/[pcre]/code/trunk/doc/html/pcrepattern.html
ViewVC logotype

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

Parent Directory Parent Directory | Revision Log Revision Log


Revision 77 - (show annotations)
Sat Feb 24 21:40:45 2007 UTC (12 years, 6 months ago) by nigel
File MIME type: text/html
File size: 65817 byte(s)
Error occurred while calculating annotation data.
Load pcre-6.0 into code/trunk.
1 <html>
2 <head>
3 <title>pcrepattern specification</title>
4 </head>
5 <body bgcolor="#FFFFFF" text="#00005A" link="#0066FF" alink="#3399FF" vlink="#2222BB">
6 <h1>pcrepattern man page</h1>
7 <p>
8 Return to the <a href="index.html">PCRE index page</a>.
9 </p>
10 <p>
11 This page is part of the PCRE HTML documentation. It was generated automatically
12 from the original man page. If there is any nonsense in it, please consult the
13 man page, in case the conversion went wrong.
14 <br>
15 <ul>
16 <li><a name="TOC1" href="#SEC1">PCRE REGULAR EXPRESSION DETAILS</a>
17 <li><a name="TOC2" href="#SEC2">BACKSLASH</a>
18 <li><a name="TOC3" href="#SEC3">CIRCUMFLEX AND DOLLAR</a>
19 <li><a name="TOC4" href="#SEC4">FULL STOP (PERIOD, DOT)</a>
20 <li><a name="TOC5" href="#SEC5">MATCHING A SINGLE BYTE</a>
21 <li><a name="TOC6" href="#SEC6">SQUARE BRACKETS AND CHARACTER CLASSES</a>
22 <li><a name="TOC7" href="#SEC7">POSIX CHARACTER CLASSES</a>
23 <li><a name="TOC8" href="#SEC8">VERTICAL BAR</a>
24 <li><a name="TOC9" href="#SEC9">INTERNAL OPTION SETTING</a>
25 <li><a name="TOC10" href="#SEC10">SUBPATTERNS</a>
26 <li><a name="TOC11" href="#SEC11">NAMED SUBPATTERNS</a>
27 <li><a name="TOC12" href="#SEC12">REPETITION</a>
28 <li><a name="TOC13" href="#SEC13">ATOMIC GROUPING AND POSSESSIVE QUANTIFIERS</a>
29 <li><a name="TOC14" href="#SEC14">BACK REFERENCES</a>
30 <li><a name="TOC15" href="#SEC15">ASSERTIONS</a>
31 <li><a name="TOC16" href="#SEC16">CONDITIONAL SUBPATTERNS</a>
32 <li><a name="TOC17" href="#SEC17">COMMENTS</a>
33 <li><a name="TOC18" href="#SEC18">RECURSIVE PATTERNS</a>
34 <li><a name="TOC19" href="#SEC19">SUBPATTERNS AS SUBROUTINES</a>
35 <li><a name="TOC20" href="#SEC20">CALLOUTS</a>
36 </ul>
37 <br><a name="SEC1" href="#TOC1">PCRE REGULAR EXPRESSION DETAILS</a><br>
38 <P>
39 The syntax and semantics of the regular expressions supported by PCRE are
40 described below. Regular expressions are also described in the Perl
41 documentation and in a number of books, some of which have copious examples.
42 Jeffrey Friedl's "Mastering Regular Expressions", published by O'Reilly, covers
43 regular expressions in great detail. This description of PCRE's regular
44 expressions is intended as reference material.
45 </P>
46 <P>
47 The original operation of PCRE was on strings of one-byte characters. However,
48 there is now also support for UTF-8 character strings. To use this, you must
49 build PCRE to include UTF-8 support, and then call <b>pcre_compile()</b> with
50 the PCRE_UTF8 option. How this affects pattern matching is mentioned in several
51 places below. There is also a summary of UTF-8 features in the
52 <a href="pcre.html#utf8support">section on UTF-8 support</a>
53 in the main
54 <a href="pcre.html"><b>pcre</b></a>
55 page.
56 </P>
57 <P>
58 The remainder of this document discusses the patterns that are supported by
59 PCRE when its main matching function, <b>pcre_exec()</b>, is used.
60 From release 6.0, PCRE offers a second matching function,
61 <b>pcre_dfa_exec()</b>, which matches using a different algorithm that is not
62 Perl-compatible. The advantages and disadvantages of the alternative function,
63 and how it differs from the normal function, are discussed in the
64 <a href="pcrematching.html"><b>pcrematching</b></a>
65 page.
66 </P>
67 <P>
68 A regular expression is a pattern that is matched against a subject string from
69 left to right. Most characters stand for themselves in a pattern, and match the
70 corresponding characters in the subject. As a trivial example, the pattern
71 <pre>
72 The quick brown fox
73 </pre>
74 matches a portion of a subject string that is identical to itself. When
75 caseless matching is specified (the PCRE_CASELESS option), letters are matched
76 independently of case. In UTF-8 mode, PCRE always understands the concept of
77 case for characters whose values are less than 128, so caseless matching is
78 always possible. For characters with higher values, the concept of case is
79 supported if PCRE is compiled with Unicode property support, but not otherwise.
80 If you want to use caseless matching for characters 128 and above, you must
81 ensure that PCRE is compiled with Unicode property support as well as with
82 UTF-8 support.
83 </P>
84 <P>
85 The power of regular expressions comes from the ability to include alternatives
86 and repetitions in the pattern. These are encoded in the pattern by the use of
87 <i>metacharacters</i>, which do not stand for themselves but instead are
88 interpreted in some special way.
89 </P>
90 <P>
91 There are two different sets of metacharacters: those that are recognized
92 anywhere in the pattern except within square brackets, and those that are
93 recognized in square brackets. Outside square brackets, the metacharacters are
94 as follows:
95 <pre>
96 \ general escape character with several uses
97 ^ assert start of string (or line, in multiline mode)
98 $ assert end of string (or line, in multiline mode)
99 . match any character except newline (by default)
100 [ start character class definition
101 | start of alternative branch
102 ( start subpattern
103 ) end subpattern
104 ? extends the meaning of (
105 also 0 or 1 quantifier
106 also quantifier minimizer
107 * 0 or more quantifier
108 + 1 or more quantifier
109 also "possessive quantifier"
110 { start min/max quantifier
111 </pre>
112 Part of a pattern that is in square brackets is called a "character class". In
113 a character class the only metacharacters are:
114 <pre>
115 \ general escape character
116 ^ negate the class, but only if the first character
117 - indicates character range
118 [ POSIX character class (only if followed by POSIX syntax)
119 ] terminates the character class
120 </pre>
121 The following sections describe the use of each of the metacharacters.
122 </P>
123 <br><a name="SEC2" href="#TOC1">BACKSLASH</a><br>
124 <P>
125 The backslash character has several uses. Firstly, if it is followed by a
126 non-alphanumeric character, it takes away any special meaning that character may
127 have. This use of backslash as an escape character applies both inside and
128 outside character classes.
129 </P>
130 <P>
131 For example, if you want to match a * character, you write \* in the pattern.
132 This escaping action applies whether or not the following character would
133 otherwise be interpreted as a metacharacter, so it is always safe to precede a
134 non-alphanumeric with backslash to specify that it stands for itself. In
135 particular, if you want to match a backslash, you write \\.
136 </P>
137 <P>
138 If a pattern is compiled with the PCRE_EXTENDED option, whitespace in the
139 pattern (other than in a character class) and characters between a # outside
140 a character class and the next newline character are ignored. An escaping
141 backslash can be used to include a whitespace or # character as part of the
142 pattern.
143 </P>
144 <P>
145 If you want to remove the special meaning from a sequence of characters, you
146 can do so by putting them between \Q and \E. This is different from Perl in
147 that $ and @ are handled as literals in \Q...\E sequences in PCRE, whereas in
148 Perl, $ and @ cause variable interpolation. Note the following examples:
149 <pre>
150 Pattern PCRE matches Perl matches
151
152 \Qabc$xyz\E abc$xyz abc followed by the contents of $xyz
153 \Qabc\$xyz\E abc\$xyz abc\$xyz
154 \Qabc\E\$\Qxyz\E abc$xyz abc$xyz
155 </pre>
156 The \Q...\E sequence is recognized both inside and outside character classes.
157 <a name="digitsafterbackslash"></a></P>
158 <br><b>
159 Non-printing characters
160 </b><br>
161 <P>
162 A second use of backslash provides a way of encoding non-printing characters
163 in patterns in a visible manner. There is no restriction on the appearance of
164 non-printing characters, apart from the binary zero that terminates a pattern,
165 but when a pattern is being prepared by text editing, it is usually easier to
166 use one of the following escape sequences than the binary character it
167 represents:
168 <pre>
169 \a alarm, that is, the BEL character (hex 07)
170 \cx "control-x", where x is any character
171 \e escape (hex 1B)
172 \f formfeed (hex 0C)
173 \n newline (hex 0A)
174 \r carriage return (hex 0D)
175 \t tab (hex 09)
176 \ddd character with octal code ddd, or backreference
177 \xhh character with hex code hh
178 \x{hhh..} character with hex code hhh... (UTF-8 mode only)
179 </pre>
180 The precise effect of \cx is as follows: if x is a lower case letter, it
181 is converted to upper case. Then bit 6 of the character (hex 40) is inverted.
