Commit | Line | Data |
---|---|---|
a0d0e21e LW |
1 | =head1 NAME |
2 | ||
3 | perlre - Perl regular expressions | |
4 | ||
5 | =head1 DESCRIPTION | |
6 | ||
cb1a09d0 | 7 | This page describes the syntax of regular expressions in Perl. For a |
5f05dabc | 8 | description of how to I<use> regular expressions in matching |
19799a22 | 9 | operations, plus various examples of the same, see discussions |
1e66bd83 | 10 | of C<m//>, C<s///>, C<qr//> and C<??> in L<perlop/"Regexp Quote-Like Operators">. |
cb1a09d0 | 11 | |
19799a22 | 12 | Matching operations can have various modifiers. Modifiers |
5a964f20 | 13 | that relate to the interpretation of the regular expression inside |
19799a22 GS |
14 | are listed below. Modifiers that alter the way a regular expression |
15 | is used by Perl are detailed in L<perlop/"Regexp Quote-Like Operators"> and | |
1e66bd83 | 16 | L<perlop/"Gory details of parsing quoted constructs">. |
a0d0e21e | 17 | |
55497cff | 18 | =over 4 |
19 | ||
20 | =item i | |
21 | ||
22 | Do case-insensitive pattern matching. | |
23 | ||
a034a98d DD |
24 | If C<use locale> is in effect, the case map is taken from the current |
25 | locale. See L<perllocale>. | |
26 | ||
54310121 | 27 | =item m |
55497cff | 28 | |
29 | Treat string as multiple lines. That is, change "^" and "$" from matching | |
14218588 | 30 | the start or end of the string to matching the start or end of any |
7f761169 | 31 | line anywhere within the string. |
55497cff | 32 | |
54310121 | 33 | =item s |
55497cff | 34 | |
35 | Treat string as single line. That is, change "." to match any character | |
19799a22 | 36 | whatsoever, even a newline, which normally it would not match. |
55497cff | 37 | |
19799a22 GS |
38 | The C</s> and C</m> modifiers both override the C<$*> setting. That |
39 | is, no matter what C<$*> contains, C</s> without C</m> will force | |
40 | "^" to match only at the beginning of the string and "$" to match | |
41 | only at the end (or just before a newline at the end) of the string. | |
42 | Together, as /ms, they let the "." match any character whatsoever, | |
fb55449c | 43 | while still allowing "^" and "$" to match, respectively, just after |
19799a22 | 44 | and just before newlines within the string. |
7b8d334a | 45 | |
54310121 | 46 | =item x |
55497cff | 47 | |
48 | Extend your pattern's legibility by permitting whitespace and comments. | |
49 | ||
50 | =back | |
a0d0e21e LW |
51 | |
52 | These are usually written as "the C</x> modifier", even though the delimiter | |
14218588 | 53 | in question might not really be a slash. Any of these |
a0d0e21e | 54 | modifiers may also be embedded within the regular expression itself using |
14218588 | 55 | the C<(?...)> construct. See below. |
a0d0e21e | 56 | |
4633a7c4 | 57 | The C</x> modifier itself needs a little more explanation. It tells |
55497cff | 58 | the regular expression parser to ignore whitespace that is neither |
59 | backslashed nor within a character class. You can use this to break up | |
4633a7c4 | 60 | your regular expression into (slightly) more readable parts. The C<#> |
54310121 | 61 | character is also treated as a metacharacter introducing a comment, |
55497cff | 62 | just as in ordinary Perl code. This also means that if you want real |
14218588 | 63 | whitespace or C<#> characters in the pattern (outside a character |
5a964f20 | 64 | class, where they are unaffected by C</x>), that you'll either have to |
55497cff | 65 | escape them or encode them using octal or hex escapes. Taken together, |
66 | these features go a long way towards making Perl's regular expressions | |
0c815be9 HS |
67 | more readable. Note that you have to be careful not to include the |
68 | pattern delimiter in the comment--perl has no way of knowing you did | |
5a964f20 | 69 | not intend to close the pattern early. See the C-comment deletion code |
0c815be9 | 70 | in L<perlop>. |
a0d0e21e LW |
71 | |
72 | =head2 Regular Expressions | |
73 | ||
