3 perlreapi - perl regular expression plugin interface
7 As of Perl 5.9.5 there is a new interface for plugging and using other
8 regular expression engines than the default one.
10 Each engine is supposed to provide access to a constant structure of the
13 typedef struct regexp_engine {
14 REGEXP* (*comp) (pTHX_ const SV * const pattern, const U32 flags);
15 I32 (*exec) (pTHX_ REGEXP * const rx, char* stringarg, char* strend,
16 char* strbeg, I32 minend, SV* screamer,
17 void* data, U32 flags);
18 char* (*intuit) (pTHX_ REGEXP * const rx, SV *sv, char *strpos,
19 char *strend, U32 flags,
20 struct re_scream_pos_data_s *data);
21 SV* (*checkstr) (pTHX_ REGEXP * const rx);
22 void (*free) (pTHX_ REGEXP * const rx);
23 void (*numbered_buff_FETCH) (pTHX_ REGEXP * const rx, const I32 paren,
25 void (*numbered_buff_STORE) (pTHX_ REGEXP * const rx, const I32 paren,
26 SV const * const value);
27 I32 (*numbered_buff_LENGTH) (pTHX_ REGEXP * const rx, const SV * const sv,
29 SV* (*named_buff) (pTHX_ REGEXP * const rx, SV * const key,
30 SV * const value, U32 flags);
31 SV* (*named_buff_iter) (pTHX_ REGEXP * const rx, const SV * const lastkey,
33 SV* (*qr_package)(pTHX_ REGEXP * const rx);
35 void* (*dupe) (pTHX_ REGEXP * const rx, CLONE_PARAMS *param);
38 When a regexp is compiled, its C<engine> field is then set to point at
39 the appropriate structure, so that when it needs to be used Perl can find
40 the right routines to do so.
42 In order to install a new regexp handler, C<$^H{regcomp}> is set
43 to an integer which (when casted appropriately) resolves to one of these
44 structures. When compiling, the C<comp> method is executed, and the
45 resulting regexp structure's engine field is expected to point back at
48 The pTHX_ symbol in the definition is a macro used by perl under threading
49 to provide an extra argument to the routine holding a pointer back to
50 the interpreter that is executing the regexp. So under threading all
51 routines get an extra argument.
57 REGEXP* comp(pTHX_ const SV * const pattern, const U32 flags);
59 Compile the pattern stored in C<pattern> using the given C<flags> and
60 return a pointer to a prepared C<REGEXP> structure that can perform
61 the match. See L</The REGEXP structure> below for an explanation of
62 the individual fields in the REGEXP struct.
64 The C<pattern> parameter is the scalar that was used as the
65 pattern. previous versions of perl would pass two C<char*> indicating
66 the start and end of the stringified pattern, the following snippet can
67 be used to get the old parameters:
70 char* exp = SvPV(pattern, plen);
71 char* xend = exp + plen;
73 Since any scalar can be passed as a pattern it's possible to implement
74 an engine that does something with an array (C<< "ook" =~ [ qw/ eek
75 hlagh / ] >>) or with the non-stringified form of a compiled regular
76 expression (C<< "ook" =~ qr/eek/ >>). perl's own engine will always
77 stringify everything using the snippet above but that doesn't mean
78 other engines have to.
80 The C<flags> parameter is a bitfield which indicates which of the
81 C<msixp> flags the regex was compiled with. It also contains
82 additional info such as whether C<use locale> is in effect.
84 The C<eogc> flags are stripped out before being passed to the comp
85 routine. The regex engine does not need to know whether any of these
86 are set as those flags should only affect what perl does with the
87 pattern and its match variables, not how it gets compiled and
90 By the time the comp callback is called, some of these flags have
91 already had effect (noted below where applicable). However most of
92 their effect occurs after the comp callback has run in routines that
93 read the C<< rx->extflags >> field which it populates.
