ViewVC logotype

Contents of /code/trunk/doc/pcrepattern.3

Parent Directory Parent Directory | Revision Log Revision Log

Revision 514 - (show annotations)
Mon May 3 12:54:22 2010 UTC (11 years, 4 months ago) by ph10
File size: 106275 byte(s)
Add support for \N.
3 PCRE - Perl-compatible regular expressions
5 .rs
6 .sp
7 The syntax and semantics of the regular expressions that are supported by PCRE
8 are described in detail below. There is a quick-reference syntax summary in the
9 .\" HREF
10 \fBpcresyntax\fP
11 .\"
12 page. PCRE tries to match Perl syntax and semantics as closely as it can. PCRE
13 also supports some alternative regular expression syntax (which does not
14 conflict with the Perl syntax) in order to provide some compatibility with
15 regular expressions in Python, .NET, and Oniguruma.
16 .P
17 Perl's regular expressions are described in its own documentation, and
18 regular expressions in general are covered in a number of books, some of which
19 have copious examples. Jeffrey Friedl's "Mastering Regular Expressions",
20 published by O'Reilly, covers regular expressions in great detail. This
21 description of PCRE's regular expressions is intended as reference material.
22 .P
23 The original operation of PCRE was on strings of one-byte characters. However,
24 there is now also support for UTF-8 character strings. To use this,
25 PCRE must be built to include UTF-8 support, and you must call
26 \fBpcre_compile()\fP or \fBpcre_compile2()\fP with the PCRE_UTF8 option. There
27 is also a special sequence that can be given at the start of a pattern:
28 .sp
29 (*UTF8)
30 .sp
31 Starting a pattern with this sequence is equivalent to setting the PCRE_UTF8
32 option. This feature is not Perl-compatible. How setting UTF-8 mode affects
33 pattern matching is mentioned in several places below. There is also a summary
34 of UTF-8 features in the
35 .\" HTML <a href="pcre.html#utf8support">
36 .\" </a>
37 section on UTF-8 support
38 .\"
39 in the main
40 .\" HREF
41 \fBpcre\fP
42 .\"
43 page.
44 .P
45 The remainder of this document discusses the patterns that are supported by
46 PCRE when its main matching function, \fBpcre_exec()\fP, is used.
47 From release 6.0, PCRE offers a second matching function,
48 \fBpcre_dfa_exec()\fP, which matches using a different algorithm that is not
49 Perl-compatible. Some of the features discussed below are not available when
50 \fBpcre_dfa_exec()\fP is used. The advantages and disadvantages of the
51 alternative function, and how it differs from the normal function, are
52 discussed in the
53 .\" HREF
54 \fBpcrematching\fP
55 .\"
56 page.
57 .
58 .
60 .rs
61 .sp
62 PCRE supports five different conventions for indicating line breaks in
63 strings: a single CR (carriage return) character, a single LF (linefeed)
64 character, the two-character sequence CRLF, any of the three preceding, or any
65 Unicode newline sequence. The
66 .\" HREF
67 \fBpcreapi\fP
68 .\"
69 page has
70 .\" HTML <a href="pcreapi.html#newlines">
71 .\" </a>
72 further discussion
73 .\"
74 about newlines, and shows how to set the newline convention in the
75 \fIoptions\fP arguments for the compiling and matching functions.
76 .P
77 It is also possible to specify a newline convention by starting a pattern
78 string with one of the following five sequences:
79 .sp
80 (*CR) carriage return
81 (*LF) linefeed
82 (*CRLF) carriage return, followed by linefeed
83 (*ANYCRLF) any of the three above
84 (*ANY) all Unicode newline sequences
85 .sp
86 These override the default and the options given to \fBpcre_compile()\fP or
87 \fBpcre_compile2()\fP. For example, on a Unix system where LF is the default
88 newline sequence, the pattern
89 .sp
90 (*CR)a.b
91 .sp
92 changes the convention to CR. That pattern matches "a\enb" because LF is no
93 longer a newline. Note that these special settings, which are not
94 Perl-compatible, are recognized only at the very start of a pattern, and that
95 they must be in upper case. If more than one of them is present, the last one
96 is used.
97 .P
98 The newline convention affects the interpretation of the dot metacharacter when
99 PCRE_DOTALL is not set, and also the behaviour of \eN. However, it does not
100 affect what the \eR escape sequence matches. By default, this is any Unicode
101 newline sequence, for Perl compatibility. However, this can be changed; see the
102 description of \eR in the section entitled
103 .\" HTML <a href="#newlineseq">
104 .\" </a>
105 "Newline sequences"
106 .\"
107 below. A change of \eR setting can be combined with a change of newline
108 convention.
109 .
110 .
112 .rs
113 .sp
114 A regular expression is a pattern that is matched against a subject string from
115 left to right. Most characters stand for themselves in a pattern, and match the
116 corresponding characters in the subject. As a trivial example, the pattern
117 .sp
118 The quick brown fox
119 .sp
120 matches a portion of a subject string that is identical to itself. When
121 caseless matching is specified (the PCRE_CASELESS option), letters are matched
122 independently of case. In UTF-8 mode, PCRE always understands the concept of
123 case for characters whose values are less than 128, so caseless matching is
124 always possible. For characters with higher values, the concept of case is
125 supported if PCRE is compiled with Unicode property support, but not otherwise.
126 If you want to use caseless matching for characters 128 and above, you must
127 ensure that PCRE is compiled with Unicode property support as well as with
128 UTF-8 support.
129 .P
130 The power of regular expressions comes from the ability to include alternatives
131 and repetitions in the pattern. These are encoded in the pattern by the use of
132 \fImetacharacters\fP, which do not stand for themselves but instead are
133 interpreted in some special way.
134 .P
135 There are two different sets of metacharacters: those that are recognized
136 anywhere in the pattern except within square brackets, and those that are
137 recognized within square brackets. Outside square brackets, the metacharacters
138 are as follows:
139 .sp
140 \e general escape character with several uses
141 ^ assert start of string (or line, in multiline mode)
142 $ assert end of string (or line, in multiline mode)
143 . match any character except newline (by default)
144 [ start character class definition
145 | start of alternative branch
146 ( start subpattern
147 ) end subpattern
148 ? extends the meaning of (
149 also 0 or 1 quantifier
150 also quantifier minimizer
151 * 0 or more quantifier
152 + 1 or more quantifier
153 also "possessive quantifier"
154 { start min/max quantifier
155 .sp
156 Part of a pattern that is in square brackets is called a "character class". In
157 a character class the only metacharacters are:
158 .sp
159 \e general escape character
160 ^ negate the class, but only if the first character
161 - indicates character range
162 .\" JOIN
163 [ POSIX character class (only if followed by POSIX
164 syntax)
165 ] terminates the character class
166 .sp
167 The following sections describe the use of each of the metacharacters.
168 .
169 .
171 .rs
172 .sp
173 The backslash character has several uses. Firstly, if it is followed by a
174 non-alphanumeric character, it takes away any special meaning that character
175 may have. This use of backslash as an escape character applies both inside and
176 outside character classes.
177 .P
178 For example, if you want to match a * character, you write \e* in the pattern.
179 This escaping action applies whether or not the following character would
180 otherwise be interpreted as a metacharacter, so it is always safe to precede a
181 non-alphanumeric with backslash to specify that it stands for itself. In
182 particular, if you want to match a backslash, you write \e\e.
183 .P
184 If a pattern is compiled with the PCRE_EXTENDED option, whitespace in the
185 pattern (other than in a character class) and characters between a # outside
186 a character class and the next newline are ignored. An escaping backslash can
187 be used to include a whitespace or # character as part of the pattern.
188 .P
189 If you want to remove the special meaning from a sequence of characters, you
190 can do so by putting them between \eQ and \eE. This is different from Perl in
191 that $ and @ are handled as literals in \eQ...\eE sequences in PCRE, whereas in
192 Perl, $ and @ cause variable interpolation. Note the following examples:
193 .sp
194 Pattern PCRE matches Perl matches
195 .sp
196 .\" JOIN
197 \eQabc$xyz\eE abc$xyz abc followed by the
198 contents of $xyz
199 \eQabc\e$xyz\eE abc\e$xyz abc\e$xyz
200 \eQabc\eE\e$\eQxyz\eE abc$xyz abc$xyz
201 .sp
202 The \eQ...\eE sequence is recognized both inside and outside character classes.
203 .
204 .
205 .\" HTML <a name="digitsafterbackslash"></a>
206 .SS "Non-printing characters"
207 .rs
208 .sp
209 A second use of backslash provides a way of encoding non-printing characters
210 in patterns in a visible manner. There is no restriction on the appearance of
211 non-printing characters, apart from the binary zero that terminates a pattern,
212 but when a pattern is being prepared by text editing, it is often easier to use
213 one of the following escape sequences than the binary character it represents:
214 .sp
215 \ea alarm, that is, the BEL character (hex 07)
216 \ecx "control-x", where x is any character
217 \ee escape (hex 1B)
218 \ef formfeed (hex 0C)
219 \en linefeed (hex 0A)
220 \er carriage return (hex 0D)
221 \et tab (hex 09)
222 \eddd character with octal code ddd, or back reference
223 \exhh character with hex code hh
224 \ex{hhh..} character with hex code hhh..
225 .sp
226 The precise effect of \ecx is as follows: if x is a lower case letter, it
227 is converted to upper case. Then bit 6 of the character (hex 40) is inverted.
228 Thus \ecz becomes hex 1A, but \ec{ becomes hex 3B, while \ec; becomes hex
229 7B.
230 .P
231 After \ex, from zero to two hexadecimal digits are read (letters can be in
232 upper or lower case). Any number of hexadecimal digits may appear between \ex{
233 and }, but the value of the character code must be less than 256 in non-UTF-8
234 mode, and less than 2**31 in UTF-8 mode. That is, the maximum value in
235 hexadecimal is 7FFFFFFF. Note that this is bigger than the largest Unicode code
236 point, which is 10FFFF.
237 .P
238 If characters other than hexadecimal digits appear between \ex{ and }, or if
239 there is no terminating }, this form of escape is not recognized. Instead, the
240 initial \ex will be interpreted as a basic hexadecimal escape, with no
241 following digits, giving a character whose value is zero.
242 .P
243 Characters whose value is less than 256 can be defined by either of the two
244 syntaxes for \ex. There is no difference in the way they are handled. For
245 example, \exdc is exactly the same as \ex{dc}.
246 .P
247 After \e0 up to two further octal digits are read. If there are fewer than two
248 digits, just those that are present are used. Thus the sequence \e0\ex\e07
249 specifies two binary zeros followed by a BEL character (code value 7). Make
250 sure you supply two digits after the initial zero if the pattern character that
251 follows is itself an octal digit.
252 .P
253 The handling of a backslash followed by a digit other than 0 is complicated.
254 Outside a character class, PCRE reads it and any following digits as a decimal
255 number. If the number is less than 10, or if there have been at least that many
256 previous capturing left parentheses in the expression, the entire sequence is
257 taken as a \fIback reference\fP. A description of how this works is given
258 .\" HTML <a href="#backreferences">
259 .\" </a>
260 later,
261 .\"
262 following the discussion of
263 .\" HTML <a href="#subpattern">
264 .\" </a>
265 parenthesized subpatterns.
266 .\"
267 .P
268 Inside a character class, or if the decimal number is greater than 9 and there
269 have not been that many capturing subpatterns, PCRE re-reads up to three octal
270 digits following the backslash, and uses them to generate a data character. Any
271 subsequent digits stand for themselves. In non-UTF-8 mode, the value of a
272 character specified in octal must be less than \e400. In UTF-8 mode, values up
273 to \e777 are permitted. For example:
274 .sp
275 \e040 is another way of writing a space
276 .\" JOIN
277 \e40 is the same, provided there are fewer than 40
278 previous capturing subpatterns
279 \e7 is always a back reference
280 .\" JOIN
281 \e11 might be a back reference, or another way of
282 writing a tab
283 \e011 is always a tab
284 \e0113 is a tab followed by the character "3"
285 .\" JOIN
286 \e113 might be a back reference, otherwise the
287 character with octal code 113
288 .\" JOIN
289 \e377 might be a back reference, otherwise
290 the byte consisting entirely of 1 bits
291 .\" JOIN
292 \e81 is either a back reference, or a binary zero
293 followed by the two characters "8" and "1"
294 .sp
295 Note that octal values of 100 or greater must not be introduced by a leading
296 zero, because no more than three octal digits are ever read.
297 .P
298 All the sequences that define a single character value can be used both inside
299 and outside character classes. In addition, inside a character class, the
300 sequence \eb is interpreted as the backspace character (hex 08). The sequences
301 \eB, \eN, \eR, and \eX are not special inside a character class. Like any other
302 unrecognized escape sequences, they are treated as the literal characters "B",
303 "N", "R", and "X" by default, but cause an error if the PCRE_EXTRA option is
304 set. Outside a character class, these sequences have different meanings.
305 .
306 .
307 .SS "Absolute and relative back references"
308 .rs
309 .sp
310 The sequence \eg followed by an unsigned or a negative number, optionally
311 enclosed in braces, is an absolute or relative back reference. A named back
312 reference can be coded as \eg{name}. Back references are discussed
313 .\" HTML <a href="#backreferences">
314 .\" </a>
315 later,
316 .\"
317 following the discussion of
318 .\" HTML <a href="#subpattern">
319 .\" </a>
320 parenthesized subpatterns.
321 .\"
322 .
323 .
324 .SS "Absolute and relative subroutine calls"
325 .rs
326 .sp
327 For compatibility with Oniguruma, the non-Perl syntax \eg followed by a name or
328 a number enclosed either in angle brackets or single quotes, is an alternative
329 syntax for referencing a subpattern as a "subroutine". Details are discussed
330 .\" HTML <a href="#onigurumasubroutines">
331 .\" </a>
332 later.
333 .\"
334 Note that \eg{...} (Perl syntax) and \eg<...> (Oniguruma syntax) are \fInot\fP
335 synonymous. The former is a back reference; the latter is a
336 .\" HTML <a href="#subpatternsassubroutines">
337 .\" </a>
338 subroutine
339 .\"
340 call.
341 .
342 .
343 .SS "Generic character types"
344 .rs
345 .sp
346 Another use of backslash is for specifying generic character types:
347 .sp
348 \ed any decimal digit
349 \eD any character that is not a decimal digit
350 \eh any horizontal whitespace character
351 \eH any character that is not a horizontal whitespace character
352 \es any whitespace character
353 \eS any character that is not a whitespace character
354 \ev any vertical whitespace character
355 \eV any character that is not a vertical whitespace character
356 \ew any "word" character
357 \eW any "non-word" character
358 .sp
359 There is also the single sequence \eN, which matches a non-newline character.
360 This is the same as
361 .\" HTML <a href="#fullstopdot">
362 .\" </a>
363 the "." metacharacter
364 .\"
365 when PCRE_DOTALL is not set.
366 .P
367 Each pair of lower and upper case escape sequences partitions the complete set
368 of characters into two disjoint sets. Any given character matches one, and only
369 one, of each pair.
370 .P
371 These character type sequences can appear both inside and outside character
372 classes. They each match one character of the appropriate type. If the current
373 matching point is at the end of the subject string, all of them fail, since
374 there is no character to match.
375 .P
376 For compatibility with Perl, \es does not match the VT character (code 11).
377 This makes it different from the the POSIX "space" class. The \es characters
378 are HT (9), LF (10), FF (12), CR (13), and space (32). If "use locale;" is
379 included in a Perl script, \es may match the VT character. In PCRE, it never
380 does.
381 .P
382 In UTF-8 mode, characters with values greater than 128 never match \ed, \es, or
383 \ew, and always match \eD, \eS, and \eW. This is true even when Unicode
384 character property support is available. These sequences retain their original
385 meanings from before UTF-8 support was available, mainly for efficiency
386 reasons. Note that this also affects \eb, because it is defined in terms of \ew
387 and \eW.
388 .P
389 The sequences \eh, \eH, \ev, and \eV are Perl 5.10 features. In contrast to the
390 other sequences, these do match certain high-valued codepoints in UTF-8 mode.
391 The horizontal space characters are:
392 .sp
393 U+0009 Horizontal tab
394 U+0020 Space
395 U+00A0 Non-break space
396 U+1680 Ogham space mark
397 U+180E Mongolian vowel separator
398 U+2000 En quad
399 U+2001 Em quad
400 U+2002 En space
401 U+2003 Em space
402 U+2004 Three-per-em space
403 U+2005 Four-per-em space
404 U+2006 Six-per-em space
405 U+2007 Figure space
406 U+2008 Punctuation space
407 U+2009 Thin space
408 U+200A Hair space
409 U+202F Narrow no-break space
410 U+205F Medium mathematical space
411 U+3000 Ideographic space
412 .sp
413 The vertical space characters are:
414 .sp
415 U+000A Linefeed
416 U+000B Vertical tab
417 U+000C Formfeed
418 U+000D Carriage return
419 U+0085 Next line
420 U+2028 Line separator
421 U+2029 Paragraph separator
422 .P
423 A "word" character is an underscore or any character less than 256 that is a
424 letter or digit. The definition of letters and digits is controlled by PCRE's
425 low-valued character tables, and may vary if locale-specific matching is taking
426 place (see
427 .\" HTML <a href="pcreapi.html#localesupport">
428 .\" </a>
429 "Locale support"
430 .\"
431 in the
432 .\" HREF
433 \fBpcreapi\fP
434 .\"
435 page). For example, in a French locale such as "fr_FR" in Unix-like systems,
436 or "french" in Windows, some character codes greater than 128 are used for
437 accented letters, and these are matched by \ew. The use of locales with Unicode
438 is discouraged.
439 .
440 .
441 .\" HTML <a name="newlineseq"></a>
442 .SS "Newline sequences"
443 .rs
444 .sp
445 Outside a character class, by default, the escape sequence \eR matches any
446 Unicode newline sequence. This is a Perl 5.10 feature. In non-UTF-8 mode \eR is
447 equivalent to the following:
448 .sp
449 (?>\er\en|\en|\ex0b|\ef|\er|\ex85)
450 .sp
451 This is an example of an "atomic group", details of which are given
452 .\" HTML <a href="#atomicgroup">
453 .\" </a>
454 below.
455 .\"
456 This particular group matches either the two-character sequence CR followed by
457 LF, or one of the single characters LF (linefeed, U+000A), VT (vertical tab,
458 U+000B), FF (formfeed, U+000C), CR (carriage return, U+000D), or NEL (next
459 line, U+0085). The two-character sequence is treated as a single unit that
460 cannot be split.
461 .P
462 In UTF-8 mode, two additional characters whose codepoints are greater than 255
463 are added: LS (line separator, U+2028) and PS (paragraph separator, U+2029).
464 Unicode character property support is not needed for these characters to be
465 recognized.
466 .P
467 It is possible to restrict \eR to match only CR, LF, or CRLF (instead of the
468 complete set of Unicode line endings) by setting the option PCRE_BSR_ANYCRLF
469 either at compile time or when the pattern is matched. (BSR is an abbrevation
470 for "backslash R".) This can be made the default when PCRE is built; if this is
471 the case, the other behaviour can be requested via the PCRE_BSR_UNICODE option.
472 It is also possible to specify these settings by starting a pattern string with
473 one of the following sequences:
474 .sp
475 (*BSR_ANYCRLF) CR, LF, or CRLF only
476 (*BSR_UNICODE) any Unicode newline sequence
477 .sp
478 These override the default and the options given to \fBpcre_compile()\fP or
479 \fBpcre_compile2()\fP, but they can be overridden by options given to
480 \fBpcre_exec()\fP or \fBpcre_dfa_exec()\fP. Note that these special settings,
481 which are not Perl-compatible, are recognized only at the very start of a
482 pattern, and that they must be in upper case. If more than one of them is
483 present, the last one is used. They can be combined with a change of newline
484 convention, for example, a pattern can start with:
485 .sp
487 .sp
488 Inside a character class, \eR is treated as an unrecognized escape sequence,
489 and so matches the letter "R" by default, but causes an error if PCRE_EXTRA is
490 set.
491 .
492 .
493 .\" HTML <a name="uniextseq"></a>
494 .SS Unicode character properties
495 .rs
496 .sp
497 When PCRE is built with Unicode character property support, three additional
498 escape sequences that match characters with specific properties are available.
499 When not in UTF-8 mode, these sequences are of course limited to testing
500 characters whose codepoints are less than 256, but they do work in this mode.
501 The extra escape sequences are:
502 .sp
503 \ep{\fIxx\fP} a character with the \fIxx\fP property
504 \eP{\fIxx\fP} a character without the \fIxx\fP property
505 \eX an extended Unicode sequence
506 .sp
507 The property names represented by \fIxx\fP above are limited to the Unicode
508 script names, the general category properties, and "Any", which matches any
509 character (including newline). Other properties such as "InMusicalSymbols" are
510 not currently supported by PCRE. Note that \eP{Any} does not match any
511 characters, so always causes a match failure.
512 .P
513 Sets of Unicode characters are defined as belonging to certain scripts. A
514 character from one of these sets can be matched using a script name. For
515 example:
516 .sp
517 \ep{Greek}
518 \eP{Han}
519 .sp
520 Those that are not part of an identified script are lumped together as
521 "Common". The current list of scripts is:
522 .P
523 Arabic,
524 Armenian,
525 Avestan,
526 Balinese,
527 Bamum,
528 Bengali,
529 Bopomofo,
530 Braille,
531 Buginese,
532 Buhid,
533 Canadian_Aboriginal,
534 Carian,
535 Cham,
536 Cherokee,
537 Common,
538 Coptic,
539 Cuneiform,
540 Cypriot,
541 Cyrillic,
542 Deseret,
543 Devanagari,
544 Egyptian_Hieroglyphs,
545 Ethiopic,
546 Georgian,
547 Glagolitic,
548 Gothic,
549 Greek,
550 Gujarati,
551 Gurmukhi,
552 Han,
553 Hangul,
554 Hanunoo,
555 Hebrew,
556 Hiragana,
557 Imperial_Aramaic,
558 Inherited,
559 Inscriptional_Pahlavi,
560 Inscriptional_Parthian,
561 Javanese,
562 Kaithi,
563 Kannada,
564 Katakana,
565 Kayah_Li,
566 Kharoshthi,
567 Khmer,
568 Lao,
569 Latin,
570 Lepcha,
571 Limbu,
572 Linear_B,
573 Lisu,
574 Lycian,
575 Lydian,
576 Malayalam,
577 Meetei_Mayek,
578 Mongolian,
579 Myanmar,
580 New_Tai_Lue,
581 Nko,
582 Ogham,
583 Old_Italic,
584 Old_Persian,
585 Old_South_Arabian,
586 Old_Turkic,
587 Ol_Chiki,
588 Oriya,
589 Osmanya,
590 Phags_Pa,
591 Phoenician,
592 Rejang,
593 Runic,
594 Samaritan,
595 Saurashtra,
596 Shavian,
597 Sinhala,
598 Sundanese,
599 Syloti_Nagri,
600 Syriac,
601 Tagalog,
602 Tagbanwa,
603 Tai_Le,
604 Tai_Tham,
605 Tai_Viet,
606 Tamil,
607 Telugu,
608 Thaana,
609 Thai,
610 Tibetan,
611 Tifinagh,
612 Ugaritic,
613 Vai,
614 Yi.
615 .P
616 Each character has exactly one general category property, specified by a
617 two-letter abbreviation. For compatibility with Perl, negation can be specified
618 by including a circumflex between the opening brace and the property name. For
619 example, \ep{^Lu} is the same as \eP{Lu}.
620 .P
621 If only one letter is specified with \ep or \eP, it includes all the general
622 category properties that start with that letter. In this case, in the absence
623 of negation, the curly brackets in the escape sequence are optional; these two
624 examples have the same effect:
625 .sp
626 \ep{L}
627 \epL
628 .sp
629 The following general category property codes are supported:
630 .sp
631 C Other
632 Cc Control
633 Cf Format
634 Cn Unassigned
635 Co Private use
636 Cs Surrogate
637 .sp
638 L Letter
639 Ll Lower case letter
640 Lm Modifier letter
641 Lo Other letter
642 Lt Title case letter
643 Lu Upper case letter
644 .sp
645 M Mark
646 Mc Spacing mark
647 Me Enclosing mark
648 Mn Non-spacing mark
649 .sp
650 N Number
651 Nd Decimal number
652 Nl Letter number
653 No Other number
654 .sp
655 P Punctuation
656 Pc Connector punctuation
657 Pd Dash punctuation
658 Pe Close punctuation
659 Pf Final punctuation
660 Pi Initial punctuation
661 Po Other punctuation
662 Ps Open punctuation
663 .sp
664 S Symbol
665 Sc Currency symbol
666 Sk Modifier symbol
667 Sm Mathematical symbol
668 So Other symbol
669 .sp
670 Z Separator
671 Zl Line separator
672 Zp Paragraph separator
673 Zs Space separator
674 .sp
675 The special property L& is also supported: it matches a character that has
676 the Lu, Ll, or Lt property, in other words, a letter that is not classified as
677 a modifier or "other".
678 .P
679 The Cs (Surrogate) property applies only to characters in the range U+D800 to
680 U+DFFF. Such characters are not valid in UTF-8 strings (see RFC 3629) and so
681 cannot be tested by PCRE, unless UTF-8 validity checking has been turned off
682 (see the discussion of PCRE_NO_UTF8_CHECK in the
683 .\" HREF
684 \fBpcreapi\fP
685 .\"
686 page). Perl does not support the Cs property.
687 .P
688 The long synonyms for property names that Perl supports (such as \ep{Letter})
689 are not supported by PCRE, nor is it permitted to prefix any of these
690 properties with "Is".
691 .P
692 No character that is in the Unicode table has the Cn (unassigned) property.
693 Instead, this property is assumed for any code point that is not in the
694 Unicode table.
695 .P
696 Specifying caseless matching does not affect these escape sequences. For
697 example, \ep{Lu} always matches only upper case letters.
698 .P
699 The \eX escape matches any number of Unicode characters that form an extended
700 Unicode sequence. \eX is equivalent to
701 .sp
702 (?>\ePM\epM*)
703 .sp
704 That is, it matches a character without the "mark" property, followed by zero
705 or more characters with the "mark" property, and treats the sequence as an
706 atomic group
707 .\" HTML <a href="#atomicgroup">
708 .\" </a>
709 (see below).
710 .\"
711 Characters with the "mark" property are typically accents that affect the
712 preceding character. None of them have codepoints less than 256, so in
713 non-UTF-8 mode \eX matches any one character.
714 .P
715 Matching characters by Unicode property is not fast, because PCRE has to search
716 a structure that contains data for over fifteen thousand characters. That is
717 why the traditional escape sequences such as \ed and \ew do not use Unicode
718 properties in PCRE.
719 .
720 .
721 .\" HTML <a name="resetmatchstart"></a>
722 .SS "Resetting the match start"
723 .rs
724 .sp
725 The escape sequence \eK, which is a Perl 5.10 feature, causes any previously
726 matched characters not to be included in the final matched sequence. For
727 example, the pattern:
728 .sp
729 foo\eKbar
730 .sp
731 matches "foobar", but reports that it has matched "bar". This feature is
732 similar to a lookbehind assertion
733 .\" HTML <a href="#lookbehind">
734 .\" </a>
735 (described below).
736 .\"
737 However, in this case, the part of the subject before the real match does not
738 have to be of fixed length, as lookbehind assertions do. The use of \eK does
739 not interfere with the setting of
740 .\" HTML <a href="#subpattern">
741 .\" </a>
742 captured substrings.
743 .\"
744 For example, when the pattern
745 .sp
746 (foo)\eKbar
747 .sp
748 matches "foobar", the first substring is still set to "foo".
749 .P
750 Perl documents that the use of \eK within assertions is "not well defined". In
751 PCRE, \eK is acted upon when it occurs inside positive assertions, but is
752 ignored in negative assertions.
753 .
754 .
755 .\" HTML <a name="smallassertions"></a>
756 .SS "Simple assertions"
757 .rs
758 .sp
759 The final use of backslash is for certain simple assertions. An assertion
760 specifies a condition that has to be met at a particular point in a match,
761 without consuming any characters from the subject string. The use of
762 subpatterns for more complicated assertions is described
763 .\" HTML <a href="#bigassertions">
764 .\" </a>
765 below.
766 .\"
767 The backslashed assertions are:
768 .sp
769 \eb matches at a word boundary
770 \eB matches when not at a word boundary
771 \eA matches at the start of the subject
772 \eZ matches at the end of the subject
773 also matches before a newline at the end of the subject
774 \ez matches only at the end of the subject
775 \eG matches at the first matching position in the subject
776 .sp
777 Inside a character class, \eb has a different meaning; it matches the backspace
778 character. If any other of these assertions appears in a character class, by
779 default it matches the corresponding literal character (for example, \eB
780 matches the letter B). However, if the PCRE_EXTRA option is set, an "invalid
781 escape sequence" error is generated instead.
782 .P
783 A word boundary is a position in the subject string where the current character
784 and the previous character do not both match \ew or \eW (i.e. one matches
785 \ew and the other matches \eW), or the start or end of the string if the
786 first or last character matches \ew, respectively. Neither PCRE nor Perl has a
787 separte "start of word" or "end of word" metasequence. However, whatever
788 follows \eb normally determines which it is. For example, the fragment
789 \eba matches "a" at the start of a word.
790 .P
791 The \eA, \eZ, and \ez assertions differ from the traditional circumflex and
792 dollar (described in the next section) in that they only ever match at the very
793 start and end of the subject string, whatever options are set. Thus, they are
794 independent of multiline mode. These three assertions are not affected by the
795 PCRE_NOTBOL or PCRE_NOTEOL options, which affect only the behaviour of the
796 circumflex and dollar metacharacters. However, if the \fIstartoffset\fP
797 argument of \fBpcre_exec()\fP is non-zero, indicating that matching is to start
798 at a point other than the beginning of the subject, \eA can never match. The
799 difference between \eZ and \ez is that \eZ matches before a newline at the end
800 of the string as well as at the very end, whereas \ez matches only at the end.
801 .P
802 The \eG assertion is true only when the current matching position is at the
803 start point of the match, as specified by the \fIstartoffset\fP argument of
804 \fBpcre_exec()\fP. It differs from \eA when the value of \fIstartoffset\fP is
805 non-zero. By calling \fBpcre_exec()\fP multiple times with appropriate
806 arguments, you can mimic Perl's /g option, and it is in this kind of
807 implementation where \eG can be useful.
808 .P
809 Note, however, that PCRE's interpretation of \eG, as the start of the current
810 match, is subtly different from Perl's, which defines it as the end of the
811 previous match. In Perl, these can be different when the previously matched
812 string was empty. Because PCRE does just one match at a time, it cannot
813 reproduce this behaviour.
814 .P
815 If all the alternatives of a pattern begin with \eG, the expression is anchored
816 to the starting match position, and the "anchored" flag is set in the compiled
817 regular expression.
818 .
819 .
821 .rs
822 .sp
823 Outside a character class, in the default matching mode, the circumflex
824 character is an assertion that is true only if the current matching point is
825 at the start of the subject string. If the \fIstartoffset\fP argument of
826 \fBpcre_exec()\fP is non-zero, circumflex can never match if the PCRE_MULTILINE
827 option is unset. Inside a character class, circumflex has an entirely different
828 meaning
829 .\" HTML <a href="#characterclass">
830 .\" </a>
831 (see below).
832 .\"
833 .P
834 Circumflex need not be the first character of the pattern if a number of
835 alternatives are involved, but it should be the first thing in each alternative
836 in which it appears if the pattern is ever to match that branch. If all
837 possible alternatives start with a circumflex, that is, if the pattern is
838 constrained to match only at the start of the subject, it is said to be an
839 "anchored" pattern. (There are also other constructs that can cause a pattern
840 to be anchored.)
841 .P
842 A dollar character is an assertion that is true only if the current matching
843 point is at the end of the subject string, or immediately before a newline
844 at the end of the string (by default). Dollar need not be the last character of
845 the pattern if a number of alternatives are involved, but it should be the last
846 item in any branch in which it appears. Dollar has no special meaning in a
847 character class.
848 .P
849 The meaning of dollar can be changed so that it matches only at the very end of
850 the string, by setting the PCRE_DOLLAR_ENDONLY option at compile time. This
851 does not affect the \eZ assertion.
852 .P
853 The meanings of the circumflex and dollar characters are changed if the
854 PCRE_MULTILINE option is set. When this is the case, a circumflex matches
855 immediately after internal newlines as well as at the start of the subject
856 string. It does not match after a newline that ends the string. A dollar
857 matches before any newlines in the string, as well as at the very end, when
858 PCRE_MULTILINE is set. When newline is specified as the two-character
859 sequence CRLF, isolated CR and LF characters do not indicate newlines.
860 .P
861 For example, the pattern /^abc$/ matches the subject string "def\enabc" (where
862 \en represents a newline) in multiline mode, but not otherwise. Consequently,
863 patterns that are anchored in single line mode because all branches start with
864 ^ are not anchored in multiline mode, and a match for circumflex is possible
865 when the \fIstartoffset\fP argument of \fBpcre_exec()\fP is non-zero. The
866 PCRE_DOLLAR_ENDONLY option is ignored if PCRE_MULTILINE is set.
867 .P
868 Note that the sequences \eA, \eZ, and \ez can be used to match the start and
869 end of the subject in both modes, and if all branches of a pattern start with
870 \eA it is always anchored, whether or not PCRE_MULTILINE is set.
871 .
872 .
873 .\" HTML <a name="fullstopdot"></a>
875 .rs
876 .sp
877 Outside a character class, a dot in the pattern matches any one character in
878 the subject string except (by default) a character that signifies the end of a
879 line. In UTF-8 mode, the matched character may be more than one byte long.
880 .P
881 When a line ending is defined as a single character, dot never matches that
882 character; when the two-character sequence CRLF is used, dot does not match CR
883 if it is immediately followed by LF, but otherwise it matches all characters
884 (including isolated CRs and LFs). When any Unicode line endings are being
885 recognized, dot does not match CR or LF or any of the other line ending
886 characters.
887 .P
888 The behaviour of dot with regard to newlines can be changed. If the PCRE_DOTALL
889 option is set, a dot matches any one character, without exception. If the
890 two-character sequence CRLF is present in the subject string, it takes two dots
891 to match it.
892 .P
893 The handling of dot is entirely independent of the handling of circumflex and
894 dollar, the only relationship being that they both involve newlines. Dot has no
895 special meaning in a character class.
896 .P
897 The escape sequence \eN always behaves as a dot does when PCRE_DOTALL is not
898 set. In other words, it matches any one character except one that signifies the
899 end of a line.
900 .
901 .
903 .rs
904 .sp
905 Outside a character class, the escape sequence \eC matches any one byte, both
906 in and out of UTF-8 mode. Unlike a dot, it always matches any line-ending
907 characters. The feature is provided in Perl in order to match individual bytes
908 in UTF-8 mode. Because it breaks up UTF-8 characters into individual bytes,
909 what remains in the string may be a malformed UTF-8 string. For this reason,
910 the \eC escape sequence is best avoided.
911 .P
912 PCRE does not allow \eC to appear in lookbehind assertions
913 .\" HTML <a href="#lookbehind">
914 .\" </a>
915 (described below),
916 .\"
917 because in UTF-8 mode this would make it impossible to calculate the length of
918 the lookbehind.
919 .
920 .
921 .\" HTML <a name="characterclass"></a>
923 .rs
924 .sp
925 An opening square bracket introduces a character class, terminated by a closing
926 square bracket. A closing square bracket on its own is not special by default.
927 However, if the PCRE_JAVASCRIPT_COMPAT option is set, a lone closing square
928 bracket causes a compile-time error. If a closing square bracket is required as
929 a member of the class, it should be the first data character in the class
930 (after an initial circumflex, if present) or escaped with a backslash.
931 .P
932 A character class matches a single character in the subject. In UTF-8 mode, the
933 character may be more than one byte long. A matched character must be in the
934 set of characters defined by the class, unless the first character in the class
935 definition is a circumflex, in which case the subject character must not be in
936 the set defined by the class. If a circumflex is actually required as a member
937 of the class, ensure it is not the first character, or escape it with a
938 backslash.
939 .P
940 For example, the character class [aeiou] matches any lower case vowel, while
941 [^aeiou] matches any character that is not a lower case vowel. Note that a
942 circumflex is just a convenient notation for specifying the characters that
943 are in the class by enumerating those that are not. A class that starts with a
944 circumflex is not an assertion; it still consumes a character from the subject
945 string, and therefore it fails if the current pointer is at the end of the
946 string.
947 .P
948 In UTF-8 mode, characters with values greater than 255 can be included in a
949 class as a literal string of bytes, or by using the \ex{ escaping mechanism.
950 .P
951 When caseless matching is set, any letters in a class represent both their
952 upper case and lower case versions, so for example, a caseless [aeiou] matches
953 "A" as well as "a", and a caseless [^aeiou] does not match "A", whereas a
954 caseful version would. In UTF-8 mode, PCRE always understands the concept of
955 case for characters whose values are less than 128, so caseless matching is
956 always possible. For characters with higher values, the concept of case is
957 supported if PCRE is compiled with Unicode property support, but not otherwise.
958 If you want to use caseless matching in UTF8-mode for characters 128 and above,
959 you must ensure that PCRE is compiled with Unicode property support as well as
960 with UTF-8 support.
961 .P
962 Characters that might indicate line breaks are never treated in any special way
963 when matching character classes, whatever line-ending sequence is in use, and
964 whatever setting of the PCRE_DOTALL and PCRE_MULTILINE options is used. A class
965 such as [^a] always matches one of these characters.
966 .P
967 The minus (hyphen) character can be used to specify a range of characters in a
968 character class. For example, [d-m] matches any letter between d and m,
969 inclusive. If a minus character is required in a class, it must be escaped with
970 a backslash or appear in a position where it cannot be interpreted as
971 indicating a range, typically as the first or last character in the class.
972 .P
973 It is not possible to have the literal character "]" as the end character of a
974 range. A pattern such as [W-]46] is interpreted as a class of two characters
975 ("W" and "-") followed by a literal string "46]", so it would match "W46]" or
976 "-46]". However, if the "]" is escaped with a backslash it is interpreted as
977 the end of range, so [W-\e]46] is interpreted as a class containing a range
978 followed by two other characters. The octal or hexadecimal representation of
979 "]" can also be used to end a range.
980 .P
981 Ranges operate in the collating sequence of character values. They can also be
982 used for characters specified numerically, for example [\e000-\e037]. In UTF-8
983 mode, ranges can include characters whose values are greater than 255, for
984 example [\ex{100}-\ex{2ff}].
985 .P
986 If a range that includes letters is used when caseless matching is set, it
987 matches the letters in either case. For example, [W-c] is equivalent to
988 [][\e\e^_`wxyzabc], matched caselessly, and in non-UTF-8 mode, if character
989 tables for a French locale are in use, [\exc8-\excb] matches accented E
990 characters in both cases. In UTF-8 mode, PCRE supports the concept of case for
991 characters with values greater than 128 only when it is compiled with Unicode
992 property support.
993 .P
994 The character types \ed, \eD, \ep, \eP, \es, \eS, \ew, and \eW may also appear
995 in a character class, and add the characters that they match to the class. For
996 example, [\edABCDEF] matches any hexadecimal digit. A circumflex can
997 conveniently be used with the upper case character types to specify a more
998 restricted set of characters than the matching lower case type. For example,
999 the class [^\eW_] matches any letter or digit, but not underscore.
1000 .P
1001 The only metacharacters that are recognized in character classes are backslash,
1002 hyphen (only where it can be interpreted as specifying a range), circumflex
1003 (only at the start), opening square bracket (only when it can be interpreted as
1004 introducing a POSIX class name - see the next section), and the terminating
1005 closing square bracket. However, escaping other non-alphanumeric characters
1006 does no harm.
1007 .
1008 .
1010 .rs
1011 .sp
1012 Perl supports the POSIX notation for character classes. This uses names
1013 enclosed by [: and :] within the enclosing square brackets. PCRE also supports
1014 this notation. For example,
1015 .sp
1016 [01[:alpha:]%]
1017 .sp
1018 matches "0", "1", any alphabetic character, or "%". The supported class names
1019 are
1020 .sp
1021 alnum letters and digits
1022 alpha letters
1023 ascii character codes 0 - 127
1024 blank space or tab only
1025 cntrl control characters
1026 digit decimal digits (same as \ed)
1027 graph printing characters, excluding space
1028 lower lower case letters
1029 print printing characters, including space
1030 punct printing characters, excluding letters and digits
1031 space white space (not quite the same as \es)
1032 upper upper case letters
1033 word "word" characters (same as \ew)
1034 xdigit hexadecimal digits
1035 .sp
1036 The "space" characters are HT (9), LF (10), VT (11), FF (12), CR (13), and
1037 space (32). Notice that this list includes the VT character (code 11). This
1038 makes "space" different to \es, which does not include VT (for Perl
1039 compatibility).
1040 .P
1041 The name "word" is a Perl extension, and "blank" is a GNU extension from Perl
1042 5.8. Another Perl extension is negation, which is indicated by a ^ character
1043 after the colon. For example,
1044 .sp
1045 [12[:^digit:]]
1046 .sp
1047 matches "1", "2", or any non-digit. PCRE (and Perl) also recognize the POSIX
1048 syntax [.ch.] and [=ch=] where "ch" is a "collating element", but these are not
1049 supported, and an error is given if they are encountered.
1050 .P
1051 In UTF-8 mode, characters with values greater than 128 do not match any of
1052 the POSIX character classes.
1053 .
1054 .
1056 .rs
1057 .sp
1058 Vertical bar characters are used to separate alternative patterns. For example,
1059 the pattern
1060 .sp
1061 gilbert|sullivan
1062 .sp
1063 matches either "gilbert" or "sullivan". Any number of alternatives may appear,
1064 and an empty alternative is permitted (matching the empty string). The matching
1065 process tries each alternative in turn, from left to right, and the first one
1066 that succeeds is used. If the alternatives are within a subpattern
1067 .\" HTML <a href="#subpattern">
1068 .\" </a>
1069 (defined below),
1070 .\"
1071 "succeeds" means matching the rest of the main pattern as well as the
1072 alternative in the subpattern.
1073 .
1074 .
1076 .rs
1077 .sp
1078 The settings of the PCRE_CASELESS, PCRE_MULTILINE, PCRE_DOTALL, and
1079 PCRE_EXTENDED options (which are Perl-compatible) can be changed from within
1080 the pattern by a sequence of Perl option letters enclosed between "(?" and ")".
1081 The option letters are
1082 .sp
1083 i for PCRE_CASELESS
1085 s for PCRE_DOTALL
1086 x for PCRE_EXTENDED
1087 .sp
1088 For example, (?im) sets caseless, multiline matching. It is also possible to
1089 unset these options by preceding the letter with a hyphen, and a combined
1090 setting and unsetting such as (?im-sx), which sets PCRE_CASELESS and
1091 PCRE_MULTILINE while unsetting PCRE_DOTALL and PCRE_EXTENDED, is also
1092 permitted. If a letter appears both before and after the hyphen, the option is
1093 unset.
1094 .P
1095 The PCRE-specific options PCRE_DUPNAMES, PCRE_UNGREEDY, and PCRE_EXTRA can be
1096 changed in the same way as the Perl-compatible options by using the characters
1097 J, U and X respectively.
1098 .P
1099 When one of these option changes occurs at top level (that is, not inside
1100 subpattern parentheses), the change applies to the remainder of the pattern
1101 that follows. If the change is placed right at the start of a pattern, PCRE
1102 extracts it into the global options (and it will therefore show up in data
1103 extracted by the \fBpcre_fullinfo()\fP function).
1104 .P
1105 An option change within a subpattern (see below for a description of
1106 subpatterns) affects only that part of the current pattern that follows it, so
1107 .sp
1108 (a(?i)b)c
1109 .sp
1110 matches abc and aBc and no other strings (assuming PCRE_CASELESS is not used).
1111 By this means, options can be made to have different settings in different
1112 parts of the pattern. Any changes made in one alternative do carry on
1113 into subsequent branches within the same subpattern. For example,
1114 .sp
1115 (a(?i)b|c)
1116 .sp
1117 matches "ab", "aB", "c", and "C", even though when matching "C" the first
1118 branch is abandoned before the option setting. This is because the effects of
1119 option settings happen at compile time. There would be some very weird
1120 behaviour otherwise.
1121 .P
1122 \fBNote:\fP There are other PCRE-specific options that can be set by the
1123 application when the compile or match functions are called. In some cases the
1124 pattern can contain special leading sequences such as (*CRLF) to override what
1125 the application has set or what has been defaulted. Details are given in the
1126 section entitled
1127 .\" HTML <a href="#newlineseq">
1128 .\" </a>
1129 "Newline sequences"
1130 .\"
1131 above. There is also the (*UTF8) leading sequence that can be used to set UTF-8
1132 mode; this is equivalent to setting the PCRE_UTF8 option.
1133 .
1134 .
1135 .\" HTML <a name="subpattern"></a>
1137 .rs
1138 .sp
1139 Subpatterns are delimited by parentheses (round brackets), which can be nested.
1140 Turning part of a pattern into a subpattern does two things:
1141 .sp
1142 1. It localizes a set of alternatives. For example, the pattern
1143 .sp
1144 cat(aract|erpillar|)
1145 .sp
1146 matches one of the words "cat", "cataract", or "caterpillar". Without the
1147 parentheses, it would match "cataract", "erpillar" or an empty string.
1148 .sp
1149 2. It sets up the subpattern as a capturing subpattern. This means that, when
1150 the whole pattern matches, that portion of the subject string that matched the
1151 subpattern is passed back to the caller via the \fIovector\fP argument of
1152 \fBpcre_exec()\fP. Opening parentheses are counted from left to right (starting
1153 from 1) to obtain numbers for the capturing subpatterns.
1154 .P
1155 For example, if the string "the red king" is matched against the pattern
1156 .sp
1157 the ((red|white) (king|queen))
1158 .sp
1159 the captured substrings are "red king", "red", and "king", and are numbered 1,
1160 2, and 3, respectively.
1161 .P
1162 The fact that plain parentheses fulfil two functions is not always helpful.
1163 There are often times when a grouping subpattern is required without a
1164 capturing requirement. If an opening parenthesis is followed by a question mark
1165 and a colon, the subpattern does not do any capturing, and is not counted when
1166 computing the number of any subsequent capturing subpatterns. For example, if
1167 the string "the white queen" is matched against the pattern
1168 .sp
1169 the ((?:red|white) (king|queen))
1170 .sp
1171 the captured substrings are "white queen" and "queen", and are numbered 1 and
1172 2. The maximum number of capturing subpatterns is 65535.
1173 .P
1174 As a convenient shorthand, if any option settings are required at the start of
1175 a non-capturing subpattern, the option letters may appear between the "?" and
1176 the ":". Thus the two patterns
1177 .sp
1178 (?i:saturday|sunday)
1179 (?:(?i)saturday|sunday)
1180 .sp
1181 match exactly the same set of strings. Because alternative branches are tried
1182 from left to right, and options are not reset until the end of the subpattern
1183 is reached, an option setting in one branch does affect subsequent branches, so
1184 the above patterns match "SUNDAY" as well as "Saturday".
1185 .
1186 .
1187 .\" HTML <a name="dupsubpatternnumber"></a>
1189 .rs
1190 .sp
1191 Perl 5.10 introduced a feature whereby each alternative in a subpattern uses
1192 the same numbers for its capturing parentheses. Such a subpattern starts with
1193 (?| and is itself a non-capturing subpattern. For example, consider this
1194 pattern:
1195 .sp
1196 (?|(Sat)ur|(Sun))day
1197 .sp
1198 Because the two alternatives are inside a (?| group, both sets of capturing
1199 parentheses are numbered one. Thus, when the pattern matches, you can look
1200 at captured substring number one, whichever alternative matched. This construct
1201 is useful when you want to capture part, but not all, of one of a number of
1202 alternatives. Inside a (?| group, parentheses are numbered as usual, but the
1203 number is reset at the start of each branch. The numbers of any capturing
1204 buffers that follow the subpattern start after the highest number used in any
1205 branch. The following example is taken from the Perl documentation.
1206 The numbers underneath show in which buffer the captured content will be
1207 stored.
1208 .sp
1209 # before ---------------branch-reset----------- after
1210 / ( a ) (?| x ( y ) z | (p (q) r) | (t) u (v) ) ( z ) /x
1211 # 1 2 2 3 2 3 4
1212 .sp
1213 A back reference to a numbered subpattern uses the most recent value that is
1214 set for that number by any subpattern. The following pattern matches "abcabc"
1215 or "defdef":
1216 .sp
1217 /(?|(abc)|(def))\e1/
1218 .sp
1219 In contrast, a recursive or "subroutine" call to a numbered subpattern always
1220 refers to the first one in the pattern with the given number. The following
1221 pattern matches "abcabc" or "defabc":
1222 .sp
1223 /(?|(abc)|(def))(?1)/
1224 .sp
1225 If a
1226 .\" HTML <a href="#conditions">
1227 .\" </a>
1228 condition test
1229 .\"
1230 for a subpattern's having matched refers to a non-unique number, the test is
1231 true if any of the subpatterns of that number have matched.
1232 .P
1233 An alternative approach to using this "branch reset" feature is to use
1234 duplicate named subpatterns, as described in the next section.
1235 .
1236 .
1238 .rs
1239 .sp
1240 Identifying capturing parentheses by number is simple, but it can be very hard
1241 to keep track of the numbers in complicated regular expressions. Furthermore,
1242 if an expression is modified, the numbers may change. To help with this
1243 difficulty, PCRE supports the naming of subpatterns. This feature was not
1244 added to Perl until release 5.10. Python had the feature earlier, and PCRE
1245 introduced it at release 4.0, using the Python syntax. PCRE now supports both
1246 the Perl and the Python syntax. Perl allows identically numbered subpatterns to
1247 have different names, but PCRE does not.
1248 .P
1249 In PCRE, a subpattern can be named in one of three ways: (?<name>...) or
1250 (?'name'...) as in Perl, or (?P<name>...) as in Python. References to capturing
1251 parentheses from other parts of the pattern, such as
1252 .\" HTML <a href="#backreferences">
1253 .\" </a>
1254 back references,
1255 .\"
1256 .\" HTML <a href="#recursion">
1257 .\" </a>
1258 recursion,
1259 .\"
1260 and
1261 .\" HTML <a href="#conditions">
1262 .\" </a>
1263 conditions,
1264 .\"
1265 can be made by name as well as by number.
1266 .P
1267 Names consist of up to 32 alphanumeric characters and underscores. Named
1268 capturing parentheses are still allocated numbers as well as names, exactly as
1269 if the names were not present. The PCRE API provides function calls for
1270 extracting the name-to-number translation table from a compiled pattern. There
1271 is also a convenience function for extracting a captured substring by name.
1272 .P
1273 By default, a name must be unique within a pattern, but it is possible to relax
1274 this constraint by setting the PCRE_DUPNAMES option at compile time. (Duplicate
1275 names are also always permitted for subpatterns with the same number, set up as
1276 described in the previous section.) Duplicate names can be useful for patterns
1277 where only one instance of the named parentheses can match. Suppose you want to
1278 match the name of a weekday, either as a 3-letter abbreviation or as the full
1279 name, and in both cases you want to extract the abbreviation. This pattern
1280 (ignoring the line breaks) does the job:
1281 .sp
1282 (?<DN>Mon|Fri|Sun)(?:day)?|
1283 (?<DN>Tue)(?:sday)?|
1284 (?<DN>Wed)(?:nesday)?|
1285 (?<DN>Thu)(?:rsday)?|
1286 (?<DN>Sat)(?:urday)?
1287 .sp
1288 There are five capturing substrings, but only one is ever set after a match.
1289 (An alternative way of solving this problem is to use a "branch reset"
1290 subpattern, as described in the previous section.)
1291 .P
1292 The convenience function for extracting the data by name returns the substring
1293 for the first (and in this example, the only) subpattern of that name that
1294 matched. This saves searching to find which numbered subpattern it was.
1295 .P
1296 If you make a back reference to a non-unique named subpattern from elsewhere in
1297 the pattern, the one that corresponds to the first occurrence of the name is
1298 used. In the absence of duplicate numbers (see the previous section) this is
1299 the one with the lowest number. If you use a named reference in a condition
1300 test (see the
1301 .\"
1302 .\" HTML <a href="#conditions">
1303 .\" </a>
1304 section about conditions
1305 .\"
1306 below), either to check whether a subpattern has matched, or to check for
1307 recursion, all subpatterns with the same name are tested. If the condition is
1308 true for any one of them, the overall condition is true. This is the same
1309 behaviour as testing by number. For further details of the interfaces for
1310 handling named subpatterns, see the
1311 .\" HREF
1312 \fBpcreapi\fP
1313 .\"
1314 documentation.
1315 .P
1316 \fBWarning:\fP You cannot use different names to distinguish between two
1317 subpatterns with the same number because PCRE uses only the numbers when
1318 matching. For this reason, an error is given at compile time if different names
1319 are given to subpatterns with the same number. However, you can give the same
1320 name to subpatterns with the same number, even when PCRE_DUPNAMES is not set.
1321 .
1322 .
1324 .rs
1325 .sp
1326 Repetition is specified by quantifiers, which can follow any of the following
1327 items:
1328 .sp
1329 a literal data character
1330 the dot metacharacter
1331 the \eC escape sequence
1332 the \eX escape sequence (in UTF-8 mode with Unicode properties)
1333 the \eR escape sequence
1334 an escape such as \ed that matches a single character
1335 a character class
1336 a back reference (see next section)
1337 a parenthesized subpattern (unless it is an assertion)
1338 a recursive or "subroutine" call to a subpattern
1339 .sp
1340 The general repetition quantifier specifies a minimum and maximum number of
1341 permitted matches, by giving the two numbers in curly brackets (braces),
1342 separated by a comma. The numbers must be less than 65536, and the first must
1343 be less than or equal to the second. For example:
1344 .sp
1345 z{2,4}
1346 .sp
1347 matches "zz", "zzz", or "zzzz". A closing brace on its own is not a special
1348 character. If the second number is omitted, but the comma is present, there is
1349 no upper limit; if the second number and the comma are both omitted, the
1350 quantifier specifies an exact number of required matches. Thus
1351 .sp
1352 [aeiou]{3,}
1353 .sp
1354 matches at least 3 successive vowels, but may match many more, while
1355 .sp
1356 \ed{8}
1357 .sp
1358 matches exactly 8 digits. An opening curly bracket that appears in a position
1359 where a quantifier is not allowed, or one that does not match the syntax of a
1360 quantifier, is taken as a literal character. For example, {,6} is not a
1361 quantifier, but a literal string of four characters.
1362 .P
1363 In UTF-8 mode, quantifiers apply to UTF-8 characters rather than to individual
1364 bytes. Thus, for example, \ex{100}{2} matches two UTF-8 characters, each of
1365 which is represented by a two-byte sequence. Similarly, when Unicode property
1366 support is available, \eX{3} matches three Unicode extended sequences, each of
1367 which may be several bytes long (and they may be of different lengths).
1368 .P
1369 The quantifier {0} is permitted, causing the expression to behave as if the
1370 previous item and the quantifier were not present. This may be useful for
1371 subpatterns that are referenced as
1372 .\" HTML <a href="#subpatternsassubroutines">
1373 .\" </a>
1374 subroutines
1375 .\"
1376 from elsewhere in the pattern. Items other than subpatterns that have a {0}
1377 quantifier are omitted from the compiled pattern.
1378 .P
1379 For convenience, the three most common quantifiers have single-character
1380 abbreviations:
1381 .sp
1382 * is equivalent to {0,}
1383 + is equivalent to {1,}
1384 ? is equivalent to {0,1}
1385 .sp
1386 It is possible to construct infinite loops by following a subpattern that can
1387 match no characters with a quantifier that has no upper limit, for example:
1388 .sp
1389 (a?)*
1390 .sp
1391 Earlier versions of Perl and PCRE used to give an error at compile time for
1392 such patterns. However, because there are cases where this can be useful, such
1393 patterns are now accepted, but if any repetition of the subpattern does in fact
1394 match no characters, the loop is forcibly broken.
1395 .P
1396 By default, the quantifiers are "greedy", that is, they match as much as
1397 possible (up to the maximum number of permitted times), without causing the
1398 rest of the pattern to fail. The classic example of where this gives problems
1399 is in trying to match comments in C programs. These appear between /* and */
1400 and within the comment, individual * and / characters may appear. An attempt to
1401 match C comments by applying the pattern
1402 .sp
1403 /\e*.*\e*/
1404 .sp
1405 to the string
1406 .sp
1407 /* first comment */ not comment /* second comment */
1408 .sp
1409 fails, because it matches the entire string owing to the greediness of the .*
1410 item.
1411 .P
1412 However, if a quantifier is followed by a question mark, it ceases to be
1413 greedy, and instead matches the minimum number of times possible, so the
1414 pattern
1415 .sp
1416 /\e*.*?\e*/
1417 .sp
1418 does the right thing with the C comments. The meaning of the various
1419 quantifiers is not otherwise changed, just the preferred number of matches.
1420 Do not confuse this use of question mark with its use as a quantifier in its
1421 own right. Because it has two uses, it can sometimes appear doubled, as in
1422 .sp
1423 \ed??\ed
1424 .sp
1425 which matches one digit by preference, but can match two if that is the only
1426 way the rest of the pattern matches.
1427 .P
1428 If the PCRE_UNGREEDY option is set (an option that is not available in Perl),
1429 the quantifiers are not greedy by default, but individual ones can be made
1430 greedy by following them with a question mark. In other words, it inverts the
1431 default behaviour.
1432 .P
1433 When a parenthesized subpattern is quantified with a minimum repeat count that
1434 is greater than 1 or with a limited maximum, more memory is required for the
1435 compiled pattern, in proportion to the size of the minimum or maximum.
1436 .P
1437 If a pattern starts with .* or .{0,} and the PCRE_DOTALL option (equivalent
1438 to Perl's /s) is set, thus allowing the dot to match newlines, the pattern is
1439 implicitly anchored, because whatever follows will be tried against every
1440 character position in the subject string, so there is no point in retrying the
1441 overall match at any position after the first. PCRE normally treats such a
1442 pattern as though it were preceded by \eA.
1443 .P
1444 In cases where it is known that the subject string contains no newlines, it is
1445 worth setting PCRE_DOTALL in order to obtain this optimization, or
1446 alternatively using ^ to indicate anchoring explicitly.
1447 .P
1448 However, there is one situation where the optimization cannot be used. When .*
1449 is inside capturing parentheses that are the subject of a back reference
1450 elsewhere in the pattern, a match at the start may fail where a later one
1451 succeeds. Consider, for example:
1452 .sp
1453 (.*)abc\e1
1454 .sp
1455 If the subject is "xyz123abc123" the match point is the fourth character. For
1456 this reason, such a pattern is not implicitly anchored.
1457 .P
1458 When a capturing subpattern is repeated, the value captured is the substring
1459 that matched the final iteration. For example, after
1460 .sp
1461 (tweedle[dume]{3}\es*)+
1462 .sp
1463 has matched "tweedledum tweedledee" the value of the captured substring is
1464 "tweedledee". However, if there are nested capturing subpatterns, the
1465 corresponding captured values may have been set in previous iterations. For
1466 example, after
1467 .sp
1468 /(a|(b))+/
1469 .sp
1470 matches "aba" the value of the second captured substring is "b".
1471 .
1472 .
1473 .\" HTML <a name="atomicgroup"></a>
1475 .rs
1476 .sp
1477 With both maximizing ("greedy") and minimizing ("ungreedy" or "lazy")
1478 repetition, failure of what follows normally causes the repeated item to be
1479 re-evaluated to see if a different number of repeats allows the rest of the
1480 pattern to match. Sometimes it is useful to prevent this, either to change the
1481 nature of the match, or to cause it fail earlier than it otherwise might, when
1482 the author of the pattern knows there is no point in carrying on.
1483 .P
1484 Consider, for example, the pattern \ed+foo when applied to the subject line
1485 .sp
1486 123456bar
1487 .sp
1488 After matching all 6 digits and then failing to match "foo", the normal
1489 action of the matcher is to try again with only 5 digits matching the \ed+
1490 item, and then with 4, and so on, before ultimately failing. "Atomic grouping"
1491 (a term taken from Jeffrey Friedl's book) provides the means for specifying
1492 that once a subpattern has matched, it is not to be re-evaluated in this way.
1493 .P
1494 If we use atomic grouping for the previous example, the matcher gives up
1495 immediately on failing to match "foo" the first time. The notation is a kind of
1496 special parenthesis, starting with (?> as in this example:
1497 .sp
1498 (?>\ed+)foo
1499 .sp
1500 This kind of parenthesis "locks up" the part of the pattern it contains once
1501 it has matched, and a failure further into the pattern is prevented from
1502 backtracking into it. Backtracking past it to previous items, however, works as
1503 normal.
1504 .P
1505 An alternative description is that a subpattern of this type matches the string
1506 of characters that an identical standalone pattern would match, if anchored at
1507 the current point in the subject string.
1508 .P
1509 Atomic grouping subpatterns are not capturing subpatterns. Simple cases such as
1510 the above example can be thought of as a maximizing repeat that must swallow
1511 everything it can. So, while both \ed+ and \ed+? are prepared to adjust the
1512 number of digits they match in order to make the rest of the pattern match,
1513 (?>\ed+) can only match an entire sequence of digits.
1514 .P
1515 Atomic groups in general can of course contain arbitrarily complicated
1516 subpatterns, and can be nested. However, when the subpattern for an atomic
1517 group is just a single repeated item, as in the example above, a simpler
1518 notation, called a "possessive quantifier" can be used. This consists of an
1519 additional + character following a quantifier. Using this notation, the
1520 previous example can be rewritten as
1521 .sp
1522 \ed++foo
1523 .sp
1524 Note that a possessive quantifier can be used with an entire group, for
1525 example:
1526 .sp
1527 (abc|xyz){2,3}+
1528 .sp
1529 Possessive quantifiers are always greedy; the setting of the PCRE_UNGREEDY
1530 option is ignored. They are a convenient notation for the simpler forms of
1531 atomic group. However, there is no difference in the meaning of a possessive
1532 quantifier and the equivalent atomic group, though there may be a performance
1533 difference; possessive quantifiers should be slightly faster.
1534 .P
1535 The possessive quantifier syntax is an extension to the Perl 5.8 syntax.
1536 Jeffrey Friedl originated the idea (and the name) in the first edition of his
1537 book. Mike McCloskey liked it, so implemented it when he built Sun's Java
1538 package, and PCRE copied it from there. It ultimately found its way into Perl
1539 at release 5.10.
1540 .P
1541 PCRE has an optimization that automatically "possessifies" certain simple
1542 pattern constructs. For example, the sequence A+B is treated as A++B because
1543 there is no point in backtracking into a sequence of A's when B must follow.
1544 .P
1545 When a pattern contains an unlimited repeat inside a subpattern that can itself
1546 be repeated an unlimited number of times, the use of an atomic group is the
1547 only way to avoid some failing matches taking a very long time indeed. The
1548 pattern
1549 .sp
1550 (\eD+|<\ed+>)*[!?]
1551 .sp
1552 matches an unlimited number of substrings that either consist of non-digits, or
1553 digits enclosed in <>, followed by either ! or ?. When it matches, it runs
1554 quickly. However, if it is applied to
1555 .sp
1556 aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa
1557 .sp
1558 it takes a long time before reporting failure. This is because the string can
1559 be divided between the internal \eD+ repeat and the external * repeat in a
1560 large number of ways, and all have to be tried. (The example uses [!?] rather
1561 than a single character at the end, because both PCRE and Perl have an
1562 optimization that allows for fast failure when a single character is used. They
1563 remember the last single character that is required for a match, and fail early
1564 if it is not present in the string.) If the pattern is changed so that it uses
1565 an atomic group, like this:
1566 .sp
1567 ((?>\eD+)|<\ed+>)*[!?]
1568 .sp
1569 sequences of non-digits cannot be broken, and failure happens quickly.
1570 .
1571 .
1572 .\" HTML <a name="backreferences"></a>
1574 .rs
1575 .sp
1576 Outside a character class, a backslash followed by a digit greater than 0 (and
1577 possibly further digits) is a back reference to a capturing subpattern earlier
1578 (that is, to its left) in the pattern, provided there have been that many
1579 previous capturing left parentheses.
1580 .P
1581 However, if the decimal number following the backslash is less than 10, it is
1582 always taken as a back reference, and causes an error only if there are not
1583 that many capturing left parentheses in the entire pattern. In other words, the
1584 parentheses that are referenced need not be to the left of the reference for
1585 numbers less than 10. A "forward back reference" of this type can make sense
1586 when a repetition is involved and the subpattern to the right has participated
1587 in an earlier iteration.
1588 .P
1589 It is not possible to have a numerical "forward back reference" to a subpattern
1590 whose number is 10 or more using this syntax because a sequence such as \e50 is
1591 interpreted as a character defined in octal. See the subsection entitled
1592 "Non-printing characters"
1593 .\" HTML <a href="#digitsafterbackslash">
1594 .\" </a>
1595 above
1596 .\"
1597 for further details of the handling of digits following a backslash. There is
1598 no such problem when named parentheses are used. A back reference to any
1599 subpattern is possible using named parentheses (see below).
1600 .P
1601 Another way of avoiding the ambiguity inherent in the use of digits following a
1602 backslash is to use the \eg escape sequence, which is a feature introduced in
1603 Perl 5.10. This escape must be followed by an unsigned number or a negative
1604 number, optionally enclosed in braces. These examples are all identical:
1605 .sp
1606 (ring), \e1
1607 (ring), \eg1
1608 (ring), \eg{1}
1609 .sp
1610 An unsigned number specifies an absolute reference without the ambiguity that
1611 is present in the older syntax. It is also useful when literal digits follow
1612 the reference. A negative number is a relative reference. Consider this
1613 example:
1614 .sp
1615 (abc(def)ghi)\eg{-1}
1616 .sp
1617 The sequence \eg{-1} is a reference to the most recently started capturing
1618 subpattern before \eg, that is, is it equivalent to \e2. Similarly, \eg{-2}
1619 would be equivalent to \e1. The use of relative references can be helpful in
1620 long patterns, and also in patterns that are created by joining together
1621 fragments that contain references within themselves.
1622 .P
1623 A back reference matches whatever actually matched the capturing subpattern in
1624 the current subject string, rather than anything matching the subpattern
1625 itself (see
1626 .\" HTML <a href="#subpatternsassubroutines">
1627 .\" </a>
1628 "Subpatterns as subroutines"
1629 .\"
1630 below for a way of doing that). So the pattern
1631 .sp
1632 (sens|respons)e and \e1ibility
1633 .sp
1634 matches "sense and sensibility" and "response and responsibility", but not
1635 "sense and responsibility". If caseful matching is in force at the time of the
1636 back reference, the case of letters is relevant. For example,
1637 .sp
1638 ((?i)rah)\es+\e1
1639 .sp
1640 matches "rah rah" and "RAH RAH", but not "RAH rah", even though the original
1641 capturing subpattern is matched caselessly.
1642 .P
1643 There are several different ways of writing back references to named
1644 subpatterns. The .NET syntax \ek{name} and the Perl syntax \ek<name> or
1645 \ek'name' are supported, as is the Python syntax (?P=name). Perl 5.10's unified
1646 back reference syntax, in which \eg can be used for both numeric and named
1647 references, is also supported. We could rewrite the above example in any of
1648 the following ways:
1649 .sp
1650 (?<p1>(?i)rah)\es+\ek<p1>
1651 (?'p1'(?i)rah)\es+\ek{p1}
1652 (?P<p1>(?i)rah)\es+(?P=p1)
1653 (?<p1>(?i)rah)\es+\eg{p1}
1654 .sp
1655 A subpattern that is referenced by name may appear in the pattern before or
1656 after the reference.
1657 .P
1658 There may be more than one back reference to the same subpattern. If a
1659 subpattern has not actually been used in a particular match, any back
1660 references to it always fail by default. For example, the pattern
1661 .sp
1662 (a|(bc))\e2
1663 .sp
1664 always fails if it starts to match "a" rather than "bc". However, if the
1665 PCRE_JAVASCRIPT_COMPAT option is set at compile time, a back reference to an
1666 unset value matches an empty string.
1667 .P
1668 Because there may be many capturing parentheses in a pattern, all digits
1669 following a backslash are taken as part of a potential back reference number.
1670 If the pattern continues with a digit character, some delimiter must be used to
1671 terminate the back reference. If the PCRE_EXTENDED option is set, this can be
1672 whitespace. Otherwise, the \eg{ syntax or an empty comment (see
1673 .\" HTML <a href="#comments">
1674 .\" </a>
1675 "Comments"
1676 .\"
1677 below) can be used.
1678 .
1679 .SS "Recursive back references"
1680 .rs
1681 .sp
1682 A back reference that occurs inside the parentheses to which it refers fails
1683 when the subpattern is first used, so, for example, (a\e1) never matches.
1684 However, such references can be useful inside repeated subpatterns. For
1685 example, the pattern
1686 .sp
1687 (a|b\e1)+
1688 .sp
1689 matches any number of "a"s and also "aba", "ababbaa" etc. At each iteration of
1690 the subpattern, the back reference matches the character string corresponding
1691 to the previous iteration. In order for this to work, the pattern must be such
1692 that the first iteration does not need to match the back reference. This can be
1693 done using alternation, as in the example above, or by a quantifier with a
1694 minimum of zero.
1695 .P
1696 Back references of this type cause the group that they reference to be treated
1697 as an
1698 .\" HTML <a href="#atomicgroup">
1699 .\" </a>
1700 atomic group.
1701 .\"
1702 Once the whole group has been matched, a subsequent matching failure cannot
1703 cause backtracking into the middle of the group.
1704 .
1705 .
1706 .\" HTML <a name="bigassertions"></a>
1708 .rs
1709 .sp
1710 An assertion is a test on the characters following or preceding the current
1711 matching point that does not actually consume any characters. The simple
1712 assertions coded as \eb, \eB, \eA, \eG, \eZ, \ez, ^ and $ are described
1713 .\" HTML <a href="#smallassertions">
1714 .\" </a>
1715 above.
1716 .\"
1717 .P
1718 More complicated assertions are coded as subpatterns. There are two kinds:
1719 those that look ahead of the current position in the subject string, and those
1720 that look behind it. An assertion subpattern is matched in the normal way,
1721 except that it does not cause the current matching position to be changed.
1722 .P
1723 Assertion subpatterns are not capturing subpatterns, and may not be repeated,
1724 because it makes no sense to assert the same thing several times. If any kind
1725 of assertion contains capturing subpatterns within it, these are counted for
1726 the purposes of numbering the capturing subpatterns in the whole pattern.
1727 However, substring capturing is carried out only for positive assertions,
1728 because it does not make sense for negative assertions.
1729 .
1730 .
1731 .SS "Lookahead assertions"
1732 .rs
1733 .sp
1734 Lookahead assertions start with (?= for positive assertions and (?! for
1735 negative assertions. For example,
1736 .sp
1737 \ew+(?=;)
1738 .sp
1739 matches a word followed by a semicolon, but does not include the semicolon in
1740 the match, and
1741 .sp
1742 foo(?!bar)
1743 .sp
1744 matches any occurrence of "foo" that is not followed by "bar". Note that the
1745 apparently similar pattern
1746 .sp
1747 (?!foo)bar
1748 .sp
1749 does not find an occurrence of "bar" that is preceded by something other than
1750 "foo"; it finds any occurrence of "bar" whatsoever, because the assertion
1751 (?!foo) is always true when the next three characters are "bar". A
1752 lookbehind assertion is needed to achieve the other effect.
1753 .P
1754 If you want to force a matching failure at some point in a pattern, the most
1755 convenient way to do it is with (?!) because an empty string always matches, so
1756 an assertion that requires there not to be an empty string must always fail.
1757 The Perl 5.10 backtracking control verb (*FAIL) or (*F) is essentially a
1758 synonym for (?!).
1759 .
1760 .
1761 .\" HTML <a name="lookbehind"></a>
1762 .SS "Lookbehind assertions"
1763 .rs
1764 .sp
1765 Lookbehind assertions start with (?<= for positive assertions and (?<! for
1766 negative assertions. For example,
1767 .sp
1768 (?<!foo)bar
1769 .sp
1770 does find an occurrence of "bar" that is not preceded by "foo". The contents of
1771 a lookbehind assertion are restricted such that all the strings it matches must
1772 have a fixed length. However, if there are several top-level alternatives, they
1773 do not all have to have the same fixed length. Thus
1774 .sp
1775 (?<=bullock|donkey)
1776 .sp
1777 is permitted, but
1778 .sp
1779 (?<!dogs?|cats?)
1780 .sp
1781 causes an error at compile time. Branches that match different length strings
1782 are permitted only at the top level of a lookbehind assertion. This is an
1783 extension compared with Perl (5.8 and 5.10), which requires all branches to
1784 match the same length of string. An assertion such as
1785 .sp
1786 (?<=ab(c|de))
1787 .sp
1788 is not permitted, because its single top-level branch can match two different
1789 lengths, but it is acceptable to PCRE if rewritten to use two top-level
1790 branches:
1791 .sp
1792 (?<=abc|abde)
1793 .sp
1794 In some cases, the Perl 5.10 escape sequence \eK
1795 .\" HTML <a href="#resetmatchstart">
1796 .\" </a>
1797 (see above)
1798 .\"
1799 can be used instead of a lookbehind assertion to get round the fixed-length
1800 restriction.
1801 .P
1802 The implementation of lookbehind assertions is, for each alternative, to
1803 temporarily move the current position back by the fixed length and then try to
1804 match. If there are insufficient characters before the current position, the
1805 assertion fails.
1806 .P
1807 PCRE does not allow the \eC escape (which matches a single byte in UTF-8 mode)
1808 to appear in lookbehind assertions, because it makes it impossible to calculate
1809 the length of the lookbehind. The \eX and \eR escapes, which can match
1810 different numbers of bytes, are also not permitted.
1811 .P
1812 .\" HTML <a href="#subpatternsassubroutines">
1813 .\" </a>
1814 "Subroutine"
1815 .\"
1816 calls (see below) such as (?2) or (?&X) are permitted in lookbehinds, as long
1817 as the subpattern matches a fixed-length string.
1818 .\" HTML <a href="#recursion">
1819 .\" </a>
1820 Recursion,
1821 .\"
1822 however, is not supported.
1823 .P
1824 Possessive quantifiers can be used in conjunction with lookbehind assertions to
1825 specify efficient matching of fixed-length strings at the end of subject
1826 strings. Consider a simple pattern such as
1827 .sp
1828 abcd$
1829 .sp
1830 when applied to a long string that does not match. Because matching proceeds
1831 from left to right, PCRE will look for each "a" in the subject and then see if
1832 what follows matches the rest of the pattern. If the pattern is specified as
1833 .sp
1834 ^.*abcd$
1835 .sp
1836 the initial .* matches the entire string at first, but when this fails (because
1837 there is no following "a"), it backtracks to match all but the last character,
1838 then all but the last two characters, and so on. Once again the search for "a"
1839 covers the entire string, from right to left, so we are no better off. However,
1840 if the pattern is written as
1841 .sp
1842 ^.*+(?<=abcd)
1843 .sp
1844 there can be no backtracking for the .*+ item; it can match only the entire
1845 string. The subsequent lookbehind assertion does a single test on the last four
1846 characters. If it fails, the match fails immediately. For long strings, this
1847 approach makes a significant difference to the processing time.
1848 .
1849 .
1850 .SS "Using multiple assertions"
1851 .rs
1852 .sp
1853 Several assertions (of any sort) may occur in succession. For example,
1854 .sp
1855 (?<=\ed{3})(?<!999)foo
1856 .sp
1857 matches "foo" preceded by three digits that are not "999". Notice that each of
1858 the assertions is applied independently at the same point in the subject
1859 string. First there is a check that the previous three characters are all
1860 digits, and then there is a check that the same three characters are not "999".
1861 This pattern does \fInot\fP match "foo" preceded by six characters, the first
1862 of which are digits and the last three of which are not "999". For example, it
1863 doesn't match "123abcfoo". A pattern to do that is
1864 .sp
1865 (?<=\ed{3}...)(?<!999)foo
1866 .sp
1867 This time the first assertion looks at the preceding six characters, checking
1868 that the first three are digits, and then the second assertion checks that the
1869 preceding three characters are not "999".
1870 .P
1871 Assertions can be nested in any combination. For example,
1872 .sp
1873 (?<=(?<!foo)bar)baz
1874 .sp
1875 matches an occurrence of "baz" that is preceded by "bar" which in turn is not
1876 preceded by "foo", while
1877 .sp
1878 (?<=\ed{3}(?!999)...)foo
1879 .sp
1880 is another pattern that matches "foo" preceded by three digits and any three
1881 characters that are not "999".
1882 .
1883 .
1884 .\" HTML <a name="conditions"></a>
1886 .rs
1887 .sp
1888 It is possible to cause the matching process to obey a subpattern
1889 conditionally or to choose between two alternative subpatterns, depending on
1890 the result of an assertion, or whether a specific capturing subpattern has
1891 already been matched. The two possible forms of conditional subpattern are:
1892 .sp
1893 (?(condition)yes-pattern)
1894 (?(condition)yes-pattern|no-pattern)
1895 .sp
1896 If the condition is satisfied, the yes-pattern is used; otherwise the
1897 no-pattern (if present) is used. If there are more than two alternatives in the
1898 subpattern, a compile-time error occurs.
1899 .P
1900 There are four kinds of condition: references to subpatterns, references to
1901 recursion, a pseudo-condition called DEFINE, and assertions.
1902 .
1903 .SS "Checking for a used subpattern by number"
1904 .rs
1905 .sp
1906 If the text between the parentheses consists of a sequence of digits, the
1907 condition is true if a capturing subpattern of that number has previously
1908 matched. If there is more than one capturing subpattern with the same number
1909 (see the earlier
1910 .\"
1911 .\" HTML <a href="#recursion">
1912 .\" </a>
1913 section about duplicate subpattern numbers),
1914 .\"
1915 the condition is true if any of them have been set. An alternative notation is
1916 to precede the digits with a plus or minus sign. In this case, the subpattern
1917 number is relative rather than absolute. The most recently opened parentheses
1918 can be referenced by (?(-1), the next most recent by (?(-2), and so on. In
1919 looping constructs it can also make sense to refer to subsequent groups with
1920 constructs such as (?(+2).
1921 .P
1922 Consider the following pattern, which contains non-significant white space to
1923 make it more readable (assume the PCRE_EXTENDED option) and to divide it into
1924 three parts for ease of discussion:
1925 .sp
1926 ( \e( )? [^()]+ (?(1) \e) )
1927 .sp
1928 The first part matches an optional opening parenthesis, and if that
1929 character is present, sets it as the first captured substring. The second part
1930 matches one or more characters that are not parentheses. The third part is a
1931 conditional subpattern that tests whether the first set of parentheses matched
1932 or not. If they did, that is, if subject started with an opening parenthesis,
1933 the condition is true, and so the yes-pattern is executed and a closing
1934 parenthesis is required. Otherwise, since no-pattern is not present, the
1935 subpattern matches nothing. In other words, this pattern matches a sequence of
1936 non-parentheses, optionally enclosed in parentheses.
1937 .P
1938 If you were embedding this pattern in a larger one, you could use a relative
1939 reference:
1940 .sp
1941 ...other stuff... ( \e( )? [^()]+ (?(-1) \e) ) ...
1942 .sp
1943 This makes the fragment independent of the parentheses in the larger pattern.
1944 .
1945 .SS "Checking for a used subpattern by name"
1946 .rs
1947 .sp
1948 Perl uses the syntax (?(<name>)...) or (?('name')...) to test for a used
1949 subpattern by name. For compatibility with earlier versions of PCRE, which had
1950 this facility before Perl, the syntax (?(name)...) is also recognized. However,
1951 there is a possible ambiguity with this syntax, because subpattern names may
1952 consist entirely of digits. PCRE looks first for a named subpattern; if it
1953 cannot find one and the name consists entirely of digits, PCRE looks for a
1954 subpattern of that number, which must be greater than zero. Using subpattern
1955 names that consist entirely of digits is not recommended.
1956 .P
1957 Rewriting the above example to use a named subpattern gives this:
1958 .sp
1959 (?<OPEN> \e( )? [^()]+ (?(<OPEN>) \e) )
1960 .sp
1961 If the name used in a condition of this kind is a duplicate, the test is
1962 applied to all subpatterns of the same name, and is true if any one of them has
1963 matched.
1964 .
1965 .SS "Checking for pattern recursion"
1966 .rs
1967 .sp
1968 If the condition is the string (R), and there is no subpattern with the name R,
1969 the condition is true if a recursive call to the whole pattern or any
1970 subpattern has been made. If digits or a name preceded by ampersand follow the
1971 letter R, for example:
1972 .sp
1973 (?(R3)...) or (?(R&name)...)
1974 .sp
1975 the condition is true if the most recent recursion is into a subpattern whose
1976 number or name is given. This condition does not check the entire recursion
1977 stack. If the name used in a condition of this kind is a duplicate, the test is
1978 applied to all subpatterns of the same name, and is true if any one of them is
1979 the most recent recursion.
1980 .P
1981 At "top level", all these recursion test conditions are false.
1982 .\" HTML <a href="#recursion">
1983 .\" </a>
1984 The syntax for recursive patterns
1985 .\"
1986 is described below.
1987 .
1988 .SS "Defining subpatterns for use by reference only"
1989 .rs
1990 .sp
1991 If the condition is the string (DEFINE), and there is no subpattern with the
1992 name DEFINE, the condition is always false. In this case, there may be only one
1993 alternative in the subpattern. It is always skipped if control reaches this
1994 point in the pattern; the idea of DEFINE is that it can be used to define
1995 "subroutines" that can be referenced from elsewhere. (The use of
1996 .\" HTML <a href="#subpatternsassubroutines">
1997 .\" </a>
1998 "subroutines"
1999 .\"
2000 is described below.) For example, a pattern to match an IPv4 address could be
2001 written like this (ignore whitespace and line breaks):
2002 .sp
2003 (?(DEFINE) (?<byte> 2[0-4]\ed | 25[0-5] | 1\ed\ed | [1-9]?\ed) )
2004 \eb (?&byte) (\e.(?&byte)){3} \eb
2005 .sp
2006 The first part of the pattern is a DEFINE group inside which a another group
2007 named "byte" is defined. This matches an individual component of an IPv4
2008 address (a number less than 256). When matching takes place, this part of the
2009 pattern is skipped because DEFINE acts like a false condition. The rest of the
2010 pattern uses references to the named group to match the four dot-separated
2011 components of an IPv4 address, insisting on a word boundary at each end.