19799a22 | 74 | The patterns used in Perl pattern matching derive from supplied in |
14218588 | 75 | the Version 8 regex routines. (The routines are derived |
19799a22 GS |
76 | (distantly) from Henry Spencer's freely redistributable reimplementation |
77 | of the V8 routines.) See L<Version 8 Regular Expressions> for | |
78 | details. | |
a0d0e21e LW |
79 | |
80 | In particular the following metacharacters have their standard I<egrep>-ish | |
81 | meanings: | |
82 | ||
54310121 | 83 | \ Quote the next metacharacter |
a0d0e21e LW |
84 | ^ Match the beginning of the line |
85 | . Match any character (except newline) | |
c07a80fd | 86 | $ Match the end of the line (or before newline at the end) |
a0d0e21e LW |
87 | | Alternation |
88 | () Grouping | |
89 | [] Character class | |
90 | ||
14218588 GS |
91 | By default, the "^" character is guaranteed to match only the |
92 | beginning of the string, the "$" character only the end (or before the | |
93 | newline at the end), and Perl does certain optimizations with the | |
a0d0e21e LW |
94 | assumption that the string contains only one line. Embedded newlines |
95 | will not be matched by "^" or "$". You may, however, wish to treat a | |
4a6725af | 96 | string as a multi-line buffer, such that the "^" will match after any |
a0d0e21e LW |
97 | newline within the string, and "$" will match before any newline. At the |
98 | cost of a little more overhead, you can do this by using the /m modifier | |
99 | on the pattern match operator. (Older programs did this by setting C<$*>, | |
5f05dabc | 100 | but this practice is now deprecated.) |
a0d0e21e | 101 | |
14218588 | 102 | To simplify multi-line substitutions, the "." character never matches a |
55497cff | 103 | newline unless you use the C</s> modifier, which in effect tells Perl to pretend |
a0d0e21e LW |
104 | the string is a single line--even if it isn't. The C</s> modifier also |
105 | overrides the setting of C<$*>, in case you have some (badly behaved) older | |
106 | code that sets it in another module. | |
107 | ||
108 | The following standard quantifiers are recognized: | |
109 | ||
110 | * Match 0 or more times | |
111 | + Match 1 or more times | |
112 | ? Match 1 or 0 times | |
113 | {n} Match exactly n times | |
114 | {n,} Match at least n times | |
115 | {n,m} Match at least n but not more than m times | |
116 | ||
117 | (If a curly bracket occurs in any other context, it is treated | |
118 | as a regular character.) The "*" modifier is equivalent to C<{0,}>, the "+" | |
25f94b33 | 119 | modifier to C<{1,}>, and the "?" modifier to C<{0,1}>. n and m are limited |
9c79236d GS |
120 | to integral values less than a preset limit defined when perl is built. |
121 | This is usually 32766 on the most common platforms. The actual limit can | |
122 | be seen in the error message generated by code such as this: | |
123 | ||
820475bd | 124 | $_ **= $_ , / {$_} / for 2 .. 42; |
a0d0e21e | 125 | |
54310121 | 126 | By default, a quantified subpattern is "greedy", that is, it will match as |
127 | many times as possible (given a particular starting location) while still | |
128 | allowing the rest of the pattern to match. If you want it to match the | |
129 | minimum number of times possible, follow the quantifier with a "?". Note | |
130 | that the meanings don't change, just the "greediness": | |
a0d0e21e LW |
131 | |
132 | *? Match 0 or more times | |
133 | +? Match 1 or more times | |
134 | ?? Match 0 or 1 time | |
135 | {n}? Match exactly n times | |
136 | {n,}? Match at least n times | |
137 | {n,m}? Match at least n but not more than m times | |
138 | ||
5f05dabc | 139 | Because patterns are processed as double quoted strings, the following |
a0d0e21e LW |
140 | also work: |
141 | ||
0f36ee90 | 142 | \t tab (HT, TAB) |
143 | \n newline (LF, NL) | |
144 | \r return (CR) | |
145 | \f form feed (FF) | |
146 | \a alarm (bell) (BEL) | |
147 | \e escape (think troff) (ESC) | |
cb1a09d0 AD |
