ViewVC logotype

Contents of /code/tags/pcre-6.3/doc/pcrepattern.3

Parent Directory Parent Directory | Revision Log Revision Log

Revision 84 - (show annotations)
Sat Feb 24 21:41:08 2007 UTC (14 years, 3 months ago) by nigel
File size: 63013 byte(s)
Error occurred while calculating annotation data.
Tag code/trunk as code/tags/pcre-6.3.
3 PCRE - Perl-compatible regular expressions
5 .rs
6 .sp
7 The syntax and semantics of the regular expressions supported by PCRE are
8 described below. Regular expressions are also described in the Perl
9 documentation and in a number of books, some of which have copious examples.
10 Jeffrey Friedl's "Mastering Regular Expressions", published by O'Reilly, covers
11 regular expressions in great detail. This description of PCRE's regular
12 expressions is intended as reference material.
13 .P
14 The original operation of PCRE was on strings of one-byte characters. However,
15 there is now also support for UTF-8 character strings. To use this, you must
16 build PCRE to include UTF-8 support, and then call \fBpcre_compile()\fP with
17 the PCRE_UTF8 option. How this affects pattern matching is mentioned in several
18 places below. There is also a summary of UTF-8 features in the
19 .\" HTML <a href="pcre.html#utf8support">
20 .\" </a>
21 section on UTF-8 support
22 .\"
23 in the main
24 .\" HREF
25 \fBpcre\fP
26 .\"
27 page.
28 .P
29 The remainder of this document discusses the patterns that are supported by
30 PCRE when its main matching function, \fBpcre_exec()\fP, is used.
31 From release 6.0, PCRE offers a second matching function,
32 \fBpcre_dfa_exec()\fP, which matches using a different algorithm that is not
33 Perl-compatible. The advantages and disadvantages of the alternative function,
34 and how it differs from the normal function, are discussed in the
35 .\" HREF
36 \fBpcrematching\fP
37 .\"
38 page.
39 .P
40 A regular expression is a pattern that is matched against a subject string from
41 left to right. Most characters stand for themselves in a pattern, and match the
42 corresponding characters in the subject. As a trivial example, the pattern
43 .sp
44 The quick brown fox
45 .sp
46 matches a portion of a subject string that is identical to itself. When
47 caseless matching is specified (the PCRE_CASELESS option), letters are matched
48 independently of case. In UTF-8 mode, PCRE always understands the concept of
49 case for characters whose values are less than 128, so caseless matching is
50 always possible. For characters with higher values, the concept of case is
51 supported if PCRE is compiled with Unicode property support, but not otherwise.
52 If you want to use caseless matching for characters 128 and above, you must
53 ensure that PCRE is compiled with Unicode property support as well as with
54 UTF-8 support.
55 .P
56 The power of regular expressions comes from the ability to include alternatives
57 and repetitions in the pattern. These are encoded in the pattern by the use of
58 \fImetacharacters\fP, which do not stand for themselves but instead are
59 interpreted in some special way.
60 .P
61 There are two different sets of metacharacters: those that are recognized
62 anywhere in the pattern except within square brackets, and those that are
63 recognized in square brackets. Outside square brackets, the metacharacters are
64 as follows:
65 .sp
66 \e general escape character with several uses
67 ^ assert start of string (or line, in multiline mode)
68 $ assert end of string (or line, in multiline mode)
69 . match any character except newline (by default)
70 [ start character class definition
71 | start of alternative branch
72 ( start subpattern
73 ) end subpattern
74 ? extends the meaning of (
75 also 0 or 1 quantifier
76 also quantifier minimizer
77 * 0 or more quantifier
78 + 1 or more quantifier
79 also "possessive quantifier"
80 { start min/max quantifier
81 .sp
82 Part of a pattern that is in square brackets is called a "character class". In
83 a character class the only metacharacters are:
84 .sp
85 \e general escape character
86 ^ negate the class, but only if the first character
87 - indicates character range
88 .\" JOIN
89 [ POSIX character class (only if followed by POSIX
90 syntax)
91 ] terminates the character class
92 .sp
93 The following sections describe the use of each of the metacharacters.
94 .
96 .rs
97 .sp
98 The backslash character has several uses. Firstly, if it is followed by a
99 non-alphanumeric character, it takes away any special meaning that character may
100 have. This use of backslash as an escape character applies both inside and
101 outside character classes.
102 .P
103 For example, if you want to match a * character, you write \e* in the pattern.
104 This escaping action applies whether or not the following character would
105 otherwise be interpreted as a metacharacter, so it is always safe to precede a
106 non-alphanumeric with backslash to specify that it stands for itself. In
107 particular, if you want to match a backslash, you write \e\e.
108 .P
109 If a pattern is compiled with the PCRE_EXTENDED option, whitespace in the
110 pattern (other than in a character class) and characters between a # outside
111 a character class and the next newline character are ignored. An escaping
112 backslash can be used to include a whitespace or # character as part of the
113 pattern.
114 .P
115 If you want to remove the special meaning from a sequence of characters, you
116 can do so by putting them between \eQ and \eE. This is different from Perl in
117 that $ and @ are handled as literals in \eQ...\eE sequences in PCRE, whereas in
118 Perl, $ and @ cause variable interpolation. Note the following examples:
119 .sp
120 Pattern PCRE matches Perl matches
121 .sp
122 .\" JOIN
123 \eQabc$xyz\eE abc$xyz abc followed by the
124 contents of $xyz
125 \eQabc\e$xyz\eE abc\e$xyz abc\e$xyz
126 \eQabc\eE\e$\eQxyz\eE abc$xyz abc$xyz
127 .sp
128 The \eQ...\eE sequence is recognized both inside and outside character classes.
129 .
130 .
131 .\" HTML <a name="digitsafterbackslash"></a>
132 .SS "Non-printing characters"
133 .rs
134 .sp
135 A second use of backslash provides a way of encoding non-printing characters
136 in patterns in a visible manner. There is no restriction on the appearance of
137 non-printing characters, apart from the binary zero that terminates a pattern,
138 but when a pattern is being prepared by text editing, it is usually easier to
139 use one of the following escape sequences than the binary character it
140 represents:
141 .sp
142 \ea alarm, that is, the BEL character (hex 07)
143 \ecx "control-x", where x is any character
144 \ee escape (hex 1B)
145 \ef formfeed (hex 0C)
146 \en newline (hex 0A)
147 \er carriage return (hex 0D)
148 \et tab (hex 09)
149 \eddd character with octal code ddd, or backreference
150 \exhh character with hex code hh
151 \ex{hhh..} character with hex code hhh... (UTF-8 mode only)
152 .sp
153 The precise effect of \ecx is as follows: if x is a lower case letter, it
154 is converted to upper case. Then bit 6 of the character (hex 40) is inverted.
155 Thus \ecz becomes hex 1A, but \ec{ becomes hex 3B, while \ec; becomes hex
156 7B.
157 .P
158 After \ex, from zero to two hexadecimal digits are read (letters can be in
159 upper or lower case). In UTF-8 mode, any number of hexadecimal digits may
160 appear between \ex{ and }, but the value of the character code must be less
161 than 2**31 (that is, the maximum hexadecimal value is 7FFFFFFF). If characters
162 other than hexadecimal digits appear between \ex{ and }, or if there is no
163 terminating }, this form of escape is not recognized. Instead, the initial
164 \ex will be interpreted as a basic hexadecimal escape, with no following
165 digits, giving a character whose value is zero.
166 .P
167 Characters whose value is less than 256 can be defined by either of the two
168 syntaxes for \ex when PCRE is in UTF-8 mode. There is no difference in the
169 way they are handled. For example, \exdc is exactly the same as \ex{dc}.
170 .P
171 After \e0 up to two further octal digits are read. In both cases, if there
172 are fewer than two digits, just those that are present are used. Thus the
173 sequence \e0\ex\e07 specifies two binary zeros followed by a BEL character
174 (code value 7). Make sure you supply two digits after the initial zero if the
175 pattern character that follows is itself an octal digit.