2012 .
2013 .SS "Assertion conditions"
2014 .rs
2015 .sp
2016 If the condition is not in any of the above formats, it must be an assertion.
2017 This may be a positive or negative lookahead or lookbehind assertion. Consider
2018 this pattern, again containing non-significant white space, and with the two
2019 alternatives on the second line:
2020 .sp
2021 (?(?=[^a-z]*[a-z])
2022 \ed{2}-[a-z]{3}-\ed{2} | \ed{2}-\ed{2}-\ed{2} )
2023 .sp
2024 The condition is a positive lookahead assertion that matches an optional
2025 sequence of non-letters followed by a letter. In other words, it tests for the
2026 presence of at least one letter in the subject. If a letter is found, the
2027 subject is matched against the first alternative; otherwise it is matched
2028 against the second. This pattern matches strings in one of the two forms
2029 dd-aaa-dd or dd-dd-dd, where aaa are letters and dd are digits.
2030 .
2031 .
2032 .\" HTML <a name="comments"></a>
2034 .rs
2035 .sp
2036 The sequence (?# marks the start of a comment that continues up to the next
2037 closing parenthesis. Nested parentheses are not permitted. The characters
2038 that make up a comment play no part in the pattern matching at all.
2039 .P
2040 If the PCRE_EXTENDED option is set, an unescaped # character outside a
2041 character class introduces a comment that continues to immediately after the
2042 next newline in the pattern.
2043 .
2044 .
2045 .\" HTML <a name="recursion"></a>
2047 .rs
2048 .sp
2049 Consider the problem of matching a string in parentheses, allowing for
2050 unlimited nested parentheses. Without the use of recursion, the best that can
2051 be done is to use a pattern that matches up to some fixed depth of nesting. It
2052 is not possible to handle an arbitrary nesting depth.
2053 .P
2054 For some time, Perl has provided a facility that allows regular expressions to
2055 recurse (amongst other things). It does this by interpolating Perl code in the
2056 expression at run time, and the code can refer to the expression itself. A Perl
2057 pattern using code interpolation to solve the parentheses problem can be
2058 created like this:
2059 .sp
2060 $re = qr{\e( (?: (?>[^()]+) | (?p{$re}) )* \e)}x;
2061 .sp
2062 The (?p{...}) item interpolates Perl code at run time, and in this case refers
2063 recursively to the pattern in which it appears.
2064 .P
2065 Obviously, PCRE cannot support the interpolation of Perl code. Instead, it
2066 supports special syntax for recursion of the entire pattern, and also for
2067 individual subpattern recursion. After its introduction in PCRE and Python,
2068 this kind of recursion was subsequently introduced into Perl at release 5.10.
2069 .P
2070 A special item that consists of (? followed by a number greater than zero and a
2071 closing parenthesis is a recursive call of the subpattern of the given number,
2072 provided that it occurs inside that subpattern. (If not, it is a
2073 .\" HTML <a href="#subpatternsassubroutines">
2074 .\" </a>
2075 "subroutine"
2076 .\"
2077 call, which is described in the next section.) The special item (?R) or (?0) is
2078 a recursive call of the entire regular expression.
2079 .P
2080 This PCRE pattern solves the nested parentheses problem (assume the
2081 PCRE_EXTENDED option is set so that white space is ignored):
2082 .sp
2083 \e( ( [^()]++ | (?R) )* \e)
2084 .sp
2085 First it matches an opening parenthesis. Then it matches any number of
2086 substrings which can either be a sequence of non-parentheses, or a recursive
2087 match of the pattern itself (that is, a correctly parenthesized substring).
2088 Finally there is a closing parenthesis. Note the use of a possessive quantifier
2089 to avoid backtracking into sequences of non-parentheses.
2090 .P
2091 If this were part of a larger pattern, you would not want to recurse the entire
2092 pattern, so instead you could use this:
2093 .sp
2094 ( \e( ( [^()]++ | (?1) )* \e) )
2095 .sp
2096 We have put the pattern into parentheses, and caused the recursion to refer to
2097 them instead of the whole pattern.
2098 .P
2099 In a larger pattern, keeping track of parenthesis numbers can be tricky. This
2100 is made easier by the use of relative references (a Perl 5.10 feature).
2101 Instead of (?1) in the pattern above you can write (?-2) to refer to the second
2102 most recently opened parentheses preceding the recursion. In other words, a
2103 negative number counts capturing parentheses leftwards from the point at which
2104 it is encountered.
2105 .P
2106 It is also possible to refer to subsequently opened parentheses, by writing
2107 references such as (?+2). However, these cannot be recursive because the
2108 reference is not inside the parentheses that are referenced. They are always
2109 .\" HTML <a href="#subpatternsassubroutines">
2110 .\" </a>
2111 "subroutine"
2112 .\"
2113 calls, as described in the next section.
2114 .P
2115 An alternative approach is to use named parentheses instead. The Perl syntax
2116 for this is (?&name); PCRE's earlier syntax (?P>name) is also supported. We
2117 could rewrite the above example as follows:
2118 .sp
2119 (?<pn> \e( ( [^()]++ | (?&pn) )* \e) )
2120 .sp
2121 If there is more than one subpattern with the same name, the earliest one is
2122 used.
2123 .P
2124 This particular example pattern that we have been looking at contains nested
2125 unlimited repeats, and so the use of a possessive quantifier for matching
2126 strings of non-parentheses is important when applying the pattern to strings
2127 that do not match. For example, when this pattern is applied to
2128 .sp
2129 (aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa()
2130 .sp
2131 it yields "no match" quickly. However, if a possessive quantifier is not used,
2132 the match runs for a very long time indeed because there are so many different
2133 ways the + and * repeats can carve up the subject, and all have to be tested
2134 before failure can be reported.
2135 .P
2136 At the end of a match, the values of capturing parentheses are those from
2137 the outermost level. If you want to obtain intermediate values, a callout
2138 function can be used (see below and the
2139 .\" HREF
2140 \fBpcrecallout\fP
2141 .\"
2142 documentation). If the pattern above is matched against
2143 .sp
2144 (ab(cd)ef)
2145 .sp
2146 the value for the inner capturing parentheses (numbered 2) is "ef", which is
2147 the last value taken on at the top level. If a capturing subpattern is not
2148 matched at the top level, its final value is unset, even if it is (temporarily)
2149 set at a deeper level.
2150 .P
2151 If there are more than 15 capturing parentheses in a pattern, PCRE has to
2152 obtain extra memory to store data during a recursion, which it does by using
2153 \fBpcre_malloc\fP, freeing it via \fBpcre_free\fP afterwards. If no memory can
2154 be obtained, the match fails with the PCRE_ERROR_NOMEMORY error.
2155 .P
2156 Do not confuse the (?R) item with the condition (R), which tests for recursion.
2157 Consider this pattern, which matches text in angle brackets, allowing for
2158 arbitrary nesting. Only digits are allowed in nested brackets (that is, when
2159 recursing), whereas any characters are permitted at the outer level.
2160 .sp
2161 < (?: (?(R) \ed++ | [^<>]*+) | (?R)) * >
2162 .sp
2163 In this pattern, (?(R) is the start of a conditional subpattern, with two
2164 different alternatives for the recursive and non-recursive cases. The (?R) item
2165 is the actual recursive call.
2166 .
2167 .
2168 .\" HTML <a name="recursiondifference"></a>
2169 .SS "Recursion difference from Perl"
2170 .rs
2171 .sp
2172 In PCRE (like Python, but unlike Perl), a recursive subpattern call is always
2173 treated as an atomic group. That is, once it has matched some of the subject
2174 string, it is never re-entered, even if it contains untried alternatives and
2175 there is a subsequent matching failure. This can be illustrated by the
2176 following pattern, which purports to match a palindromic string that contains
2177 an odd number of characters (for example, "a", "aba", "abcba", "abcdcba"):
2178 .sp
2179 ^(.|(.)(?1)\e2)$
2180 .sp
2181 The idea is that it either matches a single character, or two identical
2182 characters surrounding a sub-palindrome. In Perl, this pattern works; in PCRE
2183 it does not if the pattern is longer than three characters. Consider the
2184 subject string "abcba":
2185 .P
2186 At the top level, the first character is matched, but as it is not at the end
2187 of the string, the first alternative fails; the second alternative is taken
2188 and the recursion kicks in. The recursive call to subpattern 1 successfully
2189 matches the next character ("b"). (Note that the beginning and end of line
2190 tests are not part of the recursion).
2191 .P
2192 Back at the top level, the next character ("c") is compared with what
2193 subpattern 2 matched, which was "a". This fails. Because the recursion is
2194 treated as an atomic group, there are now no backtracking points, and so the
2195 entire match fails. (Perl is able, at this point, to re-enter the recursion and
2196 try the second alternative.) However, if the pattern is written with the
2197 alternatives in the other order, things are different:
2198 .sp
2199 ^((.)(?1)\e2|.)$
2200 .sp
2201 This time, the recursing alternative is tried first, and continues to recurse
2202 until it runs out of characters, at which point the recursion fails. But this
2203 time we do have another alternative to try at the higher level. That is the big
2204 difference: in the previous case the remaining alternative is at a deeper
2205 recursion level, which PCRE cannot use.
2206 .P
2207 To change the pattern so that matches all palindromic strings, not just those
2208 with an odd number of characters, it is tempting to change the pattern to this:
2209 .sp
2210 ^((.)(?1)\e2|.?)$
2211 .sp
2212 Again, this works in Perl, but not in PCRE, and for the same reason. When a
2213 deeper recursion has matched a single character, it cannot be entered again in
2214 order to match an empty string. The solution is to separate the two cases, and
2215 write out the odd and even cases as alternatives at the higher level:
2216 .sp
2217 ^(?:((.)(?1)\e2|)|((.)(?3)\e4|.))
2218 .sp
2219 If you want to match typical palindromic phrases, the pattern has to ignore all
2220 non-word characters, which can be done like this:
2221 .sp
2222 ^\eW*+(?:((.)\eW*+(?1)\eW*+\e2|)|((.)\eW*+(?3)\eW*+\e4|\eW*+.\eW*+))\eW*+$
2223 .sp
2224 If run with the PCRE_CASELESS option, this pattern matches phrases such as "A
2225 man, a plan, a canal: Panama!" and it works well in both PCRE and Perl. Note
2226 the use of the possessive quantifier *+ to avoid backtracking into sequences of
2227 non-word characters. Without this, PCRE takes a great deal longer (ten times or
2228 more) to match typical phrases, and Perl takes so long that you think it has
2229 gone into a loop.
2230 .P
2231 \fBWARNING\fP: The palindrome-matching patterns above work only if the subject
2232 string does not start with a palindrome that is shorter than the entire string.
2233 For example, although "abcba" is correctly matched, if the subject is "ababa",
2234 PCRE finds the palindrome "aba" at the start, then fails at top level because
2235 the end of the string does not follow. Once again, it cannot jump back into the
2236 recursion to try other alternatives, so the entire match fails.
2237 .
2238 .
2239 .\" HTML <a name="subpatternsassubroutines"></a>
2241 .rs
2242 .sp
2243 If the syntax for a recursive subpattern reference (either by number or by
2244 name) is used outside the parentheses to which it refers, it operates like a
2245 subroutine in a programming language. The "called" subpattern may be defined
2246 before or after the reference. A numbered reference can be absolute or
2247 relative, as in these examples:
2248 .sp
2249 (...(absolute)...)...(?2)...
2250 (...(relative)...)...(?-1)...
2251 (...(?+1)...(relative)...
2252 .sp
2253 An earlier example pointed out that the pattern
2254 .sp
2255 (sens|respons)e and \e1ibility
2256 .sp
2257 matches "sense and sensibility" and "response and responsibility", but not
2258 "sense and responsibility". If instead the pattern
2259 .sp
2260 (sens|respons)e and (?1)ibility
2261 .sp
2262 is used, it does match "sense and responsibility" as well as the other two
2263 strings. Another example is given in the discussion of DEFINE above.
2264 .P
2265 Like recursive subpatterns, a subroutine call is always treated as an atomic
2266 group. That is, once it has matched some of the subject string, it is never
2267 re-entered, even if it contains untried alternatives and there is a subsequent
2268 matching failure. Any capturing parentheses that are set during the subroutine
2269 call revert to their previous values afterwards.
2270 .P
2271 When a subpattern is used as a subroutine, processing options such as
2272 case-independence are fixed when the subpattern is defined. They cannot be
2273 changed for different calls. For example, consider this pattern:
2274 .sp
2275 (abc)(?i:(?-1))
2276 .sp
2277 It matches "abcabc". It does not match "abcABC" because the change of
2278 processing option does not affect the called subpattern.
2279 .
2280 .
2281 .\" HTML <a name="onigurumasubroutines"></a>
2283 .rs
2284 .sp
2285 For compatibility with Oniguruma, the non-Perl syntax \eg followed by a name or
2286 a number enclosed either in angle brackets or single quotes, is an alternative
2287 syntax for referencing a subpattern as a subroutine, possibly recursively. Here
2288 are two of the examples used above, rewritten using this syntax:
2289 .sp
2290 (?<pn> \e( ( (?>[^()]+) | \eg<pn> )* \e) )
2291 (sens|respons)e and \eg'1'ibility
2292 .sp
2293 PCRE supports an extension to Oniguruma: if a number is preceded by a
2294 plus or a minus sign it is taken as a relative reference. For example:
2295 .sp
2296 (abc)(?i:\eg<-1>)
2297 .sp
2298 Note that \eg{...} (Perl syntax) and \eg<...> (Oniguruma syntax) are \fInot\fP
2299 synonymous. The former is a back reference; the latter is a subroutine call.