148 | \033 octal char (think of a PDP-11) |
149 | \x1B hex char | |
a0ed51b3 | 150 | \x{263a} wide hex char (Unicode SMILEY) |
a0d0e21e | 151 | \c[ control char |
4a2d328f | 152 | \N{name} named char |
cb1a09d0 AD |
153 | \l lowercase next char (think vi) |
154 | \u uppercase next char (think vi) | |
155 | \L lowercase till \E (think vi) | |
156 | \U uppercase till \E (think vi) | |
157 | \E end case modification (think vi) | |
5a964f20 | 158 | \Q quote (disable) pattern metacharacters till \E |
a0d0e21e | 159 | |
a034a98d | 160 | If C<use locale> is in effect, the case map used by C<\l>, C<\L>, C<\u> |
423cee85 | 161 | and C<\U> is taken from the current locale. See L<perllocale>. For |
4a2d328f | 162 | documentation of C<\N{name}>, see L<charnames>. |
a034a98d | 163 | |
1d2dff63 GS |
164 | You cannot include a literal C<$> or C<@> within a C<\Q> sequence. |
165 | An unescaped C<$> or C<@> interpolates the corresponding variable, | |
166 | while escaping will cause the literal string C<\$> to be matched. | |
167 | You'll need to write something like C<m/\Quser\E\@\Qhost/>. | |
168 | ||
a0d0e21e LW |
169 | In addition, Perl defines the following: |
170 | ||
171 | \w Match a "word" character (alphanumeric plus "_") | |
36bbe248 | 172 | \W Match a non-"word" character |
a0d0e21e LW |
173 | \s Match a whitespace character |
174 | \S Match a non-whitespace character | |
175 | \d Match a digit character | |
176 | \D Match a non-digit character | |
a0ed51b3 LW |
177 | \pP Match P, named property. Use \p{Prop} for longer names. |
178 | \PP Match non-P | |
f244e06d GS |
179 | \X Match eXtended Unicode "combining character sequence", |
180 | equivalent to C<(?:\PM\pM*)> | |
4a2d328f | 181 | \C Match a single C char (octet) even under utf8. |
a0d0e21e | 182 | |
36bbe248 | 183 | A C<\w> matches a single alphanumeric character or C<_>, not a whole word. |
14218588 GS |
184 | Use C<\w+> to match a string of Perl-identifier characters (which isn't |
185 | the same as matching an English word). If C<use locale> is in effect, the | |
186 | list of alphabetic characters generated by C<\w> is taken from the | |
187 | current locale. See L<perllocale>. You may use C<\w>, C<\W>, C<\s>, C<\S>, | |
1209ba90 JH |
188 | C<\d>, and C<\D> within character classes, but if you try to use them |
189 | as endpoints of a range, that's not a range, the "-" is understood literally. | |
190 | See L<utf8> for details about C<\pP>, C<\PP>, and C<\X>. | |
a0d0e21e | 191 | |
b8c5462f JH |
192 | The POSIX character class syntax |
193 | ||
820475bd | 194 | [:class:] |
b8c5462f | 195 | |
26b44a0a JH |
196 | is also available. The available classes and their backslash |
197 | equivalents (if available) are as follows: | |
b8c5462f JH |
198 | |
199 | alpha | |
200 | alnum | |
201 | ascii | |
aaa51d5e | 202 | blank [1] |
b8c5462f JH |
203 | cntrl |
204 | digit \d | |
205 | graph | |
206 | lower | |
207 | ||
208 | punct | |
aaa51d5e | 209 | space \s [2] |
b8c5462f | 210 | upper |
aaa51d5e | 211 | word \w [3] |
b8c5462f JH |
212 | xdigit |
213 | ||
aaa51d5e JF |
214 | [1] A GNU extension equivalent to C<[ \t]>, `all horizontal whitespace'. |
215 | [2] Not I<exactly equivalent> to C<\s> since the C<[[:space:]]> includes | |
216 | also the (very rare) `vertical tabulator', "\ck", chr(11). | |
217 | [3] A Perl extension. | |
218 | ||
26b44a0a | 219 | For example use C<[:upper:]> to match all the uppercase characters. |
aaa51d5e JF |
220 | Note that the C<[]> are part of the C<[::]> construct, not part of the |
221 | whole character class. For example: | |
b8c5462f | 222 | |
820475bd | 223 | [01[:alpha:]%] |
b8c5462f | 224 | |
593df60c | 225 | matches zero, one, any alphabetic character, and the percentage sign. |
b8c5462f | 226 | |
26b44a0a | 227 | If the C<utf8> pragma is used, the following equivalences to Unicode |
3bec3564 JH |
228 | \p{} constructs and equivalent backslash character classes (if available), |
229 | will hold: | |
b8c5462f JH |