176 .P
177 The handling of a backslash followed by a digit other than 0 is complicated.
178 Outside a character class, PCRE reads it and any following digits as a decimal
179 number. If the number is less than 10, or if there have been at least that many
180 previous capturing left parentheses in the expression, the entire sequence is
181 taken as a \fIback reference\fP. A description of how this works is given
182 .\" HTML <a href="#backreferences">
183 .\" </a>
184 later,
185 .\"
186 following the discussion of
187 .\" HTML <a href="#subpattern">
188 .\" </a>
189 parenthesized subpatterns.
190 .\"
191 .P
192 Inside a character class, or if the decimal number is greater than 9 and there
193 have not been that many capturing subpatterns, PCRE re-reads up to three octal
194 digits following the backslash, and generates a single byte from the least
195 significant 8 bits of the value. Any subsequent digits stand for themselves.
196 For example:
197 .sp
198 \e040 is another way of writing a space
199 .\" JOIN
200 \e40 is the same, provided there are fewer than 40
201 previous capturing subpatterns
202 \e7 is always a back reference
203 .\" JOIN
204 \e11 might be a back reference, or another way of
205 writing a tab
206 \e011 is always a tab
207 \e0113 is a tab followed by the character "3"
208 .\" JOIN
209 \e113 might be a back reference, otherwise the
210 character with octal code 113
211 .\" JOIN
212 \e377 might be a back reference, otherwise
213 the byte consisting entirely of 1 bits
214 .\" JOIN
215 \e81 is either a back reference, or a binary zero
216 followed by the two characters "8" and "1"
217 .sp
218 Note that octal values of 100 or greater must not be introduced by a leading
219 zero, because no more than three octal digits are ever read.
220 .P
221 All the sequences that define a single byte value or a single UTF-8 character
222 (in UTF-8 mode) can be used both inside and outside character classes. In
223 addition, inside a character class, the sequence \eb is interpreted as the
224 backspace character (hex 08), and the sequence \eX is interpreted as the
225 character "X". Outside a character class, these sequences have different
226 meanings
227 .\" HTML <a href="#uniextseq">
228 .\" </a>
229 (see below).
230 .\"
231 .
232 .
233 .SS "Generic character types"
234 .rs
235 .sp
236 The third use of backslash is for specifying generic character types. The
237 following are always recognized:
238 .sp
239 \ed any decimal digit
240 \eD any character that is not a decimal digit
241 \es any whitespace character
242 \eS any character that is not a whitespace character
243 \ew any "word" character
244 \eW any "non-word" character
245 .sp
246 Each pair of escape sequences partitions the complete set of characters into
247 two disjoint sets. Any given character matches one, and only one, of each pair.
248 .P
249 These character type sequences can appear both inside and outside character
250 classes. They each match one character of the appropriate type. If the current
251 matching point is at the end of the subject string, all of them fail, since
252 there is no character to match.
253 .P
254 For compatibility with Perl, \es does not match the VT character (code 11).
255 This makes it different from the the POSIX "space" class. The \es characters
256 are HT (9), LF (10), FF (12), CR (13), and space (32).
257 .P
258 A "word" character is an underscore or any character less than 256 that is a
259 letter or digit. The definition of letters and digits is controlled by PCRE's
260 low-valued character tables, and may vary if locale-specific matching is taking
261 place (see
262 .\" HTML <a href="pcreapi.html#localesupport">
263 .\" </a>
264 "Locale support"
265 .\"
266 in the
267 .\" HREF
268 \fBpcreapi\fP
269 .\"
270 page). For example, in the "fr_FR" (French) locale, some character codes
271 greater than 128 are used for accented letters, and these are matched by \ew.
272 .P
273 In UTF-8 mode, characters with values greater than 128 never match \ed, \es, or
274 \ew, and always match \eD, \eS, and \eW. This is true even when Unicode
275 character property support is available.
276 .
277 .
278 .\" HTML <a name="uniextseq"></a>
279 .SS Unicode character properties
280 .rs
281 .sp
282 When PCRE is built with Unicode character property support, three additional
283 escape sequences to match generic character types are available when UTF-8 mode
284 is selected. They are:
285 .sp
286 \ep{\fIxx\fP} a character with the \fIxx\fP property
287 \eP{\fIxx\fP} a character without the \fIxx\fP property
288 \eX an extended Unicode sequence
289 .sp
290 The property names represented by \fIxx\fP above are limited to the
291 Unicode general category properties. Each character has exactly one such
292 property, specified by a two-letter abbreviation. For compatibility with Perl,
293 negation can be specified by including a circumflex between the opening brace
294 and the property name. For example, \ep{^Lu} is the same as \eP{Lu}.
295 .P
296 If only one letter is specified with \ep or \eP, it includes all the properties
297 that start with that letter. In this case, in the absence of negation, the
298 curly brackets in the escape sequence are optional; these two examples have
299 the same effect:
300 .sp
301 \ep{L}
302 \epL
303 .sp
304 The following property codes are supported:
305 .sp
306 C Other
307 Cc Control
308 Cf Format
309 Cn Unassigned
310 Co Private use
311 Cs Surrogate
312 .sp
313 L Letter
314 Ll Lower case letter
315 Lm Modifier letter
316 Lo Other letter
317 Lt Title case letter
318 Lu Upper case letter
319 .sp
320 M Mark
321 Mc Spacing mark
322 Me Enclosing mark
323 Mn Non-spacing mark
324 .sp
325 N Number
326 Nd Decimal number
327 Nl Letter number
328 No Other number
329 .sp
330 P Punctuation
331 Pc Connector punctuation
332 Pd Dash punctuation
333 Pe Close punctuation
334 Pf Final punctuation
335 Pi Initial punctuation
336 Po Other punctuation
337 Ps Open punctuation
338 .sp
339 S Symbol
340 Sc Currency symbol
341 Sk Modifier symbol
342 Sm Mathematical symbol
343 So Other symbol
344 .sp
345 Z Separator
346 Zl Line separator
347 Zp Paragraph separator
348 Zs Space separator
349 .sp
350 Extended properties such as "Greek" or "InMusicalSymbols" are not supported by
351 PCRE.
352 .P
353 Specifying caseless matching does not affect these escape sequences. For
354 example, \ep{Lu} always matches only upper case letters.
355 .P
356 The \eX escape matches any number of Unicode characters that form an extended
357 Unicode sequence. \eX is equivalent to
358 .sp
359 (?>\ePM\epM*)
360 .sp
361 That is, it matches a character without the "mark" property, followed by zero
362 or more characters with the "mark" property, and treats the sequence as an
363 atomic group
364 .\" HTML <a href="#atomicgroup">
365 .\" </a>
366 (see below).
367 .\"
368 Characters with the "mark" property are typically accents that affect the
369 preceding character.
370 .P
371 Matching characters by Unicode property is not fast, because PCRE has to search
372 a structure that contains data for over fifteen thousand characters. That is
373 why the traditional escape sequences such as \ed and \ew do not use Unicode
374 properties in PCRE.
375 .
376 .
377 .\" HTML <a name="smallassertions"></a>
378 .SS "Simple assertions"
379 .rs
380 .sp
381 The fourth use of backslash is for certain simple assertions. An assertion
382 specifies a condition that has to be met at a particular point in a match,
383 without consuming any characters from the subject string. The use of
384 subpatterns for more complicated assertions is described
385 .\" HTML <a href="#bigassertions">
386 .\" </a>
387 below.
388 .\"
389 The backslashed
390 assertions are:
391 .sp
392 \eb matches at a word boundary
393 \eB matches when not at a word boundary
394 \eA matches at start of subject
395 \eZ matches at end of subject or before newline at end
396 \ez matches at end of subject
397 \eG matches at first matching position in subject
398 .sp
399 These assertions may not appear in character classes (but note that \eb has a
400 different meaning, namely the backspace character, inside a character class).