95 In general the flags should be preserved in C<< rx->extflags >> after
96 compilation, although the regex engine might want to add or delete
97 some of them to invoke or disable some special behavior in perl. The
98 flags along with any special behavior they cause are documented below:
100 The pattern modifiers:
104 =item C</m> - RXf_PMf_MULTILINE
106 If this is in C<< rx->extflags >> it will be passed to
107 C<Perl_fbm_instr> by C<pp_split> which will treat the subject string
108 as a multi-line string.
110 =item C</s> - RXf_PMf_SINGLELINE
112 =item C</i> - RXf_PMf_FOLD
114 =item C</x> - RXf_PMf_EXTENDED
116 If present on a regex C<#> comments will be handled differently by the
117 tokenizer in some cases.
119 TODO: Document those cases.
121 =item C</p> - RXf_PMf_KEEPCOPY
131 Set if C<use locale> is in effect. If present in C<< rx->extflags >>
132 C<split> will use the locale dependent definition of whitespace under
133 when RXf_SKIPWHITE or RXf_WHITE are in effect. Under ASCII whitespace
134 is defined as per L<isSPACE|perlapi/ISSPACE>, and by the internal
135 macros C<is_utf8_space> under UTF-8 and C<isSPACE_LC> under C<use
140 Set if the pattern is L<SvUTF8()|perlapi/SvUTF8>, set by Perl_pmruntime.
142 A regex engine may want to set or disable this flag during
143 compilation. The perl engine for instance may upgrade non-UTF-8
144 strings to UTF-8 if the pattern includes constructs such as C<\x{...}>
145 that can only match Unicode values.
149 If C<split> is invoked as C<split ' '> or with no arguments (which
150 really means C<split(' ', $_)>, see L<split|perlfunc/split>), perl will
151 set this flag. The regex engine can then check for it and set the
152 SKIPWHITE and WHITE extflags. To do this the perl engine does:
154 if (flags & RXf_SPLIT && r->prelen == 1 && r->precomp[0] == ' ')
155 r->extflags |= (RXf_SKIPWHITE|RXf_WHITE);
159 These flags can be set during compilation to enable optimizations in
160 the C<split> operator.
166 If the flag is present in C<< rx->extflags >> C<split> will delete
167 whitespace from the start of the subject string before it's operated
168 on. What is considered whitespace depends on whether the subject is a
169 UTF-8 string and whether the C<RXf_PMf_LOCALE> flag is set.
171 If RXf_WHITE is set in addition to this flag C<split> will behave like
172 C<split " "> under the perl engine.
176 Tells the split operator to split the target string on newlines
177 (C<\n>) without invoking the regex engine.
179 Perl's engine sets this if the pattern is C</^/> (C<plen == 1 && *exp
180 == '^'>), even under C</^/s>, see L<split|perlfunc>. Of course a
181 different regex engine might want to use the same optimizations
182 with a different syntax.
186 Tells the split operator to split the target string on whitespace
187 without invoking the regex engine. The definition of whitespace varies
188 depending on whether the target string is a UTF-8 string and on
189 whether RXf_PMf_LOCALE is set.
191 Perl's engine sets this flag if the pattern is C<\s+>.
195 Tells the split operator to split the target string on
196 characters. The definition of character varies depending on whether
197 the target string is a UTF-8 string.
199 Perl's engine sets this flag on empty patterns, this optimization
200 makes C<split //> much faster than it would otherwise be. It's even
201 faster than C<unpack>.
207 I32 exec(pTHX_ REGEXP * const rx,
208 char *stringarg, char* strend, char* strbeg,
209 I32 minend, SV* screamer,
210 void* data, U32 flags);
216 char* intuit(pTHX_ REGEXP * const rx,
217 SV *sv, char *strpos, char *strend,
218 const U32 flags, struct re_scream_pos_data_s *data);
220 Find the start position where a regex match should be attempted,
221 or possibly whether the regex engine should not be run because the
222 pattern can't match. This is called as appropriate by the core
223 depending on the values of the extflags member of the regexp
228 SV* checkstr(pTHX_ REGEXP * const rx);
230 Return a SV containing a string that must appear in the pattern. Used
231 by C<split> for optimising matches.