182 Thus \cz becomes hex 1A, but \c{ becomes hex 3B, while \c; becomes hex
183 7B.
184 </P>
185 <P>
186 After \x, from zero to two hexadecimal digits are read (letters can be in
187 upper or lower case). In UTF-8 mode, any number of hexadecimal digits may
188 appear between \x{ and }, but the value of the character code must be less
189 than 2**31 (that is, the maximum hexadecimal value is 7FFFFFFF). If characters
190 other than hexadecimal digits appear between \x{ and }, or if there is no
191 terminating }, this form of escape is not recognized. Instead, the initial
192 \x will be interpreted as a basic hexadecimal escape, with no following
193 digits, giving a character whose value is zero.
194 </P>
195 <P>
196 Characters whose value is less than 256 can be defined by either of the two
197 syntaxes for \x when PCRE is in UTF-8 mode. There is no difference in the
198 way they are handled. For example, \xdc is exactly the same as \x{dc}.
199 </P>
200 <P>
201 After \0 up to two further octal digits are read. In both cases, if there
202 are fewer than two digits, just those that are present are used. Thus the
203 sequence \0\x\07 specifies two binary zeros followed by a BEL character
204 (code value 7). Make sure you supply two digits after the initial zero if the
205 pattern character that follows is itself an octal digit.
206 </P>
207 <P>
208 The handling of a backslash followed by a digit other than 0 is complicated.
209 Outside a character class, PCRE reads it and any following digits as a decimal
210 number. If the number is less than 10, or if there have been at least that many
211 previous capturing left parentheses in the expression, the entire sequence is
212 taken as a <i>back reference</i>. A description of how this works is given
213 <a href="#backreferences">later,</a>
214 following the discussion of
215 <a href="#subpattern">parenthesized subpatterns.</a>
216 </P>
217 <P>
218 Inside a character class, or if the decimal number is greater than 9 and there
219 have not been that many capturing subpatterns, PCRE re-reads up to three octal
220 digits following the backslash, and generates a single byte from the least
221 significant 8 bits of the value. Any subsequent digits stand for themselves.
222 For example:
223 <pre>
224 \040 is another way of writing a space
225 \40 is the same, provided there are fewer than 40 previous capturing subpatterns
226 \7 is always a back reference
227 \11 might be a back reference, or another way of writing a tab
228 \011 is always a tab
229 \0113 is a tab followed by the character "3"
230 \113 might be a back reference, otherwise the character with octal code 113
231 \377 might be a back reference, otherwise the byte consisting entirely of 1 bits
232 \81 is either a back reference, or a binary zero followed by the two characters "8" and "1"
233 </pre>
234 Note that octal values of 100 or greater must not be introduced by a leading
235 zero, because no more than three octal digits are ever read.
236 </P>
237 <P>
238 All the sequences that define a single byte value or a single UTF-8 character
239 (in UTF-8 mode) can be used both inside and outside character classes. In
240 addition, inside a character class, the sequence \b is interpreted as the
241 backspace character (hex 08), and the sequence \X is interpreted as the
242 character "X". Outside a character class, these sequences have different
243 meanings
244 <a href="#uniextseq">(see below).</a>
245 </P>
246 <br><b>
247 Generic character types
248 </b><br>
249 <P>
250 The third use of backslash is for specifying generic character types. The
251 following are always recognized:
252 <pre>
253 \d any decimal digit
254 \D any character that is not a decimal digit
255 \s any whitespace character
256 \S any character that is not a whitespace character
257 \w any "word" character
258 \W any "non-word" character
259 </pre>
260 Each pair of escape sequences partitions the complete set of characters into
261 two disjoint sets. Any given character matches one, and only one, of each pair.
262 </P>
263 <P>
264 These character type sequences can appear both inside and outside character
265 classes. They each match one character of the appropriate type. If the current
266 matching point is at the end of the subject string, all of them fail, since
267 there is no character to match.
268 </P>
269 <P>
270 For compatibility with Perl, \s does not match the VT character (code 11).
271 This makes it different from the the POSIX "space" class. The \s characters
272 are HT (9), LF (10), FF (12), CR (13), and space (32).
273 </P>
274 <P>
275 A "word" character is an underscore or any character less than 256 that is a
276 letter or digit. The definition of letters and digits is controlled by PCRE's
277 low-valued character tables, and may vary if locale-specific matching is taking
278 place (see
279 <a href="pcreapi.html#localesupport">"Locale support"</a>
280 in the
281 <a href="pcreapi.html"><b>pcreapi</b></a>
282 page). For example, in the "fr_FR" (French) locale, some character codes
283 greater than 128 are used for accented letters, and these are matched by \w.
284 </P>
285 <P>
286 In UTF-8 mode, characters with values greater than 128 never match \d, \s, or
287 \w, and always match \D, \S, and \W. This is true even when Unicode
288 character property support is available.
289 <a name="uniextseq"></a></P>
290 <br><b>
291 Unicode character properties
292 </b><br>
293 <P>
294 When PCRE is built with Unicode character property support, three additional
295 escape sequences to match generic character types are available when UTF-8 mode
296 is selected. They are:
297 <pre>
298 \p{<i>xx</i>} a character with the <i>xx</i> property
299 \P{<i>xx</i>} a character without the <i>xx</i> property
300 \X an extended Unicode sequence
301 </pre>
302 The property names represented by <i>xx</i> above are limited to the
303 Unicode general category properties. Each character has exactly one such
304 property, specified by a two-letter abbreviation. For compatibility with Perl,
305 negation can be specified by including a circumflex between the opening brace
306 and the property name. For example, \p{^Lu} is the same as \P{Lu}.
307 </P>
308 <P>
309 If only one letter is specified with \p or \P, it includes all the properties
310 that start with that letter. In this case, in the absence of negation, the
311 curly brackets in the escape sequence are optional; these two examples have
312 the same effect:
313 <pre>
314 \p{L}
315 \pL
316 </pre>
317 The following property codes are supported:
318 <pre>
319 C Other
320 Cc Control
321 Cf Format
322 Cn Unassigned
323 Co Private use
324 Cs Surrogate
325
326 L Letter
327 Ll Lower case letter
328 Lm Modifier letter
329 Lo Other letter
330 Lt Title case letter
331 Lu Upper case letter
332
333 M Mark
334 Mc Spacing mark
335 Me Enclosing mark
336 Mn Non-spacing mark
337
338 N Number
339 Nd Decimal number
340 Nl Letter number
341 No Other number
342
343 P Punctuation
344 Pc Connector punctuation
345 Pd Dash punctuation
346 Pe Close punctuation
347 Pf Final punctuation
348 Pi Initial punctuation
349 Po Other punctuation
350 Ps Open punctuation
351
352 S Symbol
353 Sc Currency symbol
354 Sk Modifier symbol
355 Sm Mathematical symbol
356 So Other symbol
357
358 Z Separator
359 Zl Line separator
360 Zp Paragraph separator
361 Zs Space separator
362 </pre>
363 Extended properties such as "Greek" or "InMusicalSymbols" are not supported by
364 PCRE.
365 </P>
366 <P>
367 Specifying caseless matching does not affect these escape sequences. For
368 example, \p{Lu} always matches only upper case letters.
369 </P>
370 <P>
371 The \X escape matches any number of Unicode characters that form an extended
372 Unicode sequence. \X is equivalent to
373 <pre>
374 (?&#62;\PM\pM*)
375 </pre>
376 That is, it matches a character without the "mark" property, followed by zero
377 or more characters with the "mark" property, and treats the sequence as an
378 atomic group
379 <a href="#atomicgroup">(see below).</a>
380 Characters with the "mark" property are typically accents that affect the
381 preceding character.
382 </P>
383 <P>
384 Matching characters by Unicode property is not fast, because PCRE has to search
385 a structure that contains data for over fifteen thousand characters. That is
386 why the traditional escape sequences such as \d and \w do not use Unicode
387 properties in PCRE.
388 <a name="smallassertions"></a></P>
389 <br><b>
390 Simple assertions
391 </b><br>
392 <P>
393 The fourth use of backslash is for certain simple assertions. An assertion
394 specifies a condition that has to be met at a particular point in a match,
395 without consuming any characters from the subject string. The use of
396 subpatterns for more complicated assertions is described
397 <a href="#bigassertions">below.</a>
398 The backslashed
399 assertions are:
400 <pre>
401 \b matches at a word boundary
402 \B matches when not at a word boundary
403 \A matches at start of subject
404 \Z matches at end of subject or before newline at end
405 \z matches at end of subject
406 \G matches at first matching position in subject
407 </pre>
408 These assertions may not appear in character classes (but note that \b has a
409 different meaning, namely the backspace character, inside a character class).
410 </P>
411 <P>
412 A word boundary is a position in the subject string where the current character
413 and the previous character do not both match \w or \W (i.e. one matches
414 \w and the other matches \W), or the start or end of the string if the
415 first or last character matches \w, respectively.
416 </P>
417 <P>
418 The \A, \Z, and \z assertions differ from the traditional circumflex and
419 dollar (described in the next section) in that they only ever match at the very
420 start and end of the subject string, whatever options are set. Thus, they are
421 independent of multiline mode. These three assertions are not affected by the
422 PCRE_NOTBOL or PCRE_NOTEOL options, which affect only the behaviour of the
423 circumflex and dollar metacharacters. However, if the <i>startoffset</i>
424 argument of <b>pcre_exec()</b> is non-zero, indicating that matching is to start
425 at a point other than the beginning of the subject, \A can never match. The
426 difference between \Z and \z is that \Z matches before a newline that is the
427 last character of the string as well as at the end of the string, whereas \z
428 matches only at the end.
429 </P>
430 <P>
431 The \G assertion is true only when the current matching position is at the
432 start point of the match, as specified by the <i>startoffset</i> argument of
433 <b>pcre_exec()</b>. It differs from \A when the value of <i>startoffset</i> is
434 non-zero. By calling <b>pcre_exec()</b> multiple times with appropriate
435 arguments, you can mimic Perl's /g option, and it is in this kind of
436 implementation where \G can be useful.
437 </P>
438 <P>
439 Note, however, that PCRE's interpretation of \G, as the start of the current
440 match, is subtly different from Perl's, which defines it as the end of the
441 previous match. In Perl, these can be different when the previously matched
442 string was empty. Because PCRE does just one match at a time, it cannot
443 reproduce this behaviour.
444 </P>
445 <P>
446 If all the alternatives of a pattern begin with \G, the expression is anchored
447 to the starting match position, and the "anchored" flag is set in the compiled
448 regular expression.
449 </P>
450 <br><a name="SEC3" href="#TOC1">CIRCUMFLEX AND DOLLAR</a><br>
451 <P>
452 Outside a character class, in the default matching mode, the circumflex
453 character is an assertion that is true only if the current matching point is
454 at the start of the subject string. If the <i>startoffset</i> argument of
455 <b>pcre_exec()</b> is non-zero, circumflex can never match if the PCRE_MULTILINE
456 option is unset. Inside a character class, circumflex has an entirely different
457 meaning
458 <a href="#characterclass">(see below).</a>
459 </P>
460 <P>
461 Circumflex need not be the first character of the pattern if a number of
462 alternatives are involved, but it should be the first thing in each alternative
463 in which it appears if the pattern is ever to match that branch. If all
464 possible alternatives start with a circumflex, that is, if the pattern is
465 constrained to match only at the start of the subject, it is said to be an
466 "anchored" pattern. (There are also other constructs that can cause a pattern
467 to be anchored.)
468 </P>
469 <P>
470 A dollar character is an assertion that is true only if the current matching
471 point is at the end of the subject string, or immediately before a newline
472 character that is the last character in the string (by default). Dollar need
473 not be the last character of the pattern if a number of alternatives are
474 involved, but it should be the last item in any branch in which it appears.
475 Dollar has no special meaning in a character class.
476 </P>
477 <P>
478 The meaning of dollar can be changed so that it matches only at the very end of
479 the string, by setting the PCRE_DOLLAR_ENDONLY option at compile time. This
480 does not affect the \Z assertion.
481 </P>
482 <P>
483 The meanings of the circumflex and dollar characters are changed if the
484 PCRE_MULTILINE option is set. When this is the case, they match immediately
485 after and immediately before an internal newline character, respectively, in
486 addition to matching at the start and end of the subject string. For example,
487 the pattern /^abc$/ matches the subject string "def\nabc" (where \n
488 represents a newline character) in multiline mode, but not otherwise.
489 Consequently, patterns that are anchored in single line mode because all
490 branches start with ^ are not anchored in multiline mode, and a match for
491 circumflex is possible when the <i>startoffset</i> argument of <b>pcre_exec()</b>
492 is non-zero. The PCRE_DOLLAR_ENDONLY option is ignored if PCRE_MULTILINE is
493 set.
494 </P>
495 <P>
496 Note that the sequences \A, \Z, and \z can be used to match the start and
497 end of the subject in both modes, and if all branches of a pattern start with
498 \A it is always anchored, whether PCRE_MULTILINE is set or not.
499 </P>
500 <br><a name="SEC4" href="#TOC1">FULL STOP (PERIOD, DOT)</a><br>
501 <P>
502 Outside a character class, a dot in the pattern matches any one character in
503 the subject, including a non-printing character, but not (by default) newline.
504 In UTF-8 mode, a dot matches any UTF-8 character, which might be more than one
505 byte long, except (by default) newline. If the PCRE_DOTALL option is set,
506 dots match newlines as well. The handling of dot is entirely independent of the
507 handling of circumflex and dollar, the only relationship being that they both
508 involve newline characters. Dot has no special meaning in a character class.
509 </P>
510 <br><a name="SEC5" href="#TOC1">MATCHING A SINGLE BYTE</a><br>
511 <P>
512 Outside a character class, the escape sequence \C matches any one byte, both
513 in and out of UTF-8 mode. Unlike a dot, it can match a newline. The feature is
514 provided in Perl in order to match individual bytes in UTF-8 mode. Because it
515 breaks up UTF-8 characters into individual bytes, what remains in the string
516 may be a malformed UTF-8 string. For this reason, the \C escape sequence is
517 best avoided.
518 </P>
519 <P>
520 PCRE does not allow \C to appear in lookbehind assertions
521 <a href="#lookbehind">(described below),</a>
522 because in UTF-8 mode this would make it impossible to calculate the length of
523 the lookbehind.
524 <a name="characterclass"></a></P>
525 <br><a name="SEC6" href="#TOC1">SQUARE BRACKETS AND CHARACTER CLASSES</a><br>
526 <P>
527 An opening square bracket introduces a character class, terminated by a closing
528 square bracket. A closing square bracket on its own is not special. If a
529 closing square bracket is required as a member of the class, it should be the
530 first data character in the class (after an initial circumflex, if present) or
531 escaped with a backslash.
532 </P>
533 <P>
534 A character class matches a single character in the subject. In UTF-8 mode, the
535 character may occupy more than one byte. A matched character must be in the set
536 of characters defined by the class, unless the first character in the class
537 definition is a circumflex, in which case the subject character must not be in
538 the set defined by the class. If a circumflex is actually required as a member
539 of the class, ensure it is not the first character, or escape it with a
540 backslash.
541 </P>
542 <P>
543 For example, the character class [aeiou] matches any lower case vowel, while
544 [^aeiou] matches any character that is not a lower case vowel. Note that a
545 circumflex is just a convenient notation for specifying the characters that
546 are in the class by enumerating those that are not. A class that starts with a
547 circumflex is not an assertion: it still consumes a character from the subject
548 string, and therefore it fails if the current pointer is at the end of the
549 string.
550 </P>
551 <P>
552 In UTF-8 mode, characters with values greater than 255 can be included in a
553 class as a literal string of bytes, or by using the \x{ escaping mechanism.
554 </P>
555 <P>
556 When caseless matching is set, any letters in a class represent both their
557 upper case and lower case versions, so for example, a caseless [aeiou] matches
558 "A" as well as "a", and a caseless [^aeiou] does not match "A", whereas a
559 caseful version would. In UTF-8 mode, PCRE always understands the concept of
560 case for characters whose values are less than 128, so caseless matching is
561 always possible. For characters with higher values, the concept of case is
562 supported if PCRE is compiled with Unicode property support, but not otherwise.
563 If you want to use caseless matching for characters 128 and above, you must
564 ensure that PCRE is compiled with Unicode property support as well as with
565 UTF-8 support.
566 </P>
567 <P>
568 The newline character is never treated in any special way in character classes,
569 whatever the setting of the PCRE_DOTALL or PCRE_MULTILINE options is. A class
570 such as [^a] will always match a newline.
571 </P>
572 <P>
573 The minus (hyphen) character can be used to specify a range of characters in a
574 character class. For example, [d-m] matches any letter between d and m,
575 inclusive. If a minus character is required in a class, it must be escaped with
576 a backslash or appear in a position where it cannot be interpreted as
577 indicating a range, typically as the first or last character in the class.
578 </P>
579 <P>
580 It is not possible to have the literal character "]" as the end character of a
581 range. A pattern such as [W-]46] is interpreted as a class of two characters
582 ("W" and "-") followed by a literal string "46]", so it would match "W46]" or
583 "-46]". However, if the "]" is escaped with a backslash it is interpreted as
584 the end of range, so [W-\]46] is interpreted as a class containing a range
585 followed by two other characters. The octal or hexadecimal representation of
586 "]" can also be used to end a range.
587 </P>
588 <P>
589 Ranges operate in the collating sequence of character values. They can also be
590 used for characters specified numerically, for example [\000-\037]. In UTF-8
591 mode, ranges can include characters whose values are greater than 255, for
592 example [\x{100}-\x{2ff}].
593 </P>
594 <P>
595 If a range that includes letters is used when caseless matching is set, it
596 matches the letters in either case. For example, [W-c] is equivalent to
597 [][\\^_`wxyzabc], matched caselessly, and in non-UTF-8 mode, if character
598 tables for the "fr_FR" locale are in use, [\xc8-\xcb] matches accented E
599 characters in both cases. In UTF-8 mode, PCRE supports the concept of case for
600 characters with values greater than 128 only when it is compiled with Unicode
601 property support.
602 </P>
603 <P>
604 The character types \d, \D, \p, \P, \s, \S, \w, and \W may also appear
605 in a character class, and add the characters that they match to the class. For
606 example, [\dABCDEF] matches any hexadecimal digit. A circumflex can
607 conveniently be used with the upper case character types to specify a more
608 restricted set of characters than the matching lower case type. For example,
609 the class [^\W_] matches any letter or digit, but not underscore.
610 </P>
611 <P>
612 The only metacharacters that are recognized in character classes are backslash,
613 hyphen (only where it can be interpreted as specifying a range), circumflex
614 (only at the start), opening square bracket (only when it can be interpreted as
615 introducing a POSIX class name - see the next section), and the terminating
616 closing square bracket. However, escaping other non-alphanumeric characters
617 does no harm.
618 </P>
619 <br><a name="SEC7" href="#TOC1">POSIX CHARACTER CLASSES</a><br>
620 <P>
621 Perl supports the POSIX notation for character classes. This uses names
622 enclosed by [: and :] within the enclosing square brackets. PCRE also supports
623 this notation. For example,
624 <pre>
625 [01[:alpha:]%]
626 </pre>
627 matches "0", "1", any alphabetic character, or "%". The supported class names
628 are
629 <pre>
630 alnum letters and digits
631 alpha letters
632 ascii character codes 0 - 127
633 blank space or tab only
634 cntrl control characters
635 digit decimal digits (same as \d)
636 graph printing characters, excluding space
637 lower lower case letters
638 print printing characters, including space
639 punct printing characters, excluding letters and digits
640 space white space (not quite the same as \s)
641 upper upper case letters
642 word "word" characters (same as \w)
643 xdigit hexadecimal digits
644 </pre>
645 The "space" characters are HT (9), LF (10), VT (11), FF (12), CR (13), and
646 space (32). Notice that this list includes the VT character (code 11). This
647 makes "space" different to \s, which does not include VT (for Perl
648 compatibility).
649 </P>
650 <P>
651 The name "word" is a Perl extension, and "blank" is a GNU extension from Perl
652 5.8. Another Perl extension is negation, which is indicated by a ^ character
653 after the colon. For example,
654 <pre>
655 [12[:^digit:]]
656 </pre>
657 matches "1", "2", or any non-digit. PCRE (and Perl) also recognize the POSIX
658 syntax [.ch.] and [=ch=] where "ch" is a "collating element", but these are not
659 supported, and an error is given if they are encountered.
660 </P>
661 <P>
662 In UTF-8 mode, characters with values greater than 128 do not match any of
663 the POSIX character classes.
664 </P>
665 <br><a name="SEC8" href="#TOC1">VERTICAL BAR</a><br>
666 <P>
667 Vertical bar characters are used to separate alternative patterns. For example,
668 the pattern
669 <pre>
670 gilbert|sullivan
671 </pre>
672 matches either "gilbert" or "sullivan". Any number of alternatives may appear,
673 and an empty alternative is permitted (matching the empty string).
674 The matching process tries each alternative in turn, from left to right,
675 and the first one that succeeds is used. If the alternatives are within a
676 subpattern
677 <a href="#subpattern">(defined below),</a>
678 "succeeds" means matching the rest of the main pattern as well as the
679 alternative in the subpattern.
680 </P>
681 <br><a name="SEC9" href="#TOC1">INTERNAL OPTION SETTING</a><br>
682 <P>
683 The settings of the PCRE_CASELESS, PCRE_MULTILINE, PCRE_DOTALL, and
684 PCRE_EXTENDED options can be changed from within the pattern by a sequence of
685 Perl option letters enclosed between "(?" and ")". The option letters are
686 <pre>
687 i for PCRE_CASELESS
688 m for PCRE_MULTILINE
689 s for PCRE_DOTALL
690 x for PCRE_EXTENDED
691 </pre>
692 For example, (?im) sets caseless, multiline matching. It is also possible to
693 unset these options by preceding the letter with a hyphen, and a combined
694 setting and unsetting such as (?im-sx), which sets PCRE_CASELESS and
695 PCRE_MULTILINE while unsetting PCRE_DOTALL and PCRE_EXTENDED, is also
696 permitted. If a letter appears both before and after the hyphen, the option is
697 unset.
698 </P>
699 <P>
700 When an option change occurs at top level (that is, not inside subpattern
701 parentheses), the change applies to the remainder of the pattern that follows.
702 If the change is placed right at the start of a pattern, PCRE extracts it into
703 the global options (and it will therefore show up in data extracted by the
704 <b>pcre_fullinfo()</b> function).
705 </P>
706 <P>
707 An option change within a subpattern affects only that part of the current
708 pattern that follows it, so
709 <pre>
710 (a(?i)b)c
711 </pre>
712 matches abc and aBc and no other strings (assuming PCRE_CASELESS is not used).
713 By this means, options can be made to have different settings in different
714 parts of the pattern. Any changes made in one alternative do carry on
715 into subsequent branches within the same subpattern. For example,
716 <pre>
717 (a(?i)b|c)
718 </pre>
719 matches "ab", "aB", "c", and "C", even though when matching "C" the first
720 branch is abandoned before the option setting. This is because the effects of
721 option settings happen at compile time. There would be some very weird
722 behaviour otherwise.
723 </P>
724 <P>
725 The PCRE-specific options PCRE_UNGREEDY and PCRE_EXTRA can be changed in the
726 same way as the Perl-compatible options by using the characters U and X
727 respectively. The (?X) flag setting is special in that it must always occur
728 earlier in the pattern than any of the additional features it turns on, even
729 when it is at top level. It is best to put it at the start.
730 <a name="subpattern"></a></P>
731 <br><a name="SEC10" href="#TOC1">SUBPATTERNS</a><br>
732 <P>
733 Subpatterns are delimited by parentheses (round brackets), which can be nested.
734 Turning part of a pattern into a subpattern does two things:
735 <br>
736 <br>
737 1. It localizes a set of alternatives. For example, the pattern
738 <pre>
739 cat(aract|erpillar|)
740 </pre>
741 matches one of the words "cat", "cataract", or "caterpillar". Without the
742 parentheses, it would match "cataract", "erpillar" or the empty string.
743 <br>
744 <br>
745 2. It sets up the subpattern as a capturing subpattern. This means that, when
746 the whole pattern matches, that portion of the subject string that matched the
747 subpattern is passed back to the caller via the <i>ovector</i> argument of
748 <b>pcre_exec()</b>. Opening parentheses are counted from left to right (starting
749 from 1) to obtain numbers for the capturing subpatterns.