230 | |
231 | alpha IsAlpha | |
232 | alnum IsAlnum | |
233 | ascii IsASCII | |
aaa51d5e | 234 | blank IsSpace |
b8c5462f | 235 | cntrl IsCntrl |
3bec3564 | 236 | digit IsDigit \d |
b8c5462f JH |
237 | graph IsGraph |
238 | lower IsLower | |
239 | print IsPrint | |
240 | punct IsPunct | |
241 | space IsSpace | |
3bec3564 | 242 | IsSpacePerl \s |
b8c5462f JH |
243 | upper IsUpper |
244 | word IsWord | |
245 | xdigit IsXDigit | |
246 | ||
26b44a0a | 247 | For example C<[:lower:]> and C<\p{IsLower}> are equivalent. |
b8c5462f JH |
248 | |
249 | If the C<utf8> pragma is not used but the C<locale> pragma is, the | |
aaa51d5e JF |
250 | classes correlate with the usual isalpha(3) interface (except for |
251 | `word' and `blank'). | |
b8c5462f JH |
252 | |
253 | The assumedly non-obviously named classes are: | |
254 | ||
255 | =over 4 | |
256 | ||
257 | =item cntrl | |
258 | ||
820475bd GS |
259 | Any control character. Usually characters that don't produce output as |
260 | such but instead control the terminal somehow: for example newline and | |
261 | backspace are control characters. All characters with ord() less than | |
593df60c JH |
262 | 32 are most often classified as control characters (assuming ASCII, |
263 | the ISO Latin character sets, and Unicode). | |
b8c5462f JH |
264 | |
265 | =item graph | |
266 | ||
f1cbbd6e | 267 | Any alphanumeric or punctuation (special) character. |
b8c5462f JH |
268 | |
269 | =item print | |
270 | ||
f1cbbd6e | 271 | Any alphanumeric or punctuation (special) character or space. |
b8c5462f JH |
272 | |
273 | =item punct | |
274 | ||
f1cbbd6e | 275 | Any punctuation (special) character. |
b8c5462f JH |
276 | |
277 | =item xdigit | |
278 | ||
593df60c | 279 | Any hexadecimal digit. Though this may feel silly ([0-9A-Fa-f] would |
820475bd | 280 | work just fine) it is included for completeness. |
b8c5462f | 281 | |
b8c5462f JH |
282 | =back |
283 | ||
284 | You can negate the [::] character classes by prefixing the class name | |
285 | with a '^'. This is a Perl extension. For example: | |
286 | ||
93733859 JH |
287 | POSIX trad. Perl utf8 Perl |
288 | ||
289 | [:^digit:] \D \P{IsDigit} | |
290 | [:^space:] \S \P{IsSpace} | |
291 | [:^word:] \W \P{IsWord} | |
b8c5462f | 292 | |
26b44a0a JH |
293 | The POSIX character classes [.cc.] and [=cc=] are recognized but |
294 | B<not> supported and trying to use them will cause an error. | |
b8c5462f | 295 | |
a0d0e21e LW |
296 | Perl defines the following zero-width assertions: |
297 | ||
298 | \b Match a word boundary | |
299 | \B Match a non-(word boundary) | |
b85d18e9 IZ |
300 | \A Match only at beginning of string |
301 | \Z Match only at end of string, or before newline at the end | |
302 | \z Match only at end of string | |
9da458fc IZ |
303 | \G Match only at pos() (e.g. at the end-of-match position |
304 | of prior m//g) | |
a0d0e21e | 305 | |
14218588 | 306 | A word boundary (C<\b>) is a spot between two characters |
19799a22 GS |
307 | that has a C<\w> on one side of it and a C<\W> on the other side |
308 | of it (in either order), counting the imaginary characters off the | |
309 | beginning and end of the string as matching a C<\W>. (Within | |
310 | character classes C<\b> represents backspace rather than a word | |
311 | boundary, just as it normally does in any double-quoted string.) | |
312 | The C<\A> and C<\Z> are just like "^" and "$", except that they | |
313 | won't match multiple times when the C</m> modifier is used, while | |
314 | "^" and "$" will match at every internal line boundary. To match | |
315 | the actual end of the string and not ignore an optional trailing | |
316 | newline, use C<\z>. | |
317 | ||
318 | The C<\G> assertion can be used to chain global matches (using | |
319 | C<m//g>), as described in L<perlop/"Regexp Quote-Like Operators">. | |
320 | It is also useful when writing C<lex>-like scanners, when you have | |
321 | several patterns that you want to match against consequent substrings | |
