node) + EXTRA_STEP_2ARGS, next);
16846 else if ( op == PLUS || op == STAR) {
16847 DUMPUNTIL(NEXTOPER(node), NEXTOPER(node) + 1);
16849 else if (PL_regkind[(U8)op] == ANYOF) {
16850 /* arglen 1 + class block */
16851 node += 1 + ((ANYOF_FLAGS(node) & ANYOF_POSIXL)
16852 ? ANYOF_POSIXL_SKIP
16854 node = NEXTOPER(node);
16856 else if (PL_regkind[(U8)op] == EXACT) {
16857 /* Literal string, where present. */
16858 node += NODE_SZ_STR(node) - 1;
16859 node = NEXTOPER(node);
16862 node = NEXTOPER(node);
16863 node += regarglen[(U8)op];
16865 if (op == CURLYX || op == OPEN)
16869 #ifdef DEBUG_DUMPUNTIL
16870 PerlIO_printf(Perl_debug_log, "--- %d\n", (int)indent);
16875 #endif /* DEBUGGING */
16879 * c-indentation-style: bsd
16880 * c-basic-offset: 4
16881 * indent-tabs-mode: nil
16884 * ex: set ts=8 sts=4 sw=4 et: