ViewVC logotype

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

Parent Directory Parent Directory | Revision Log Revision Log

Revision 562 - (show annotations)
Sun Oct 31 14:06:43 2010 UTC (10 years, 4 months ago) by ph10
File size: 111913 byte(s)
Clarify documentation about comments in patterns.
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 Another special sequence that may appear at the start of a pattern or in
46 combination with (*UTF8) is:
47 .sp
48 (*UCP)
49 .sp
50 This has the same effect as setting the PCRE_UCP option: it causes sequences
51 such as \ed and \ew to use Unicode properties to determine character types,
52 instead of recognizing only characters with codes less than 128 via a lookup
53 table.
54 .P
55 The remainder of this document discusses the patterns that are supported by
56 PCRE when its main matching function, \fBpcre_exec()\fP, is used.
57 From release 6.0, PCRE offers a second matching function,
58 \fBpcre_dfa_exec()\fP, which matches using a different algorithm that is not
59 Perl-compatible. Some of the features discussed below are not available when
60 \fBpcre_dfa_exec()\fP is used. The advantages and disadvantages of the
61 alternative function, and how it differs from the normal function, are
62 discussed in the
63 .\" HREF
64 \fBpcrematching\fP
65 .\"
66 page.
67 .
68 .
69 .\" HTML <a name="newlines"></a>
71 .rs
72 .sp
73 PCRE supports five different conventions for indicating line breaks in
74 strings: a single CR (carriage return) character, a single LF (linefeed)
75 character, the two-character sequence CRLF, any of the three preceding, or any
76 Unicode newline sequence. The
77 .\" HREF
78 \fBpcreapi\fP
79 .\"
80 page has
81 .\" HTML <a href="pcreapi.html#newlines">
82 .\" </a>
83 further discussion
84 .\"
85 about newlines, and shows how to set the newline convention in the
86 \fIoptions\fP arguments for the compiling and matching functions.
87 .P
88 It is also possible to specify a newline convention by starting a pattern
89 string with one of the following five sequences:
90 .sp
91 (*CR) carriage return
92 (*LF) linefeed
93 (*CRLF) carriage return, followed by linefeed
94 (*ANYCRLF) any of the three above
95 (*ANY) all Unicode newline sequences
96 .sp
97 These override the default and the options given to \fBpcre_compile()\fP or
98 \fBpcre_compile2()\fP. For example, on a Unix system where LF is the default
99 newline sequence, the pattern
100 .sp
101 (*CR)a.b
102 .sp
103 changes the convention to CR. That pattern matches "a\enb" because LF is no
104 longer a newline. Note that these special settings, which are not
105 Perl-compatible, are recognized only at the very start of a pattern, and that
106 they must be in upper case. If more than one of them is present, the last one
107 is used.
108 .P
109 The newline convention affects the interpretation of the dot metacharacter when
110 PCRE_DOTALL is not set, and also the behaviour of \eN. However, it does not
111 affect what the \eR escape sequence matches. By default, this is any Unicode
112 newline sequence, for Perl compatibility. However, this can be changed; see the
113 description of \eR in the section entitled
114 .\" HTML <a href="#newlineseq">
115 .\" </a>
116 "Newline sequences"
117 .\"
118 below. A change of \eR setting can be combined with a change of newline
119 convention.
120 .
121 .
123 .rs
124 .sp
125 A regular expression is a pattern that is matched against a subject string from
126 left to right. Most characters stand for themselves in a pattern, and match the
127 corresponding characters in the subject. As a trivial example, the pattern
128 .sp
129 The quick brown fox
130 .sp
131 matches a portion of a subject string that is identical to itself. When
132 caseless matching is specified (the PCRE_CASELESS option), letters are matched
133 independently of case. In UTF-8 mode, PCRE always understands the concept of
134 case for characters whose values are less than 128, so caseless matching is
135 always possible. For characters with higher values, the concept of case is
136 supported if PCRE is compiled with Unicode property support, but not otherwise.
137 If you want to use caseless matching for characters 128 and above, you must
138 ensure that PCRE is compiled with Unicode property support as well as with
139 UTF-8 support.
140 .P
141 The power of regular expressions comes from the ability to include alternatives
142 and repetitions in the pattern. These are encoded in the pattern by the use of
143 \fImetacharacters\fP, which do not stand for themselves but instead are
144 interpreted in some special way.
145 .P
146 There are two different sets of metacharacters: those that are recognized
147 anywhere in the pattern except within square brackets, and those that are
148 recognized within square brackets. Outside square brackets, the metacharacters
149 are as follows:
150 .sp
151 \e general escape character with several uses
152 ^ assert start of string (or line, in multiline mode)
153 $ assert end of string (or line, in multiline mode)
154 . match any character except newline (by default)
155 [ start character class definition
156 | start of alternative branch
157 ( start subpattern
158 ) end subpattern
159 ? extends the meaning of (
160 also 0 or 1 quantifier
161 also quantifier minimizer
162 * 0 or more quantifier
163 + 1 or more quantifier
164 also "possessive quantifier"
165 { start min/max quantifier
166 .sp
167 Part of a pattern that is in square brackets is called a "character class". In
168 a character class the only metacharacters are:
169 .sp
170 \e general escape character
171 ^ negate the class, but only if the first character
172 - indicates character range
173 .\" JOIN
174 [ POSIX character class (only if followed by POSIX
175 syntax)
176 ] terminates the character class
177 .sp
178 The following sections describe the use of each of the metacharacters.
179 .
180 .
182 .rs
183 .sp
184 The backslash character has several uses. Firstly, if it is followed by a
185 non-alphanumeric character, it takes away any special meaning that character
186 may have. This use of backslash as an escape character applies both inside and
187 outside character classes.
188 .P
189 For example, if you want to match a * character, you write \e* in the pattern.
190 This escaping action applies whether or not the following character would
191 otherwise be interpreted as a metacharacter, so it is always safe to precede a
192 non-alphanumeric with backslash to specify that it stands for itself. In
193 particular, if you want to match a backslash, you write \e\e.
194 .P
195 If a pattern is compiled with the PCRE_EXTENDED option, whitespace in the
196 pattern (other than in a character class) and characters between a # outside
197 a character class and the next newline are ignored. An escaping backslash can
198 be used to include a whitespace or # character as part of the pattern.
199 .P
200 If you want to remove the special meaning from a sequence of characters, you
201 can do so by putting them between \eQ and \eE. This is different from Perl in
202 that $ and @ are handled as literals in \eQ...\eE sequences in PCRE, whereas in
203 Perl, $ and @ cause variable interpolation. Note the following examples:
204 .sp
205 Pattern PCRE matches Perl matches
206 .sp
207 .\" JOIN
208 \eQabc$xyz\eE abc$xyz abc followed by the
209 contents of $xyz
210 \eQabc\e$xyz\eE abc\e$xyz abc\e$xyz
211 \eQabc\eE\e$\eQxyz\eE abc$xyz abc$xyz
212 .sp
213 The \eQ...\eE sequence is recognized both inside and outside character classes.
214 An isolated \eE that is not preceded by \eQ is ignored.
215 .
216 .
217 .\" HTML <a name="digitsafterbackslash"></a>
218 .SS "Non-printing characters"
219 .rs
220 .sp
221 A second use of backslash provides a way of encoding non-printing characters
222 in patterns in a visible manner. There is no restriction on the appearance of
223 non-printing characters, apart from the binary zero that terminates a pattern,
224 but when a pattern is being prepared by text editing, it is often easier to use
225 one of the following escape sequences than the binary character it represents:
226 .sp
227 \ea alarm, that is, the BEL character (hex 07)
228 \ecx "control-x", where x is any character
229 \ee escape (hex 1B)
230 \ef formfeed (hex 0C)
231 \en linefeed (hex 0A)
232 \er carriage return (hex 0D)
233 \et tab (hex 09)
234 \eddd character with octal code ddd, or back reference
235 \exhh character with hex code hh
236 \ex{hhh..} character with hex code hhh..
237 .sp
238 The precise effect of \ecx is as follows: if x is a lower case letter, it
239 is converted to upper case. Then bit 6 of the character (hex 40) is inverted.
240 Thus \ecz becomes hex 1A, but \ec{ becomes hex 3B, while \ec; becomes hex
241 7B.
242 .P
243 After \ex, from zero to two hexadecimal digits are read (letters can be in
244 upper or lower case). Any number of hexadecimal digits may appear between \ex{
245 and }, but the value of the character code must be less than 256 in non-UTF-8
246 mode, and less than 2**31 in UTF-8 mode. That is, the maximum value in
247 hexadecimal is 7FFFFFFF. Note that this is bigger than the largest Unicode code
248 point, which is 10FFFF.
249 .P
250 If characters other than hexadecimal digits appear between \ex{ and }, or if
251 there is no terminating }, this form of escape is not recognized. Instead, the
252 initial \ex will be interpreted as a basic hexadecimal escape, with no
253 following digits, giving a character whose value is zero.
254 .P
255 Characters whose value is less than 256 can be defined by either of the two
256 syntaxes for \ex. There is no difference in the way they are handled. For
257 example, \exdc is exactly the same as \ex{dc}.
258 .P
259 After \e0 up to two further octal digits are read. If there are fewer than two
260 digits, just those that are present are used. Thus the sequence \e0\ex\e07
261 specifies two binary zeros followed by a BEL character (code value 7). Make
262 sure you supply two digits after the initial zero if the pattern character that
263 follows is itself an octal digit.
264 .P
265 The handling of a backslash followed by a digit other than 0 is complicated.
266 Outside a character class, PCRE reads it and any following digits as a decimal
267 number. If the number is less than 10, or if there have been at least that many
268 previous capturing left parentheses in the expression, the entire sequence is
269 taken as a \fIback reference\fP. A description of how this works is given
270 .\" HTML <a href="#backreferences">
271 .\" </a>
272 later,
273 .\"
274 following the discussion of
275 .\" HTML <a href="#subpattern">
276 .\" </a>
277 parenthesized subpatterns.
278 .\"
279 .P
280 Inside a character class, or if the decimal number is greater than 9 and there
281 have not been that many capturing subpatterns, PCRE re-reads up to three octal
282 digits following the backslash, and uses them to generate a data character. Any
283 subsequent digits stand for themselves. In non-UTF-8 mode, the value of a
284 character specified in octal must be less than \e400. In UTF-8 mode, values up
285 to \e777 are permitted. For example:
286 .sp
287 \e040 is another way of writing a space
288 .\" JOIN
289 \e40 is the same, provided there are fewer than 40
290 previous capturing subpatterns
291 \e7 is always a back reference
292 .\" JOIN
293 \e11 might be a back reference, or another way of
294 writing a tab
295 \e011 is always a tab
296 \e0113 is a tab followed by the character "3"
297 .\" JOIN
298 \e113 might be a back reference, otherwise the
299 character with octal code 113
300 .\" JOIN
301 \e377 might be a back reference, otherwise
302 the byte consisting entirely of 1 bits
303 .\" JOIN
304 \e81 is either a back reference, or a binary zero
305 followed by the two characters "8" and "1"
306 .sp
307 Note that octal values of 100 or greater must not be introduced by a leading
308 zero, because no more than three octal digits are ever read.
309 .P
310 All the sequences that define a single character value can be used both inside
311 and outside character classes. In addition, inside a character class, the
312 sequence \eb is interpreted as the backspace character (hex 08). The sequences
313 \eB, \eN, \eR, and \eX are not special inside a character class. Like any other
314 unrecognized escape sequences, they are treated as the literal characters "B",
315 "N", "R", and "X" by default, but cause an error if the PCRE_EXTRA option is
316 set. Outside a character class, these sequences have different meanings.
317 .
318 .
319 .SS "Absolute and relative back references"
320 .rs
321 .sp
322 The sequence \eg followed by an unsigned or a negative number, optionally
323 enclosed in braces, is an absolute or relative back reference. A named back
324 reference can be coded as \eg{name}. Back references are discussed
325 .\" HTML <a href="#backreferences">
326 .\" </a>
327 later,
328 .\"
329 following the discussion of
330 .\" HTML <a href="#subpattern">
331 .\" </a>
332 parenthesized subpatterns.
333 .\"
334 .
335 .
336 .SS "Absolute and relative subroutine calls"
337 .rs
338 .sp
339 For compatibility with Oniguruma, the non-Perl syntax \eg followed by a name or
340 a number enclosed either in angle brackets or single quotes, is an alternative
341 syntax for referencing a subpattern as a "subroutine". Details are discussed
342 .\" HTML <a href="#onigurumasubroutines">
343 .\" </a>
344 later.
345 .\"
346 Note that \eg{...} (Perl syntax) and \eg<...> (Oniguruma syntax) are \fInot\fP
347 synonymous. The former is a back reference; the latter is a
348 .\" HTML <a href="#subpatternsassubroutines">
349 .\" </a>
350 subroutine
351 .\"
352 call.
353 .
354 .
355 .\" HTML <a name="genericchartypes"></a>
356 .SS "Generic character types"
357 .rs
358 .sp
359 Another use of backslash is for specifying generic character types:
360 .sp
361 \ed any decimal digit
362 \eD any character that is not a decimal digit
363 \eh any horizontal whitespace character
364 \eH any character that is not a horizontal whitespace character
365 \es any whitespace character
366 \eS any character that is not a whitespace character
367 \ev any vertical whitespace character
368 \eV any character that is not a vertical whitespace character
369 \ew any "word" character
370 \eW any "non-word" character
371 .sp
372 There is also the single sequence \eN, which matches a non-newline character.
373 This is the same as
374 .\" HTML <a href="#fullstopdot">
375 .\" </a>
376 the "." metacharacter
377 .\"
378 when PCRE_DOTALL is not set.
379 .P
380 Each pair of lower and upper case escape sequences partitions the complete set
381 of characters into two disjoint sets. Any given character matches one, and only
382 one, of each pair. The sequences can appear both inside and outside character
383 classes. They each match one character of the appropriate type. If the current
384 matching point is at the end of the subject string, all of them fail, because
385 there is no character to match.
386 .P
387 For compatibility with Perl, \es does not match the VT character (code 11).
388 This makes it different from the the POSIX "space" class. The \es characters
389 are HT (9), LF (10), FF (12), CR (13), and space (32). If "use locale;" is
390 included in a Perl script, \es may match the VT character. In PCRE, it never
391 does.
