3 perlreapi - perl regular expression plugin interface
7 As of Perl 5.9.5 there is a new interface for using other regexp
8 engines than the default one. Each engine is supposed to provide
9 access to a constant structure of the following format:
11 typedef struct regexp_engine {
12 REGEXP* (*comp) (pTHX_ const SV * const pattern, const U32 flags);
13 I32 (*exec) (pTHX_ REGEXP * const rx, char* stringarg, char* strend,
14 char* strbeg, I32 minend, SV* screamer,
15 void* data, U32 flags);
16 char* (*intuit) (pTHX_ REGEXP * const rx, SV *sv, char *strpos,
17 char *strend, U32 flags,
18 struct re_scream_pos_data_s *data);
19 SV* (*checkstr) (pTHX_ REGEXP * const rx);
20 void (*free) (pTHX_ REGEXP * const rx);
21 void (*numbered_buff_FETCH) (pTHX_ REGEXP * const rx, const I32 paren,
23 void (*numbered_buff_STORE) (pTHX_ REGEXP * const rx, const I32 paren,
24 SV const * const value);
25 I32 (*numbered_buff_LENGTH) (pTHX_ REGEXP * const rx, const SV * const sv,
27 SV* (*named_buff) (pTHX_ REGEXP * const rx, SV * const key,
28 SV * const value, U32 flags);
29 SV* (*named_buff_iter) (pTHX_ REGEXP * const rx, const SV * const lastkey,
31 SV* (*qr_package)(pTHX_ REGEXP * const rx);
33 void* (*dupe) (pTHX_ REGEXP * const rx, CLONE_PARAMS *param);
36 When a regexp is compiled, its C<engine> field is then set to point at
37 the appropriate structure so that when it needs to be used Perl can find
38 the right routines to do so.
40 In order to install a new regexp handler, C<$^H{regcomp}> is set
41 to an integer which (when casted appropriately) resolves to one of these
42 structures. When compiling, the C<comp> method is executed, and the
43 resulting regexp structure's engine field is expected to point back at
46 The pTHX_ symbol in the definition is a macro used by perl under threading
47 to provide an extra argument to the routine holding a pointer back to
48 the interpreter that is executing the regexp. So under threading all
49 routines get an extra argument.
55 REGEXP* comp(pTHX_ const SV * const pattern, const U32 flags);
57 Compile the pattern stored in C<pattern> using the given C<flags> and
58 return a pointer to a prepared C<REGEXP> structure that can perform
59 the match. See L</The REGEXP structure> below for an explanation of
60 the individual fields in the REGEXP struct.
62 The C<pattern> parameter is the scalar that was used as the
63 pattern. previous versions of perl would pass two C<char*> indicating
64 the start and end of the stringifed pattern, the following snippet can
65 be used to get the old parameters:
68 char* exp = SvPV(pattern, plen);
69 char* xend = exp + plen;
71 Since any scalar can be passed as a pattern it's possible to implement
72 an engine that does something with an array (C<< "ook" =~ [ qw/ eek
73 hlagh / ] >>) or with the non-stringified form of a compiled regular
74 expression (C<< "ook" =~ qr/eek/ >>). perl's own engine will always
75 stringify everything using the snippet above but that doesn't mean
76 other engines have to.
78 The C<flags> paramater is a bitfield which indicates which of the
79 C<msixp> flags the regex was compiled with. It also contains
80 additional info such as whether C<use locale> is in effect.
82 The C<eogc> flags are stripped out before being passed to the comp
83 routine. The regex engine does not need to know whether any of these
84 are set as those flags should only affect what perl does with the
85 pattern and its match variables, not how it gets compiled and
88 By the time the comp callback is called, some of these flags have
89 already had effect (noted below where applicable). However most of
90 their effect occurs after the comp callback has run in routines that
91 read the C<< rx->extflags >> field which it populates.
93 In general the flags should be preserved in C<< rx->extflags >> after
94 compilation, although the regex engine might want to add or delete
95 some of them to invoke or disable some special behavior in perl. The
96 flags along with any special behavior they cause are documented below:
98 The pattern modifiers:
102 =item C</m> - RXf_PMf_MULTILINE
104 If this is in C<< rx->extflags >> it will be passed to
105 C<Perl_fbm_instr> by C<pp_split> which will treat the subject string
106 as a multi-line string.
108 =item C</s> - RXf_PMf_SINGLELINE
110 =item C</i> - RXf_PMf_FOLD
112 =item C</x> - RXf_PMf_EXTENDED
114 If present on a regex C<#> comments will be handled differently by the
115 tokenizer in some cases.
117 TODO: Document those cases.
119 =item C</p> - RXf_PMf_KEEPCOPY
129 Set if C<use locale> is in effect. If present in C<< rx->extflags >>
130 C<split> will use the locale dependant definition of whitespace under
131 when RXf_SKIPWHITE or RXf_WHITE are in effect. Under ASCII whitespace
132 is defined as per L<isSPACE|perlapi/ISSPACE>, and by the internal
133 macros C<is_utf8_space> under UTF-8 and C<isSPACE_LC> under C<use
138 Set if the pattern is L<SvUTF8()|perlapi/SvUTF8>, set by Perl_pmruntime.
140 A regex engine may want to set or disable this flag during
141 compilation. The perl engine for instance may upgrade non-UTF-8
142 strings to UTF-8 if the pattern includes constructs such as C<\x{...}>
143 that can only match Unicode values.
147 If C<split> is invoked as C<split ' '> or with no arguments (which
148 really means C<split(' ', $_)>, see L<split|perlfunc/split>), perl will
149 set this flag. The regex engine can then check for it and set the
150 SKIPWHITE and WHITE extflags. To do this the perl engine does:
152 if (flags & RXf_SPLIT && r->prelen == 1 && r->precomp[0] == ' ')
153 r->extflags |= (RXf_SKIPWHITE|RXf_WHITE);
157 These flags can be set during compilation to enable optimizations in
158 the C<split> operator.
164 If the flag is present in C<< rx->extflags >> C<split> will delete
165 whitespace from the start of the subject string before it's operated
166 on. What is considered whitespace depends on whether the subject is a
167 UTF-8 string and whether the C<RXf_PMf_LOCALE> flag is set.
169 If RXf_WHITE is set in addition to this flag C<split> will behave like
170 C<split " "> under the perl engine.
174 Tells the split operator to split the target string on newlines
175 (C<\n>) without invoking the regex engine.
177 Perl's engine sets this if the pattern is C</^/> (C<plen == 1 && *exp
178 == '^'>), even under C</^/s>, see L<split|perlfunc>. Of course a
179 different regex engine might want to use the same optimizations
180 with a different syntax.
184 Tells the split operator to split the target string on whitespace
185 without invoking the regex engine. The definition of whitespace varies
186 depending on whether the target string is a UTF-8 string and on
187 whether RXf_PMf_LOCALE is set.
189 Perl's engine sets this flag if the pattern is C<\s+>.
193 Tells the split operatior to split the target string on
194 characters. The definition of character varies depending on whether
195 the target string is a UTF-8 string.
197 Perl's engine sets this flag on empty patterns, this optimization
198 makes C<split //> much faster than it would otherwise be, it's even
199 faster than C<unpack>.
205 I32 exec(pTHX_ REGEXP * const rx,
206 char *stringarg, char* strend, char* strbeg,
207 I32 minend, SV* screamer,
208 void* data, U32 flags);
214 char* intuit(pTHX_ REGEXP * const rx,
215 SV *sv, char *strpos, char *strend,
216 const U32 flags, struct re_scream_pos_data_s *data);
218 Find the start position where a regex match should be attempted,
219 or possibly whether the regex engine should not be run because the
220 pattern can't match. This is called as appropriate by the core
221 depending on the values of the extflags member of the regexp
226 SV* checkstr(pTHX_ REGEXP * const rx);
228 Return a SV containing a string that must appear in the pattern. Used
229 by C<split> for optimising matches.