750 </P>
751 <P>
752 For example, if the string "the red king" is matched against the pattern
753 <pre>
754 the ((red|white) (king|queen))
755 </pre>
756 the captured substrings are "red king", "red", and "king", and are numbered 1,
757 2, and 3, respectively.
758 </P>
759 <P>
760 The fact that plain parentheses fulfil two functions is not always helpful.
761 There are often times when a grouping subpattern is required without a
762 capturing requirement. If an opening parenthesis is followed by a question mark
763 and a colon, the subpattern does not do any capturing, and is not counted when
764 computing the number of any subsequent capturing subpatterns. For example, if
765 the string "the white queen" is matched against the pattern
766 <pre>
767 the ((?:red|white) (king|queen))
768 </pre>
769 the captured substrings are "white queen" and "queen", and are numbered 1 and
770 2. The maximum number of capturing subpatterns is 65535, and the maximum depth
771 of nesting of all subpatterns, both capturing and non-capturing, is 200.
772 </P>
773 <P>
774 As a convenient shorthand, if any option settings are required at the start of
775 a non-capturing subpattern, the option letters may appear between the "?" and
776 the ":". Thus the two patterns
777 <pre>
778 (?i:saturday|sunday)
779 (?:(?i)saturday|sunday)
780 </pre>
781 match exactly the same set of strings. Because alternative branches are tried
782 from left to right, and options are not reset until the end of the subpattern
783 is reached, an option setting in one branch does affect subsequent branches, so
784 the above patterns match "SUNDAY" as well as "Saturday".
785 </P>
786 <br><a name="SEC11" href="#TOC1">NAMED SUBPATTERNS</a><br>
787 <P>
788 Identifying capturing parentheses by number is simple, but it can be very hard
789 to keep track of the numbers in complicated regular expressions. Furthermore,
790 if an expression is modified, the numbers may change. To help with this
791 difficulty, PCRE supports the naming of subpatterns, something that Perl does
792 not provide. The Python syntax (?P&#60;name&#62;...) is used. Names consist of
793 alphanumeric characters and underscores, and must be unique within a pattern.
794 </P>
795 <P>
796 Named capturing parentheses are still allocated numbers as well as names. The
797 PCRE API provides function calls for extracting the name-to-number translation
798 table from a compiled pattern. There is also a convenience function for
799 extracting a captured substring by name. For further details see the
800 <a href="pcreapi.html"><b>pcreapi</b></a>
801 documentation.
802 </P>
803 <br><a name="SEC12" href="#TOC1">REPETITION</a><br>
804 <P>
805 Repetition is specified by quantifiers, which can follow any of the following
806 items:
807 <pre>
808 a literal data character
809 the . metacharacter
810 the \C escape sequence
811 the \X escape sequence (in UTF-8 mode with Unicode properties)
812 an escape such as \d that matches a single character
813 a character class
814 a back reference (see next section)
815 a parenthesized subpattern (unless it is an assertion)
816 </pre>
817 The general repetition quantifier specifies a minimum and maximum number of
818 permitted matches, by giving the two numbers in curly brackets (braces),
819 separated by a comma. The numbers must be less than 65536, and the first must
820 be less than or equal to the second. For example:
821 <pre>
822 z{2,4}
823 </pre>
824 matches "zz", "zzz", or "zzzz". A closing brace on its own is not a special
825 character. If the second number is omitted, but the comma is present, there is
826 no upper limit; if the second number and the comma are both omitted, the
827 quantifier specifies an exact number of required matches. Thus
828 <pre>
829 [aeiou]{3,}
830 </pre>
831 matches at least 3 successive vowels, but may match many more, while
832 <pre>
833 \d{8}
834 </pre>
835 matches exactly 8 digits. An opening curly bracket that appears in a position
836 where a quantifier is not allowed, or one that does not match the syntax of a
837 quantifier, is taken as a literal character. For example, {,6} is not a
838 quantifier, but a literal string of four characters.
839 </P>
840 <P>
841 In UTF-8 mode, quantifiers apply to UTF-8 characters rather than to individual
842 bytes. Thus, for example, \x{100}{2} matches two UTF-8 characters, each of
843 which is represented by a two-byte sequence. Similarly, when Unicode property
844 support is available, \X{3} matches three Unicode extended sequences, each of
845 which may be several bytes long (and they may be of different lengths).
846 </P>
847 <P>
848 The quantifier {0} is permitted, causing the expression to behave as if the
849 previous item and the quantifier were not present.
850 </P>
851 <P>
852 For convenience (and historical compatibility) the three most common
853 quantifiers have single-character abbreviations:
854 <pre>
855 * is equivalent to {0,}
856 + is equivalent to {1,}
857 ? is equivalent to {0,1}
858 </pre>
859 It is possible to construct infinite loops by following a subpattern that can
860 match no characters with a quantifier that has no upper limit, for example:
861 <pre>
862 (a?)*
863 </pre>
864 Earlier versions of Perl and PCRE used to give an error at compile time for
865 such patterns. However, because there are cases where this can be useful, such
866 patterns are now accepted, but if any repetition of the subpattern does in fact
867 match no characters, the loop is forcibly broken.
868 </P>
869 <P>
870 By default, the quantifiers are "greedy", that is, they match as much as
871 possible (up to the maximum number of permitted times), without causing the
872 rest of the pattern to fail. The classic example of where this gives problems
873 is in trying to match comments in C programs. These appear between /* and */
874 and within the comment, individual * and / characters may appear. An attempt to
875 match C comments by applying the pattern
876 <pre>
877 /\*.*\*/
878 </pre>
879 to the string
880 <pre>
881 /* first comment */ not comment /* second comment */
882 </pre>
883 fails, because it matches the entire string owing to the greediness of the .*
884 item.
885 </P>
886 <P>
887 However, if a quantifier is followed by a question mark, it ceases to be
888 greedy, and instead matches the minimum number of times possible, so the
889 pattern
890 <pre>
891 /\*.*?\*/
892 </pre>
893 does the right thing with the C comments. The meaning of the various
894 quantifiers is not otherwise changed, just the preferred number of matches.
895 Do not confuse this use of question mark with its use as a quantifier in its
896 own right. Because it has two uses, it can sometimes appear doubled, as in
897 <pre>
898 \d??\d
899 </pre>
900 which matches one digit by preference, but can match two if that is the only
901 way the rest of the pattern matches.
902 </P>
903 <P>
904 If the PCRE_UNGREEDY option is set (an option which is not available in Perl),
905 the quantifiers are not greedy by default, but individual ones can be made
906 greedy by following them with a question mark. In other words, it inverts the
907 default behaviour.
908 </P>
909 <P>
910 When a parenthesized subpattern is quantified with a minimum repeat count that
911 is greater than 1 or with a limited maximum, more memory is required for the
912 compiled pattern, in proportion to the size of the minimum or maximum.
913 </P>
914 <P>
915 If a pattern starts with .* or .{0,} and the PCRE_DOTALL option (equivalent
916 to Perl's /s) is set, thus allowing the . to match newlines, the pattern is
917 implicitly anchored, because whatever follows will be tried against every
918 character position in the subject string, so there is no point in retrying the
919 overall match at any position after the first. PCRE normally treats such a
920 pattern as though it were preceded by \A.
921 </P>
922 <P>
923 In cases where it is known that the subject string contains no newlines, it is
924 worth setting PCRE_DOTALL in order to obtain this optimization, or
925 alternatively using ^ to indicate anchoring explicitly.
926 </P>
927 <P>
928 However, there is one situation where the optimization cannot be used. When .*
929 is inside capturing parentheses that are the subject of a backreference
930 elsewhere in the pattern, a match at the start may fail, and a later one
931 succeed. Consider, for example:
932 <pre>
933 (.*)abc\1
934 </pre>
935 If the subject is "xyz123abc123" the match point is the fourth character. For
936 this reason, such a pattern is not implicitly anchored.
937 </P>
938 <P>
939 When a capturing subpattern is repeated, the value captured is the substring
940 that matched the final iteration. For example, after
941 <pre>
942 (tweedle[dume]{3}\s*)+
943 </pre>
944 has matched "tweedledum tweedledee" the value of the captured substring is
945 "tweedledee". However, if there are nested capturing subpatterns, the
946 corresponding captured values may have been set in previous iterations. For
947 example, after
948 <pre>
949 /(a|(b))+/
950 </pre>
951 matches "aba" the value of the second captured substring is "b".
952 <a name="atomicgroup"></a></P>
953 <br><a name="SEC13" href="#TOC1">ATOMIC GROUPING AND POSSESSIVE QUANTIFIERS</a><br>
954 <P>
955 With both maximizing and minimizing repetition, failure of what follows
956 normally causes the repeated item to be re-evaluated to see if a different
957 number of repeats allows the rest of the pattern to match. Sometimes it is
958 useful to prevent this, either to change the nature of the match, or to cause
959 it fail earlier than it otherwise might, when the author of the pattern knows
960 there is no point in carrying on.
961 </P>
962 <P>
963 Consider, for example, the pattern \d+foo when applied to the subject line
964 <pre>
965 123456bar
966 </pre>
967 After matching all 6 digits and then failing to match "foo", the normal
968 action of the matcher is to try again with only 5 digits matching the \d+
969 item, and then with 4, and so on, before ultimately failing. "Atomic grouping"
970 (a term taken from Jeffrey Friedl's book) provides the means for specifying
971 that once a subpattern has matched, it is not to be re-evaluated in this way.
972 </P>
973 <P>
974 If we use atomic grouping for the previous example, the matcher would give up
975 immediately on failing to match "foo" the first time. The notation is a kind of
976 special parenthesis, starting with (?&#62; as in this example:
977 <pre>
978 (?&#62;\d+)foo
979 </pre>
980 This kind of parenthesis "locks up" the part of the pattern it contains once
981 it has matched, and a failure further into the pattern is prevented from
982 backtracking into it. Backtracking past it to previous items, however, works as
983 normal.
984 </P>
985 <P>
986 An alternative description is that a subpattern of this type matches the string
987 of characters that an identical standalone pattern would match, if anchored at
988 the current point in the subject string.
989 </P>
990 <P>
991 Atomic grouping subpatterns are not capturing subpatterns. Simple cases such as
992 the above example can be thought of as a maximizing repeat that must swallow
993 everything it can. So, while both \d+ and \d+? are prepared to adjust the
994 number of digits they match in order to make the rest of the pattern match,
995 (?&#62;\d+) can only match an entire sequence of digits.
996 </P>
997 <P>
998 Atomic groups in general can of course contain arbitrarily complicated
999 subpatterns, and can be nested. However, when the subpattern for an atomic
1000 group is just a single repeated item, as in the example above, a simpler
1001 notation, called a "possessive quantifier" can be used. This consists of an
1002 additional + character following a quantifier. Using this notation, the
1003 previous example can be rewritten as
1004 <pre>
1005 \d++foo
1006 </pre>
1007 Possessive quantifiers are always greedy; the setting of the PCRE_UNGREEDY
1008 option is ignored. They are a convenient notation for the simpler forms of
1009 atomic group. However, there is no difference in the meaning or processing of a
1010 possessive quantifier and the equivalent atomic group.
1011 </P>
1012 <P>
1013 The possessive quantifier syntax is an extension to the Perl syntax. It
1014 originates in Sun's Java package.
1015 </P>
1016 <P>
1017 When a pattern contains an unlimited repeat inside a subpattern that can itself
1018 be repeated an unlimited number of times, the use of an atomic group is the
1019 only way to avoid some failing matches taking a very long time indeed. The
1020 pattern
1021 <pre>
1022 (\D+|&#60;\d+&#62;)*[!?]
1023 </pre>
1024 matches an unlimited number of substrings that either consist of non-digits, or
1025 digits enclosed in &#60;&#62;, followed by either ! or ?. When it matches, it runs
1026 quickly. However, if it is applied to
1027 <pre>
1028 aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa
1029 </pre>
1030 it takes a long time before reporting failure. This is because the string can
1031 be divided between the internal \D+ repeat and the external * repeat in a
1032 large number of ways, and all have to be tried. (The example uses [!?] rather
1033 than a single character at the end, because both PCRE and Perl have an
1034 optimization that allows for fast failure when a single character is used. They
1035 remember the last single character that is required for a match, and fail early
1036 if it is not present in the string.) If the pattern is changed so that it uses
1037 an atomic group, like this:
1038 <pre>
1039 ((?&#62;\D+)|&#60;\d+&#62;)*[!?]
1040 </pre>
1041 sequences of non-digits cannot be broken, and failure happens quickly.
1042 <a name="backreferences"></a></P>
1043 <br><a name="SEC14" href="#TOC1">BACK REFERENCES</a><br>
1044 <P>
1045 Outside a character class, a backslash followed by a digit greater than 0 (and
1046 possibly further digits) is a back reference to a capturing subpattern earlier
1047 (that is, to its left) in the pattern, provided there have been that many
1048 previous capturing left parentheses.
1049 </P>
1050 <P>
1051 However, if the decimal number following the backslash is less than 10, it is
1052 always taken as a back reference, and causes an error only if there are not
1053 that many capturing left parentheses in the entire pattern. In other words, the
1054 parentheses that are referenced need not be to the left of the reference for
1055 numbers less than 10. See the subsection entitled "Non-printing characters"
1056 <a href="#digitsafterbackslash">above</a>
1057 for further details of the handling of digits following a backslash.
1058 </P>
1059 <P>
1060 A back reference matches whatever actually matched the capturing subpattern in
1061 the current subject string, rather than anything matching the subpattern
1062 itself (see
1063 <a href="#subpatternsassubroutines">"Subpatterns as subroutines"</a>
1064 below for a way of doing that). So the pattern
1065 <pre>
1066 (sens|respons)e and \1ibility
1067 </pre>
1068 matches "sense and sensibility" and "response and responsibility", but not
1069 "sense and responsibility". If caseful matching is in force at the time of the
1070 back reference, the case of letters is relevant. For example,
1071 <pre>
1072 ((?i)rah)\s+\1
1073 </pre>
1074 matches "rah rah" and "RAH RAH", but not "RAH rah", even though the original
1075 capturing subpattern is matched caselessly.
1076 </P>
1077 <P>
1078 Back references to named subpatterns use the Python syntax (?P=name). We could
1079 rewrite the above example as follows:
1080 <pre>
1081 (?&#60;p1&#62;(?i)rah)\s+(?P=p1)
1082 </pre>
1083 There may be more than one back reference to the same subpattern. If a
1084 subpattern has not actually been used in a particular match, any back
1085 references to it always fail. For example, the pattern
1086 <pre>
1087 (a|(bc))\2
1088 </pre>
1089 always fails if it starts to match "a" rather than "bc". Because there may be
1090 many capturing parentheses in a pattern, all digits following the backslash are
1091 taken as part of a potential back reference number. If the pattern continues
1092 with a digit character, some delimiter must be used to terminate the back
1093 reference. If the PCRE_EXTENDED option is set, this can be whitespace.
1094 Otherwise an empty comment (see
1095 <a href="#comments">"Comments"</a>
1096 below) can be used.
1097 </P>
1098 <P>
1099 A back reference that occurs inside the parentheses to which it refers fails
1100 when the subpattern is first used, so, for example, (a\1) never matches.
1101 However, such references can be useful inside repeated subpatterns. For
1102 example, the pattern
1103 <pre>
1104 (a|b\1)+
1105 </pre>
1106 matches any number of "a"s and also "aba", "ababbaa" etc. At each iteration of
1107 the subpattern, the back reference matches the character string corresponding
1108 to the previous iteration. In order for this to work, the pattern must be such
1109 that the first iteration does not need to match the back reference. This can be
1110 done using alternation, as in the example above, or by a quantifier with a
1111 minimum of zero.
1112 <a name="bigassertions"></a></P>
1113 <br><a name="SEC15" href="#TOC1">ASSERTIONS</a><br>
1114 <P>
1115 An assertion is a test on the characters following or preceding the current
1116 matching point that does not actually consume any characters. The simple
1117 assertions coded as \b, \B, \A, \G, \Z, \z, ^ and $ are described
1118 <a href="#smallassertions">above.</a>
1119 </P>
1120 <P>
1121 More complicated assertions are coded as subpatterns. There are two kinds:
1122 those that look ahead of the current position in the subject string, and those
1123 that look behind it. An assertion subpattern is matched in the normal way,
1124 except that it does not cause the current matching position to be changed.
1125 </P>
1126 <P>
1127 Assertion subpatterns are not capturing subpatterns, and may not be repeated,
1128 because it makes no sense to assert the same thing several times. If any kind
1129 of assertion contains capturing subpatterns within it, these are counted for
1130 the purposes of numbering the capturing subpatterns in the whole pattern.
1131 However, substring capturing is carried out only for positive assertions,
1132 because it does not make sense for negative assertions.