392 .P
393 A "word" character is an underscore or any character that is a letter or digit.
394 By default, the definition of letters and digits is controlled by PCRE's
395 low-valued character tables, and may vary if locale-specific matching is taking
396 place (see
397 .\" HTML <a href="pcreapi.html#localesupport">
398 .\" </a>
399 "Locale support"
400 .\"
401 in the
402 .\" HREF
403 \fBpcreapi\fP
404 .\"
405 page). For example, in a French locale such as "fr_FR" in Unix-like systems,
406 or "french" in Windows, some character codes greater than 128 are used for
407 accented letters, and these are then matched by \ew. The use of locales with
408 Unicode is discouraged.
409 .P
410 By default, in UTF-8 mode, characters with values greater than 128 never match
411 \ed, \es, or \ew, and always match \eD, \eS, and \eW. These sequences retain
412 their original meanings from before UTF-8 support was available, mainly for
413 efficiency reasons. However, if PCRE is compiled with Unicode property support,
414 and the PCRE_UCP option is set, the behaviour is changed so that Unicode
415 properties are used to determine character types, as follows:
416 .sp
417 \ed any character that \ep{Nd} matches (decimal digit)
418 \es any character that \ep{Z} matches, plus HT, LF, FF, CR
419 \ew any character that \ep{L} or \ep{N} matches, plus underscore
420 .sp
421 The upper case escapes match the inverse sets of characters. Note that \ed
422 matches only decimal digits, whereas \ew matches any Unicode digit, as well as
423 any Unicode letter, and underscore. Note also that PCRE_UCP affects \eb, and
424 \eB because they are defined in terms of \ew and \eW. Matching these sequences
425 is noticeably slower when PCRE_UCP is set.
426 .P
427 The sequences \eh, \eH, \ev, and \eV are Perl 5.10 features. In contrast to the
428 other sequences, which match only ASCII characters by default, these always
429 match certain high-valued codepoints in UTF-8 mode, whether or not PCRE_UCP is
430 set. The horizontal space characters are:
431 .sp
432 U+0009 Horizontal tab
433 U+0020 Space
434 U+00A0 Non-break space
435 U+1680 Ogham space mark
436 U+180E Mongolian vowel separator
437 U+2000 En quad
438 U+2001 Em quad
439 U+2002 En space
440 U+2003 Em space
441 U+2004 Three-per-em space
442 U+2005 Four-per-em space
443 U+2006 Six-per-em space
444 U+2007 Figure space
445 U+2008 Punctuation space
446 U+2009 Thin space
447 U+200A Hair space
448 U+202F Narrow no-break space
449 U+205F Medium mathematical space
450 U+3000 Ideographic space
451 .sp
452 The vertical space characters are:
453 .sp
454 U+000A Linefeed
455 U+000B Vertical tab
456 U+000C Formfeed
457 U+000D Carriage return
458 U+0085 Next line
459 U+2028 Line separator
460 U+2029 Paragraph separator
461 .
462 .
463 .\" HTML <a name="newlineseq"></a>
464 .SS "Newline sequences"
465 .rs
466 .sp
467 Outside a character class, by default, the escape sequence \eR matches any
468 Unicode newline sequence. This is a Perl 5.10 feature. In non-UTF-8 mode \eR is
469 equivalent to the following:
470 .sp
471 (?>\er\en|\en|\ex0b|\ef|\er|\ex85)
472 .sp
473 This is an example of an "atomic group", details of which are given
474 .\" HTML <a href="#atomicgroup">
475 .\" </a>
476 below.
477 .\"
478 This particular group matches either the two-character sequence CR followed by
479 LF, or one of the single characters LF (linefeed, U+000A), VT (vertical tab,
480 U+000B), FF (formfeed, U+000C), CR (carriage return, U+000D), or NEL (next
481 line, U+0085). The two-character sequence is treated as a single unit that
482 cannot be split.
483 .P
484 In UTF-8 mode, two additional characters whose codepoints are greater than 255
485 are added: LS (line separator, U+2028) and PS (paragraph separator, U+2029).
486 Unicode character property support is not needed for these characters to be
487 recognized.
488 .P
489 It is possible to restrict \eR to match only CR, LF, or CRLF (instead of the
490 complete set of Unicode line endings) by setting the option PCRE_BSR_ANYCRLF
491 either at compile time or when the pattern is matched. (BSR is an abbrevation
492 for "backslash R".) This can be made the default when PCRE is built; if this is
493 the case, the other behaviour can be requested via the PCRE_BSR_UNICODE option.
494 It is also possible to specify these settings by starting a pattern string with
495 one of the following sequences:
496 .sp
497 (*BSR_ANYCRLF) CR, LF, or CRLF only
498 (*BSR_UNICODE) any Unicode newline sequence
499 .sp
500 These override the default and the options given to \fBpcre_compile()\fP or
501 \fBpcre_compile2()\fP, but they can be overridden by options given to
502 \fBpcre_exec()\fP or \fBpcre_dfa_exec()\fP. Note that these special settings,
503 which are not Perl-compatible, are recognized only at the very start of a
504 pattern, and that they must be in upper case. If more than one of them is
505 present, the last one is used. They can be combined with a change of newline
506 convention; for example, a pattern can start with:
507 .sp
509 .sp
510 They can also be combined with the (*UTF8) or (*UCP) special sequences. Inside
511 a character class, \eR is treated as an unrecognized escape sequence, and so
512 matches the letter "R" by default, but causes an error if PCRE_EXTRA is set.
513 .
514 .
515 .\" HTML <a name="uniextseq"></a>
516 .SS Unicode character properties
517 .rs
518 .sp
519 When PCRE is built with Unicode character property support, three additional
520 escape sequences that match characters with specific properties are available.
521 When not in UTF-8 mode, these sequences are of course limited to testing
522 characters whose codepoints are less than 256, but they do work in this mode.
523 The extra escape sequences are:
524 .sp
525 \ep{\fIxx\fP} a character with the \fIxx\fP property
526 \eP{\fIxx\fP} a character without the \fIxx\fP property
527 \eX an extended Unicode sequence
528 .sp
529 The property names represented by \fIxx\fP above are limited to the Unicode
530 script names, the general category properties, "Any", which matches any
531 character (including newline), and some special PCRE properties (described
532 in the
533 .\" HTML <a href="#extraprops">
534 .\" </a>
535 next section).
536 .\"
537 Other Perl properties such as "InMusicalSymbols" are not currently supported by
538 PCRE. Note that \eP{Any} does not match any characters, so always causes a
539 match failure.
540 .P
541 Sets of Unicode characters are defined as belonging to certain scripts. A
542 character from one of these sets can be matched using a script name. For
543 example:
544 .sp
545 \ep{Greek}
546 \eP{Han}
547 .sp
548 Those that are not part of an identified script are lumped together as
549 "Common". The current list of scripts is:
550 .P
551 Arabic,
552 Armenian,
553 Avestan,
554 Balinese,
555 Bamum,
556 Bengali,
557 Bopomofo,
558 Braille,
559 Buginese,
560 Buhid,
561 Canadian_Aboriginal,
562 Carian,
563 Cham,
564 Cherokee,
565 Common,
566 Coptic,
567 Cuneiform,
568 Cypriot,
569 Cyrillic,
570 Deseret,
571 Devanagari,
572 Egyptian_Hieroglyphs,
573 Ethiopic,
574 Georgian,
575 Glagolitic,
576 Gothic,
577 Greek,
578 Gujarati,
579 Gurmukhi,
580 Han,
581 Hangul,
582 Hanunoo,
583 Hebrew,
584 Hiragana,
585 Imperial_Aramaic,
586 Inherited,
587 Inscriptional_Pahlavi,
588 Inscriptional_Parthian,
589 Javanese,
590 Kaithi,
591 Kannada,
592 Katakana,
593 Kayah_Li,
594 Kharoshthi,
595 Khmer,
596 Lao,
597 Latin,
598 Lepcha,
599 Limbu,
600 Linear_B,
601 Lisu,
602 Lycian,
603 Lydian,
604 Malayalam,
605 Meetei_Mayek,
606 Mongolian,
607 Myanmar,
608 New_Tai_Lue,
609 Nko,
610 Ogham,
611 Old_Italic,
612 Old_Persian,
613 Old_South_Arabian,
614 Old_Turkic,
615 Ol_Chiki,
616 Oriya,
617 Osmanya,
618 Phags_Pa,
619 Phoenician,
620 Rejang,
621 Runic,
622 Samaritan,
623 Saurashtra,
624 Shavian,
625 Sinhala,
626 Sundanese,
627 Syloti_Nagri,
628 Syriac,
629 Tagalog,
630 Tagbanwa,
631 Tai_Le,
632 Tai_Tham,
633 Tai_Viet,
634 Tamil,
635 Telugu,
636 Thaana,
637 Thai,
638 Tibetan,
639 Tifinagh,
640 Ugaritic,
641 Vai,
642 Yi.
643 .P
644 Each character has exactly one Unicode general category property, specified by
645 a two-letter abbreviation. For compatibility with Perl, negation can be
646 specified by including a circumflex between the opening brace and the property
647 name. For example, \ep{^Lu} is the same as \eP{Lu}.
648 .P
649 If only one letter is specified with \ep or \eP, it includes all the general
650 category properties that start with that letter. In this case, in the absence
651 of negation, the curly brackets in the escape sequence are optional; these two
652 examples have the same effect:
653 .sp
654 \ep{L}
655 \epL
656 .sp
657 The following general category property codes are supported:
658 .sp
659 C Other
660 Cc Control
661 Cf Format
662 Cn Unassigned
663 Co Private use
664 Cs Surrogate
665 .sp
666 L Letter
667 Ll Lower case letter
668 Lm Modifier letter
669 Lo Other letter
670 Lt Title case letter
671 Lu Upper case letter
672 .sp
673 M Mark
674 Mc Spacing mark
675 Me Enclosing mark
676 Mn Non-spacing mark
677 .sp
678 N Number
679 Nd Decimal number
680 Nl Letter number
681 No Other number
682 .sp
683 P Punctuation
684 Pc Connector punctuation
685 Pd Dash punctuation
686 Pe Close punctuation
687 Pf Final punctuation
688 Pi Initial punctuation
689 Po Other punctuation
690 Ps Open punctuation
691 .sp
692 S Symbol
693 Sc Currency symbol
694 Sk Modifier symbol
695 Sm Mathematical symbol
696 So Other symbol
697 .sp
698 Z Separator
699 Zl Line separator
700 Zp Paragraph separator
701 Zs Space separator
702 .sp
703 The special property L& is also supported: it matches a character that has
704 the Lu, Ll, or Lt property, in other words, a letter that is not classified as
705 a modifier or "other".
706 .P
707 The Cs (Surrogate) property applies only to characters in the range U+D800 to
708 U+DFFF. Such characters are not valid in UTF-8 strings (see RFC 3629) and so
709 cannot be tested by PCRE, unless UTF-8 validity checking has been turned off
710 (see the discussion of PCRE_NO_UTF8_CHECK in the
711 .\" HREF
712 \fBpcreapi\fP
713 .\"
714 page). Perl does not support the Cs property.
715 .P
716 The long synonyms for property names that Perl supports (such as \ep{Letter})
717 are not supported by PCRE, nor is it permitted to prefix any of these
718 properties with "Is".
719 .P
720 No character that is in the Unicode table has the Cn (unassigned) property.
721 Instead, this property is assumed for any code point that is not in the
722 Unicode table.
723 .P
724 Specifying caseless matching does not affect these escape sequences. For
725 example, \ep{Lu} always matches only upper case letters.
726 .P
727 The \eX escape matches any number of Unicode characters that form an extended
728 Unicode sequence. \eX is equivalent to
729 .sp
730 (?>\ePM\epM*)
731 .sp
732 That is, it matches a character without the "mark" property, followed by zero
733 or more characters with the "mark" property, and treats the sequence as an
734 atomic group
735 .\" HTML <a href="#atomicgroup">
736 .\" </a>
737 (see below).
738 .\"
739 Characters with the "mark" property are typically accents that affect the
740 preceding character. None of them have codepoints less than 256, so in
741 non-UTF-8 mode \eX matches any one character.
742 .P
743 Matching characters by Unicode property is not fast, because PCRE has to search
744 a structure that contains data for over fifteen thousand characters. That is
745 why the traditional escape sequences such as \ed and \ew do not use Unicode
746 properties in PCRE by default, though you can make them do so by setting the
747 PCRE_UCP option for \fBpcre_compile()\fP or by starting the pattern with
748 (*UCP).
749 .
750 .
751 .\" HTML <a name="extraprops"></a>
752 .SS PCRE's additional properties
753 .rs
754 .sp
755 As well as the standard Unicode properties described in the previous
756 section, PCRE supports four more that make it possible to convert traditional
757 escape sequences such as \ew and \es and POSIX character classes to use Unicode
758 properties. PCRE uses these non-standard, non-Perl properties internally when
759 PCRE_UCP is set. They are:
760 .sp
761 Xan Any alphanumeric character
762 Xps Any POSIX space character
763 Xsp Any Perl space character
764 Xwd Any Perl "word" character
765 .sp
766 Xan matches characters that have either the L (letter) or the N (number)
767 property. Xps matches the characters tab, linefeed, vertical tab, formfeed, or
768 carriage return, and any other character that has the Z (separator) property.
769 Xsp is the same as Xps, except that vertical tab is excluded. Xwd matches the
770 same characters as Xan, plus underscore.
771 .
772 .
773 .\" HTML <a name="resetmatchstart"></a>
774 .SS "Resetting the match start"
775 .rs
776 .sp
777 The escape sequence \eK, which is a Perl 5.10 feature, causes any previously
778 matched characters not to be included in the final matched sequence. For
779 example, the pattern:
780 .sp
781 foo\eKbar
782 .sp
783 matches "foobar", but reports that it has matched "bar". This feature is
784 similar to a lookbehind assertion
785 .\" HTML <a href="#lookbehind">
786 .\" </a>
787 (described below).
788 .\"
789 However, in this case, the part of the subject before the real match does not
790 have to be of fixed length, as lookbehind assertions do. The use of \eK does
791 not interfere with the setting of
792 .\" HTML <a href="#subpattern">
793 .\" </a>
794 captured substrings.
795 .\"
796 For example, when the pattern
797 .sp
798 (foo)\eKbar
799 .sp
800 matches "foobar", the first substring is still set to "foo".
801 .P
802 Perl documents that the use of \eK within assertions is "not well defined". In
803 PCRE, \eK is acted upon when it occurs inside positive assertions, but is
804 ignored in negative assertions.
805 .
806 .
807 .\" HTML <a name="smallassertions"></a>
808 .SS "Simple assertions"
809 .rs
810 .sp
811 The final use of backslash is for certain simple assertions. An assertion
812 specifies a condition that has to be met at a particular point in a match,
813 without consuming any characters from the subject string. The use of
814 subpatterns for more complicated assertions is described
815 .\" HTML <a href="#bigassertions">
816 .\" </a>
817 below.
818 .\"
819 The backslashed assertions are:
820 .sp
821 \eb matches at a word boundary
822 \eB matches when not at a word boundary
823 \eA matches at the start of the subject
824 \eZ matches at the end of the subject
825 also matches before a newline at the end of the subject
826 \ez matches only at the end of the subject
827 \eG matches at the first matching position in the subject
828 .sp
829 Inside a character class, \eb has a different meaning; it matches the backspace
830 character. If any other of these assertions appears in a character class, by
831 default it matches the corresponding literal character (for example, \eB
832 matches the letter B). However, if the PCRE_EXTRA option is set, an "invalid
833 escape sequence" error is generated instead.
834 .P
835 A word boundary is a position in the subject string where the current character
836 and the previous character do not both match \ew or \eW (i.e. one matches
837 \ew and the other matches \eW), or the start or end of the string if the
838 first or last character matches \ew, respectively. In UTF-8 mode, the meanings
839 of \ew and \eW can be changed by setting the PCRE_UCP option. When this is
840 done, it also affects \eb and \eB. Neither PCRE nor Perl has a separate "start
841 of word" or "end of word" metasequence. However, whatever follows \eb normally
842 determines which it is. For example, the fragment \eba matches "a" at the start
843 of a word.
844 .P
845 The \eA, \eZ, and \ez assertions differ from the traditional circumflex and
846 dollar (described in the next section) in that they only ever match at the very
847 start and end of the subject string, whatever options are set. Thus, they are
848 independent of multiline mode. These three assertions are not affected by the
849 PCRE_NOTBOL or PCRE_NOTEOL options, which affect only the behaviour of the
850 circumflex and dollar metacharacters. However, if the \fIstartoffset\fP
851 argument of \fBpcre_exec()\fP is non-zero, indicating that matching is to start
852 at a point other than the beginning of the subject, \eA can never match. The
853 difference between \eZ and \ez is that \eZ matches before a newline at the end
854 of the string as well as at the very end, whereas \ez matches only at the end.
855 .P
856 The \eG assertion is true only when the current matching position is at the
857 start point of the match, as specified by the \fIstartoffset\fP argument of
858 \fBpcre_exec()\fP. It differs from \eA when the value of \fIstartoffset\fP is
859 non-zero. By calling \fBpcre_exec()\fP multiple times with appropriate
860 arguments, you can mimic Perl's /g option, and it is in this kind of
861 implementation where \eG can be useful.
862 .P
863 Note, however, that PCRE's interpretation of \eG, as the start of the current
864 match, is subtly different from Perl's, which defines it as the end of the
865 previous match. In Perl, these can be different when the previously matched
866 string was empty. Because PCRE does just one match at a time, it cannot
867 reproduce this behaviour.
868 .P
869 If all the alternatives of a pattern begin with \eG, the expression is anchored
870 to the starting match position, and the "anchored" flag is set in the compiled
871 regular expression.
872 .
873 .
875 .rs
876 .sp
877 Outside a character class, in the default matching mode, the circumflex
878 character is an assertion that is true only if the current matching point is
879 at the start of the subject string. If the \fIstartoffset\fP argument of
880 \fBpcre_exec()\fP is non-zero, circumflex can never match if the PCRE_MULTILINE
881 option is unset. Inside a character class, circumflex has an entirely different
882 meaning
883 .\" HTML <a href="#characterclass">
884 .\" </a>
885 (see below).
886 .\"
887 .P
888 Circumflex need not be the first character of the pattern if a number of
889 alternatives are involved, but it should be the first thing in each alternative
890 in which it appears if the pattern is ever to match that branch. If all
891 possible alternatives start with a circumflex, that is, if the pattern is
892 constrained to match only at the start of the subject, it is said to be an
893 "anchored" pattern. (There are also other constructs that can cause a pattern
894 to be anchored.)
895 .P
896 A dollar character is an assertion that is true only if the current matching
897 point is at the end of the subject string, or immediately before a newline
898 at the end of the string (by default). Dollar need not be the last character of
899 the pattern if a number of alternatives are involved, but it should be the last
900 item in any branch in which it appears. Dollar has no special meaning in a
901 character class.
902 .P
903 The meaning of dollar can be changed so that it matches only at the very end of
904 the string, by setting the PCRE_DOLLAR_ENDONLY option at compile time. This
905 does not affect the \eZ assertion.
906 .P
907 The meanings of the circumflex and dollar characters are changed if the
908 PCRE_MULTILINE option is set. When this is the case, a circumflex matches
909 immediately after internal newlines as well as at the start of the subject
910 string. It does not match after a newline that ends the string. A dollar
911 matches before any newlines in the string, as well as at the very end, when
912 PCRE_MULTILINE is set. When newline is specified as the two-character
913 sequence CRLF, isolated CR and LF characters do not indicate newlines.
914 .P
915 For example, the pattern /^abc$/ matches the subject string "def\enabc" (where
916 \en represents a newline) in multiline mode, but not otherwise. Consequently,
917 patterns that are anchored in single line mode because all branches start with
918 ^ are not anchored in multiline mode, and a match for circumflex is possible
919 when the \fIstartoffset\fP argument of \fBpcre_exec()\fP is non-zero. The
920 PCRE_DOLLAR_ENDONLY option is ignored if PCRE_MULTILINE is set.
921 .P
922 Note that the sequences \eA, \eZ, and \ez can be used to match the start and
923 end of the subject in both modes, and if all branches of a pattern start with
924 \eA it is always anchored, whether or not PCRE_MULTILINE is set.
925 .
926 .
927 .\" HTML <a name="fullstopdot"></a>
929 .rs
930 .sp
931 Outside a character class, a dot in the pattern matches any one character in
932 the subject string except (by default) a character that signifies the end of a
933 line. In UTF-8 mode, the matched character may be more than one byte long.
934 .P
935 When a line ending is defined as a single character, dot never matches that
936 character; when the two-character sequence CRLF is used, dot does not match CR
937 if it is immediately followed by LF, but otherwise it matches all characters
938 (including isolated CRs and LFs). When any Unicode line endings are being
939 recognized, dot does not match CR or LF or any of the other line ending
940 characters.
941 .P
942 The behaviour of dot with regard to newlines can be changed. If the PCRE_DOTALL
943 option is set, a dot matches any one character, without exception. If the
944 two-character sequence CRLF is present in the subject string, it takes two dots
945 to match it.
946 .P
947 The handling of dot is entirely independent of the handling of circumflex and
948 dollar, the only relationship being that they both involve newlines. Dot has no
949 special meaning in a character class.
950 .P
951 The escape sequence \eN always behaves as a dot does when PCRE_DOTALL is not
952 set. In other words, it matches any one character except one that signifies the
953 end of a line.
954 .
955 .
957 .rs
958 .sp
959 Outside a character class, the escape sequence \eC matches any one byte, both
960 in and out of UTF-8 mode. Unlike a dot, it always matches any line-ending
961 characters. The feature is provided in Perl in order to match individual bytes
962 in UTF-8 mode. Because it breaks up UTF-8 characters into individual bytes,
963 what remains in the string may be a malformed UTF-8 string. For this reason,
964 the \eC escape sequence is best avoided.
965 .P
966 PCRE does not allow \eC to appear in lookbehind assertions
967 .\" HTML <a href="#lookbehind">
968 .\" </a>
969 (described below),
970 .\"
971 because in UTF-8 mode this would make it impossible to calculate the length of
972 the lookbehind.
973 .
974 .
975 .\" HTML <a name="characterclass"></a>
977 .rs
978 .sp
979 An opening square bracket introduces a character class, terminated by a closing
980 square bracket. A closing square bracket on its own is not special by default.
981 However, if the PCRE_JAVASCRIPT_COMPAT option is set, a lone closing square
982 bracket causes a compile-time error. If a closing square bracket is required as
983 a member of the class, it should be the first data character in the class
984 (after an initial circumflex, if present) or escaped with a backslash.
985 .P
986 A character class matches a single character in the subject. In UTF-8 mode, the
987 character may be more than one byte long. A matched character must be in the
988 set of characters defined by the class, unless the first character in the class
989 definition is a circumflex, in which case the subject character must not be in
990 the set defined by the class. If a circumflex is actually required as a member
991 of the class, ensure it is not the first character, or escape it with a
992 backslash.
993 .P
994 For example, the character class [aeiou] matches any lower case vowel, while
995 [^aeiou] matches any character that is not a lower case vowel. Note that a
996 circumflex is just a convenient notation for specifying the characters that
997 are in the class by enumerating those that are not. A class that starts with a
998 circumflex is not an assertion; it still consumes a character from the subject
999 string, and therefore it fails if the current pointer is at the end of the
1000 string.
1001 .P
1002 In UTF-8 mode, characters with values greater than 255 can be included in a
1003 class as a literal string of bytes, or by using the \ex{ escaping mechanism.
1004 .P
1005 When caseless matching is set, any letters in a class represent both their
1006 upper case and lower case versions, so for example, a caseless [aeiou] matches
1007 "A" as well as "a", and a caseless [^aeiou] does not match "A", whereas a
1008 caseful version would. In UTF-8 mode, PCRE always understands the concept of
1009 case for characters whose values are less than 128, so caseless matching is
1010 always possible. For characters with higher values, the concept of case is
1011 supported if PCRE is compiled with Unicode property support, but not otherwise.
1012 If you want to use caseless matching in UTF8-mode for characters 128 and above,
1013 you must ensure that PCRE is compiled with Unicode property support as well as
1014 with UTF-8 support.
1015 .P
1016 Characters that might indicate line breaks are never treated in any special way
1017 when matching character classes, whatever line-ending sequence is in use, and
1018 whatever setting of the PCRE_DOTALL and PCRE_MULTILINE options is used. A class
1019 such as [^a] always matches one of these characters.
1020 .P
1021 The minus (hyphen) character can be used to specify a range of characters in a
1022 character class. For example, [d-m] matches any letter between d and m,
1023 inclusive. If a minus character is required in a class, it must be escaped with
1024 a backslash or appear in a position where it cannot be interpreted as
1025 indicating a range, typically as the first or last character in the class.
1026 .P
1027 It is not possible to have the literal character "]" as the end character of a
1028 range. A pattern such as [W-]46] is interpreted as a class of two characters
1029 ("W" and "-") followed by a literal string "46]", so it would match "W46]" or
1030 "-46]". However, if the "]" is escaped with a backslash it is interpreted as
1031 the end of range, so [W-\e]46] is interpreted as a class containing a range
1032 followed by two other characters. The octal or hexadecimal representation of
1033 "]" can also be used to end a range.
1034 .P
1035 Ranges operate in the collating sequence of character values. They can also be
1036 used for characters specified numerically, for example [\e000-\e037]. In UTF-8
1037 mode, ranges can include characters whose values are greater than 255, for
1038 example [\ex{100}-\ex{2ff}].
1039 .P
1040 If a range that includes letters is used when caseless matching is set, it
1041 matches the letters in either case. For example, [W-c] is equivalent to
1042 [][\e\e^_`wxyzabc], matched caselessly, and in non-UTF-8 mode, if character
1043 tables for a French locale are in use, [\exc8-\excb] matches accented E
1044 characters in both cases. In UTF-8 mode, PCRE supports the concept of case for
1045 characters with values greater than 128 only when it is compiled with Unicode
1046 property support.
1047 .P
1048 The character types \ed, \eD, \eh, \eH, \ep, \eP, \es, \eS, \ev, \eV, \ew, and
1049 \eW may also appear in a character class, and add the characters that they
1050 match to the class. For example, [\edABCDEF] matches any hexadecimal digit. A
1051 circumflex can conveniently be used with the upper case character types to
1052 specify a more restricted set of characters than the matching lower case type.
1053 For example, the class [^\eW_] matches any letter or digit, but not underscore.
1054 .P
1055 The only metacharacters that are recognized in character classes are backslash,
1056 hyphen (only where it can be interpreted as specifying a range), circumflex
1057 (only at the start), opening square bracket (only when it can be interpreted as
1058 introducing a POSIX class name - see the next section), and the terminating
1059 closing square bracket. However, escaping other non-alphanumeric characters
1060 does no harm.
1061 .
1062 .
1064 .rs
1065 .sp
1066 Perl supports the POSIX notation for character classes. This uses names
1067 enclosed by [: and :] within the enclosing square brackets. PCRE also supports
1068 this notation. For example,
1069 .sp
1070 [01[:alpha:]%]
1071 .sp
1072 matches "0", "1", any alphabetic character, or "%". The supported class names
1073 are:
1074 .sp
1075 alnum letters and digits
1076 alpha letters
1077 ascii character codes 0 - 127
1078 blank space or tab only
1079 cntrl control characters
1080 digit decimal digits (same as \ed)
1081 graph printing characters, excluding space
1082 lower lower case letters
1083 print printing characters, including space
1084 punct printing characters, excluding letters and digits and space
1085 space white space (not quite the same as \es)
1086 upper upper case letters
1087 word "word" characters (same as \ew)
1088 xdigit hexadecimal digits
1089 .sp
1090 The "space" characters are HT (9), LF (10), VT (11), FF (12), CR (13), and
1091 space (32). Notice that this list includes the VT character (code 11). This
1092 makes "space" different to \es, which does not include VT (for Perl
1093 compatibility).
1094 .P
1095 The name "word" is a Perl extension, and "blank" is a GNU extension from Perl
1096 5.8. Another Perl extension is negation, which is indicated by a ^ character
1097 after the colon. For example,
1098 .sp
1099 [12[:^digit:]]
1100 .sp
1101 matches "1", "2", or any non-digit. PCRE (and Perl) also recognize the POSIX
1102 syntax [.ch.] and [=ch=] where "ch" is a "collating element", but these are not
1103 supported, and an error is given if they are encountered.
1104 .P
1105 By default, in UTF-8 mode, characters with values greater than 128 do not match
1106 any of the POSIX character classes. However, if the PCRE_UCP option is passed
1107 to \fBpcre_compile()\fP, some of the classes are changed so that Unicode
1108 character properties are used. This is achieved by replacing the POSIX classes
1109 by other sequences, as follows:
1110 .sp
1111 [:alnum:] becomes \ep{Xan}
1112 [:alpha:] becomes \ep{L}
1113 [:blank:] becomes \eh
1114 [:digit:] becomes \ep{Nd}
1115 [:lower:] becomes \ep{Ll}
1116 [:space:] becomes \ep{Xps}
1117 [:upper:] becomes \ep{Lu}
1118 [:word:] becomes \ep{Xwd}
1119 .sp
1120 Negated versions, such as [:^alpha:] use \eP instead of \ep. The other POSIX
1121 classes are unchanged, and match only characters with code points less than
1122 128.
1123 .
1124 .
1126 .rs
1127 .sp
1128 Vertical bar characters are used to separate alternative patterns. For example,
1129 the pattern
1130 .sp
1131 gilbert|sullivan
1132 .sp
1133 matches either "gilbert" or "sullivan". Any number of alternatives may appear,
1134 and an empty alternative is permitted (matching the empty string). The matching
1135 process tries each alternative in turn, from left to right, and the first one
1136 that succeeds is used. If the alternatives are within a subpattern
1137 .\" HTML <a href="#subpattern">
1138 .\" </a>
1139 (defined below),
1140 .\"
1141 "succeeds" means matching the rest of the main pattern as well as the
1142 alternative in the subpattern.
1143 .
1144 .
1146 .rs
1147 .sp
1148 The settings of the PCRE_CASELESS, PCRE_MULTILINE, PCRE_DOTALL, and
1149 PCRE_EXTENDED options (which are Perl-compatible) can be changed from within
1150 the pattern by a sequence of Perl option letters enclosed between "(?" and ")".
1151 The option letters are
1152 .sp
1153 i for PCRE_CASELESS
1155 s for PCRE_DOTALL
1156 x for PCRE_EXTENDED
1157 .sp
1158 For example, (?im) sets caseless, multiline matching. It is also possible to
1159 unset these options by preceding the letter with a hyphen, and a combined
1160 setting and unsetting such as (?im-sx), which sets PCRE_CASELESS and
1161 PCRE_MULTILINE while unsetting PCRE_DOTALL and PCRE_EXTENDED, is also
1162 permitted. If a letter appears both before and after the hyphen, the option is
1163 unset.
1164 .P
1165 The PCRE-specific options PCRE_DUPNAMES, PCRE_UNGREEDY, and PCRE_EXTRA can be
1166 changed in the same way as the Perl-compatible options by using the characters
1167 J, U and X respectively.
1168 .P
1169 When one of these option changes occurs at top level (that is, not inside
1170 subpattern parentheses), the change applies to the remainder of the pattern
1171 that follows. If the change is placed right at the start of a pattern, PCRE
1172 extracts it into the global options (and it will therefore show up in data
1173 extracted by the \fBpcre_fullinfo()\fP function).
1174 .P
1175 An option change within a subpattern (see below for a description of
1176 subpatterns) affects only that part of the current pattern that follows it, so
1177 .sp
1178 (a(?i)b)c
1179 .sp
1180 matches abc and aBc and no other strings (assuming PCRE_CASELESS is not used).
1181 By this means, options can be made to have different settings in different
1182 parts of the pattern. Any changes made in one alternative do carry on
1183 into subsequent branches within the same subpattern. For example,
1184 .sp
1185 (a(?i)b|c)
1186 .sp
1187 matches "ab", "aB", "c", and "C", even though when matching "C" the first
1188 branch is abandoned before the option setting. This is because the effects of
1189 option settings happen at compile time. There would be some very weird
1190 behaviour otherwise.
1191 .P
1192 \fBNote:\fP There are other PCRE-specific options that can be set by the
1193 application when the compile or match functions are called. In some cases the
1194 pattern can contain special leading sequences such as (*CRLF) to override what
1195 the application has set or what has been defaulted. Details are given in the
1196 section entitled
1197 .\" HTML <a href="#newlineseq">
1198 .\" </a>
1199 "Newline sequences"
1200 .\"
1201 above. There are also the (*UTF8) and (*UCP) leading sequences that can be used
1202 to set UTF-8 and Unicode property modes; they are equivalent to setting the
1203 PCRE_UTF8 and the PCRE_UCP options, respectively.
1204 .
1205 .
1206 .\" HTML <a name="subpattern"></a>
1208 .rs
1209 .sp
1210 Subpatterns are delimited by parentheses (round brackets), which can be nested.
1211 Turning part of a pattern into a subpattern does two things:
1212 .sp
1213 1. It localizes a set of alternatives. For example, the pattern
1214 .sp
1215 cat(aract|erpillar|)
1216 .sp
1217 matches one of the words "cat", "cataract", or "caterpillar". Without the
1218 parentheses, it would match "cataract", "erpillar" or an empty string.
1219 .sp
1220 2. It sets up the subpattern as a capturing subpattern. This means that, when
1221 the whole pattern matches, that portion of the subject string that matched the
1222 subpattern is passed back to the caller via the \fIovector\fP argument of
1223 \fBpcre_exec()\fP. Opening parentheses are counted from left to right (starting
1224 from 1) to obtain numbers for the capturing subpatterns.
1225 .P
1226 For example, if the string "the red king" is matched against the pattern
1227 .sp
1228 the ((red|white) (king|queen))
1229 .sp
1230 the captured substrings are "red king", "red", and "king", and are numbered 1,
1231 2, and 3, respectively.
1232 .P
1233 The fact that plain parentheses fulfil two functions is not always helpful.
1234 There are often times when a grouping subpattern is required without a
1235 capturing requirement. If an opening parenthesis is followed by a question mark
1236 and a colon, the subpattern does not do any capturing, and is not counted when
1237 computing the number of any subsequent capturing subpatterns. For example, if
1238 the string "the white queen" is matched against the pattern
1239 .sp
1240 the ((?:red|white) (king|queen))
1241 .sp
1242 the captured substrings are "white queen" and "queen", and are numbered 1 and
1243 2. The maximum number of capturing subpatterns is 65535.
1244 .P
1245 As a convenient shorthand, if any option settings are required at the start of
1246 a non-capturing subpattern, the option letters may appear between the "?" and
1247 the ":". Thus the two patterns
1248 .sp
1249 (?i:saturday|sunday)
1250 (?:(?i)saturday|sunday)
1251 .sp
1252 match exactly the same set of strings. Because alternative branches are tried
1253 from left to right, and options are not reset until the end of the subpattern
1254 is reached, an option setting in one branch does affect subsequent branches, so
1255 the above patterns match "SUNDAY" as well as "Saturday".
1256 .
1257 .
1258 .\" HTML <a name="dupsubpatternnumber"></a>
1260 .rs
1261 .sp
1262 Perl 5.10 introduced a feature whereby each alternative in a subpattern uses
1263 the same numbers for its capturing parentheses. Such a subpattern starts with
1264 (?| and is itself a non-capturing subpattern. For example, consider this
1265 pattern:
1266 .sp
1267 (?|(Sat)ur|(Sun))day
1268 .sp
1269 Because the two alternatives are inside a (?| group, both sets of capturing
1270 parentheses are numbered one. Thus, when the pattern matches, you can look
1271 at captured substring number one, whichever alternative matched. This construct
1272 is useful when you want to capture part, but not all, of one of a number of
1273 alternatives. Inside a (?| group, parentheses are numbered as usual, but the
1274 number is reset at the start of each branch. The numbers of any capturing
1275 buffers that follow the subpattern start after the highest number used in any
1276 branch. The following example is taken from the Perl documentation.
1277 The numbers underneath show in which buffer the captured content will be
1278 stored.
1279 .sp
1280 # before ---------------branch-reset----------- after
1281 / ( a ) (?| x ( y ) z | (p (q) r) | (t) u (v) ) ( z ) /x
1282 # 1 2 2 3 2 3 4
1283 .sp
1284 A back reference to a numbered subpattern uses the most recent value that is
1285 set for that number by any subpattern. The following pattern matches "abcabc"
1286 or "defdef":
1287 .sp
1288 /(?|(abc)|(def))\e1/
1289 .sp
1290 In contrast, a recursive or "subroutine" call to a numbered subpattern always
1291 refers to the first one in the pattern with the given number. The following
1292 pattern matches "abcabc" or "defabc":
1293 .sp
1294 /(?|(abc)|(def))(?1)/
1295 .sp
1296 If a
1297 .\" HTML <a href="#conditions">
1298 .\" </a>
1299 condition test
1300 .\"
1301 for a subpattern's having matched refers to a non-unique number, the test is
1302 true if any of the subpatterns of that number have matched.
1303 .P
1304 An alternative approach to using this "branch reset" feature is to use
1305 duplicate named subpatterns, as described in the next section.
1306 .
1307 .
1309 .rs
1310 .sp
1311 Identifying capturing parentheses by number is simple, but it can be very hard
1312 to keep track of the numbers in complicated regular expressions. Furthermore,
1313 if an expression is modified, the numbers may change. To help with this
1314 difficulty, PCRE supports the naming of subpatterns. This feature was not
1315 added to Perl until release 5.10. Python had the feature earlier, and PCRE
1316 introduced it at release 4.0, using the Python syntax. PCRE now supports both
1317 the Perl and the Python syntax. Perl allows identically numbered subpatterns to
1318 have different names, but PCRE does not.
1319 .P
1320 In PCRE, a subpattern can be named in one of three ways: (?<name>...) or
1321 (?'name'...) as in Perl, or (?P<name>...) as in Python. References to capturing
1322 parentheses from other parts of the pattern, such as
1323 .\" HTML <a href="#backreferences">
1324 .\" </a>
1325 back references,
1326 .\"
1327 .\" HTML <a href="#recursion">
1328 .\" </a>
1329 recursion,
1330 .\"
1331 and
1332 .\" HTML <a href="#conditions">
1333 .\" </a>
1334 conditions,
1335 .\"
1336 can be made by name as well as by number.
1337 .P
1338 Names consist of up to 32 alphanumeric characters and underscores. Named
1339 capturing parentheses are still allocated numbers as well as names, exactly as
1340 if the names were not present. The PCRE API provides function calls for
1341 extracting the name-to-number translation table from a compiled pattern. There
1342 is also a convenience function for extracting a captured substring by name.
1343 .P
1344 By default, a name must be unique within a pattern, but it is possible to relax
1345 this constraint by setting the PCRE_DUPNAMES option at compile time. (Duplicate
1346 names are also always permitted for subpatterns with the same number, set up as
1347 described in the previous section.) Duplicate names can be useful for patterns
1348 where only one instance of the named parentheses can match. Suppose you want to
1349 match the name of a weekday, either as a 3-letter abbreviation or as the full
1350 name, and in both cases you want to extract the abbreviation. This pattern
1351 (ignoring the line breaks) does the job:
1352 .sp
1353 (?<DN>Mon|Fri|Sun)(?:day)?|
1354 (?<DN>Tue)(?:sday)?|
1355 (?<DN>Wed)(?:nesday)?|
1356 (?<DN>Thu)(?:rsday)?|
1357 (?<DN>Sat)(?:urday)?
1358 .sp
1359 There are five capturing substrings, but only one is ever set after a match.
1360 (An alternative way of solving this problem is to use a "branch reset"
1361 subpattern, as described in the previous section.)
1362 .P
1363 The convenience function for extracting the data by name returns the substring
1364 for the first (and in this example, the only) subpattern of that name that
1365 matched. This saves searching to find which numbered subpattern it was.
1366 .P
1367 If you make a back reference to a non-unique named subpattern from elsewhere in
1368 the pattern, the one that corresponds to the first occurrence of the name is
1369 used. In the absence of duplicate numbers (see the previous section) this is
1370 the one with the lowest number. If you use a named reference in a condition
1371 test (see the
1372 .\"
1373 .\" HTML <a href="#conditions">
1374 .\" </a>
1375 section about conditions
1376 .\"
1377 below), either to check whether a subpattern has matched, or to check for
1378 recursion, all subpatterns with the same name are tested. If the condition is
1379 true for any one of them, the overall condition is true. This is the same
1380 behaviour as testing by number. For further details of the interfaces for
1381 handling named subpatterns, see the
1382 .\" HREF
1383 \fBpcreapi\fP
1384 .\"
1385 documentation.
1386 .P
1387 \fBWarning:\fP You cannot use different names to distinguish between two
1388 subpatterns with the same number because PCRE uses only the numbers when
1389 matching. For this reason, an error is given at compile time if different names
1390 are given to subpatterns with the same number. However, you can give the same
1391 name to subpatterns with the same number, even when PCRE_DUPNAMES is not set.
1392 .
1393 .
1395 .rs
1396 .sp
1397 Repetition is specified by quantifiers, which can follow any of the following
1398 items:
1399 .sp
1400 a literal data character
1401 the dot metacharacter
1402 the \eC escape sequence
1403 the \eX escape sequence (in UTF-8 mode with Unicode properties)
1404 the \eR escape sequence
1405 an escape such as \ed that matches a single character
1406 a character class
1407 a back reference (see next section)
1408 a parenthesized subpattern (unless it is an assertion)
1409 a recursive or "subroutine" call to a subpattern
1410 .sp
1411 The general repetition quantifier specifies a minimum and maximum number of
1412 permitted matches, by giving the two numbers in curly brackets (braces),
1413 separated by a comma. The numbers must be less than 65536, and the first must
1414 be less than or equal to the second. For example:
1415 .sp
1416 z{2,4}
1417 .sp
1418 matches "zz", "zzz", or "zzzz". A closing brace on its own is not a special
1419 character. If the second number is omitted, but the comma is present, there is
1420 no upper limit; if the second number and the comma are both omitted, the
1421 quantifier specifies an exact number of required matches. Thus
1422 .sp
1423 [aeiou]{3,}
1424 .sp
1425 matches at least 3 successive vowels, but may match many more, while
1426 .sp
1427 \ed{8}
1428 .sp
1429 matches exactly 8 digits. An opening curly bracket that appears in a position
1430 where a quantifier is not allowed, or one that does not match the syntax of a
1431 quantifier, is taken as a literal character. For example, {,6} is not a
1432 quantifier, but a literal string of four characters.
1433 .P
1434 In UTF-8 mode, quantifiers apply to UTF-8 characters rather than to individual
1435 bytes. Thus, for example, \ex{100}{2} matches two UTF-8 characters, each of
1436 which is represented by a two-byte sequence. Similarly, when Unicode property
1437 support is available, \eX{3} matches three Unicode extended sequences, each of
1438 which may be several bytes long (and they may be of different lengths).
1439 .P
1440 The quantifier {0} is permitted, causing the expression to behave as if the
1441 previous item and the quantifier were not present. This may be useful for
1442 subpatterns that are referenced as
1443 .\" HTML <a href="#subpatternsassubroutines">
1444 .\" </a>
1445 subroutines
1446 .\"
1447 from elsewhere in the pattern. Items other than subpatterns that have a {0}
1448 quantifier are omitted from the compiled pattern.
1449 .P
1450 For convenience, the three most common quantifiers have single-character
1451 abbreviations:
1452 .sp
1453 * is equivalent to {0,}
1454 + is equivalent to {1,}
1455 ? is equivalent to {0,1}
1456 .sp
1457 It is possible to construct infinite loops by following a subpattern that can
1458 match no characters with a quantifier that has no upper limit, for example:
1459 .sp
1460 (a?)*
1461 .sp
1462 Earlier versions of Perl and PCRE used to give an error at compile time for
1463 such patterns. However, because there are cases where this can be useful, such
1464 patterns are now accepted, but if any repetition of the subpattern does in fact
1465 match no characters, the loop is forcibly broken.
1466 .P
1467 By default, the quantifiers are "greedy", that is, they match as much as
1468 possible (up to the maximum number of permitted times), without causing the
1469 rest of the pattern to fail. The classic example of where this gives problems
1470 is in trying to match comments in C programs. These appear between /* and */
1471 and within the comment, individual * and / characters may appear. An attempt to
1472 match C comments by applying the pattern
1473 .sp
1474 /\e*.*\e*/
1475 .sp
1476 to the string
1477 .sp
1478 /* first comment */ not comment /* second comment */
1479 .sp
1480 fails, because it matches the entire string owing to the greediness of the .*
1481 item.
1482 .P
1483 However, if a quantifier is followed by a question mark, it ceases to be
1484 greedy, and instead matches the minimum number of times possible, so the
1485 pattern
1486 .sp
1487 /\e*.*?\e*/
1488 .sp
1489 does the right thing with the C comments. The meaning of the various
1490 quantifiers is not otherwise changed, just the preferred number of matches.
1491 Do not confuse this use of question mark with its use as a quantifier in its
1492 own right. Because it has two uses, it can sometimes appear doubled, as in
1493 .sp
1494 \ed??\ed
1495 .sp
1496 which matches one digit by preference, but can match two if that is the only
1497 way the rest of the pattern matches.
1498 .P
1499 If the PCRE_UNGREEDY option is set (an option that is not available in Perl),
1500 the quantifiers are not greedy by default, but individual ones can be made
1501 greedy by following them with a question mark. In other words, it inverts the
1502 default behaviour.
1503 .P
1504 When a parenthesized subpattern is quantified with a minimum repeat count that
1505 is greater than 1 or with a limited maximum, more memory is required for the
1506 compiled pattern, in proportion to the size of the minimum or maximum.
1507 .P
1508 If a pattern starts with .* or .{0,} and the PCRE_DOTALL option (equivalent
1509 to Perl's /s) is set, thus allowing the dot to match newlines, the pattern is
1510 implicitly anchored, because whatever follows will be tried against every
1511 character position in the subject string, so there is no point in retrying the
1512 overall match at any position after the first. PCRE normally treats such a
1513 pattern as though it were preceded by \eA.
1514 .P
1515 In cases where it is known that the subject string contains no newlines, it is
1516 worth setting PCRE_DOTALL in order to obtain this optimization, or
1517 alternatively using ^ to indicate anchoring explicitly.
1518 .P
1519 However, there is one situation where the optimization cannot be used. When .*
1520 is inside capturing parentheses that are the subject of a back reference
1521 elsewhere in the pattern, a match at the start may fail where a later one
1522 succeeds. Consider, for example:
1523 .sp
1524 (.*)abc\e1
1525 .sp
1526 If the subject is "xyz123abc123" the match point is the fourth character. For
1527 this reason, such a pattern is not implicitly anchored.
1528 .P
1529 When a capturing subpattern is repeated, the value captured is the substring
1530 that matched the final iteration. For example, after
1531 .sp
1532 (tweedle[dume]{3}\es*)+
1533 .sp
1534 has matched "tweedledum tweedledee" the value of the captured substring is
1535 "tweedledee". However, if there are nested capturing subpatterns, the
1536 corresponding captured values may have been set in previous iterations. For
1537 example, after
1538 .sp
1539 /(a|(b))+/
1540 .sp
1541 matches "aba" the value of the second captured substring is "b".
1542 .
1543 .
1544 .\" HTML <a name="atomicgroup"></a>
1546 .rs
1547 .sp
1548 With both maximizing ("greedy") and minimizing ("ungreedy" or "lazy")
1549 repetition, failure of what follows normally causes the repeated item to be
1550 re-evaluated to see if a different number of repeats allows the rest of the
1551 pattern to match. Sometimes it is useful to prevent this, either to change the
1552 nature of the match, or to cause it fail earlier than it otherwise might, when
1553 the author of the pattern knows there is no point in carrying on.
1554 .P
1555 Consider, for example, the pattern \ed+foo when applied to the subject line
1556 .sp
1557 123456bar
1558 .sp
1559 After matching all 6 digits and then failing to match "foo", the normal
1560 action of the matcher is to try again with only 5 digits matching the \ed+
1561 item, and then with 4, and so on, before ultimately failing. "Atomic grouping"
1562 (a term taken from Jeffrey Friedl's book) provides the means for specifying
1563 that once a subpattern has matched, it is not to be re-evaluated in this way.
1564 .P
1565 If we use atomic grouping for the previous example, the matcher gives up
1566 immediately on failing to match "foo" the first time. The notation is a kind of
1567 special parenthesis, starting with (?> as in this example:
1568 .sp
1569 (?>\ed+)foo
1570 .sp
1571 This kind of parenthesis "locks up" the part of the pattern it contains once
1572 it has matched, and a failure further into the pattern is prevented from
1573 backtracking into it. Backtracking past it to previous items, however, works as
1574 normal.
1575 .P
1576 An alternative description is that a subpattern of this type matches the string
1577 of characters that an identical standalone pattern would match, if anchored at
1578 the current point in the subject string.
1579 .P
1580 Atomic grouping subpatterns are not capturing subpatterns. Simple cases such as
1581 the above example can be thought of as a maximizing repeat that must swallow
1582 everything it can. So, while both \ed+ and \ed+? are prepared to adjust the
1583 number of digits they match in order to make the rest of the pattern match,
1584 (?>\ed+) can only match an entire sequence of digits.
1585 .P
1586 Atomic groups in general can of course contain arbitrarily complicated
1587 subpatterns, and can be nested. However, when the subpattern for an atomic
1588 group is just a single repeated item, as in the example above, a simpler
1589 notation, called a "possessive quantifier" can be used. This consists of an
1590 additional + character following a quantifier. Using this notation, the
1591 previous example can be rewritten as
1592 .sp
1593 \ed++foo
1594 .sp
1595 Note that a possessive quantifier can be used with an entire group, for
1596 example:
1597 .sp
1598 (abc|xyz){2,3}+
1599 .sp
1600 Possessive quantifiers are always greedy; the setting of the PCRE_UNGREEDY
1601 option is ignored. They are a convenient notation for the simpler forms of
1602 atomic group. However, there is no difference in the meaning of a possessive
1603 quantifier and the equivalent atomic group, though there may be a performance
1604 difference; possessive quantifiers should be slightly faster.
1605 .P
1606 The possessive quantifier syntax is an extension to the Perl 5.8 syntax.
1607 Jeffrey Friedl originated the idea (and the name) in the first edition of his
1608 book. Mike McCloskey liked it, so implemented it when he built Sun's Java
1609 package, and PCRE copied it from there. It ultimately found its way into Perl
1610 at release 5.10.
1611 .P
1612 PCRE has an optimization that automatically "possessifies" certain simple
1613 pattern constructs. For example, the sequence A+B is treated as A++B because
1614 there is no point in backtracking into a sequence of A's when B must follow.
1615 .P
1616 When a pattern contains an unlimited repeat inside a subpattern that can itself
1617 be repeated an unlimited number of times, the use of an atomic group is the
1618 only way to avoid some failing matches taking a very long time indeed. The
1619 pattern
1620 .sp
1621 (\eD+|<\ed+>)*[!?]
1622 .sp
1623 matches an unlimited number of substrings that either consist of non-digits, or
1624 digits enclosed in <>, followed by either ! or ?. When it matches, it runs
1625 quickly. However, if it is applied to
1626 .sp
1627 aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa
1628 .sp
1629 it takes a long time before reporting failure. This is because the string can
1630 be divided between the internal \eD+ repeat and the external * repeat in a
1631 large number of ways, and all have to be tried. (The example uses [!?] rather
1632 than a single character at the end, because both PCRE and Perl have an
1633 optimization that allows for fast failure when a single character is used. They
1634 remember the last single character that is required for a match, and fail early
1635 if it is not present in the string.) If the pattern is changed so that it uses
1636 an atomic group, like this:
1637 .sp
1638 ((?>\eD+)|<\ed+>)*[!?]
1639 .sp
1640 sequences of non-digits cannot be broken, and failure happens quickly.
1641 .
1642 .
1643 .\" HTML <a name="backreferences"></a>
1645 .rs
1646 .sp
1647 Outside a character class, a backslash followed by a digit greater than 0 (and
1648 possibly further digits) is a back reference to a capturing subpattern earlier
1649 (that is, to its left) in the pattern, provided there have been that many
1650 previous capturing left parentheses.
1651 .P
1652 However, if the decimal number following the backslash is less than 10, it is
1653 always taken as a back reference, and causes an error only if there are not
1654 that many capturing left parentheses in the entire pattern. In other words, the
1655 parentheses that are referenced need not be to the left of the reference for
1656 numbers less than 10. A "forward back reference" of this type can make sense
1657 when a repetition is involved and the subpattern to the right has participated
1658 in an earlier iteration.
1659 .P
1660 It is not possible to have a numerical "forward back reference" to a subpattern
1661 whose number is 10 or more using this syntax because a sequence such as \e50 is
1662 interpreted as a character defined in octal. See the subsection entitled
1663 "Non-printing characters"
1664 .\" HTML <a href="#digitsafterbackslash">
1665 .\" </a>
1666 above
1667 .\"
1668 for further details of the handling of digits following a backslash. There is
1669 no such problem when named parentheses are used. A back reference to any
1670 subpattern is possible using named parentheses (see below).
1671 .P
1672 Another way of avoiding the ambiguity inherent in the use of digits following a
1673 backslash is to use the \eg escape sequence, which is a feature introduced in
1674 Perl 5.10. This escape must be followed by an unsigned number or a negative
1675 number, optionally enclosed in braces. These examples are all identical:
1676 .sp
1677 (ring), \e1
1678 (ring), \eg1
1679 (ring), \eg{1}
1680 .sp
1681 An unsigned number specifies an absolute reference without the ambiguity that
1682 is present in the older syntax. It is also useful when literal digits follow
1683 the reference. A negative number is a relative reference. Consider this
1684 example:
1685 .sp
1686 (abc(def)ghi)\eg{-1}
1687 .sp
1688 The sequence \eg{-1} is a reference to the most recently started capturing
1689 subpattern before \eg, that is, is it equivalent to \e2. Similarly, \eg{-2}
1690 would be equivalent to \e1. The use of relative references can be helpful in
1691 long patterns, and also in patterns that are created by joining together
1692 fragments that contain references within themselves.
1693 .P
1694 A back reference matches whatever actually matched the capturing subpattern in
1695 the current subject string, rather than anything matching the subpattern
1696 itself (see
1697 .\" HTML <a href="#subpatternsassubroutines">
1698 .\" </a>
1699 "Subpatterns as subroutines"
1700 .\"
1701 below for a way of doing that). So the pattern
1702 .sp
1703 (sens|respons)e and \e1ibility
1704 .sp
1705 matches "sense and sensibility" and "response and responsibility", but not
1706 "sense and responsibility". If caseful matching is in force at the time of the
1707 back reference, the case of letters is relevant. For example,
1708 .sp
1709 ((?i)rah)\es+\e1
1710 .sp
1711 matches "rah rah" and "RAH RAH", but not "RAH rah", even though the original
1712 capturing subpattern is matched caselessly.
1713 .P
1714 There are several different ways of writing back references to named
1715 subpatterns. The .NET syntax \ek{name} and the Perl syntax \ek<name> or
1716 \ek'name' are supported, as is the Python syntax (?P=name). Perl 5.10's unified
1717 back reference syntax, in which \eg can be used for both numeric and named
1718 references, is also supported. We could rewrite the above example in any of
1719 the following ways:
1720 .sp
1721 (?<p1>(?i)rah)\es+\ek<p1>
1722 (?'p1'(?i)rah)\es+\ek{p1}
1723 (?P<p1>(?i)rah)\es+(?P=p1)
1724 (?<p1>(?i)rah)\es+\eg{p1}
1725 .sp
1726 A subpattern that is referenced by name may appear in the pattern before or
1727 after the reference.
1728 .P
1729 There may be more than one back reference to the same subpattern. If a
1730 subpattern has not actually been used in a particular match, any back
1731 references to it always fail by default. For example, the pattern
1732 .sp
1733 (a|(bc))\e2
1734 .sp
1735 always fails if it starts to match "a" rather than "bc". However, if the
1736 PCRE_JAVASCRIPT_COMPAT option is set at compile time, a back reference to an
1737 unset value matches an empty string.
1738 .P
1739 Because there may be many capturing parentheses in a pattern, all digits
1740 following a backslash are taken as part of a potential back reference number.
1741 If the pattern continues with a digit character, some delimiter must be used to
1742 terminate the back reference. If the PCRE_EXTENDED option is set, this can be
1743 whitespace. Otherwise, the \eg{ syntax or an empty comment (see
1744 .\" HTML <a href="#comments">
1745 .\" </a>
1746 "Comments"
1747 .\"
1748 below) can be used.
1749 .
1750 .SS "Recursive back references"
1751 .rs
1752 .sp
1753 A back reference that occurs inside the parentheses to which it refers fails
1754 when the subpattern is first used, so, for example, (a\e1) never matches.
1755 However, such references can be useful inside repeated subpatterns. For
1756 example, the pattern
1757 .sp
1758 (a|b\e1)+
1759 .sp
1760 matches any number of "a"s and also "aba", "ababbaa" etc. At each iteration of
1761 the subpattern, the back reference matches the character string corresponding
1762 to the previous iteration. In order for this to work, the pattern must be such
1763 that the first iteration does not need to match the back reference. This can be
1764 done using alternation, as in the example above, or by a quantifier with a
1765 minimum of zero.
1766 .P
1767 Back references of this type cause the group that they reference to be treated
1768 as an
1769 .\" HTML <a href="#atomicgroup">
1770 .\" </a>
1771 atomic group.
1772 .\"
1773 Once the whole group has been matched, a subsequent matching failure cannot
1774 cause backtracking into the middle of the group.
1775 .
1776 .
1777 .\" HTML <a name="bigassertions"></a>
1779 .rs
1780 .sp
1781 An assertion is a test on the characters following or preceding the current
1782 matching point that does not actually consume any characters. The simple
1783 assertions coded as \eb, \eB, \eA, \eG, \eZ, \ez, ^ and $ are described
1784 .\" HTML <a href="#smallassertions">
1785 .\" </a>
1786 above.
1787 .\"
1788 .P
1789 More complicated assertions are coded as subpatterns. There are two kinds:
1790 those that look ahead of the current position in the subject string, and those
1791 that look behind it. An assertion subpattern is matched in the normal way,
1792 except that it does not cause the current matching position to be changed.
1793 .P
1794 Assertion subpatterns are not capturing subpatterns, and may not be repeated,
1795 because it makes no sense to assert the same thing several times. If any kind
1796 of assertion contains capturing subpatterns within it, these are counted for
1797 the purposes of numbering the capturing subpatterns in the whole pattern.
1798 However, substring capturing is carried out only for positive assertions,
1799 because it does not make sense for negative assertions.
1800 .
1801 .
1802 .SS "Lookahead assertions"
1803 .rs
1804 .sp
1805 Lookahead assertions start with (?= for positive assertions and (?! for
1806 negative assertions. For example,
1807 .sp
1808 \ew+(?=;)
1809 .sp
1810 matches a word followed by a semicolon, but does not include the semicolon in
1811 the match, and
1812 .sp
1813 foo(?!bar)
1814 .sp
1815 matches any occurrence of "foo" that is not followed by "bar". Note that the
1816 apparently similar pattern
1817 .sp
1818 (?!foo)bar
1819 .sp
1820 does not find an occurrence of "bar" that is preceded by something other than
1821 "foo"; it finds any occurrence of "bar" whatsoever, because the assertion
1822 (?!foo) is always true when the next three characters are "bar". A
1823 lookbehind assertion is needed to achieve the other effect.
1824 .P
1825 If you want to force a matching failure at some point in a pattern, the most
1826 convenient way to do it is with (?!) because an empty string always matches, so
1827 an assertion that requires there not to be an empty string must always fail.
1828 The Perl 5.10 backtracking control verb (*FAIL) or (*F) is essentially a
1829 synonym for (?!).
1830 .
1831 .
1832 .\" HTML <a name="lookbehind"></a>
1833 .SS "Lookbehind assertions"
1834 .rs
1835 .sp
1836 Lookbehind assertions start with (?<= for positive assertions and (?<! for
1837 negative assertions. For example,
1838 .sp
1839 (?<!foo)bar
1840 .sp
1841 does find an occurrence of "bar" that is not preceded by "foo". The contents of
1842 a lookbehind assertion are restricted such that all the strings it matches must
1843 have a fixed length. However, if there are several top-level alternatives, they
1844 do not all have to have the same fixed length. Thus
1845 .sp
1846 (?<=bullock|donkey)
1847 .sp
1848 is permitted, but
1849 .sp
1850 (?<!dogs?|cats?)
1851 .sp
1852 causes an error at compile time. Branches that match different length strings
1853 are permitted only at the top level of a lookbehind assertion. This is an
1854 extension compared with Perl (5.8 and 5.10), which requires all branches to
1855 match the same length of string. An assertion such as
1856 .sp
1857 (?<=ab(c|de))
1858 .sp
1859 is not permitted, because its single top-level branch can match two different
1860 lengths, but it is acceptable to PCRE if rewritten to use two top-level
1861 branches:
1862 .sp
1863 (?<=abc|abde)
1864 .sp
1865 In some cases, the Perl 5.10 escape sequence \eK
1866 .\" HTML <a href="#resetmatchstart">
1867 .\" </a>
1868 (see above)
1869 .\"
1870 can be used instead of a lookbehind assertion to get round the fixed-length
1871 restriction.
1872 .P
1873 The implementation of lookbehind assertions is, for each alternative, to
1874 temporarily move the current position back by the fixed length and then try to
1875 match. If there are insufficient characters before the current position, the
1876 assertion fails.
1877 .P
1878 PCRE does not allow the \eC escape (which matches a single byte in UTF-8 mode)
1879 to appear in lookbehind assertions, because it makes it impossible to calculate
1880 the length of the lookbehind. The \eX and \eR escapes, which can match
1881 different numbers of bytes, are also not permitted.
1882 .P
1883 .\" HTML <a href="#subpatternsassubroutines">
1884 .\" </a>
1885 "Subroutine"
1886 .\"
1887 calls (see below) such as (?2) or (?&X) are permitted in lookbehinds, as long
1888 as the subpattern matches a fixed-length string.
1889 .\" HTML <a href="#recursion">
1890 .\" </a>
1891 Recursion,
1892 .\"
1893 however, is not supported.
1894 .P
1895 Possessive quantifiers can be used in conjunction with lookbehind assertions to
1896 specify efficient matching of fixed-length strings at the end of subject
1897 strings. Consider a simple pattern such as
1898 .sp
1899 abcd$
1900 .sp
1901 when applied to a long string that does not match. Because matching proceeds
1902 from left to right, PCRE will look for each "a" in the subject and then see if
1903 what follows matches the rest of the pattern. If the pattern is specified as
1904 .sp
1905 ^.*abcd$
1906 .sp
1907 the initial .* matches the entire string at first, but when this fails (because
1908 there is no following "a"), it backtracks to match all but the last character,
1909 then all but the last two characters, and so on. Once again the search for "a"
1910 covers the entire string, from right to left, so we are no better off. However,
1911 if the pattern is written as
1912 .sp
1913 ^.*+(?<=abcd)
1914 .sp
1915 there can be no backtracking for the .*+ item; it can match only the entire
1916 string. The subsequent lookbehind assertion does a single test on the last four
1917 characters. If it fails, the match fails immediately. For long strings, this
1918 approach makes a significant difference to the processing time.
1919 .
1920 .
1921 .SS "Using multiple assertions"
1922 .rs
1923 .sp
1924 Several assertions (of any sort) may occur in succession. For example,
1925 .sp
1926 (?<=\ed{3})(?<!999)foo
1927 .sp
1928 matches "foo" preceded by three digits that are not "999". Notice that each of
1929 the assertions is applied independently at the same point in the subject
1930 string. First there is a check that the previous three characters are all
1931 digits, and then there is a check that the same three characters are not "999".
1932 This pattern does \fInot\fP match "foo" preceded by six characters, the first
1933 of which are digits and the last three of which are not "999". For example, it
1934 doesn't match "123abcfoo". A pattern to do that is
1935 .sp
1936 (?<=\ed{3}...)(?<!999)foo
1937 .sp
1938 This time the first assertion looks at the preceding six characters, checking
1939 that the first three are digits, and then the second assertion checks that the
1940 preceding three characters are not "999".
1941 .P
1942 Assertions can be nested in any combination. For example,
1943 .sp
1944 (?<=(?<!foo)bar)baz
1945 .sp
1946 matches an occurrence of "baz" that is preceded by "bar" which in turn is not
1947 preceded by "foo", while
1948 .sp
1949 (?<=\ed{3}(?!999)...)foo
1950 .sp
1951 is another pattern that matches "foo" preceded by three digits and any three
1952 characters that are not "999".
1953 .
1954 .
1955 .\" HTML <a name="conditions"></a>
1957 .rs
1958 .sp
1959 It is possible to cause the matching process to obey a subpattern
1960 conditionally or to choose between two alternative subpatterns, depending on
1961 the result of an assertion, or whether a specific capturing subpattern has
1962 already been matched. The two possible forms of conditional subpattern are:
1963 .sp
1964 (?(condition)yes-pattern)
1965 (?(condition)yes-pattern|no-pattern)
1966 .sp
1967 If the condition is satisfied, the yes-pattern is used; otherwise the
1968 no-pattern (if present) is used. If there are more than two alternatives in the
1969 subpattern, a compile-time error occurs. Each of the two alternatives may
1970 itself contain nested subpatterns of any form, including conditional
1971 subpatterns; the restriction to two alternatives applies only at the level of
1972 the condition. This pattern fragment is an example where the alternatives are
1973 complex:
1974 .sp
1975 (?(1) (A|B|C) | (D | (?(2)E|F) | E) )
1976 .sp
1977 .P
1978 There are four kinds of condition: references to subpatterns, references to
1979 recursion, a pseudo-condition called DEFINE, and assertions.
1980 .
1981 .SS "Checking for a used subpattern by number"
1982 .rs
1983 .sp
1984 If the text between the parentheses consists of a sequence of digits, the
1985 condition is true if a capturing subpattern of that number has previously
1986 matched. If there is more than one capturing subpattern with the same number
1987 (see the earlier
1988 .\"
1989 .\" HTML <a href="#recursion">
1990 .\" </a>
1991 section about duplicate subpattern numbers),
1992 .\"
1993 the condition is true if any of them have been set. An alternative notation is
1994 to precede the digits with a plus or minus sign. In this case, the subpattern
1995 number is relative rather than absolute. The most recently opened parentheses
1996 can be referenced by (?(-1), the next most recent by (?(-2), and so on. In
1997 looping constructs it can also make sense to refer to subsequent groups with
1998 constructs such as (?(+2).
1999 .P
2000 Consider the following pattern, which contains non-significant white space to
2001 make it more readable (assume the PCRE_EXTENDED option) and to divide it into
2002 three parts for ease of discussion:
2003 .sp
2004 ( \e( )? [^()]+ (?(1) \e) )
2005 .sp
2006 The first part matches an optional opening parenthesis, and if that
2007 character is present, sets it as the first captured substring. The second part
2008 matches one or more characters that are not parentheses. The third part is a
2009 conditional subpattern that tests whether the first set of parentheses matched
2010 or not. If they did, that is, if subject started with an opening parenthesis,
2011 the condition is true, and so the yes-pattern is executed and a closing
2012 parenthesis is required. Otherwise, since no-pattern is not present, the
2013 subpattern matches nothing. In other words, this pattern matches a sequence of
2014 non-parentheses, optionally enclosed in parentheses.
2015 .P
2016 If you were embedding this pattern in a larger one, you could use a relative
2017 reference:
2018 .sp
2019 ...other stuff... ( \e( )? [^()]+ (?(-1) \e) ) ...
2020 .sp
2021 This makes the fragment independent of the parentheses in the larger pattern.
2022 .
2023 .SS "Checking for a used subpattern by name"
2024 .rs
2025 .sp
2026 Perl uses the syntax (?(<name>)...) or (?('name')...) to test for a used
2027 subpattern by name. For compatibility with earlier versions of PCRE, which had
2028 this facility before Perl, the syntax (?(name)...) is also recognized. However,
2029 there is a possible ambiguity with this syntax, because subpattern names may
2030 consist entirely of digits. PCRE looks first for a named subpattern; if it
2031 cannot find one and the name consists entirely of digits, PCRE looks for a
2032 subpattern of that number, which must be greater than zero. Using subpattern
2033 names that consist entirely of digits is not recommended.
2034 .P
2035 Rewriting the above example to use a named subpattern gives this:
2036 .sp
2037 (?<OPEN> \e( )? [^()]+ (?(<OPEN>) \e) )
2038 .sp
2039 If the name used in a condition of this kind is a duplicate, the test is
2040 applied to all subpatterns of the same name, and is true if any one of them has
2041 matched.
2042 .
2043 .SS "Checking for pattern recursion"
2044 .rs
2045 .sp
2046 If the condition is the string (R), and there is no subpattern with the name R,
2047 the condition is true if a recursive call to the whole pattern or any
2048 subpattern has been made. If digits or a name preceded by ampersand follow the
2049 letter R, for example:
2050 .sp
2051 (?(R3)...) or (?(R&name)...)
2052 .sp
2053 the condition is true if the most recent recursion is into a subpattern whose
2054 number or name is given. This condition does not check the entire recursion
2055 stack. If the name used in a condition of this kind is a duplicate, the test is
2056 applied to all subpatterns of the same name, and is true if any one of them is
2057 the most recent recursion.
2058 .P
2059 At "top level", all these recursion test conditions are false.
2060 .\" HTML <a href="#recursion">
2061 .\" </a>
2062 The syntax for recursive patterns
2063 .\"
2064 is described below.
2065 .
2066 .SS "Defining subpatterns for use by reference only"
2067 .rs
2068 .sp
2069 If the condition is the string (DEFINE), and there is no subpattern with the
2070 name DEFINE, the condition is always false. In this case, there may be only one
2071 alternative in the subpattern. It is always skipped if control reaches this
2072 point in the pattern; the idea of DEFINE is that it can be used to define
2073 "subroutines" that can be referenced from elsewhere. (The use of
2074 .\" HTML <a href="#subpatternsassubroutines">
2075 .\" </a>
2076 "subroutines"
2077 .\"
2078 is described below.) For example, a pattern to match an IPv4 address could be
2079 written like this (ignore whitespace and line breaks):
2080 .sp
2081 (?(DEFINE) (?<byte> 2[0-4]\ed | 25[0-5] | 1\ed\ed | [1-9]?\ed) )
2082 \eb (?&byte) (\e.(?&byte)){3} \eb
2083 .sp
2084 The first part of the pattern is a DEFINE group inside which a another group
2085 named "byte" is defined. This matches an individual component of an IPv4
2086 address (a number less than 256). When matching takes place, this part of the
2087 pattern is skipped because DEFINE acts like a false condition. The rest of the
2088 pattern uses references to the named group to match the four dot-separated
2089 components of an IPv4 address, insisting on a word boundary at each end.
2090 .
2091 .SS "Assertion conditions"
2092 .rs
2093 .sp
2094 If the condition is not in any of the above formats, it must be an assertion.
2095 This may be a positive or negative lookahead or lookbehind assertion. Consider
2096 this pattern, again containing non-significant white space, and with the two
2097 alternatives on the second line:
2098 .sp
2099 (?(?=[^a-z]*[a-z])
2100 \ed{2}-[a-z]{3}-\ed{2} | \ed{2}-\ed{2}-\ed{2} )
2101 .sp
2102 The condition is a positive lookahead assertion that matches an optional
2103 sequence of non-letters followed by a letter. In other words, it tests for the
2104 presence of at least one letter in the subject. If a letter is found, the
2105 subject is matched against the first alternative; otherwise it is matched
2106 against the second. This pattern matches strings in one of the two forms
2107 dd-aaa-dd or dd-dd-dd, where aaa are letters and dd are digits.
2108 .
2109 .
2110 .\" HTML <a name="comments"></a>
2112 .rs
2113 .sp
2114 There are two ways of including comments in patterns that are processed by
2115 PCRE. In both cases, the start of the comment must not be in a character class,
2116 nor in the middle of any other sequence of related characters such as (?: or a
2117 subpattern name or number. The characters that make up a comment play no part
2118 in the pattern matching.
2119 .P
2120 The sequence (?# marks the start of a comment that continues up to the next
2121 closing parenthesis. Nested parentheses are not permitted. If the PCRE_EXTENDED
2122 option is set, an unescaped # character also introduces a comment, which in
2123 this case continues to immediately after the next newline character or
2124 character sequence in the pattern. Which characters are interpreted as newlines
2125 is controlled by the options passed to \fBpcre_compile()\fP or by a special
2126 sequence at the start of the pattern, as described in the section entitled
2127 .\" HTML <a href="#recursion">
2128 .\" </a>
2129 "Newline conventions"
2130 .\"
2131 above. Note that end of this type of comment is a literal newline sequence in
2132 the pattern; escape sequences that happen to represent a newline do not count.
2133 For example, consider this pattern when PCRE_EXTENDED is set, and the default
2134 newline convention is in force:
2135 .sp
2136 abc #comment \en still comment
2137 .sp
2138 On encountering the # character, \fBpcre_compile()\fP skips along, looking for
2139 a newline in the pattern. The sequence \en is still literal at this stage, so
2140 it does not terminate the comment. Only an actual character with the code value
2141 0x0a (the default newline) does so.
2142 .
2143 .
2144 .\" HTML <a name="recursion"></a>
2146 .rs
2147 .sp
2148 Consider the problem of matching a string in parentheses, allowing for
2149 unlimited nested parentheses. Without the use of recursion, the best that can
2150 be done is to use a pattern that matches up to some fixed depth of nesting. It
2151 is not possible to handle an arbitrary nesting depth.
2152 .P
2153 For some time, Perl has provided a facility that allows regular expressions to
2154 recurse (amongst other things). It does this by interpolating Perl code in the
2155 expression at run time, and the code can refer to the expression itself. A Perl
2156 pattern using code interpolation to solve the parentheses problem can be
2157 created like this:
2158 .sp
2159 $re = qr{\e( (?: (?>[^()]+) | (?p{$re}) )* \e)}x;
2160 .sp
2161 The (?p{...}) item interpolates Perl code at run time, and in this case refers
2162 recursively to the pattern in which it appears.
2163 .P
2164 Obviously, PCRE cannot support the interpolation of Perl code. Instead, it
2165 supports special syntax for recursion of the entire pattern, and also for
2166 individual subpattern recursion. After its introduction in PCRE and Python,
2167 this kind of recursion was subsequently introduced into Perl at release 5.10.
2168 .P
2169 A special item that consists of (? followed by a number greater than zero and a
2170 closing parenthesis is a recursive call of the subpattern of the given number,
2171 provided that it occurs inside that subpattern. (If not, it is a
2172 .\" HTML <a href="#subpatternsassubroutines">
2173 .\" </a>
2174 "subroutine"
2175 .\"
2176 call, which is described in the next section.) The special item (?R) or (?0) is
2177 a recursive call of the entire regular expression.
2178 .P
2179 This PCRE pattern solves the nested parentheses problem (assume the
2180 PCRE_EXTENDED option is set so that white space is ignored):
2181 .sp
2182 \e( ( [^()]++ | (?R) )* \e)
2183 .sp
2184 First it matches an opening parenthesis. Then it matches any number of
2185 substrings which can either be a sequence of non-parentheses, or a recursive
2186 match of the pattern itself (that is, a correctly parenthesized substring).
2187 Finally there is a closing parenthesis. Note the use of a possessive quantifier
2188 to avoid backtracking into sequences of non-parentheses.
2189 .P
2190 If this were part of a larger pattern, you would not want to recurse the entire
2191 pattern, so instead you could use this:
2192 .sp
2193 ( \e( ( [^()]++ | (?1) )* \e) )
2194 .sp
2195 We have put the pattern into parentheses, and caused the recursion to refer to
2196 them instead of the whole pattern.
2197 .P
2198 In a larger pattern, keeping track of parenthesis numbers can be tricky. This
2199 is made easier by the use of relative references (a Perl 5.10 feature).
2200 Instead of (?1) in the pattern above you can write (?-2) to refer to the second
2201 most recently opened parentheses preceding the recursion. In other words, a
2202 negative number counts capturing parentheses leftwards from the point at which
2203 it is encountered.
2204 .P
2205 It is also possible to refer to subsequently opened parentheses, by writing
2206 references such as (?+2). However, these cannot be recursive because the
2207 reference is not inside the parentheses that are referenced. They are always
2208 .\" HTML <a href="#subpatternsassubroutines">
2209 .\" </a>
2210 "subroutine"
2211 .\"
2212 calls, as described in the next section.
2213 .P
2214 An alternative approach is to use named parentheses instead. The Perl syntax
2215 for this is (?&name); PCRE's earlier syntax (?P>name) is also supported. We
2216 could rewrite the above example as follows:
2217 .sp
2218 (?<pn> \e( ( [^()]++ | (?&pn) )* \e) )
2219 .sp
2220 If there is more than one subpattern with the same name, the earliest one is
2221 used.
2222 .P
2223 This particular example pattern that we have been looking at contains nested
2224 unlimited repeats, and so the use of a possessive quantifier for matching
2225 strings of non-parentheses is important when applying the pattern to strings
2226 that do not match. For example, when this pattern is applied to
2227 .sp
2228 (aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa()
2229 .sp
2230 it yields "no match" quickly. However, if a possessive quantifier is not used,
2231 the match runs for a very long time indeed because there are so many different
2232 ways the + and * repeats can carve up the subject, and all have to be tested
2233 before failure can be reported.
2234 .P
2235 At the end of a match, the values of capturing parentheses are those from
2236 the outermost level. If you want to obtain intermediate values, a callout
2237 function can be used (see below and the
2238 .\" HREF
2239 \fBpcrecallout\fP
2240 .\"
2241 documentation). If the pattern above is matched against
2242 .sp
2243 (ab(cd)ef)
2244 .sp
2245 the value for the inner capturing parentheses (numbered 2) is "ef", which is
2246 the last value taken on at the top level. If a capturing subpattern is not
2247 matched at the top level, its final value is unset, even if it is (temporarily)
2248 set at a deeper level.
2249 .P
2250 If there are more than 15 capturing parentheses in a pattern, PCRE has to
2251 obtain extra memory to store data during a recursion, which it does by using
2252 \fBpcre_malloc\fP, freeing it via \fBpcre_free\fP afterwards. If no memory can
2253 be obtained, the match fails with the PCRE_ERROR_NOMEMORY error.
2254 .P
2255 Do not confuse the (?R) item with the condition (R), which tests for recursion.
2256 Consider this pattern, which matches text in angle brackets, allowing for
2257 arbitrary nesting. Only digits are allowed in nested brackets (that is, when
2258 recursing), whereas any characters are permitted at the outer level.
2259 .sp
2260 < (?: (?(R) \ed++ | [^<>]*+) | (?R)) * >
2261 .sp
2262 In this pattern, (?(R) is the start of a conditional subpattern, with two
2263 different alternatives for the recursive and non-recursive cases. The (?R) item
2264 is the actual recursive call.
2265 .
2266 .
2267 .\" HTML <a name="recursiondifference"></a>
2268 .SS "Recursion difference from Perl"
2269 .rs
2270 .sp
2271 In PCRE (like Python, but unlike Perl), a recursive subpattern call is always
2272 treated as an atomic group. That is, once it has matched some of the subject
2273 string, it is never re-entered, even if it contains untried alternatives and
2274 there is a subsequent matching failure. This can be illustrated by the
2275 following pattern, which purports to match a palindromic string that contains
2276 an odd number of characters (for example, "a", "aba", "abcba", "abcdcba"):
2277 .sp
2278 ^(.|(.)(?1)\e2)$
2279 .sp
2280 The idea is that it either matches a single character, or two identical
2281 characters surrounding a sub-palindrome. In Perl, this pattern works; in PCRE
2282 it does not if the pattern is longer than three characters. Consider the
2283 subject string "abcba":
2284 .P
2285 At the top level, the first character is matched, but as it is not at the end
2286 of the string, the first alternative fails; the second alternative is taken
2287 and the recursion kicks in. The recursive call to subpattern 1 successfully
2288 matches the next character ("b"). (Note that the beginning and end of line
2289 tests are not part of the recursion).
2290 .P
2291 Back at the top level, the next character ("c") is compared with what
2292 subpattern 2 matched, which was "a". This fails. Because the recursion is
2293 treated as an atomic group, there are now no backtracking points, and so the
2294 entire match fails. (Perl is able, at this point, to re-enter the recursion and
2295 try the second alternative.) However, if the pattern is written with the
2296 alternatives in the other order, things are different:
2297 .sp
2298 ^((.)(?1)\e2|.)$
2299 .sp
2300 This time, the recursing alternative is tried first, and continues to recurse
2301 until it runs out of characters, at which point the recursion fails. But this
2302 time we do have another alternative to try at the higher level. That is the big
2303 difference: in the previous case the remaining alternative is at a deeper
2304 recursion level, which PCRE cannot use.
2305 .P
2306 To change the pattern so that matches all palindromic strings, not just those
2307 with an odd number of characters, it is tempting to change the pattern to this:
2308 .sp
2309 ^((.)(?1)\e2|.?)$
2310 .sp
2311 Again, this works in Perl, but not in PCRE, and for the same reason. When a
2312 deeper recursion has matched a single character, it cannot be entered again in
2313 order to match an empty string. The solution is to separate the two cases, and
2314 write out the odd and even cases as alternatives at the higher level:
2315 .sp
2316 ^(?:((.)(?1)\e2|)|((.)(?3)\e4|.))
2317 .sp
2318 If you want to match typical palindromic phrases, the pattern has to ignore all
2319 non-word characters, which can be done like this:
2320 .sp
2321 ^\eW*+(?:((.)\eW*+(?1)\eW*+\e2|)|((.)\eW*+(?3)\eW*+\e4|\eW*+.\eW*+))\eW*+$
2322 .sp
2323 If run with the PCRE_CASELESS option, this pattern matches phrases such as "A
2324 man, a plan, a canal: Panama!" and it works well in both PCRE and Perl. Note
2325 the use of the possessive quantifier *+ to avoid backtracking into sequences of
2326 non-word characters. Without this, PCRE takes a great deal longer (ten times or
2327 more) to match typical phrases, and Perl takes so long that you think it has
2328 gone into a loop.
2329 .P
2330 \fBWARNING\fP: The palindrome-matching patterns above work only if the subject
2331 string does not start with a palindrome that is shorter than the entire string.
2332 For example, although "abcba" is correctly matched, if the subject is "ababa",
2333 PCRE finds the palindrome "aba" at the start, then fails at top level because
2334 the end of the string does not follow. Once again, it cannot jump back into the
2335 recursion to try other alternatives, so the entire match fails.
2336 .
2337 .
2338 .\" HTML <a name="subpatternsassubroutines"></a>
2340 .rs
2341 .sp
2342 If the syntax for a recursive subpattern reference (either by number or by
2343 name) is used outside the parentheses to which it refers, it operates like a
2344 subroutine in a programming language. The "called" subpattern may be defined
2345 before or after the reference. A numbered reference can be absolute or
2346 relative, as in these examples:
2347 .sp
2348 (...(absolute)...)...(?2)...
2349 (...(relative)...)...(?-1)...
2350 (...(?+1)...(relative)...
2351 .sp
2352 An earlier example pointed out that the pattern
2353 .sp
2354 (sens|respons)e and \e1ibility
2355 .sp
2356 matches "sense and sensibility" and "response and responsibility", but not
2357 "sense and responsibility". If instead the pattern
2358 .sp
2359 (sens|respons)e and (?1)ibility
2360 .sp
2361 is used, it does match "sense and responsibility" as well as the other two
2362 strings. Another example is given in the discussion of DEFINE above.
2363 .P
2364 Like recursive subpatterns, a subroutine call is always treated as an atomic
2365 group. That is, once it has matched some of the subject string, it is never
2366 re-entered, even if it contains untried alternatives and there is a subsequent
2367 matching failure. Any capturing parentheses that are set during the subroutine
2368 call revert to their previous values afterwards.
2369 .P
2370 When a subpattern is used as a subroutine, processing options such as
2371 case-independence are fixed when the subpattern is defined. They cannot be
2372 changed for different calls. For example, consider this pattern:
2373 .sp
2374 (abc)(?i:(?-1))
2375 .sp
2376 It matches "abcabc". It does not match "abcABC" because the change of
2377 processing option does not affect the called subpattern.
2378 .
2379 .
2380 .\" HTML <a name="onigurumasubroutines"></a>
2382 .rs
2383 .sp
2384 For compatibility with Oniguruma, the non-Perl syntax \eg followed by a name or
2385 a number enclosed either in angle brackets or single quotes, is an alternative
2386 syntax for referencing a subpattern as a subroutine, possibly recursively. Here
2387 are two of the examples used above, rewritten using this syntax:
2388 .sp
2389 (?<pn> \e( ( (?>[^()]+) | \eg<pn> )* \e) )
2390 (sens|respons)e and \eg'1'ibility
2391 .sp
2392 PCRE supports an extension to Oniguruma: if a number is preceded by a
2393 plus or a minus sign it is taken as a relative reference. For example:
2394 .sp
2395 (abc)(?i:\eg<-1>)
2396 .sp
2397 Note that \eg{...} (Perl syntax) and \eg<...> (Oniguruma syntax) are \fInot\fP
2398 synonymous. The former is a back reference; the latter is a subroutine call.
2399 .
2400 .
2402 .rs
2403 .sp
2404 Perl has a feature whereby using the sequence (?{...}) causes arbitrary Perl
2405 code to be obeyed in the middle of matching a regular expression. This makes it
2406 possible, amongst other things, to extract different substrings that match the
2407 same pair of parentheses when there is a repetition.