233 void free(pTHX_ REGEXP * const rx);
235 Called by perl when it is freeing a regexp pattern so that the engine
236 can release any resources pointed to by the C<pprivate> member of the
237 regexp structure. This is only responsible for freeing private data;
238 perl will handle releasing anything else contained in the regexp structure.
240 =head2 Numbered capture callbacks
242 Called to get/set the value of C<$`>, C<$'>, C<$&> and their named
243 equivalents, ${^PREMATCH}, ${^POSTMATCH} and $^{MATCH}, as well as the
244 numbered capture buffers (C<$1>, C<$2>, ...).
246 The C<paren> paramater will be C<-2> for C<$`>, C<-1> for C<$'>, C<0>
247 for C<$&>, C<1> for C<$1> and so forth.
249 The names have been chosen by analogy with L<Tie::Scalar> methods
250 names with an additional B<LENGTH> callback for efficiency. However
251 named capture variables are currently not tied internally but
252 implemented via magic.
254 =head3 numbered_buff_FETCH
256 void numbered_buff_FETCH(pTHX_ REGEXP * const rx, const I32 paren,
259 Fetch a specified numbered capture. C<sv> should be set to the scalar
260 to return, the scalar is passed as an argument rather than being
261 returned from the function because when it's called perl already has a
262 scalar to store the value, creating another one would be
263 redundant. The scalar can be set with C<sv_setsv>, C<sv_setpvn> and
264 friends, see L<perlapi>.
266 This callback is where perl untaints its own capture variables under
267 taint mode (see L<perlsec>). See the C<Perl_reg_numbered_buff_fetch>
268 function in F<regcomp.c> for how to untaint capture variables if
269 that's something you'd like your engine to do as well.
271 =head3 numbered_buff_STORE
273 void (*numbered_buff_STORE) (pTHX_ REGEXP * const rx, const I32 paren,
274 SV const * const value);
276 Set the value of a numbered capture variable. C<value> is the scalar
277 that is to be used as the new value. It's up to the engine to make
278 sure this is used as the new value (or reject it).
282 if ("ook" =~ /(o*)/) {
283 # `paren' will be `1' and `value' will be `ee'
287 Perl's own engine will croak on any attempt to modify the capture
288 variables, to do this in another engine use the following callack
289 (copied from C<Perl_reg_numbered_buff_store>):
292 Example_reg_numbered_buff_store(pTHX_ REGEXP * const rx, const I32 paren,
293 SV const * const value)
296 PERL_UNUSED_ARG(paren);
297 PERL_UNUSED_ARG(value);
300 Perl_croak(aTHX_ PL_no_modify);
303 Actually perl 5.10 will not I<always> croak in a statement that looks
304 like it would modify a numbered capture variable. This is because the
305 STORE callback will not be called if perl can determine that it
306 doesn't have to modify the value. This is exactly how tied variables
307 behave in the same situation:
310 use base 'Tie::Scalar';
312 sub TIESCALAR { bless [] }
314 sub STORE { die "This doesn't get called" }
318 tie my $sv => "CatptureVar";
321 Because C<$sv> is C<undef> when the C<y///> operator is applied to it
322 the transliteration won't actually execute and the program won't
323 C<die>. This is different to how 5.8 and earlier versions behaved
324 since the capture variables were READONLY variables then, now they'll
325 just die when assigned to in the default engine.
327 =head3 numbered_buff_LENGTH
329 I32 numbered_buff_LENGTH (pTHX_ REGEXP * const rx, const SV * const sv,
332 Get the C<length> of a capture variable. There's a special callback
333 for this so that perl doesn't have to do a FETCH and run C<length> on
334 the result, since the length is (in perl's case) known from an offset
335 stored in C<<rx->offs> this is much more efficient:
337 I32 s1 = rx->offs[paren].start;
338 I32 s2 = rx->offs[paren].end;
341 This is a little bit more complex in the case of UTF-8, see what
342 C<Perl_reg_numbered_buff_length> does with
343 L<is_utf8_string_loclen|perlapi/is_utf8_string_loclen>.
345 =head2 Named capture callbacks
347 Called to get/set the value of C<%+> and C<%-> as well as by some
348 utility functions in L<re>.
350 There are two callbacks, C<named_buff> is called in all the cases the
351 FETCH, STORE, DELETE, CLEAR, EXISTS and SCALAR L<Tie::Hash> callbacks
352 would be on changes to C<%+> and C<%-> and C<named_buff_iter> in the
353 same cases as FIRSTKEY and NEXTKEY.
355 The C<flags> parameter can be used to determine which of these
356 operations the callbacks should respond to, the following flags are
359 Which L<Tie::Hash> operation is being performed from the Perl level on
360 C<%+> or C<%+>, if any:
371 Whether C<%+> or C<%-> is being operated on, if any.
376 Whether this is being called as C<re::regname>, C<re::regnames> or
377 C<re::regnames_count>, if any. The first two will be combined with
378 C<RXapif_ONE> or C<RXapif_ALL>.
382 RXapif_REGNAMES_COUNT
384 Internally C<%+> and C<%-> are implemented with a real tied interface
385 via L<Tie::Hash::NamedCapture>. The methods in that package will call
386 back into these functions. However the usage of
387 L<Tie::Hash::NamedCapture> for this purpose might change in future
388 releases. For instance this might be implemented by magic instead
389 (would need an extension to mgvtbl).
393 SV* (*named_buff) (pTHX_ REGEXP * const rx, SV * const key,
394 SV * const value, U32 flags);
396 =head3 named_buff_iter
398 SV* (*named_buff_iter) (pTHX_ REGEXP * const rx, const SV * const lastkey,
403 SV* qr_package(pTHX_ REGEXP * const rx);
405 The package the qr// magic object is blessed into (as seen by C<ref
406 qr//>). It is recommended that engines change this to their package
407 name for identification regardless of whether they implement methods
410 The package this method returns should also have the internal
411 C<Regexp> package in its C<@ISA>. C<qr//->isa("Regexp")> should always
412 be true regardless of what engine is being used.
414 Example implementation might be:
417 Example_qr_package(pTHX_ REGEXP * const rx)
420 return newSVpvs("re::engine::Example");
423 Any method calls on an object created with C<qr//> will be dispatched to the
424 package as a normal object.
426 use re::engine::Example;
428 $re->meth; # dispatched to re::engine::Example::meth()
430 To retrieve the C<REGEXP> object from the scalar in an XS function use
431 the C<SvRX> macro, see L<"REGEXP Functions" in perlapi|perlapi/REGEXP
436 REGEXP * re = SvRX(sv);
440 void* dupe(pTHX_ REGEXP * const rx, CLONE_PARAMS *param);
442 On threaded builds a regexp may need to be duplicated so that the pattern
443 can be used by mutiple threads. This routine is expected to handle the
444 duplication of any private data pointed to by the C<pprivate> member of
445 the regexp structure. It will be called with the preconstructed new
446 regexp structure as an argument, the C<pprivate> member will point at
447 the B<old> private structue, and it is this routine's responsibility to
448 construct a copy and return a pointer to it (which perl will then use to
449 overwrite the field as passed to this routine.)
451 This allows the engine to dupe its private data but also if necessary
452 modify the final structure if it really must.
454 On unthreaded builds this field doesn't exist.
456 =head1 The REGEXP structure
458 The REGEXP struct is defined in F<regexp.h>. All regex engines must be able to
459 correctly build such a structure in their L</comp> routine.
461 The REGEXP structure contains all the data that perl needs to be aware of
462 to properly work with the regular expression. It includes data about
463 optimisations that perl can use to determine if the regex engine should
464 really be used, and various other control info that is needed to properly
465 execute patterns in various contexts such as is the pattern anchored in
466 some way, or what flags were used during the compile, or whether the
467 program contains special constructs that perl needs to be aware of.
469 In addition it contains two fields that are intended for the private
470 use of the regex engine that compiled the pattern. These are the
471 C<intflags> and C<pprivate> members. C<pprivate> is a void pointer to
472 an arbitrary structure whose use and management is the responsibility
473 of the compiling engine. perl will never modify either of these
476 typedef struct regexp {
477 /* what engine created this regexp? */
478 const struct regexp_engine* engine;
480 /* what re is this a lightweight copy of? */
481 struct regexp* mother_re;
483 /* Information about the match that the perl core uses to manage things */
484 U32 extflags; /* Flags used both externally and internally */
485 I32 minlen; /* mininum possible length of string to match */
486 I32 minlenret; /* mininum possible length of $& */
487 U32 gofs; /* chars left of pos that we search from */
489 /* substring data about strings that must appear
490 in the final match, used for optimisations */
491 struct reg_substr_data *substrs;
493 U32 nparens; /* number of capture buffers */
495 /* private engine specific data */
496 U32 intflags; /* Engine Specific Internal flags */
497 void *pprivate; /* Data private to the regex engine which
498 created this object. */
500 /* Data about the last/current match. These are modified during matching*/
501 U32 lastparen; /* last open paren matched */
502 U32 lastcloseparen; /* last close paren matched */
503 regexp_paren_pair *swap; /* Swap copy of *offs */
504 regexp_paren_pair *offs; /* Array of offsets for (@-) and (@+) */
506 char *subbeg; /* saved or original string so \digit works forever. */
507 SV_SAVED_COPY /* If non-NULL, SV which is COW from original */
508 I32 sublen; /* Length of string pointed by subbeg */
510 /* Information about the match that isn't often used */
511 I32 prelen; /* length of precomp */
512 const char *precomp; /* pre-compilation regular expression */
514 char *wrapped; /* wrapped version of the pattern */
515 I32 wraplen; /* length of wrapped */
517 I32 seen_evals; /* number of eval groups in the pattern - for security checks */
518 HV *paren_names; /* Optional hash of paren names */
520 /* Refcount of this regexp */
521 I32 refcnt; /* Refcount of this regexp */
524 The fields are discussed in more detail below:
528 This field points at a regexp_engine structure which contains pointers
529 to the subroutines that are to be used for performing a match. It
530 is the compiling routine's responsibility to populate this field before
531 returning the regexp object.
533 Internally this is set to C<NULL> unless a custom engine is specified in
534 C<$^H{regcomp}>, perl's own set of callbacks can be accessed in the struct
535 pointed to by C<RE_ENGINE_PTR>.
539 TODO, see L<http://www.mail-archive.com/perl5-changes@perl.org/msg17328.html>
543 This will be used by perl to see what flags the regexp was compiled
544 with, this will normally be set to the value of the flags parameter by
545 the L<comp|/comp> callback. See the L<comp|/comp> documentation for
548 =head2 C<minlen> C<minlenret>
550 The minimum string length required for the pattern to match. This is used to
551 prune the search space by not bothering to match any closer to the end of a
552 string than would allow a match. For instance there is no point in even
553 starting the regex engine if the minlen is 10 but the string is only 5
554 characters long. There is no way that the pattern can match.
556 C<minlenret> is the minimum length of the string that would be found
559 The difference between C<minlen> and C<minlenret> can be seen in the
564 where the C<minlen> would be 3 but C<minlenret> would only be 2 as the \d is
565 required to match but is not actually included in the matched content. This
566 distinction is particularly important as the substitution logic uses the
567 C<minlenret> to tell whether it can do in-place substition which can result in
568 considerable speedup.