235 void free(pTHX_ REGEXP * const rx);
237 Called by perl when it is freeing a regexp pattern so that the engine
238 can release any resources pointed to by the C<pprivate> member of the
239 regexp structure. This is only responsible for freeing private data;
240 perl will handle releasing anything else contained in the regexp structure.
242 =head2 Numbered capture callbacks
244 Called to get/set the value of C<$`>, C<$'>, C<$&> and their named
245 equivalents, ${^PREMATCH}, ${^POSTMATCH} and $^{MATCH}, as well as the
246 numbered capture buffers (C<$1>, C<$2>, ...).
248 The C<paren> parameter will be C<-2> for C<$`>, C<-1> for C<$'>, C<0>
249 for C<$&>, C<1> for C<$1> and so forth.
251 The names have been chosen by analogy with L<Tie::Scalar> methods
252 names with an additional B<LENGTH> callback for efficiency. However
253 named capture variables are currently not tied internally but
254 implemented via magic.
256 =head3 numbered_buff_FETCH
258 void numbered_buff_FETCH(pTHX_ REGEXP * const rx, const I32 paren,
261 Fetch a specified numbered capture. C<sv> should be set to the scalar
262 to return, the scalar is passed as an argument rather than being
263 returned from the function because when it's called perl already has a
264 scalar to store the value, creating another one would be
265 redundant. The scalar can be set with C<sv_setsv>, C<sv_setpvn> and
266 friends, see L<perlapi>.
268 This callback is where perl untaints its own capture variables under
269 taint mode (see L<perlsec>). See the C<Perl_reg_numbered_buff_fetch>
270 function in F<regcomp.c> for how to untaint capture variables if
271 that's something you'd like your engine to do as well.
273 =head3 numbered_buff_STORE
275 void (*numbered_buff_STORE) (pTHX_ REGEXP * const rx, const I32 paren,
276 SV const * const value);
278 Set the value of a numbered capture variable. C<value> is the scalar
279 that is to be used as the new value. It's up to the engine to make
280 sure this is used as the new value (or reject it).
284 if ("ook" =~ /(o*)/) {
285 # `paren' will be `1' and `value' will be `ee'
289 Perl's own engine will croak on any attempt to modify the capture
290 variables, to do this in another engine use the following callback
291 (copied from C<Perl_reg_numbered_buff_store>):
294 Example_reg_numbered_buff_store(pTHX_ REGEXP * const rx, const I32 paren,
295 SV const * const value)
298 PERL_UNUSED_ARG(paren);
299 PERL_UNUSED_ARG(value);
302 Perl_croak(aTHX_ PL_no_modify);
305 Actually perl will not I<always> croak in a statement that looks
306 like it would modify a numbered capture variable. This is because the
307 STORE callback will not be called if perl can determine that it
308 doesn't have to modify the value. This is exactly how tied variables
309 behave in the same situation:
312 use base 'Tie::Scalar';
314 sub TIESCALAR { bless [] }
316 sub STORE { die "This doesn't get called" }
320 tie my $sv => "CatptureVar";
323 Because C<$sv> is C<undef> when the C<y///> operator is applied to it
324 the transliteration won't actually execute and the program won't
325 C<die>. This is different to how 5.8 and earlier versions behaved
326 since the capture variables were READONLY variables then, now they'll
327 just die when assigned to in the default engine.
329 =head3 numbered_buff_LENGTH
331 I32 numbered_buff_LENGTH (pTHX_ REGEXP * const rx, const SV * const sv,
334 Get the C<length> of a capture variable. There's a special callback
335 for this so that perl doesn't have to do a FETCH and run C<length> on
336 the result, since the length is (in perl's case) known from an offset
337 stored in C<< rx->offs >> this is much more efficient:
339 I32 s1 = rx->offs[paren].start;
340 I32 s2 = rx->offs[paren].end;
343 This is a little bit more complex in the case of UTF-8, see what
344 C<Perl_reg_numbered_buff_length> does with
345 L<is_utf8_string_loclen|perlapi/is_utf8_string_loclen>.
347 =head2 Named capture callbacks
349 Called to get/set the value of C<%+> and C<%-> as well as by some
350 utility functions in L<re>.