2300 .
2301 .
2303 .rs
2304 .sp
2305 Perl has a feature whereby using the sequence (?{...}) causes arbitrary Perl
2306 code to be obeyed in the middle of matching a regular expression. This makes it
2307 possible, amongst other things, to extract different substrings that match the
2308 same pair of parentheses when there is a repetition.
2309 .P
2310 PCRE provides a similar feature, but of course it cannot obey arbitrary Perl
2311 code. The feature is called "callout". The caller of PCRE provides an external
2312 function by putting its entry point in the global variable \fIpcre_callout\fP.
2313 By default, this variable contains NULL, which disables all calling out.
2314 .P
2315 Within a regular expression, (?C) indicates the points at which the external
2316 function is to be called. If you want to identify different callout points, you
2317 can put a number less than 256 after the letter C. The default value is zero.
2318 For example, this pattern has two callout points:
2319 .sp
2320 (?C1)abc(?C2)def
2321 .sp
2322 If the PCRE_AUTO_CALLOUT flag is passed to \fBpcre_compile()\fP, callouts are
2323 automatically installed before each item in the pattern. They are all numbered
2324 255.
2325 .P
2326 During matching, when PCRE reaches a callout point (and \fIpcre_callout\fP is
2327 set), the external function is called. It is provided with the number of the
2328 callout, the position in the pattern, and, optionally, one item of data
2329 originally supplied by the caller of \fBpcre_exec()\fP. The callout function
2330 may cause matching to proceed, to backtrack, or to fail altogether. A complete
2331 description of the interface to the callout function is given in the
2332 .\" HREF
2333 \fBpcrecallout\fP
2334 .\"
2335 documentation.
2336 .
2337 .
2338 .\" HTML <a name="backtrackcontrol"></a>
2340 .rs
2341 .sp
2342 Perl 5.10 introduced a number of "Special Backtracking Control Verbs", which
2343 are described in the Perl documentation as "experimental and subject to change
2344 or removal in a future version of Perl". It goes on to say: "Their usage in
2345 production code should be noted to avoid problems during upgrades." The same
2346 remarks apply to the PCRE features described in this section.
2347 .P
2348 Since these verbs are specifically related to backtracking, most of them can be
2349 used only when the pattern is to be matched using \fBpcre_exec()\fP, which uses
2350 a backtracking algorithm. With the exception of (*FAIL), which behaves like a
2351 failing negative assertion, they cause an error if encountered by
2352 \fBpcre_dfa_exec()\fP.
2353 .P
2354 If any of these verbs are used in an assertion or subroutine subpattern
2355 (including recursive subpatterns), their effect is confined to that subpattern;
2356 it does not extend to the surrounding pattern. Note that such subpatterns are
2357 processed as anchored at the point where they are tested.
2358 .P
2359 The new verbs make use of what was previously invalid syntax: an opening
2360 parenthesis followed by an asterisk. They are generally of the form
2361 (*VERB) or (*VERB:NAME). Some may take either form, with differing behaviour,
2362 depending on whether or not an argument is present. An name is a sequence of
2363 letters, digits, and underscores. If the name is empty, that is, if the closing
2364 parenthesis immediately follows the colon, the effect is as if the colon were
2365 not there. Any number of these verbs may occur in a pattern.
2366 .P
2367 PCRE contains some optimizations that are used to speed up matching by running
2368 some checks at the start of each match attempt. For example, it may know the
2369 minimum length of matching subject, or that a particular character must be
2370 present. When one of these optimizations suppresses the running of a match, any
2371 included backtracking verbs will not, of course, be processed. You can suppress
2372 the start-of-match optimizations by setting the PCRE_NO_START_OPTIMIZE option
2373 when calling \fBpcre_exec()\fP.
2374 .
2375 .
2376 .SS "Verbs that act immediately"
2377 .rs
2378 .sp
2379 The following verbs act as soon as they are encountered. They may not be
2380 followed by a name.
2381 .sp
2382 (*ACCEPT)
2383 .sp
2384 This verb causes the match to end successfully, skipping the remainder of the
2385 pattern. When inside a recursion, only the innermost pattern is ended
2386 immediately. If (*ACCEPT) is inside capturing parentheses, the data so far is
2387 captured. (This feature was added to PCRE at release 8.00.) For example:
2388 .sp
2389 A((?:A|B(*ACCEPT)|C)D)
2390 .sp
2391 This matches "AB", "AAD", or "ACD"; when it matches "AB", "B" is captured by
2392 the outer parentheses.
2393 .sp
2394 (*FAIL) or (*F)
2395 .sp
2396 This verb causes the match to fail, forcing backtracking to occur. It is
2397 equivalent to (?!) but easier to read. The Perl documentation notes that it is
2398 probably useful only when combined with (?{}) or (??{}). Those are, of course,
2399 Perl features that are not present in PCRE. The nearest equivalent is the
2400 callout feature, as for example in this pattern:
2401 .sp
2402 a+(?C)(*FAIL)
2403 .sp
2404 A match with the string "aaaa" always fails, but the callout is taken before
2405 each backtrack happens (in this example, 10 times).
2406 .
2407 .
2408 .SS "Recording which path was taken"
2409 .rs
2410 .sp
2411 There is one verb whose main purpose is to track how a match was arrived at,
2412 though it also has a secondary use in conjunction with advancing the match
2413 starting point (see (*SKIP) below).
2414 .sp
2415 (*MARK:NAME) or (*:NAME)
2416 .sp
2417 A name is always required with this verb. There may be as many instances of
2418 (*MARK) as you like in a pattern, and their names do not have to be unique.
2419 .P
2420 When a match succeeds, the name of the last-encountered (*MARK) is passed back
2421 to the caller via the \fIpcre_extra\fP data structure, as described in the
2422 .\" HTML <a href="pcreapi.html#extradata">
2423 .\" </a>
2424 section on \fIpcre_extra\fP
2425 .\"
2426 in the
2427 .\" HREF
2428 \fBpcreapi\fP
2429 .\"
2430 documentation. No data is returned for a partial match. Here is an example of
2431 \fBpcretest\fP output, where the /K modifier requests the retrieval and
2432 outputting of (*MARK) data:
2433 .sp
2434 /X(*MARK:A)Y|X(*MARK:B)Z/K
2435 XY
2436 0: XY
2437 MK: A
2438 XZ
2439 0: XZ
2440 MK: B
2441 .sp
2442 The (*MARK) name is tagged with "MK:" in this output, and in this example it
2443 indicates which of the two alternatives matched. This is a more efficient way
2444 of obtaining this information than putting each alternative in its own
2445 capturing parentheses.
2446 .P
2447 A name may also be returned after a failed match if the final path through the
2448 pattern involves (*MARK). However, unless (*MARK) used in conjunction with
2449 (*COMMIT), this is unlikely to happen for an unanchored pattern because, as the
2450 starting point for matching is advanced, the final check is often with an empty
2451 string, causing a failure before (*MARK) is reached. For example:
2452 .sp
2453 /X(*MARK:A)Y|X(*MARK:B)Z/K
2454 XP
2455 No match
2456 .sp
2457 There are three potential starting points for this match (starting with X,
2458 starting with P, and with an empty string). If the pattern is anchored, the
2459 result is different:
2460 .sp
2461 /^X(*MARK:A)Y|^X(*MARK:B)Z/K
2462 XP
2463 No match, mark = B
2464 .sp
2465 PCRE's start-of-match optimizations can also interfere with this. For example,
2466 if, as a result of a call to \fBpcre_study()\fP, it knows the minimum
2467 subject length for a match, a shorter subject will not be scanned at all.
2468 .P
2469 Note that similar anomalies (though different in detail) exist in Perl, no
2470 doubt for the same reasons. The use of (*MARK) data after a failed match of an
2471 unanchored pattern is not recommended, unless (*COMMIT) is involved.
2472 .
2473 .
2474 .SS "Verbs that act after backtracking"
2475 .rs
2476 .sp
2477 The following verbs do nothing when they are encountered. Matching continues
2478 with what follows, but if there is no subsequent match, causing a backtrack to
2479 the verb, a failure is forced. That is, backtracking cannot pass to the left of
2480 the verb. However, when one of these verbs appears inside an atomic group, its
2481 effect is confined to that group, because once the group has been matched,
2482 there is never any backtracking into it. In this situation, backtracking can
2483 "jump back" to the left of the entire atomic group. (Remember also, as stated
2484 above, that this localization also applies in subroutine calls and assertions.)
2485 .P
2486 These verbs differ in exactly what kind of failure occurs when backtracking
2487 reaches them.
2488 .sp
2489 (*COMMIT)
2490 .sp
2491 This verb, which may not be followed by a name, causes the whole match to fail
2492 outright if the rest of the pattern does not match. Even if the pattern is
2493 unanchored, no further attempts to find a match by advancing the starting point
2494 take place. Once (*COMMIT) has been passed, \fBpcre_exec()\fP is committed to
2495 finding a match at the current starting point, or not at all. For example:
2496 .sp
2497 a+(*COMMIT)b
2498 .sp
2499 This matches "xxaab" but not "aacaab". It can be thought of as a kind of
2500 dynamic anchor, or "I've started, so I must finish." The name of the most
2501 recently passed (*MARK) in the path is passed back when (*COMMIT) forces a
2502 match failure.
2503 .P
2504 Note that (*COMMIT) at the start of a pattern is not the same as an anchor,
2505 unless PCRE's start-of-match optimizations are turned off, as shown in this
2506 \fBpcretest\fP example:
2507 .sp
2508 /(*COMMIT)abc/
2509 xyzabc
2510 0: abc
2511 xyzabc\eY
2512 No match
2513 .sp
2514 PCRE knows that any match must start with "a", so the optimization skips along
2515 the subject to "a" before running the first match attempt, which succeeds. When
2516 the optimization is disabled by the \eY escape in the second subject, the match
2517 starts at "x" and so the (*COMMIT) causes it to fail without trying any other
2518 starting points.
2519 .sp
2520 (*PRUNE) or (*PRUNE:NAME)
2521 .sp
2522 This verb causes the match to fail at the current starting position in the
2523 subject if the rest of the pattern does not match. If the pattern is
2524 unanchored, the normal "bumpalong" advance to the next starting character then
2525 happens. Backtracking can occur as usual to the left of (*PRUNE), before it is
2526 reached, or when matching to the right of (*PRUNE), but if there is no match to
2527 the right, backtracking cannot cross (*PRUNE). In simple cases, the use of
2528 (*PRUNE) is just an alternative to an atomic group or possessive quantifier,
2529 but there are some uses of (*PRUNE) that cannot be expressed in any other way.
2530 The behaviour of (*PRUNE:NAME) is the same as (*MARK:NAME)(*PRUNE) when the
2531 match fails completely; the name is passed back if this is the final attempt.
2532 (*PRUNE:NAME) does not pass back a name if the match succeeds. In an anchored
2533 pattern (*PRUNE) has the same effect as (*COMMIT).
2534 .sp
2535 (*SKIP)
2536 .sp
2537 This verb, when given without a name, is like (*PRUNE), except that if the
2538 pattern is unanchored, the "bumpalong" advance is not to the next character,
2539 but to the position in the subject where (*SKIP) was encountered. (*SKIP)
2540 signifies that whatever text was matched leading up to it cannot be part of a
2541 successful match. Consider:
2542 .sp
2543 a+(*SKIP)b
2544 .sp
2545 If the subject is "aaaac...", after the first match attempt fails (starting at
2546 the first character in the string), the starting point skips on to start the
2547 next attempt at "c". Note that a possessive quantifer does not have the same
2548 effect as this example; although it would suppress backtracking during the
2549 first match attempt, the second attempt would start at the second character
2550 instead of skipping on to "c".
2551 .sp
2552 (*SKIP:NAME)
2553 .sp
2554 When (*SKIP) has an associated name, its behaviour is modified. If the
2555 following pattern fails to match, the previous path through the pattern is
2556 searched for the most recent (*MARK) that has the same name. If one is found,
2557 the "bumpalong" advance is to the subject position that corresponds to that
2558 (*MARK) instead of to where (*SKIP) was encountered. If no (*MARK) with a
2559 matching name is found, normal "bumpalong" of one character happens (the
2560 (*SKIP) is ignored).
2561 .sp
2562 (*THEN) or (*THEN:NAME)
2563 .sp
2564 This verb causes a skip to the next alternation if the rest of the pattern does
2565 not match. That is, it cancels pending backtracking, but only within the
2566 current alternation. Its name comes from the observation that it can be used
2567 for a pattern-based if-then-else block:
2568 .sp
2569 ( COND1 (*THEN) FOO | COND2 (*THEN) BAR | COND3 (*THEN) BAZ ) ...
2570 .sp
2571 If the COND1 pattern matches, FOO is tried (and possibly further items after
2572 the end of the group if FOO succeeds); on failure the matcher skips to the
2573 second alternative and tries COND2, without backtracking into COND1. The
2574 behaviour of (*THEN:NAME) is exactly the same as (*MARK:NAME)(*THEN) if the
2575 overall match fails. If (*THEN) is not directly inside an alternation, it acts
2576 like (*PRUNE).
2577 .
2578 .
2579 .SH "SEE ALSO"
2580 .rs
2581 .sp
2582 \fBpcreapi\fP(3), \fBpcrecallout\fP(3), \fBpcrematching\fP(3),
2583 \fBpcresyntax\fP(3), \fBpcre\fP(3).
2584 .
2585 .
2587 .rs
2588 .sp
2589 .nf
2590 Philip Hazel
2591 University Computing Service
2592 Cambridge CB2 3QH, England.
2593 .fi
2594 .
2595 .
2597 .rs
2598 .sp
2599 .nf
2600 Last updated: 03 May 2010
2601 Copyright (c) 1997-2010 University of Cambridge.
2602 .fi


Name Value
svn:eol-style native
svn:keywords "Author Date Id Revision Url"

  ViewVC Help
Powered by ViewVC 1.1.5