1133 </P>
1134 <br><b>
1135 Lookahead assertions
1136 </b><br>
1137 <P>
1138 Lookahead assertions start
1139 with (?= for positive assertions and (?! for negative assertions. For example,
1140 <pre>
1141 \w+(?=;)
1142 </pre>
1143 matches a word followed by a semicolon, but does not include the semicolon in
1144 the match, and
1145 <pre>
1146 foo(?!bar)
1147 </pre>
1148 matches any occurrence of "foo" that is not followed by "bar". Note that the
1149 apparently similar pattern
1150 <pre>
1151 (?!foo)bar
1152 </pre>
1153 does not find an occurrence of "bar" that is preceded by something other than
1154 "foo"; it finds any occurrence of "bar" whatsoever, because the assertion
1155 (?!foo) is always true when the next three characters are "bar". A
1156 lookbehind assertion is needed to achieve the other effect.
1157 </P>
1158 <P>
1159 If you want to force a matching failure at some point in a pattern, the most
1160 convenient way to do it is with (?!) because an empty string always matches, so
1161 an assertion that requires there not to be an empty string must always fail.
1162 <a name="lookbehind"></a></P>
1163 <br><b>
1164 Lookbehind assertions
1165 </b><br>
1166 <P>
1167 Lookbehind assertions start with (?&#60;= for positive assertions and (?&#60;! for
1168 negative assertions. For example,
1169 <pre>
1170 (?&#60;!foo)bar
1171 </pre>
1172 does find an occurrence of "bar" that is not preceded by "foo". The contents of
1173 a lookbehind assertion are restricted such that all the strings it matches must
1174 have a fixed length. However, if there are several alternatives, they do not
1175 all have to have the same fixed length. Thus
1176 <pre>
1177 (?&#60;=bullock|donkey)
1178 </pre>
1179 is permitted, but
1180 <pre>
1181 (?&#60;!dogs?|cats?)
1182 </pre>
1183 causes an error at compile time. Branches that match different length strings
1184 are permitted only at the top level of a lookbehind assertion. This is an
1185 extension compared with Perl (at least for 5.8), which requires all branches to
1186 match the same length of string. An assertion such as
1187 <pre>
1188 (?&#60;=ab(c|de))
1189 </pre>
1190 is not permitted, because its single top-level branch can match two different
1191 lengths, but it is acceptable if rewritten to use two top-level branches:
1192 <pre>
1193 (?&#60;=abc|abde)
1194 </pre>
1195 The implementation of lookbehind assertions is, for each alternative, to
1196 temporarily move the current position back by the fixed width and then try to
1197 match. If there are insufficient characters before the current position, the
1198 match is deemed to fail.
1199 </P>
1200 <P>
1201 PCRE does not allow the \C escape (which matches a single byte in UTF-8 mode)
1202 to appear in lookbehind assertions, because it makes it impossible to calculate
1203 the length of the lookbehind. The \X escape, which can match different numbers
1204 of bytes, is also not permitted.
1205 </P>
1206 <P>
1207 Atomic groups can be used in conjunction with lookbehind assertions to specify
1208 efficient matching at the end of the subject string. Consider a simple pattern
1209 such as
1210 <pre>
1211 abcd$
1212 </pre>
1213 when applied to a long string that does not match. Because matching proceeds
1214 from left to right, PCRE will look for each "a" in the subject and then see if
1215 what follows matches the rest of the pattern. If the pattern is specified as
1216 <pre>
1217 ^.*abcd$
1218 </pre>
1219 the initial .* matches the entire string at first, but when this fails (because
1220 there is no following "a"), it backtracks to match all but the last character,
1221 then all but the last two characters, and so on. Once again the search for "a"
1222 covers the entire string, from right to left, so we are no better off. However,
1223 if the pattern is written as
1224 <pre>
1225 ^(?&#62;.*)(?&#60;=abcd)
1226 </pre>
1227 or, equivalently, using the possessive quantifier syntax,
1228 <pre>
1229 ^.*+(?&#60;=abcd)
1230 </pre>
1231 there can be no backtracking for the .* item; it can match only the entire
1232 string. The subsequent lookbehind assertion does a single test on the last four
1233 characters. If it fails, the match fails immediately. For long strings, this
1234 approach makes a significant difference to the processing time.
1235 </P>
1236 <br><b>
1237 Using multiple assertions
1238 </b><br>
1239 <P>
1240 Several assertions (of any sort) may occur in succession. For example,
1241 <pre>
1242 (?&#60;=\d{3})(?&#60;!999)foo
1243 </pre>
1244 matches "foo" preceded by three digits that are not "999". Notice that each of
1245 the assertions is applied independently at the same point in the subject
1246 string. First there is a check that the previous three characters are all
1247 digits, and then there is a check that the same three characters are not "999".
1248 This pattern does <i>not</i> match "foo" preceded by six characters, the first
1249 of which are digits and the last three of which are not "999". For example, it
1250 doesn't match "123abcfoo". A pattern to do that is
1251 <pre>
1252 (?&#60;=\d{3}...)(?&#60;!999)foo
1253 </pre>
1254 This time the first assertion looks at the preceding six characters, checking
1255 that the first three are digits, and then the second assertion checks that the
1256 preceding three characters are not "999".
1257 </P>
1258 <P>
1259 Assertions can be nested in any combination. For example,
1260 <pre>
1261 (?&#60;=(?&#60;!foo)bar)baz
1262 </pre>
1263 matches an occurrence of "baz" that is preceded by "bar" which in turn is not
1264 preceded by "foo", while
1265 <pre>
1266 (?&#60;=\d{3}(?!999)...)foo
1267 </pre>
1268 is another pattern that matches "foo" preceded by three digits and any three
1269 characters that are not "999".
1270 </P>
1271 <br><a name="SEC16" href="#TOC1">CONDITIONAL SUBPATTERNS</a><br>
1272 <P>
1273 It is possible to cause the matching process to obey a subpattern
1274 conditionally or to choose between two alternative subpatterns, depending on
1275 the result of an assertion, or whether a previous capturing subpattern matched
1276 or not. The two possible forms of conditional subpattern are
1277 <pre>
1278 (?(condition)yes-pattern)
1279 (?(condition)yes-pattern|no-pattern)
1280 </pre>
1281 If the condition is satisfied, the yes-pattern is used; otherwise the
1282 no-pattern (if present) is used. If there are more than two alternatives in the
1283 subpattern, a compile-time error occurs.
1284 </P>
1285 <P>
1286 There are three kinds of condition. If the text between the parentheses
1287 consists of a sequence of digits, the condition is satisfied if the capturing
1288 subpattern of that number has previously matched. The number must be greater
1289 than zero. Consider the following pattern, which contains non-significant white
1290 space to make it more readable (assume the PCRE_EXTENDED option) and to divide
1291 it into three parts for ease of discussion:
1292 <pre>
1293 ( \( )? [^()]+ (?(1) \) )
1294 </pre>
1295 The first part matches an optional opening parenthesis, and if that
1296 character is present, sets it as the first captured substring. The second part
1297 matches one or more characters that are not parentheses. The third part is a
1298 conditional subpattern that tests whether the first set of parentheses matched
1299 or not. If they did, that is, if subject started with an opening parenthesis,
1300 the condition is true, and so the yes-pattern is executed and a closing
1301 parenthesis is required. Otherwise, since no-pattern is not present, the
1302 subpattern matches nothing. In other words, this pattern matches a sequence of
1303 non-parentheses, optionally enclosed in parentheses.
1304 </P>
1305 <P>
1306 If the condition is the string (R), it is satisfied if a recursive call to the
1307 pattern or subpattern has been made. At "top level", the condition is false.
1308 This is a PCRE extension. Recursive patterns are described in the next section.
1309 </P>
1310 <P>
1311 If the condition is not a sequence of digits or (R), it must be an assertion.
1312 This may be a positive or negative lookahead or lookbehind assertion. Consider
1313 this pattern, again containing non-significant white space, and with the two
1314 alternatives on the second line:
1315 <pre>
1316 (?(?=[^a-z]*[a-z])
1317 \d{2}-[a-z]{3}-\d{2} | \d{2}-\d{2}-\d{2} )
1318 </pre>
1319 The condition is a positive lookahead assertion that matches an optional
1320 sequence of non-letters followed by a letter. In other words, it tests for the
1321 presence of at least one letter in the subject. If a letter is found, the
1322 subject is matched against the first alternative; otherwise it is matched
1323 against the second. This pattern matches strings in one of the two forms
1324 dd-aaa-dd or dd-dd-dd, where aaa are letters and dd are digits.