352 There are two callbacks, C<named_buff> is called in all the cases the
353 FETCH, STORE, DELETE, CLEAR, EXISTS and SCALAR L<Tie::Hash> callbacks
354 would be on changes to C<%+> and C<%-> and C<named_buff_iter> in the
355 same cases as FIRSTKEY and NEXTKEY.
357 The C<flags> parameter can be used to determine which of these
358 operations the callbacks should respond to, the following flags are
361 Which L<Tie::Hash> operation is being performed from the Perl level on
362 C<%+> or C<%+>, if any:
373 Whether C<%+> or C<%-> is being operated on, if any.
378 Whether this is being called as C<re::regname>, C<re::regnames> or
379 C<re::regnames_count>, if any. The first two will be combined with
380 C<RXapif_ONE> or C<RXapif_ALL>.
384 RXapif_REGNAMES_COUNT
386 Internally C<%+> and C<%-> are implemented with a real tied interface
387 via L<Tie::Hash::NamedCapture>. The methods in that package will call
388 back into these functions. However the usage of
389 L<Tie::Hash::NamedCapture> for this purpose might change in future
390 releases. For instance this might be implemented by magic instead
391 (would need an extension to mgvtbl).
395 SV* (*named_buff) (pTHX_ REGEXP * const rx, SV * const key,
396 SV * const value, U32 flags);
398 =head3 named_buff_iter
400 SV* (*named_buff_iter) (pTHX_ REGEXP * const rx, const SV * const lastkey,
405 SV* qr_package(pTHX_ REGEXP * const rx);
407 The package the qr// magic object is blessed into (as seen by C<ref
408 qr//>). It is recommended that engines change this to their package
409 name for identification regardless of whether they implement methods
412 The package this method returns should also have the internal
413 C<Regexp> package in its C<@ISA>. C<< qr//->isa("Regexp") >> should always
414 be true regardless of what engine is being used.
416 Example implementation might be:
419 Example_qr_package(pTHX_ REGEXP * const rx)
422 return newSVpvs("re::engine::Example");
425 Any method calls on an object created with C<qr//> will be dispatched to the
426 package as a normal object.
428 use re::engine::Example;
430 $re->meth; # dispatched to re::engine::Example::meth()
432 To retrieve the C<REGEXP> object from the scalar in an XS function use
433 the C<SvRX> macro, see L<"REGEXP Functions" in perlapi|perlapi/REGEXP
438 REGEXP * re = SvRX(sv);
442 void* dupe(pTHX_ REGEXP * const rx, CLONE_PARAMS *param);
444 On threaded builds a regexp may need to be duplicated so that the pattern
445 can be used by multiple threads. This routine is expected to handle the
446 duplication of any private data pointed to by the C<pprivate> member of
447 the regexp structure. It will be called with the preconstructed new
448 regexp structure as an argument, the C<pprivate> member will point at
449 the B<old> private structure, and it is this routine's responsibility to
450 construct a copy and return a pointer to it (which perl will then use to
451 overwrite the field as passed to this routine.)
453 This allows the engine to dupe its private data but also if necessary
454 modify the final structure if it really must.
456 On unthreaded builds this field doesn't exist.
458 =head1 The REGEXP structure
460 The REGEXP struct is defined in F<regexp.h>. All regex engines must be able to
461 correctly build such a structure in their L</comp> routine.