322 | of your string, see the previous reference. The actual location | |
323 | where C<\G> will match can also be influenced by using C<pos()> as | |
324 | an lvalue. See L<perlfunc/pos>. | |
c47ff5f1 | 325 | |
14218588 | 326 | The bracketing construct C<( ... )> creates capture buffers. To |
c47ff5f1 | 327 | refer to the digit'th buffer use \<digit> within the |
14218588 | 328 | match. Outside the match use "$" instead of "\". (The |
c47ff5f1 | 329 | \<digit> notation works in certain circumstances outside |
14218588 GS |
330 | the match. See the warning below about \1 vs $1 for details.) |
331 | Referring back to another part of the match is called a | |
332 | I<backreference>. | |
333 | ||
334 | There is no limit to the number of captured substrings that you may | |
335 | use. However Perl also uses \10, \11, etc. as aliases for \010, | |
fb55449c JH |
336 | \011, etc. (Recall that 0 means octal, so \011 is the character at |
337 | number 9 in your coded character set; which would be the 10th character, | |
338 | a horizontal tab under ASCII.) Perl resolves this | |
339 | ambiguity by interpreting \10 as a backreference only if at least 10 | |
340 | left parentheses have opened before it. Likewise \11 is a | |
341 | backreference only if at least 11 left parentheses have opened | |
342 | before it. And so on. \1 through \9 are always interpreted as | |
343 | backreferences. | |
14218588 GS |
344 | |
345 | Examples: | |
a0d0e21e LW |
346 | |
347 | s/^([^ ]*) *([^ ]*)/$2 $1/; # swap first two words | |
348 | ||
14218588 GS |
349 | if (/(.)\1/) { # find first doubled char |
350 | print "'$1' is the first doubled character\n"; | |
351 | } | |
c47ff5f1 | 352 | |
14218588 | 353 | if (/Time: (..):(..):(..)/) { # parse out values |
a0d0e21e LW |
354 | $hours = $1; |
355 | $minutes = $2; | |
356 | $seconds = $3; | |
357 | } | |
c47ff5f1 | 358 | |
14218588 GS |
359 | Several special variables also refer back to portions of the previous |
360 | match. C<$+> returns whatever the last bracket match matched. | |
361 | C<$&> returns the entire matched string. (At one point C<$0> did | |
362 | also, but now it returns the name of the program.) C<$`> returns | |
363 | everything before the matched string. And C<$'> returns everything | |
364 | after the matched string. | |
365 | ||
366 | The numbered variables ($1, $2, $3, etc.) and the related punctuation | |
262ffc8b | 367 | set (C<$+>, C<$&>, C<$`>, and C<$'>) are all dynamically scoped |
14218588 GS |
368 | until the end of the enclosing block or until the next successful |
369 | match, whichever comes first. (See L<perlsyn/"Compound Statements">.) | |
370 | ||
371 | B<WARNING>: Once Perl sees that you need one of C<$&>, C<$`>, or | |
372 | C<$'> anywhere in the program, it has to provide them for every | |
373 | pattern match. This may substantially slow your program. Perl | |
374 | uses the same mechanism to produce $1, $2, etc, so you also pay a | |
375 | price for each pattern that contains capturing parentheses. (To | |
376 | avoid this cost while retaining the grouping behaviour, use the | |
377 | extended regular expression C<(?: ... )> instead.) But if you never | |
378 | use C<$&>, C<$`> or C<$'>, then patterns I<without> capturing | |
379 | parentheses will not be penalized. So avoid C<$&>, C<$'>, and C<$`> | |
380 | if you can, but if you can't (and some algorithms really appreciate | |
381 | them), once you've used them once, use them at will, because you've | |
382 | already paid the price. As of 5.005, C<$&> is not so costly as the | |
383 | other two. | |
68dc0745 | 384 | |
19799a22 GS |
385 | Backslashed metacharacters in Perl are alphanumeric, such as C<\b>, |
386 | C<\w>, C<\n>. Unlike some other regular expression languages, there | |
387 | are no backslashed symbols that aren't alphanumeric. So anything | |
c47ff5f1 | 388 | that looks like \\, \(, \), \<, \>, \{, or \} is always |
19799a22 GS |