2408 .P
2409 PCRE provides a similar feature, but of course it cannot obey arbitrary Perl
2410 code. The feature is called "callout". The caller of PCRE provides an external
2411 function by putting its entry point in the global variable \fIpcre_callout\fP.
2412 By default, this variable contains NULL, which disables all calling out.
2413 .P
2414 Within a regular expression, (?C) indicates the points at which the external
2415 function is to be called. If you want to identify different callout points, you
2416 can put a number less than 256 after the letter C. The default value is zero.
2417 For example, this pattern has two callout points:
2418 .sp
2419 (?C1)abc(?C2)def
2420 .sp
2421 If the PCRE_AUTO_CALLOUT flag is passed to \fBpcre_compile()\fP, callouts are
2422 automatically installed before each item in the pattern. They are all numbered
2423 255.
2424 .P
2425 During matching, when PCRE reaches a callout point (and \fIpcre_callout\fP is
2426 set), the external function is called. It is provided with the number of the
2427 callout, the position in the pattern, and, optionally, one item of data
2428 originally supplied by the caller of \fBpcre_exec()\fP. The callout function
2429 may cause matching to proceed, to backtrack, or to fail altogether. A complete
2430 description of the interface to the callout function is given in the
2431 .\" HREF
2432 \fBpcrecallout\fP
2433 .\"
2434 documentation.
2435 .
2436 .
2437 .\" HTML <a name="backtrackcontrol"></a>
2439 .rs
2440 .sp
2441 Perl 5.10 introduced a number of "Special Backtracking Control Verbs", which
2442 are described in the Perl documentation as "experimental and subject to change
2443 or removal in a future version of Perl". It goes on to say: "Their usage in
2444 production code should be noted to avoid problems during upgrades." The same
2445 remarks apply to the PCRE features described in this section.
2446 .P
2447 Since these verbs are specifically related to backtracking, most of them can be
2448 used only when the pattern is to be matched using \fBpcre_exec()\fP, which uses
2449 a backtracking algorithm. With the exception of (*FAIL), which behaves like a
2450 failing negative assertion, they cause an error if encountered by
2451 \fBpcre_dfa_exec()\fP.
2452 .P
2453 If any of these verbs are used in an assertion or subroutine subpattern
2454 (including recursive subpatterns), their effect is confined to that subpattern;
2455 it does not extend to the surrounding pattern. Note that such subpatterns are
2456 processed as anchored at the point where they are tested.
2457 .P
2458 The new verbs make use of what was previously invalid syntax: an opening
2459 parenthesis followed by an asterisk. They are generally of the form
2460 (*VERB) or (*VERB:NAME). Some may take either form, with differing behaviour,
2461 depending on whether or not an argument is present. An name is a sequence of
2462 letters, digits, and underscores. If the name is empty, that is, if the closing
2463 parenthesis immediately follows the colon, the effect is as if the colon were
2464 not there. Any number of these verbs may occur in a pattern.
2465 .P
2466 PCRE contains some optimizations that are used to speed up matching by running
2467 some checks at the start of each match attempt. For example, it may know the
2468 minimum length of matching subject, or that a particular character must be
2469 present. When one of these optimizations suppresses the running of a match, any
2470 included backtracking verbs will not, of course, be processed. You can suppress
2471 the start-of-match optimizations by setting the PCRE_NO_START_OPTIMIZE option
2472 when calling \fBpcre_exec()\fP.
2473 .
2474 .
2475 .SS "Verbs that act immediately"
2476 .rs
2477 .sp
2478 The following verbs act as soon as they are encountered. They may not be
2479 followed by a name.
2480 .sp
2481 (*ACCEPT)
2482 .sp
2483 This verb causes the match to end successfully, skipping the remainder of the
2484 pattern. When inside a recursion, only the innermost pattern is ended
2485 immediately. If (*ACCEPT) is inside capturing parentheses, the data so far is
2486 captured. (This feature was added to PCRE at release 8.00.) For example:
2487 .sp
2488 A((?:A|B(*ACCEPT)|C)D)
2489 .sp
2490 This matches "AB", "AAD", or "ACD"; when it matches "AB", "B" is captured by
2491 the outer parentheses.
2492 .sp
2493 (*FAIL) or (*F)
2494 .sp
2495 This verb causes the match to fail, forcing backtracking to occur. It is
2496 equivalent to (?!) but easier to read. The Perl documentation notes that it is
2497 probably useful only when combined with (?{}) or (??{}). Those are, of course,
2498 Perl features that are not present in PCRE. The nearest equivalent is the
2499 callout feature, as for example in this pattern:
2500 .sp
2501 a+(?C)(*FAIL)
2502 .sp
2503 A match with the string "aaaa" always fails, but the callout is taken before
2504 each backtrack happens (in this example, 10 times).
2505 .
2506 .
2507 .SS "Recording which path was taken"
2508 .rs
2509 .sp
2510 There is one verb whose main purpose is to track how a match was arrived at,
2511 though it also has a secondary use in conjunction with advancing the match
2512 starting point (see (*SKIP) below).
2513 .sp
2514 (*MARK:NAME) or (*:NAME)
2515 .sp
2516 A name is always required with this verb. There may be as many instances of
2517 (*MARK) as you like in a pattern, and their names do not have to be unique.
2518 .P
2519 When a match succeeds, the name of the last-encountered (*MARK) is passed back
2520 to the caller via the \fIpcre_extra\fP data structure, as described in the
2521 .\" HTML <a href="pcreapi.html#extradata">
2522 .\" </a>
2523 section on \fIpcre_extra\fP
2524 .\"
2525 in the
2526 .\" HREF
2527 \fBpcreapi\fP
2528 .\"
2529 documentation. No data is returned for a partial match. Here is an example of
2530 \fBpcretest\fP output, where the /K modifier requests the retrieval and
2531 outputting of (*MARK) data:
2532 .sp
2533 /X(*MARK:A)Y|X(*MARK:B)Z/K
2534 XY
2535 0: XY
2536 MK: A
2537 XZ
2538 0: XZ
2539 MK: B
2540 .sp
2541 The (*MARK) name is tagged with "MK:" in this output, and in this example it
2542 indicates which of the two alternatives matched. This is a more efficient way
2543 of obtaining this information than putting each alternative in its own
2544 capturing parentheses.
2545 .P
2546 A name may also be returned after a failed match if the final path through the
2547 pattern involves (*MARK). However, unless (*MARK) used in conjunction with
2548 (*COMMIT), this is unlikely to happen for an unanchored pattern because, as the
2549 starting point for matching is advanced, the final check is often with an empty
2550 string, causing a failure before (*MARK) is reached. For example:
2551 .sp
2552 /X(*MARK:A)Y|X(*MARK:B)Z/K
2553 XP
2554 No match
2555 .sp
2556 There are three potential starting points for this match (starting with X,
2557 starting with P, and with an empty string). If the pattern is anchored, the
2558 result is different:
2559 .sp
2560 /^X(*MARK:A)Y|^X(*MARK:B)Z/K
2561 XP
2562 No match, mark = B
2563 .sp
2564 PCRE's start-of-match optimizations can also interfere with this. For example,
2565 if, as a result of a call to \fBpcre_study()\fP, it knows the minimum
2566 subject length for a match, a shorter subject will not be scanned at all.
2567 .P
2568 Note that similar anomalies (though different in detail) exist in Perl, no
2569 doubt for the same reasons. The use of (*MARK) data after a failed match of an
2570 unanchored pattern is not recommended, unless (*COMMIT) is involved.
2571 .
2572 .
2573 .SS "Verbs that act after backtracking"
2574 .rs
2575 .sp
2576 The following verbs do nothing when they are encountered. Matching continues
2577 with what follows, but if there is no subsequent match, causing a backtrack to
2578 the verb, a failure is forced. That is, backtracking cannot pass to the left of
2579 the verb. However, when one of these verbs appears inside an atomic group, its
2580 effect is confined to that group, because once the group has been matched,
2581 there is never any backtracking into it. In this situation, backtracking can
2582 "jump back" to the left of the entire atomic group. (Remember also, as stated
2583 above, that this localization also applies in subroutine calls and assertions.)
2584 .P
2585 These verbs differ in exactly what kind of failure occurs when backtracking
2586 reaches them.
2587 .sp
2588 (*COMMIT)
2589 .sp
2590 This verb, which may not be followed by a name, causes the whole match to fail
2591 outright if the rest of the pattern does not match. Even if the pattern is
2592 unanchored, no further attempts to find a match by advancing the starting point
2593 take place. Once (*COMMIT) has been passed, \fBpcre_exec()\fP is committed to
2594 finding a match at the current starting point, or not at all. For example:
2595 .sp
2596 a+(*COMMIT)b
2597 .sp
2598 This matches "xxaab" but not "aacaab". It can be thought of as a kind of
2599 dynamic anchor, or "I've started, so I must finish." The name of the most
2600 recently passed (*MARK) in the path is passed back when (*COMMIT) forces a
2601 match failure.
2602 .P
2603 Note that (*COMMIT) at the start of a pattern is not the same as an anchor,
2604 unless PCRE's start-of-match optimizations are turned off, as shown in this
2605 \fBpcretest\fP example:
2606 .sp
2607 /(*COMMIT)abc/
2608 xyzabc
2609 0: abc
2610 xyzabc\eY
2611 No match
2612 .sp
2613 PCRE knows that any match must start with "a", so the optimization skips along
2614 the subject to "a" before running the first match attempt, which succeeds. When
2615 the optimization is disabled by the \eY escape in the second subject, the match
2616 starts at "x" and so the (*COMMIT) causes it to fail without trying any other
2617 starting points.
2618 .sp
2619 (*PRUNE) or (*PRUNE:NAME)
2620 .sp
2621 This verb causes the match to fail at the current starting position in the
2622 subject if the rest of the pattern does not match. If the pattern is
2623 unanchored, the normal "bumpalong" advance to the next starting character then
2624 happens. Backtracking can occur as usual to the left of (*PRUNE), before it is
2625 reached, or when matching to the right of (*PRUNE), but if there is no match to
2626 the right, backtracking cannot cross (*PRUNE). In simple cases, the use of
2627 (*PRUNE) is just an alternative to an atomic group or possessive quantifier,
2628 but there are some uses of (*PRUNE) that cannot be expressed in any other way.
2629 The behaviour of (*PRUNE:NAME) is the same as (*MARK:NAME)(*PRUNE) when the
2630 match fails completely; the name is passed back if this is the final attempt.
2631 (*PRUNE:NAME) does not pass back a name if the match succeeds. In an anchored
2632 pattern (*PRUNE) has the same effect as (*COMMIT).
2633 .sp
2634 (*SKIP)
2635 .sp
2636 This verb, when given without a name, is like (*PRUNE), except that if the
2637 pattern is unanchored, the "bumpalong" advance is not to the next character,
2638 but to the position in the subject where (*SKIP) was encountered. (*SKIP)
2639 signifies that whatever text was matched leading up to it cannot be part of a
2640 successful match. Consider:
2641 .sp
2642 a+(*SKIP)b
2643 .sp
2644 If the subject is "aaaac...", after the first match attempt fails (starting at
2645 the first character in the string), the starting point skips on to start the
2646 next attempt at "c". Note that a possessive quantifer does not have the same
2647 effect as this example; although it would suppress backtracking during the
2648 first match attempt, the second attempt would start at the second character
2649 instead of skipping on to "c".
2650 .sp
2651 (*SKIP:NAME)
2652 .sp
2653 When (*SKIP) has an associated name, its behaviour is modified. If the
2654 following pattern fails to match, the previous path through the pattern is
2655 searched for the most recent (*MARK) that has the same name. If one is found,
2656 the "bumpalong" advance is to the subject position that corresponds to that
2657 (*MARK) instead of to where (*SKIP) was encountered. If no (*MARK) with a
2658 matching name is found, normal "bumpalong" of one character happens (the
2659 (*SKIP) is ignored).
2660 .sp
2661 (*THEN) or (*THEN:NAME)
2662 .sp
2663 This verb causes a skip to the next alternation in the innermost enclosing
2664 group if the rest of the pattern does not match. That is, it cancels pending
2665 backtracking, but only within the current alternation. Its name comes from the
2666 observation that it can be used for a pattern-based if-then-else block:
2667 .sp
2668 ( COND1 (*THEN) FOO | COND2 (*THEN) BAR | COND3 (*THEN) BAZ ) ...
2669 .sp
2670 If the COND1 pattern matches, FOO is tried (and possibly further items after
2671 the end of the group if FOO succeeds); on failure the matcher skips to the
2672 second alternative and tries COND2, without backtracking into COND1. The
2673 behaviour of (*THEN:NAME) is exactly the same as (*MARK:NAME)(*THEN) if the
2674 overall match fails. If (*THEN) is not directly inside an alternation, it acts
2675 like (*PRUNE).
2676 .
2677 .P
2678 The above verbs provide four different "strengths" of control when subsequent
2679 matching fails. (*THEN) is the weakest, carrying on the match at the next
2680 alternation. (*PRUNE) comes next, failing the match at the current starting
2681 position, but allowing an advance to the next character (for an unanchored
2682 pattern). (*SKIP) is similar, except that the advance may be more than one
2683 character. (*COMMIT) is the strongest, causing the entire match to fail.
2684 .P
2685 If more than one is present in a pattern, the "stongest" one wins. For example,
2686 consider this pattern, where A, B, etc. are complex pattern fragments:
2687 .sp
2688 (A(*COMMIT)B(*THEN)C|D)
2689 .sp
2690 Once A has matched, PCRE is committed to this match, at the current starting
2691 position. If subsequently B matches, but C does not, the normal (*THEN) action
2692 of trying the next alternation (that is, D) does not happen because (*COMMIT)
2693 overrides.
2694 .
2695 .
2696 .SH "SEE ALSO"
2697 .rs
2698 .sp
2699 \fBpcreapi\fP(3), \fBpcrecallout\fP(3), \fBpcrematching\fP(3),
2700 \fBpcresyntax\fP(3), \fBpcre\fP(3).
2701 .
2702 .
2704 .rs
2705 .sp
2706 .nf
2707 Philip Hazel
2708 University Computing Service
2709 Cambridge CB2 3QH, England.
2710 .fi
2711 .
2712 .
2714 .rs
2715 .sp
2716 .nf
2717 Last updated: 31 October 2010
2718 Copyright (c) 1997-2010 University of Cambridge.
2719 .fi


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

  ViewVC Help
Powered by ViewVC 1.1.5