1325 <a name="comments"></a></P>
1326 <br><a name="SEC17" href="#TOC1">COMMENTS</a><br>
1327 <P>
1328 The sequence (?# marks the start of a comment that continues up to the next
1329 closing parenthesis. Nested parentheses are not permitted. The characters
1330 that make up a comment play no part in the pattern matching at all.
1331 </P>
1332 <P>
1333 If the PCRE_EXTENDED option is set, an unescaped # character outside a
1334 character class introduces a comment that continues up to the next newline
1335 character in the pattern.
1336 </P>
1337 <br><a name="SEC18" href="#TOC1">RECURSIVE PATTERNS</a><br>
1338 <P>
1339 Consider the problem of matching a string in parentheses, allowing for
1340 unlimited nested parentheses. Without the use of recursion, the best that can
1341 be done is to use a pattern that matches up to some fixed depth of nesting. It
1342 is not possible to handle an arbitrary nesting depth. Perl provides a facility
1343 that allows regular expressions to recurse (amongst other things). It does this
1344 by interpolating Perl code in the expression at run time, and the code can
1345 refer to the expression itself. A Perl pattern to solve the parentheses problem
1346 can be created like this:
1347 <pre>
1348 $re = qr{\( (?: (?&#62;[^()]+) | (?p{$re}) )* \)}x;
1349 </pre>
1350 The (?p{...}) item interpolates Perl code at run time, and in this case refers
1351 recursively to the pattern in which it appears. Obviously, PCRE cannot support
1352 the interpolation of Perl code. Instead, it supports some special syntax for
1353 recursion of the entire pattern, and also for individual subpattern recursion.
1354 </P>
1355 <P>
1356 The special item that consists of (? followed by a number greater than zero and
1357 a closing parenthesis is a recursive call of the subpattern of the given
1358 number, provided that it occurs inside that subpattern. (If not, it is a
1359 "subroutine" call, which is described in the next section.) The special item
1360 (?R) is a recursive call of the entire regular expression.
1361 </P>
1362 <P>
1363 For example, this PCRE pattern solves the nested parentheses problem (assume
1364 the PCRE_EXTENDED option is set so that white space is ignored):
1365 <pre>
1366 \( ( (?&#62;[^()]+) | (?R) )* \)
1367 </pre>
1368 First it matches an opening parenthesis. Then it matches any number of
1369 substrings which can either be a sequence of non-parentheses, or a recursive
1370 match of the pattern itself (that is a correctly parenthesized substring).
1371 Finally there is a closing parenthesis.
1372 </P>
1373 <P>
1374 If this were part of a larger pattern, you would not want to recurse the entire
1375 pattern, so instead you could use this:
1376 <pre>
1377 ( \( ( (?&#62;[^()]+) | (?1) )* \) )
1378 </pre>
1379 We have put the pattern into parentheses, and caused the recursion to refer to
1380 them instead of the whole pattern. In a larger pattern, keeping track of
1381 parenthesis numbers can be tricky. It may be more convenient to use named
1382 parentheses instead. For this, PCRE uses (?P&#62;name), which is an extension to
1383 the Python syntax that PCRE uses for named parentheses (Perl does not provide
1384 named parentheses). We could rewrite the above example as follows:
1385 <pre>
1386 (?P&#60;pn&#62; \( ( (?&#62;[^()]+) | (?P&#62;pn) )* \) )
1387 </pre>
1388 This particular example pattern contains nested unlimited repeats, and so the
1389 use of atomic grouping for matching strings of non-parentheses is important
1390 when applying the pattern to strings that do not match. For example, when this
1391 pattern is applied to
1392 <pre>
1393 (aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa()
1394 </pre>
1395 it yields "no match" quickly. However, if atomic grouping is not used,
1396 the match runs for a very long time indeed because there are so many different
1397 ways the + and * repeats can carve up the subject, and all have to be tested
1398 before failure can be reported.
1399 </P>
1400 <P>
1401 At the end of a match, the values set for any capturing subpatterns are those
1402 from the outermost level of the recursion at which the subpattern value is set.
1403 If you want to obtain intermediate values, a callout function can be used (see
1404 the next section and the
1405 <a href="pcrecallout.html"><b>pcrecallout</b></a>
1406 documentation). If the pattern above is matched against
1407 <pre>
1408 (ab(cd)ef)
1409 </pre>
1410 the value for the capturing parentheses is "ef", which is the last value taken
1411 on at the top level. If additional parentheses are added, giving
1412 <pre>
1413 \( ( ( (?&#62;[^()]+) | (?R) )* ) \)
1414 ^ ^
1415 ^ ^
1416 </pre>
1417 the string they capture is "ab(cd)ef", the contents of the top level
1418 parentheses. If there are more than 15 capturing parentheses in a pattern, PCRE
1419 has to obtain extra memory to store data during a recursion, which it does by
1420 using <b>pcre_malloc</b>, freeing it via <b>pcre_free</b> afterwards. If no
1421 memory can be obtained, the match fails with the PCRE_ERROR_NOMEMORY error.
1422 </P>
1423 <P>
1424 Do not confuse the (?R) item with the condition (R), which tests for recursion.
1425 Consider this pattern, which matches text in angle brackets, allowing for
1426 arbitrary nesting. Only digits are allowed in nested brackets (that is, when
1427 recursing), whereas any characters are permitted at the outer level.
1428 <pre>
1429 &#60; (?: (?(R) \d++ | [^&#60;&#62;]*+) | (?R)) * &#62;
1430 </pre>
1431 In this pattern, (?(R) is the start of a conditional subpattern, with two
1432 different alternatives for the recursive and non-recursive cases. The (?R) item
1433 is the actual recursive call.
1434 <a name="subpatternsassubroutines"></a></P>
1435 <br><a name="SEC19" href="#TOC1">SUBPATTERNS AS SUBROUTINES</a><br>
1436 <P>
1437 If the syntax for a recursive subpattern reference (either by number or by
1438 name) is used outside the parentheses to which it refers, it operates like a
1439 subroutine in a programming language. An earlier example pointed out that the
1440 pattern
1441 <pre>
1442 (sens|respons)e and \1ibility
1443 </pre>
1444 matches "sense and sensibility" and "response and responsibility", but not
1445 "sense and responsibility". If instead the pattern
1446 <pre>
1447 (sens|respons)e and (?1)ibility
1448 </pre>
1449 is used, it does match "sense and responsibility" as well as the other two
1450 strings. Such references must, however, follow the subpattern to which they
1451 refer.
1452 </P>
1453 <br><a name="SEC20" href="#TOC1">CALLOUTS</a><br>
1454 <P>
1455 Perl has a feature whereby using the sequence (?{...}) causes arbitrary Perl
1456 code to be obeyed in the middle of matching a regular expression. This makes it
1457 possible, amongst other things, to extract different substrings that match the
1458 same pair of parentheses when there is a repetition.
1459 </P>
1460 <P>
1461 PCRE provides a similar feature, but of course it cannot obey arbitrary Perl
1462 code. The feature is called "callout". The caller of PCRE provides an external
1463 function by putting its entry point in the global variable <i>pcre_callout</i>.
1464 By default, this variable contains NULL, which disables all calling out.
1465 </P>
1466 <P>
1467 Within a regular expression, (?C) indicates the points at which the external
1468 function is to be called. If you want to identify different callout points, you
1469 can put a number less than 256 after the letter C. The default value is zero.
1470 For example, this pattern has two callout points:
1471 <pre>
1472 (?C1)\dabc(?C2)def
1473 </pre>
1474 If the PCRE_AUTO_CALLOUT flag is passed to <b>pcre_compile()</b>, callouts are
1475 automatically installed before each item in the pattern. They are all numbered
1476 255.
1477 </P>
1478 <P>
1479 During matching, when PCRE reaches a callout point (and <i>pcre_callout</i> is
1480 set), the external function is called. It is provided with the number of the
1481 callout, the position in the pattern, and, optionally, one item of data
1482 originally supplied by the caller of <b>pcre_exec()</b>. The callout function
1483 may cause matching to proceed, to backtrack, or to fail altogether. A complete
1484 description of the interface to the callout function is given in the
1485 <a href="pcrecallout.html"><b>pcrecallout</b></a>
1486 documentation.
1487 </P>
1488 <P>
1489 Last updated: 28 February 2005
1490 <br>
1491 Copyright &copy; 1997-2005 University of Cambridge.
1492 <p>
1493 Return to the <a href="index.html">PCRE index page</a>.
1494 </p>

  ViewVC Help
Powered by ViewVC 1.1.5