572 Left offset from pos() to start match at.
576 Substring data about strings that must appear in the final match. This
577 is currently only used internally by perl's engine for but might be
578 used in the future for all engines for optimisations.
580 =head2 C<nparens>, C<lasparen>, and C<lastcloseparen>
582 These fields are used to keep track of how many paren groups could be matched
583 in the pattern, which was the last open paren to be entered, and which was
584 the last close paren to be entered.
588 The engine's private copy of the flags the pattern was compiled with. Usually
589 this is the same as C<extflags> unless the engine chose to modify one of them.
593 A void* pointing to an engine-defined data structure. The perl engine uses the
594 C<regexp_internal> structure (see L<perlreguts/Base Structures>) but a custom
595 engine should use something else.
603 A C<regexp_paren_pair> structure which defines offsets into the string being
604 matched which correspond to the C<$&> and C<$1>, C<$2> etc. captures, the
605 C<regexp_paren_pair> struct is defined as follows:
607 typedef struct regexp_paren_pair {
612 If C<< ->offs[num].start >> or C<< ->offs[num].end >> is C<-1> then that
613 capture buffer did not match. C<< ->offs[0].start/end >> represents C<$&> (or
614 C<${^MATCH> under C<//p>) and C<< ->offs[paren].end >> matches C<$$paren> where
617 =head2 C<precomp> C<prelen>
619 Used for optimisations. C<precomp> holds a copy of the pattern that
620 was compiled and C<prelen> its length. When a new pattern is to be
621 compiled (such as inside a loop) the internal C<regcomp> operator
622 checks whether the last compiled C<REGEXP>'s C<precomp> and C<prelen>
623 are equivalent to the new one, and if so uses the old pattern instead
624 of compiling a new one.
626 The relevant snippet from C<Perl_pp_regcomp>:
628 if (!re || !re->precomp || re->prelen != (I32)len ||
629 memNE(re->precomp, t, len))
630 /* Compile a new pattern */
632 =head2 C<paren_names>
634 This is a hash used internally to track named capture buffers and their
635 offsets. The keys are the names of the buffers the values are dualvars,
636 with the IV slot holding the number of buffers with the given name and the
637 pv being an embedded array of I32. The values may also be contained
638 independently in the data array in cases where named backreferences are
643 Holds information on the longest string that must occur at a fixed
644 offset from the start of the pattern, and the longest string that must
645 occur at a floating offset from the start of the pattern. Used to do
646 Fast-Boyer-Moore searches on the string to find out if its worth using
647 the regex engine at all, and if so where in the string to search.
649 =head2 C<subbeg> C<sublen> C<saved_copy>
651 Used during execution phase for managing search and replace patterns.
653 =head2 C<wrapped> C<wraplen>
655 Stores the string C<qr//> stringifies to. The perl engine for example
656 stores C<(?-xism:eek)> in the case of C<qr/eek/>.
658 When using a custom engine that doesn't support the C<(?:)> construct
659 for inline modifiers, it's probably best to have C<qr//> stringify to
660 the supplied pattern, note that this will create undesired patterns in
663 my $x = qr/a|b/; # "a|b"
664 my $y = qr/c/i; # "c"
665 my $z = qr/$x$y/; # "a|bc"
667 There's no solution for this problem other than making the custom
668 engine understand a construct like C<(?:)>.
672 This stores the number of eval groups in the pattern. This is used for security
673 purposes when embedding compiled regexes into larger patterns with C<qr//>.
677 The number of times the structure is referenced. When this falls to 0 the
678 regexp is automatically freed by a call to pregfree. This should be set to 1 in
679 each engine's L</comp> routine.
683 Originally part of L<perlreguts>.
687 Originally written by Yves Orton, expanded by E<AElig>var ArnfjE<ouml>rE<eth>
692 Copyright 2006 Yves Orton and 2007 E<AElig>var ArnfjE<ouml>rE<eth> Bjarmason.
694 This program is free software; you can redistribute it and/or modify it under
695 the same terms as Perl itself.