463 The REGEXP structure contains all the data that perl needs to be aware of
464 to properly work with the regular expression. It includes data about
465 optimisations that perl can use to determine if the regex engine should
466 really be used, and various other control info that is needed to properly
467 execute patterns in various contexts such as is the pattern anchored in
468 some way, or what flags were used during the compile, or whether the
469 program contains special constructs that perl needs to be aware of.
471 In addition it contains two fields that are intended for the private
472 use of the regex engine that compiled the pattern. These are the
473 C<intflags> and C<pprivate> members. C<pprivate> is a void pointer to
474 an arbitrary structure whose use and management is the responsibility
475 of the compiling engine. perl will never modify either of these
478 typedef struct regexp {
479 /* what engine created this regexp? */
480 const struct regexp_engine* engine;
482 /* what re is this a lightweight copy of? */
483 struct regexp* mother_re;
485 /* Information about the match that the perl core uses to manage things */
486 U32 extflags; /* Flags used both externally and internally */
487 I32 minlen; /* mininum possible length of string to match */
488 I32 minlenret; /* mininum possible length of $& */
489 U32 gofs; /* chars left of pos that we search from */
491 /* substring data about strings that must appear
492 in the final match, used for optimisations */
493 struct reg_substr_data *substrs;
495 U32 nparens; /* number of capture buffers */
497 /* private engine specific data */
498 U32 intflags; /* Engine Specific Internal flags */
499 void *pprivate; /* Data private to the regex engine which
500 created this object. */
502 /* Data about the last/current match. These are modified during matching*/
503 U32 lastparen; /* last open paren matched */
504 U32 lastcloseparen; /* last close paren matched */
505 regexp_paren_pair *swap; /* Swap copy of *offs */
506 regexp_paren_pair *offs; /* Array of offsets for (@-) and (@+) */
508 char *subbeg; /* saved or original string so \digit works forever. */
509 SV_SAVED_COPY /* If non-NULL, SV which is COW from original */
510 I32 sublen; /* Length of string pointed by subbeg */
512 /* Information about the match that isn't often used */
513 I32 prelen; /* length of precomp */
514 const char *precomp; /* pre-compilation regular expression */
516 char *wrapped; /* wrapped version of the pattern */
517 I32 wraplen; /* length of wrapped */
519 I32 seen_evals; /* number of eval groups in the pattern - for security checks */
520 HV *paren_names; /* Optional hash of paren names */
522 /* Refcount of this regexp */
523 I32 refcnt; /* Refcount of this regexp */
526 The fields are discussed in more detail below:
530 This field points at a regexp_engine structure which contains pointers
531 to the subroutines that are to be used for performing a match. It
532 is the compiling routine's responsibility to populate this field before
533 returning the regexp object.
535 Internally this is set to C<NULL> unless a custom engine is specified in
536 C<$^H{regcomp}>, perl's own set of callbacks can be accessed in the struct
537 pointed to by C<RE_ENGINE_PTR>.
541 TODO, see L<http://www.mail-archive.com/perl5-changes@perl.org/msg17328.html>
545 This will be used by perl to see what flags the regexp was compiled
546 with, this will normally be set to the value of the flags parameter by
547 the L<comp|/comp> callback. See the L<comp|/comp> documentation for
550 =head2 C<minlen> C<minlenret>
552 The minimum string length required for the pattern to match. This is used to
553 prune the search space by not bothering to match any closer to the end of a
554 string than would allow a match. For instance there is no point in even
555 starting the regex engine if the minlen is 10 but the string is only 5
556 characters long. There is no way that the pattern can match.
558 C<minlenret> is the minimum length of the string that would be found
561 The difference between C<minlen> and C<minlenret> can be seen in the
566 where the C<minlen> would be 3 but C<minlenret> would only be 2 as the \d is
567 required to match but is not actually included in the matched content. This
568 distinction is particularly important as the substitution logic uses the
569 C<minlenret> to tell whether it can do in-place substitution which can result in
570 considerable speedup.
574 Left offset from pos() to start match at.
578 Substring data about strings that must appear in the final match. This
579 is currently only used internally by perl's engine for but might be
580 used in the future for all engines for optimisations.