401 .P
402 A word boundary is a position in the subject string where the current character
403 and the previous character do not both match \ew or \eW (i.e. one matches
404 \ew and the other matches \eW), or the start or end of the string if the
405 first or last character matches \ew, respectively.
406 .P
407 The \eA, \eZ, and \ez assertions differ from the traditional circumflex and
408 dollar (described in the next section) in that they only ever match at the very
409 start and end of the subject string, whatever options are set. Thus, they are
410 independent of multiline mode. These three assertions are not affected by the
411 PCRE_NOTBOL or PCRE_NOTEOL options, which affect only the behaviour of the
412 circumflex and dollar metacharacters. However, if the \fIstartoffset\fP
413 argument of \fBpcre_exec()\fP is non-zero, indicating that matching is to start
414 at a point other than the beginning of the subject, \eA can never match. The
415 difference between \eZ and \ez is that \eZ matches before a newline that is the
416 last character of the string as well as at the end of the string, whereas \ez
417 matches only at the end.
418 .P
419 The \eG assertion is true only when the current matching position is at the
420 start point of the match, as specified by the \fIstartoffset\fP argument of
421 \fBpcre_exec()\fP. It differs from \eA when the value of \fIstartoffset\fP is
422 non-zero. By calling \fBpcre_exec()\fP multiple times with appropriate
423 arguments, you can mimic Perl's /g option, and it is in this kind of
424 implementation where \eG can be useful.
425 .P
426 Note, however, that PCRE's interpretation of \eG, as the start of the current
427 match, is subtly different from Perl's, which defines it as the end of the
428 previous match. In Perl, these can be different when the previously matched
429 string was empty. Because PCRE does just one match at a time, it cannot
430 reproduce this behaviour.
431 .P
432 If all the alternatives of a pattern begin with \eG, the expression is anchored
433 to the starting match position, and the "anchored" flag is set in the compiled
434 regular expression.
435 .
436 .
438 .rs
439 .sp
440 Outside a character class, in the default matching mode, the circumflex
441 character is an assertion that is true only if the current matching point is
442 at the start of the subject string. If the \fIstartoffset\fP argument of
443 \fBpcre_exec()\fP is non-zero, circumflex can never match if the PCRE_MULTILINE
444 option is unset. Inside a character class, circumflex has an entirely different
445 meaning
446 .\" HTML <a href="#characterclass">
447 .\" </a>
448 (see below).
449 .\"
450 .P
451 Circumflex need not be the first character of the pattern if a number of
452 alternatives are involved, but it should be the first thing in each alternative
453 in which it appears if the pattern is ever to match that branch. If all
454 possible alternatives start with a circumflex, that is, if the pattern is
455 constrained to match only at the start of the subject, it is said to be an
456 "anchored" pattern. (There are also other constructs that can cause a pattern
457 to be anchored.)
458 .P
459 A dollar character is an assertion that is true only if the current matching
460 point is at the end of the subject string, or immediately before a newline
461 character that is the last character in the string (by default). Dollar need
462 not be the last character of the pattern if a number of alternatives are
463 involved, but it should be the last item in any branch in which it appears.
464 Dollar has no special meaning in a character class.
465 .P
466 The meaning of dollar can be changed so that it matches only at the very end of
467 the string, by setting the PCRE_DOLLAR_ENDONLY option at compile time. This
468 does not affect the \eZ assertion.
469 .P
470 The meanings of the circumflex and dollar characters are changed if the
471 PCRE_MULTILINE option is set. When this is the case, they match immediately
472 after and immediately before an internal newline character, respectively, in
473 addition to matching at the start and end of the subject string. For example,
474 the pattern /^abc$/ matches the subject string "def\enabc" (where \en
475 represents a newline character) in multiline mode, but not otherwise.
476 Consequently, patterns that are anchored in single line mode because all
477 branches start with ^ are not anchored in multiline mode, and a match for
478 circumflex is possible when the \fIstartoffset\fP argument of \fBpcre_exec()\fP
479 is non-zero. The PCRE_DOLLAR_ENDONLY option is ignored if PCRE_MULTILINE is
480 set.
481 .P
482 Note that the sequences \eA, \eZ, and \ez can be used to match the start and
483 end of the subject in both modes, and if all branches of a pattern start with
484 \eA it is always anchored, whether PCRE_MULTILINE is set or not.
485 .
486 .
488 .rs
489 .sp
490 Outside a character class, a dot in the pattern matches any one character in
491 the subject, including a non-printing character, but not (by default) newline.
492 In UTF-8 mode, a dot matches any UTF-8 character, which might be more than one
493 byte long, except (by default) newline. If the PCRE_DOTALL option is set,
494 dots match newlines as well. The handling of dot is entirely independent of the
495 handling of circumflex and dollar, the only relationship being that they both
496 involve newline characters. Dot has no special meaning in a character class.
497 .
498 .
500 .rs
501 .sp
502 Outside a character class, the escape sequence \eC matches any one byte, both
503 in and out of UTF-8 mode. Unlike a dot, it can match a newline. The feature is
504 provided in Perl in order to match individual bytes in UTF-8 mode. Because it
505 breaks up UTF-8 characters into individual bytes, what remains in the string
506 may be a malformed UTF-8 string. For this reason, the \eC escape sequence is
507 best avoided.
508 .P
509 PCRE does not allow \eC to appear in lookbehind assertions
510 .\" HTML <a href="#lookbehind">
511 .\" </a>
512 (described below),
513 .\"
514 because in UTF-8 mode this would make it impossible to calculate the length of
515 the lookbehind.
516 .
517 .
518 .\" HTML <a name="characterclass"></a>
520 .rs
521 .sp
522 An opening square bracket introduces a character class, terminated by a closing
523 square bracket. A closing square bracket on its own is not special. If a
524 closing square bracket is required as a member of the class, it should be the
525 first data character in the class (after an initial circumflex, if present) or
526 escaped with a backslash.
527 .P
528 A character class matches a single character in the subject. In UTF-8 mode, the
529 character may occupy more than one byte. A matched character must be in the set
530 of characters defined by the class, unless the first character in the class
531 definition is a circumflex, in which case the subject character must not be in
532 the set defined by the class. If a circumflex is actually required as a member
533 of the class, ensure it is not the first character, or escape it with a
534 backslash.
535 .P
536 For example, the character class [aeiou] matches any lower case vowel, while
537 [^aeiou] matches any character that is not a lower case vowel. Note that a
538 circumflex is just a convenient notation for specifying the characters that
539 are in the class by enumerating those that are not. A class that starts with a
540 circumflex is not an assertion: it still consumes a character from the subject
541 string, and therefore it fails if the current pointer is at the end of the
542 string.
543 .P
544 In UTF-8 mode, characters with values greater than 255 can be included in a
545 class as a literal string of bytes, or by using the \ex{ escaping mechanism.
546 .P
547 When caseless matching is set, any letters in a class represent both their
548 upper case and lower case versions, so for example, a caseless [aeiou] matches
549 "A" as well as "a", and a caseless [^aeiou] does not match "A", whereas a
550 caseful version would. In UTF-8 mode, PCRE always understands the concept of
551 case for characters whose values are less than 128, so caseless matching is
552 always possible. For characters with higher values, the concept of case is
553 supported if PCRE is compiled with Unicode property support, but not otherwise.
554 If you want to use caseless matching for characters 128 and above, you must
555 ensure that PCRE is compiled with Unicode property support as well as with
556 UTF-8 support.
557 .P
558 The newline character is never treated in any special way in character classes,
559 whatever the setting of the PCRE_DOTALL or PCRE_MULTILINE options is. A class
560 such as [^a] will always match a newline.
561 .P
562 The minus (hyphen) character can be used to specify a range of characters in a
563 character class. For example, [d-m] matches any letter between d and m,
564 inclusive. If a minus character is required in a class, it must be escaped with
565 a backslash or appear in a position where it cannot be interpreted as
566 indicating a range, typically as the first or last character in the class.
567 .P
568 It is not possible to have the literal character "]" as the end character of a
569 range. A pattern such as [W-]46] is interpreted as a class of two characters
570 ("W" and "-") followed by a literal string "46]", so it would match "W46]" or
571 "-46]". However, if the "]" is escaped with a backslash it is interpreted as
572 the end of range, so [W-\e]46] is interpreted as a class containing a range
573 followed by two other characters. The octal or hexadecimal representation of
574 "]" can also be used to end a range.
575 .P
576 Ranges operate in the collating sequence of character values. They can also be
577 used for characters specified numerically, for example [\e000-\e037]. In UTF-8
578 mode, ranges can include characters whose values are greater than 255, for
579 example [\ex{100}-\ex{2ff}].
580 .P
581 If a range that includes letters is used when caseless matching is set, it
582 matches the letters in either case. For example, [W-c] is equivalent to
583 [][\e\e^_`wxyzabc], matched caselessly, and in non-UTF-8 mode, if character
584 tables for the "fr_FR" locale are in use, [\exc8-\excb] matches accented E
585 characters in both cases. In UTF-8 mode, PCRE supports the concept of case for
586 characters with values greater than 128 only when it is compiled with Unicode
587 property support.
588 .P
589 The character types \ed, \eD, \ep, \eP, \es, \eS, \ew, and \eW may also appear
590 in a character class, and add the characters that they match to the class. For
591 example, [\edABCDEF] matches any hexadecimal digit. A circumflex can
592 conveniently be used with the upper case character types to specify a more
593 restricted set of characters than the matching lower case type. For example,
594 the class [^\eW_] matches any letter or digit, but not underscore.
595 .P
596 The only metacharacters that are recognized in character classes are backslash,
597 hyphen (only where it can be interpreted as specifying a range), circumflex
598 (only at the start), opening square bracket (only when it can be interpreted as
599 introducing a POSIX class name - see the next section), and the terminating
600 closing square bracket. However, escaping other non-alphanumeric characters
601 does no harm.
602 .
603 .
605 .rs
606 .sp
607 Perl supports the POSIX notation for character classes. This uses names
608 enclosed by [: and :] within the enclosing square brackets. PCRE also supports
609 this notation. For example,
610 .sp
611 [01[:alpha:]%]
612 .sp
613 matches "0", "1", any alphabetic character, or "%". The supported class names
614 are
615 .sp
616 alnum letters and digits
617 alpha letters
618 ascii character codes 0 - 127
619 blank space or tab only
620 cntrl control characters
621 digit decimal digits (same as \ed)
622 graph printing characters, excluding space
623 lower lower case letters
624 print printing characters, including space
625 punct printing characters, excluding letters and digits
626 space white space (not quite the same as \es)
627 upper upper case letters
628 word "word" characters (same as \ew)
629 xdigit hexadecimal digits
630 .sp
631 The "space" characters are HT (9), LF (10), VT (11), FF (12), CR (13), and
632 space (32). Notice that this list includes the VT character (code 11). This
633 makes "space" different to \es, which does not include VT (for Perl
634 compatibility).
635 .P
636 The name "word" is a Perl extension, and "blank" is a GNU extension from Perl
637 5.8. Another Perl extension is negation, which is indicated by a ^ character
638 after the colon. For example,
639 .sp
640 [12[:^digit:]]
641 .sp
642 matches "1", "2", or any non-digit. PCRE (and Perl) also recognize the POSIX
643 syntax [.ch.] and [=ch=] where "ch" is a "collating element", but these are not
644 supported, and an error is given if they are encountered.
645 .P
646 In UTF-8 mode, characters with values greater than 128 do not match any of
647 the POSIX character classes.
648 .
649 .
651 .rs
652 .sp
653 Vertical bar characters are used to separate alternative patterns. For example,
654 the pattern
655 .sp
656 gilbert|sullivan
657 .sp
658 matches either "gilbert" or "sullivan". Any number of alternatives may appear,
659 and an empty alternative is permitted (matching the empty string).
660 The matching process tries each alternative in turn, from left to right,
661 and the first one that succeeds is used. If the alternatives are within a
662 subpattern
663 .\" HTML <a href="#subpattern">
664 .\" </a>
665 (defined below),
666 .\"
667 "succeeds" means matching the rest of the main pattern as well as the
668 alternative in the subpattern.
669 .
670 .
672 .rs
673 .sp
675 PCRE_EXTENDED options can be changed from within the pattern by a sequence of
676 Perl option letters enclosed between "(?" and ")". The option letters are
677 .sp
680 s for PCRE_DOTALL
682 .sp
683 For example, (?im) sets caseless, multiline matching. It is also possible to
684 unset these options by preceding the letter with a hyphen, and a combined
685 setting and unsetting such as (?im-sx), which sets PCRE_CASELESS and
686 PCRE_MULTILINE while unsetting PCRE_DOTALL and PCRE_EXTENDED, is also
687 permitted. If a letter appears both before and after the hyphen, the option is
688 unset.
689 .P
690 When an option change occurs at top level (that is, not inside subpattern
691 parentheses), the change applies to the remainder of the pattern that follows.
692 If the change is placed right at the start of a pattern, PCRE extracts it into
693 the global options (and it will therefore show up in data extracted by the
694 \fBpcre_fullinfo()\fP function).
695 .P
696 An option change within a subpattern affects only that part of the current
697 pattern that follows it, so
698 .sp
699 (a(?i)b)c
700 .sp
701 matches abc and aBc and no other strings (assuming PCRE_CASELESS is not used).
702 By this means, options can be made to have different settings in different
703 parts of the pattern. Any changes made in one alternative do carry on
704 into subsequent branches within the same subpattern. For example,
705 .sp
706 (a(?i)b|c)
707 .sp
708 matches "ab", "aB", "c", and "C", even though when matching "C" the first
709 branch is abandoned before the option setting. This is because the effects of
710 option settings happen at compile time. There would be some very weird
711 behaviour otherwise.
712 .P
713 The PCRE-specific options PCRE_UNGREEDY and PCRE_EXTRA can be changed in the
714 same way as the Perl-compatible options by using the characters U and X
715 respectively. The (?X) flag setting is special in that it must always occur
716 earlier in the pattern than any of the additional features it turns on, even
717 when it is at top level. It is best to put it at the start.
718 .
719 .
720 .\" HTML <a name="subpattern"></a>
722 .rs
723 .sp
724 Subpatterns are delimited by parentheses (round brackets), which can be nested.
725 Turning part of a pattern into a subpattern does two things:
726 .sp
727 1. It localizes a set of alternatives. For example, the pattern
728 .sp
729 cat(aract|erpillar|)
730 .sp
731 matches one of the words "cat", "cataract", or "caterpillar". Without the
732 parentheses, it would match "cataract", "erpillar" or the empty string.
733 .sp
734 2. It sets up the subpattern as a capturing subpattern. This means that, when
735 the whole pattern matches, that portion of the subject string that matched the
736 subpattern is passed back to the caller via the \fIovector\fP argument of
737 \fBpcre_exec()\fP. Opening parentheses are counted from left to right (starting
738 from 1) to obtain numbers for the capturing subpatterns.
739 .P
740 For example, if the string "the red king" is matched against the pattern
741 .sp
742 the ((red|white) (king|queen))
743 .sp
744 the captured substrings are "red king", "red", and "king", and are numbered 1,
745 2, and 3, respectively.
746 .P
747 The fact that plain parentheses fulfil two functions is not always helpful.
748 There are often times when a grouping subpattern is required without a
749 capturing requirement. If an opening parenthesis is followed by a question mark
750 and a colon, the subpattern does not do any capturing, and is not counted when
751 computing the number of any subsequent capturing subpatterns. For example, if
752 the string "the white queen" is matched against the pattern
753 .sp
754 the ((?:red|white) (king|queen))
755 .sp
756 the captured substrings are "white queen" and "queen", and are numbered 1 and
757 2. The maximum number of capturing subpatterns is 65535, and the maximum depth
758 of nesting of all subpatterns, both capturing and non-capturing, is 200.
759 .P
760 As a convenient shorthand, if any option settings are required at the start of
761 a non-capturing subpattern, the option letters may appear between the "?" and
762 the ":". Thus the two patterns
763 .sp
764 (?i:saturday|sunday)
765 (?:(?i)saturday|sunday)
766 .sp
767 match exactly the same set of strings. Because alternative branches are tried
768 from left to right, and options are not reset until the end of the subpattern
769 is reached, an option setting in one branch does affect subsequent branches, so
770 the above patterns match "SUNDAY" as well as "Saturday".
771 .
772 .
774 .rs
775 .sp
776 Identifying capturing parentheses by number is simple, but it can be very hard
777 to keep track of the numbers in complicated regular expressions. Furthermore,
778 if an expression is modified, the numbers may change. To help with this
779 difficulty, PCRE supports the naming of subpatterns, something that Perl does
780 not provide. The Python syntax (?P<name>...) is used. Names consist of
781 alphanumeric characters and underscores, and must be unique within a pattern.
782 .P
783 Named capturing parentheses are still allocated numbers as well as names. The
784 PCRE API provides function calls for extracting the name-to-number translation
785 table from a compiled pattern. There is also a convenience function for
786 extracting a captured substring by name. For further details see the
787 .\" HREF
788 \fBpcreapi\fP
789 .\"
790 documentation.
791 .
792 .
794 .rs
795 .sp
796 Repetition is specified by quantifiers, which can follow any of the following
797 items:
798 .sp
799 a literal data character
800 the . metacharacter
801 the \eC escape sequence
802 the \eX escape sequence (in UTF-8 mode with Unicode properties)
803 an escape such as \ed that matches a single character
804 a character class
805 a back reference (see next section)
806 a parenthesized subpattern (unless it is an assertion)
807 .sp
808 The general repetition quantifier specifies a minimum and maximum number of
809 permitted matches, by giving the two numbers in curly brackets (braces),
810 separated by a comma. The numbers must be less than 65536, and the first must
811 be less than or equal to the second. For example:
812 .sp
813 z{2,4}
814 .sp
815 matches "zz", "zzz", or "zzzz". A closing brace on its own is not a special
816 character. If the second number is omitted, but the comma is present, there is
817 no upper limit; if the second number and the comma are both omitted, the
818 quantifier specifies an exact number of required matches. Thus
819 .sp
820 [aeiou]{3,}
821 .sp
822 matches at least 3 successive vowels, but may match many more, while
823 .sp
824 \ed{8}
825 .sp
826 matches exactly 8 digits. An opening curly bracket that appears in a position
827 where a quantifier is not allowed, or one that does not match the syntax of a
828 quantifier, is taken as a literal character. For example, {,6} is not a
829 quantifier, but a literal string of four characters.
830 .P
831 In UTF-8 mode, quantifiers apply to UTF-8 characters rather than to individual
832 bytes. Thus, for example, \ex{100}{2} matches two UTF-8 characters, each of
833 which is represented by a two-byte sequence. Similarly, when Unicode property
834 support is available, \eX{3} matches three Unicode extended sequences, each of
835 which may be several bytes long (and they may be of different lengths).
836 .P
837 The quantifier {0} is permitted, causing the expression to behave as if the
838 previous item and the quantifier were not present.
839 .P
840 For convenience (and historical compatibility) the three most common
841 quantifiers have single-character abbreviations:
842 .sp
843 * is equivalent to {0,}
844 + is equivalent to {1,}
845 ? is equivalent to {0,1}
846 .sp
847 It is possible to construct infinite loops by following a subpattern that can
848 match no characters with a quantifier that has no upper limit, for example:
849 .sp
850 (a?)*
851 .sp
852 Earlier versions of Perl and PCRE used to give an error at compile time for
853 such patterns. However, because there are cases where this can be useful, such
854 patterns are now accepted, but if any repetition of the subpattern does in fact
855 match no characters, the loop is forcibly broken.
856 .P
857 By default, the quantifiers are "greedy", that is, they match as much as
858 possible (up to the maximum number of permitted times), without causing the
859 rest of the pattern to fail. The classic example of where this gives problems
860 is in trying to match comments in C programs. These appear between /* and */
861 and within the comment, individual * and / characters may appear. An attempt to
862 match C comments by applying the pattern
863 .sp
864 /\e*.*\e*/
865 .sp
866 to the string
867 .sp
868 /* first comment */ not comment /* second comment */
869 .sp
870 fails, because it matches the entire string owing to the greediness of the .*
871 item.
872 .P
873 However, if a quantifier is followed by a question mark, it ceases to be
874 greedy, and instead matches the minimum number of times possible, so the
875 pattern
876 .sp
877 /\e*.*?\e*/
878 .sp
879 does the right thing with the C comments. The meaning of the various
880 quantifiers is not otherwise changed, just the preferred number of matches.
881 Do not confuse this use of question mark with its use as a quantifier in its
882 own right. Because it has two uses, it can sometimes appear doubled, as in
883 .sp
884 \ed??\ed
885 .sp
886 which matches one digit by preference, but can match two if that is the only
887 way the rest of the pattern matches.
888 .P
889 If the PCRE_UNGREEDY option is set (an option which is not available in Perl),
890 the quantifiers are not greedy by default, but individual ones can be made
891 greedy by following them with a question mark. In other words, it inverts the
892 default behaviour.
893 .P
894 When a parenthesized subpattern is quantified with a minimum repeat count that
895 is greater than 1 or with a limited maximum, more memory is required for the
896 compiled pattern, in proportion to the size of the minimum or maximum.
897 .P
898 If a pattern starts with .* or .{0,} and the PCRE_DOTALL option (equivalent
899 to Perl's /s) is set, thus allowing the . to match newlines, the pattern is
900 implicitly anchored, because whatever follows will be tried against every
901 character position in the subject string, so there is no point in retrying the
902 overall match at any position after the first. PCRE normally treats such a
903 pattern as though it were preceded by \eA.
904 .P
905 In cases where it is known that the subject string contains no newlines, it is
906 worth setting PCRE_DOTALL in order to obtain this optimization, or
907 alternatively using ^ to indicate anchoring explicitly.
908 .P
909 However, there is one situation where the optimization cannot be used. When .*
910 is inside capturing parentheses that are the subject of a backreference
911 elsewhere in the pattern, a match at the start may fail, and a later one
912 succeed. Consider, for example:
913 .sp
914 (.*)abc\e1
915 .sp
916 If the subject is "xyz123abc123" the match point is the fourth character. For
917 this reason, such a pattern is not implicitly anchored.
918 .P
919 When a capturing subpattern is repeated, the value captured is the substring
920 that matched the final iteration. For example, after
921 .sp
922 (tweedle[dume]{3}\es*)+
923 .sp
924 has matched "tweedledum tweedledee" the value of the captured substring is
925 "tweedledee". However, if there are nested capturing subpatterns, the
926 corresponding captured values may have been set in previous iterations. For
927 example, after
928 .sp
929 /(a|(b))+/
930 .sp
931 matches "aba" the value of the second captured substring is "b".
932 .
933 .
934 .\" HTML <a name="atomicgroup"></a>
936 .rs
937 .sp
938 With both maximizing and minimizing repetition, failure of what follows
939 normally causes the repeated item to be re-evaluated to see if a different
940 number of repeats allows the rest of the pattern to match. Sometimes it is
941 useful to prevent this, either to change the nature of the match, or to cause
942 it fail earlier than it otherwise might, when the author of the pattern knows
943 there is no point in carrying on.
944 .P
945 Consider, for example, the pattern \ed+foo when applied to the subject line
946 .sp
947 123456bar
948 .sp
949 After matching all 6 digits and then failing to match "foo", the normal
950 action of the matcher is to try again with only 5 digits matching the \ed+
951 item, and then with 4, and so on, before ultimately failing. "Atomic grouping"
952 (a term taken from Jeffrey Friedl's book) provides the means for specifying
953 that once a subpattern has matched, it is not to be re-evaluated in this way.
954 .P
955 If we use atomic grouping for the previous example, the matcher would give up
956 immediately on failing to match "foo" the first time. The notation is a kind of
957 special parenthesis, starting with (?> as in this example:
958 .sp
959 (?>\ed+)foo
960 .sp
961 This kind of parenthesis "locks up" the part of the pattern it contains once
962 it has matched, and a failure further into the pattern is prevented from
963 backtracking into it. Backtracking past it to previous items, however, works as
964 normal.
965 .P
966 An alternative description is that a subpattern of this type matches the string
967 of characters that an identical standalone pattern would match, if anchored at
968 the current point in the subject string.
969 .P
970 Atomic grouping subpatterns are not capturing subpatterns. Simple cases such as
971 the above example can be thought of as a maximizing repeat that must swallow
972 everything it can. So, while both \ed+ and \ed+? are prepared to adjust the
973 number of digits they match in order to make the rest of the pattern match,
974 (?>\ed+) can only match an entire sequence of digits.
975 .P
976 Atomic groups in general can of course contain arbitrarily complicated
977 subpatterns, and can be nested. However, when the subpattern for an atomic
978 group is just a single repeated item, as in the example above, a simpler
979 notation, called a "possessive quantifier" can be used. This consists of an
980 additional + character following a quantifier. Using this notation, the
981 previous example can be rewritten as
982 .sp
983 \ed++foo
984 .sp
985 Possessive quantifiers are always greedy; the setting of the PCRE_UNGREEDY
986 option is ignored. They are a convenient notation for the simpler forms of
987 atomic group. However, there is no difference in the meaning or processing of a
988 possessive quantifier and the equivalent atomic group.
989 .P
990 The possessive quantifier syntax is an extension to the Perl syntax. It
991 originates in Sun's Java package.
992 .P
993 When a pattern contains an unlimited repeat inside a subpattern that can itself
994 be repeated an unlimited number of times, the use of an atomic group is the
995 only way to avoid some failing matches taking a very long time indeed. The
996 pattern
997 .sp
998 (\eD+|<\ed+>)*[!?]
999 .sp
1000 matches an unlimited number of substrings that either consist of non-digits, or
1001 digits enclosed in <>, followed by either ! or ?. When it matches, it runs
1002 quickly. However, if it is applied to
1003 .sp
1004 aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa
1005 .sp
1006 it takes a long time before reporting failure. This is because the string can
1007 be divided between the internal \eD+ repeat and the external * repeat in a
1008 large number of ways, and all have to be tried. (The example uses [!?] rather
1009 than a single character at the end, because both PCRE and Perl have an
1010 optimization that allows for fast failure when a single character is used. They
1011 remember the last single character that is required for a match, and fail early
1012 if it is not present in the string.) If the pattern is changed so that it uses
1013 an atomic group, like this:
1014 .sp
1015 ((?>\eD+)|<\ed+>)*[!?]
1016 .sp
1017 sequences of non-digits cannot be broken, and failure happens quickly.
1018 .
1019 .
1020 .\" HTML <a name="backreferences"></a>
1022 .rs
1023 .sp
1024 Outside a character class, a backslash followed by a digit greater than 0 (and
1025 possibly further digits) is a back reference to a capturing subpattern earlier
1026 (that is, to its left) in the pattern, provided there have been that many
1027 previous capturing left parentheses.
1028 .P
1029 However, if the decimal number following the backslash is less than 10, it is
1030 always taken as a back reference, and causes an error only if there are not
1031 that many capturing left parentheses in the entire pattern. In other words, the
1032 parentheses that are referenced need not be to the left of the reference for
1033 numbers less than 10. See the subsection entitled "Non-printing characters"
1034 .\" HTML <a href="#digitsafterbackslash">
1035 .\" </a>
1036 above
1037 .\"
1038 for further details of the handling of digits following a backslash.
1039 .P
1040 A back reference matches whatever actually matched the capturing subpattern in
1041 the current subject string, rather than anything matching the subpattern
1042 itself (see
1043 .\" HTML <a href="#subpatternsassubroutines">
1044 .\" </a>
1045 "Subpatterns as subroutines"
1046 .\"
1047 below for a way of doing that). So the pattern
1048 .sp
1049 (sens|respons)e and \e1ibility
1050 .sp
1051 matches "sense and sensibility" and "response and responsibility", but not
1052 "sense and responsibility". If caseful matching is in force at the time of the
1053 back reference, the case of letters is relevant. For example,
1054 .sp
1055 ((?i)rah)\es+\e1
1056 .sp
1057 matches "rah rah" and "RAH RAH", but not "RAH rah", even though the original
1058 capturing subpattern is matched caselessly.
1059 .P
1060 Back references to named subpatterns use the Python syntax (?P=name). We could
1061 rewrite the above example as follows:
1062 .sp
1063 (?<p1>(?i)rah)\es+(?P=p1)
1064 .sp
1065 There may be more than one back reference to the same subpattern. If a
1066 subpattern has not actually been used in a particular match, any back
1067 references to it always fail. For example, the pattern
1068 .sp
1069 (a|(bc))\e2
1070 .sp
1071 always fails if it starts to match "a" rather than "bc". Because there may be
1072 many capturing parentheses in a pattern, all digits following the backslash are
1073 taken as part of a potential back reference number. If the pattern continues
1074 with a digit character, some delimiter must be used to terminate the back
1075 reference. If the PCRE_EXTENDED option is set, this can be whitespace.
1076 Otherwise an empty comment (see
1077 .\" HTML <a href="#comments">
1078 .\" </a>
1079 "Comments"
1080 .\"
1081 below) can be used.
1082 .P
1083 A back reference that occurs inside the parentheses to which it refers fails
1084 when the subpattern is first used, so, for example, (a\e1) never matches.
1085 However, such references can be useful inside repeated subpatterns. For
1086 example, the pattern
1087 .sp
1088 (a|b\e1)+
1089 .sp
1090 matches any number of "a"s and also "aba", "ababbaa" etc. At each iteration of
1091 the subpattern, the back reference matches the character string corresponding
1092 to the previous iteration. In order for this to work, the pattern must be such
1093 that the first iteration does not need to match the back reference. This can be
1094 done using alternation, as in the example above, or by a quantifier with a
1095 minimum of zero.
1096 .
1097 .
1098 .\" HTML <a name="bigassertions"></a>
1100 .rs
1101 .sp
1102 An assertion is a test on the characters following or preceding the current
1103 matching point that does not actually consume any characters. The simple
1104 assertions coded as \eb, \eB, \eA, \eG, \eZ, \ez, ^ and $ are described
1105 .\" HTML <a href="#smallassertions">
1106 .\" </a>
1107 above.
1108 .\"
1109 .P
1110 More complicated assertions are coded as subpatterns. There are two kinds:
1111 those that look ahead of the current position in the subject string, and those
1112 that look behind it. An assertion subpattern is matched in the normal way,
1113 except that it does not cause the current matching position to be changed.
1114 .P
1115 Assertion subpatterns are not capturing subpatterns, and may not be repeated,
1116 because it makes no sense to assert the same thing several times. If any kind
1117 of assertion contains capturing subpatterns within it, these are counted for
1118 the purposes of numbering the capturing subpatterns in the whole pattern.
1119 However, substring capturing is carried out only for positive assertions,
1120 because it does not make sense for negative assertions.
1121 .
1122 .
1123 .SS "Lookahead assertions"
1124 .rs
1125 .sp
1126 Lookahead assertions start
1127 with (?= for positive assertions and (?! for negative assertions. For example,
1128 .sp
1129 \ew+(?=;)
1130 .sp
1131 matches a word followed by a semicolon, but does not include the semicolon in
1132 the match, and
1133 .sp
1134 foo(?!bar)
1135 .sp
1136 matches any occurrence of "foo" that is not followed by "bar". Note that the
1137 apparently similar pattern
1138 .sp
1139 (?!foo)bar
1140 .sp
1141 does not find an occurrence of "bar" that is preceded by something other than
1142 "foo"; it finds any occurrence of "bar" whatsoever, because the assertion
1143 (?!foo) is always true when the next three characters are "bar". A
1144 lookbehind assertion is needed to achieve the other effect.
1145 .P
1146 If you want to force a matching failure at some point in a pattern, the most
1147 convenient way to do it is with (?!) because an empty string always matches, so
1148 an assertion that requires there not to be an empty string must always fail.
1149 .
1150 .
1151 .\" HTML <a name="lookbehind"></a>
1152 .SS "Lookbehind assertions"
1153 .rs
1154 .sp
1155 Lookbehind assertions start with (?<= for positive assertions and (?<! for
1156 negative assertions. For example,
1157 .sp
1158 (?<!foo)bar
1159 .sp
1160 does find an occurrence of "bar" that is not preceded by "foo". The contents of
1161 a lookbehind assertion are restricted such that all the strings it matches must
1162 have a fixed length. However, if there are several alternatives, they do not
1163 all have to have the same fixed length. Thus
1164 .sp
1165 (?<=bullock|donkey)
1166 .sp
1167 is permitted, but
1168 .sp
1169 (?<!dogs?|cats?)
1170 .sp
1171 causes an error at compile time. Branches that match different length strings
1172 are permitted only at the top level of a lookbehind assertion. This is an
1173 extension compared with Perl (at least for 5.8), which requires all branches to
1174 match the same length of string. An assertion such as
1175 .sp
1176 (?<=ab(c|de))
1177 .sp
1178 is not permitted, because its single top-level branch can match two different
1179 lengths, but it is acceptable if rewritten to use two top-level branches:
1180 .sp
1181 (?<=abc|abde)
1182 .sp
1183 The implementation of lookbehind assertions is, for each alternative, to
1184 temporarily move the current position back by the fixed width and then try to
1185 match. If there are insufficient characters before the current position, the
1186 match is deemed to fail.
1187 .P
1188 PCRE does not allow the \eC escape (which matches a single byte in UTF-8 mode)
1189 to appear in lookbehind assertions, because it makes it impossible to calculate
1190 the length of the lookbehind. The \eX escape, which can match different numbers
1191 of bytes, is also not permitted.
1192 .P
1193 Atomic groups can be used in conjunction with lookbehind assertions to specify
1194 efficient matching at the end of the subject string. Consider a simple pattern
1195 such as
1196 .sp
1197 abcd$
1198 .sp
1199 when applied to a long string that does not match. Because matching proceeds
1200 from left to right, PCRE will look for each "a" in the subject and then see if
1201 what follows matches the rest of the pattern. If the pattern is specified as
1202 .sp
1203 ^.*abcd$
1204 .sp
1205 the initial .* matches the entire string at first, but when this fails (because
1206 there is no following "a"), it backtracks to match all but the last character,
1207 then all but the last two characters, and so on. Once again the search for "a"
1208 covers the entire string, from right to left, so we are no better off. However,
1209 if the pattern is written as
1210 .sp
1211 ^(?>.*)(?<=abcd)
1212 .sp
1213 or, equivalently, using the possessive quantifier syntax,
1214 .sp
1215 ^.*+(?<=abcd)
1216 .sp
1217 there can be no backtracking for the .* item; it can match only the entire
1218 string. The subsequent lookbehind assertion does a single test on the last four
1219 characters. If it fails, the match fails immediately. For long strings, this
1220 approach makes a significant difference to the processing time.
1221 .
1222 .
1223 .SS "Using multiple assertions"
1224 .rs
1225 .sp
1226 Several assertions (of any sort) may occur in succession. For example,
1227 .sp
1228 (?<=\ed{3})(?<!999)foo
1229 .sp
1230 matches "foo" preceded by three digits that are not "999". Notice that each of
1231 the assertions is applied independently at the same point in the subject
1232 string. First there is a check that the previous three characters are all
1233 digits, and then there is a check that the same three characters are not "999".
1234 This pattern does \fInot\fP match "foo" preceded by six characters, the first
1235 of which are digits and the last three of which are not "999". For example, it
1236 doesn't match "123abcfoo". A pattern to do that is
1237 .sp
1238 (?<=\ed{3}...)(?<!999)foo
1239 .sp
1240 This time the first assertion looks at the preceding six characters, checking
1241 that the first three are digits, and then the second assertion checks that the
1242 preceding three characters are not "999".
1243 .P
1244 Assertions can be nested in any combination. For example,
1245 .sp
1246 (?<=(?<!foo)bar)baz
1247 .sp
1248 matches an occurrence of "baz" that is preceded by "bar" which in turn is not
1249 preceded by "foo", while
1250 .sp
1251 (?<=\ed{3}(?!999)...)foo
1252 .sp
1253 is another pattern that matches "foo" preceded by three digits and any three
1254 characters that are not "999".
1255 .
1256 .
1258 .rs
1259 .sp
1260 It is possible to cause the matching process to obey a subpattern
1261 conditionally or to choose between two alternative subpatterns, depending on
1262 the result of an assertion, or whether a previous capturing subpattern matched
1263 or not. The two possible forms of conditional subpattern are
1264 .sp
1265 (?(condition)yes-pattern)
1266 (?(condition)yes-pattern|no-pattern)
1267 .sp
1268 If the condition is satisfied, the yes-pattern is used; otherwise the
1269 no-pattern (if present) is used. If there are more than two alternatives in the
1270 subpattern, a compile-time error occurs.
1271 .P
1272 There are three kinds of condition. If the text between the parentheses
1273 consists of a sequence of digits, the condition is satisfied if the capturing
1274 subpattern of that number has previously matched. The number must be greater
1275 than zero. Consider the following pattern, which contains non-significant white
1276 space to make it more readable (assume the PCRE_EXTENDED option) and to divide
1277 it into three parts for ease of discussion:
1278 .sp
1279 ( \e( )? [^()]+ (?(1) \e) )
1280 .sp
1281 The first part matches an optional opening parenthesis, and if that
1282 character is present, sets it as the first captured substring. The second part
1283 matches one or more characters that are not parentheses. The third part is a
1284 conditional subpattern that tests whether the first set of parentheses matched
1285 or not. If they did, that is, if subject started with an opening parenthesis,
1286 the condition is true, and so the yes-pattern is executed and a closing
1287 parenthesis is required. Otherwise, since no-pattern is not present, the
1288 subpattern matches nothing. In other words, this pattern matches a sequence of
1289 non-parentheses, optionally enclosed in parentheses.
1290 .P
1291 If the condition is the string (R), it is satisfied if a recursive call to the
1292 pattern or subpattern has been made. At "top level", the condition is false.
1293 This is a PCRE extension. Recursive patterns are described in the next section.
1294 .P
1295 If the condition is not a sequence of digits or (R), it must be an assertion.
1296 This may be a positive or negative lookahead or lookbehind assertion. Consider
1297 this pattern, again containing non-significant white space, and with the two
1298 alternatives on the second line:
1299 .sp
1300 (?(?=[^a-z]*[a-z])
1301 \ed{2}-[a-z]{3}-\ed{2} | \ed{2}-\ed{2}-\ed{2} )
1302 .sp
1303 The condition is a positive lookahead assertion that matches an optional
1304 sequence of non-letters followed by a letter. In other words, it tests for the
1305 presence of at least one letter in the subject. If a letter is found, the
1306 subject is matched against the first alternative; otherwise it is matched
1307 against the second. This pattern matches strings in one of the two forms
1308 dd-aaa-dd or dd-dd-dd, where aaa are letters and dd are digits.
1309 .
1310 .
1311 .\" HTML <a name="comments"></a>
1313 .rs
1314 .sp
1315 The sequence (?# marks the start of a comment that continues up to the next
1316 closing parenthesis. Nested parentheses are not permitted. The characters
1317 that make up a comment play no part in the pattern matching at all.
1318 .P
1319 If the PCRE_EXTENDED option is set, an unescaped # character outside a
1320 character class introduces a comment that continues up to the next newline
1321 character in the pattern.
1322 .
1323 .
1325 .rs
1326 .sp
1327 Consider the problem of matching a string in parentheses, allowing for
1328 unlimited nested parentheses. Without the use of recursion, the best that can
1329 be done is to use a pattern that matches up to some fixed depth of nesting. It
1330 is not possible to handle an arbitrary nesting depth. Perl provides a facility
1331 that allows regular expressions to recurse (amongst other things). It does this
1332 by interpolating Perl code in the expression at run time, and the code can
1333 refer to the expression itself. A Perl pattern to solve the parentheses problem
1334 can be created like this:
1335 .sp
1336 $re = qr{\e( (?: (?>[^()]+) | (?p{$re}) )* \e)}x;
1337 .sp
1338 The (?p{...}) item interpolates Perl code at run time, and in this case refers
1339 recursively to the pattern in which it appears. Obviously, PCRE cannot support
1340 the interpolation of Perl code. Instead, it supports some special syntax for
1341 recursion of the entire pattern, and also for individual subpattern recursion.
1342 .P
1343 The special item that consists of (? followed by a number greater than zero and
1344 a closing parenthesis is a recursive call of the subpattern of the given
1345 number, provided that it occurs inside that subpattern. (If not, it is a
1346 "subroutine" call, which is described in the next section.) The special item
1347 (?R) is a recursive call of the entire regular expression.
1348 .P
1349 For example, this PCRE pattern solves the nested parentheses problem (assume
1350 the PCRE_EXTENDED option is set so that white space is ignored):
1351 .sp
1352 \e( ( (?>[^()]+) | (?R) )* \e)
1353 .sp
1354 First it matches an opening parenthesis. Then it matches any number of
1355 substrings which can either be a sequence of non-parentheses, or a recursive
1356 match of the pattern itself (that is a correctly parenthesized substring).
1357 Finally there is a closing parenthesis.
1358 .P
1359 If this were part of a larger pattern, you would not want to recurse the entire
1360 pattern, so instead you could use this:
1361 .sp
1362 ( \e( ( (?>[^()]+) | (?1) )* \e) )
1363 .sp
1364 We have put the pattern into parentheses, and caused the recursion to refer to
1365 them instead of the whole pattern. In a larger pattern, keeping track of
1366 parenthesis numbers can be tricky. It may be more convenient to use named
1367 parentheses instead. For this, PCRE uses (?P>name), which is an extension to
1368 the Python syntax that PCRE uses for named parentheses (Perl does not provide
1369 named parentheses). We could rewrite the above example as follows:
1370 .sp
1371 (?P<pn> \e( ( (?>[^()]+) | (?P>pn) )* \e) )
1372 .sp
1373 This particular example pattern contains nested unlimited repeats, and so the
1374 use of atomic grouping for matching strings of non-parentheses is important
1375 when applying the pattern to strings that do not match. For example, when this
1376 pattern is applied to
1377 .sp
1378 (aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa()
1379 .sp
1380 it yields "no match" quickly. However, if atomic grouping is not used,
1381 the match runs for a very long time indeed because there are so many different
1382 ways the + and * repeats can carve up the subject, and all have to be tested
1383 before failure can be reported.
1384 .P
1385 At the end of a match, the values set for any capturing subpatterns are those
1386 from the outermost level of the recursion at which the subpattern value is set.
1387 If you want to obtain intermediate values, a callout function can be used (see
1388 the next section and the
1389 .\" HREF
1390 \fBpcrecallout\fP
1391 .\"
1392 documentation). If the pattern above is matched against
1393 .sp
1394 (ab(cd)ef)
1395 .sp
1396 the value for the capturing parentheses is "ef", which is the last value taken
1397 on at the top level. If additional parentheses are added, giving
1398 .sp
1399 \e( ( ( (?>[^()]+) | (?R) )* ) \e)
1400 ^ ^
1401 ^ ^
1402 .sp
1403 the string they capture is "ab(cd)ef", the contents of the top level
1404 parentheses. If there are more than 15 capturing parentheses in a pattern, PCRE
1405 has to obtain extra memory to store data during a recursion, which it does by
1406 using \fBpcre_malloc\fP, freeing it via \fBpcre_free\fP afterwards. If no
1407 memory can be obtained, the match fails with the PCRE_ERROR_NOMEMORY error.
1408 .P
1409 Do not confuse the (?R) item with the condition (R), which tests for recursion.
1410 Consider this pattern, which matches text in angle brackets, allowing for
1411 arbitrary nesting. Only digits are allowed in nested brackets (that is, when
1412 recursing), whereas any characters are permitted at the outer level.
1413 .sp
1414 < (?: (?(R) \ed++ | [^<>]*+) | (?R)) * >
1415 .sp
1416 In this pattern, (?(R) is the start of a conditional subpattern, with two
1417 different alternatives for the recursive and non-recursive cases. The (?R) item
1418 is the actual recursive call.
1419 .
1420 .
1421 .\" HTML <a name="subpatternsassubroutines"></a>
1423 .rs
1424 .sp
1425 If the syntax for a recursive subpattern reference (either by number or by
1426 name) is used outside the parentheses to which it refers, it operates like a
1427 subroutine in a programming language. An earlier example pointed out that the
1428 pattern
1429 .sp
1430 (sens|respons)e and \e1ibility
1431 .sp
1432 matches "sense and sensibility" and "response and responsibility", but not
1433 "sense and responsibility". If instead the pattern
1434 .sp
1435 (sens|respons)e and (?1)ibility
1436 .sp
1437 is used, it does match "sense and responsibility" as well as the other two
1438 strings. Such references must, however, follow the subpattern to which they
1439 refer.
1440 .
1441 .
1443 .rs
1444 .sp
1445 Perl has a feature whereby using the sequence (?{...}) causes arbitrary Perl
1446 code to be obeyed in the middle of matching a regular expression. This makes it
1447 possible, amongst other things, to extract different substrings that match the
1448 same pair of parentheses when there is a repetition.
1449 .P
1450 PCRE provides a similar feature, but of course it cannot obey arbitrary Perl
1451 code. The feature is called "callout". The caller of PCRE provides an external
1452 function by putting its entry point in the global variable \fIpcre_callout\fP.
1453 By default, this variable contains NULL, which disables all calling out.
1454 .P
1455 Within a regular expression, (?C) indicates the points at which the external
1456 function is to be called. If you want to identify different callout points, you
1457 can put a number less than 256 after the letter C. The default value is zero.
1458 For example, this pattern has two callout points:
1459 .sp
1460 (?C1)\dabc(?C2)def
1461 .sp
1462 If the PCRE_AUTO_CALLOUT flag is passed to \fBpcre_compile()\fP, callouts are
1463 automatically installed before each item in the pattern. They are all numbered
1464 255.
1465 .P
1466 During matching, when PCRE reaches a callout point (and \fIpcre_callout\fP is
1467 set), the external function is called. It is provided with the number of the
1468 callout, the position in the pattern, and, optionally, one item of data
1469 originally supplied by the caller of \fBpcre_exec()\fP. The callout function
1470 may cause matching to proceed, to backtrack, or to fail altogether. A complete
1471 description of the interface to the callout function is given in the
1472 .\" HREF
1473 \fBpcrecallout\fP
1474 .\"
1475 documentation.
1476 .P
1477 .in 0
1478 Last updated: 28 February 2005
1479 .br
1480 Copyright (c) 1997-2005 University of Cambridge.

  ViewVC Help
Powered by ViewVC 1.1.5