582 =head2 C<nparens>, C<lasparen>, and C<lastcloseparen>
584 These fields are used to keep track of how many paren groups could be matched
585 in the pattern, which was the last open paren to be entered, and which was
586 the last close paren to be entered.
590 The engine's private copy of the flags the pattern was compiled with. Usually
591 this is the same as C<extflags> unless the engine chose to modify one of them.
595 A void* pointing to an engine-defined data structure. The perl engine uses the
596 C<regexp_internal> structure (see L<perlreguts/Base Structures>) but a custom
597 engine should use something else.
601 Unused. Left in for compatibility with perl 5.10.0.
605 A C<regexp_paren_pair> structure which defines offsets into the string being
606 matched which correspond to the C<$&> and C<$1>, C<$2> etc. captures, the
607 C<regexp_paren_pair> struct is defined as follows:
609 typedef struct regexp_paren_pair {
614 If C<< ->offs[num].start >> or C<< ->offs[num].end >> is C<-1> then that
615 capture buffer did not match. C<< ->offs[0].start/end >> represents C<$&> (or
616 C<${^MATCH> under C<//p>) and C<< ->offs[paren].end >> matches C<$$paren> where
619 =head2 C<precomp> C<prelen>
621 Used for optimisations. C<precomp> holds a copy of the pattern that
622 was compiled and C<prelen> its length. When a new pattern is to be
623 compiled (such as inside a loop) the internal C<regcomp> operator
624 checks whether the last compiled C<REGEXP>'s C<precomp> and C<prelen>
625 are equivalent to the new one, and if so uses the old pattern instead
626 of compiling a new one.
628 The relevant snippet from C<Perl_pp_regcomp>:
630 if (!re || !re->precomp || re->prelen != (I32)len ||
631 memNE(re->precomp, t, len))
632 /* Compile a new pattern */
634 =head2 C<paren_names>
636 This is a hash used internally to track named capture buffers and their
637 offsets. The keys are the names of the buffers the values are dualvars,
638 with the IV slot holding the number of buffers with the given name and the
639 pv being an embedded array of I32. The values may also be contained
640 independently in the data array in cases where named backreferences are
645 Holds information on the longest string that must occur at a fixed
646 offset from the start of the pattern, and the longest string that must
647 occur at a floating offset from the start of the pattern. Used to do
648 Fast-Boyer-Moore searches on the string to find out if its worth using
649 the regex engine at all, and if so where in the string to search.
651 =head2 C<subbeg> C<sublen> C<saved_copy>
653 Used during execution phase for managing search and replace patterns.
655 =head2 C<wrapped> C<wraplen>
657 Stores the string C<qr//> stringifies to. The perl engine for example
658 stores C<(?-xism:eek)> in the case of C<qr/eek/>.
660 When using a custom engine that doesn't support the C<(?:)> construct
661 for inline modifiers, it's probably best to have C<qr//> stringify to
662 the supplied pattern, note that this will create undesired patterns in
665 my $x = qr/a|b/; # "a|b"
666 my $y = qr/c/i; # "c"
667 my $z = qr/$x$y/; # "a|bc"
669 There's no solution for this problem other than making the custom
670 engine understand a construct like C<(?:)>.
674 This stores the number of eval groups in the pattern. This is used for security
675 purposes when embedding compiled regexes into larger patterns with C<qr//>.
679 The number of times the structure is referenced. When this falls to 0 the
680 regexp is automatically freed by a call to pregfree. This should be set to 1 in
681 each engine's L</comp> routine.
685 Originally part of L<perlreguts>.
689 Originally written by Yves Orton, expanded by E<AElig>var ArnfjE<ouml>rE<eth>
694 Copyright 2006 Yves Orton and 2007 E<AElig>var ArnfjE<ouml>rE<eth> Bjarmason.
696 This program is free software; you can redistribute it and/or modify it under
697 the same terms as Perl itself.