389 | interpreted as a literal character, not a metacharacter. This was |
390 | once used in a common idiom to disable or quote the special meanings | |
391 | of regular expression metacharacters in a string that you want to | |
36bbe248 | 392 | use for a pattern. Simply quote all non-"word" characters: |
a0d0e21e LW |
393 | |
394 | $pattern =~ s/(\W)/\\$1/g; | |
395 | ||
f1cbbd6e | 396 | (If C<use locale> is set, then this depends on the current locale.) |
14218588 GS |
397 | Today it is more common to use the quotemeta() function or the C<\Q> |
398 | metaquoting escape sequence to disable all metacharacters' special | |
399 | meanings like this: | |
a0d0e21e LW |
400 | |
401 | /$unquoted\Q$quoted\E$unquoted/ | |
402 | ||
9da458fc IZ |
403 | Beware that if you put literal backslashes (those not inside |
404 | interpolated variables) between C<\Q> and C<\E>, double-quotish | |
405 | backslash interpolation may lead to confusing results. If you | |
406 | I<need> to use literal backslashes within C<\Q...\E>, | |
407 | consult L<perlop/"Gory details of parsing quoted constructs">. | |
408 | ||
19799a22 GS |
409 | =head2 Extended Patterns |
410 | ||
14218588 GS |
411 | Perl also defines a consistent extension syntax for features not |
412 | found in standard tools like B<awk> and B<lex>. The syntax is a | |
413 | pair of parentheses with a question mark as the first thing within | |
414 | the parentheses. The character after the question mark indicates | |
415 | the extension. | |
19799a22 | 416 | |
14218588 GS |
417 | The stability of these extensions varies widely. Some have been |
418 | part of the core language for many years. Others are experimental | |
419 | and may change without warning or be completely removed. Check | |
420 | the documentation on an individual feature to verify its current | |
421 | status. | |
19799a22 | 422 | |
14218588 GS |
423 | A question mark was chosen for this and for the minimal-matching |
424 | construct because 1) question marks are rare in older regular | |
425 | expressions, and 2) whenever you see one, you should stop and | |
426 | "question" exactly what is going on. That's psychology... | |
a0d0e21e LW |
427 | |
428 | =over 10 | |
429 | ||
cc6b7395 | 430 | =item C<(?#text)> |
a0d0e21e | 431 | |
14218588 | 432 | A comment. The text is ignored. If the C</x> modifier enables |
19799a22 | 433 | whitespace formatting, a simple C<#> will suffice. Note that Perl closes |
259138e3 GS |
434 | the comment as soon as it sees a C<)>, so there is no way to put a literal |
435 | C<)> in the comment. | |
a0d0e21e | 436 | |
19799a22 GS |
437 | =item C<(?imsx-imsx)> |
438 | ||
439 | One or more embedded pattern-match modifiers. This is particularly | |
440 | useful for dynamic patterns, such as those read in from a configuration | |
441 | file, read in as an argument, are specified in a table somewhere, | |
442 | etc. Consider the case that some of which want to be case sensitive | |
443 | and some do not. The case insensitive ones need to include merely | |
444 | C<(?i)> at the front of the pattern. For example: | |
445 | ||
446 | $pattern = "foobar"; | |
447 | if ( /$pattern/i ) { } | |
448 | ||
449 | # more flexible: | |
450 | ||
451 | $pattern = "(?i)foobar"; | |
452 | if ( /$pattern/ ) { } | |
453 | ||
454 | Letters after a C<-> turn those modifiers off. These modifiers are | |
455 | localized inside an enclosing group (if any). For example, | |
456 | ||
457 | ( (?i) blah ) \s+ \1 | |
458 | ||
459 | will match a repeated (I<including the case>!) word C<blah> in any | |
14218588 | 460 | case, assuming C<x> modifier, and no C<i> modifier outside this |
19799a22 GS |
461 | group. |
462 | ||
5a964f20 | 463 | =item C<(?:pattern)> |
a0d0e21e | 464 | |
ca9dfc88 IZ |
465 | =item C<(?imsx-imsx:pattern)> |
466 | ||
5a964f20 TC |
467 | This is for clustering, not capturing; it groups subexpressions like |
468 | "()", but doesn't make backreferences as "()" does. So | |
a0d0e21e | 469 | |
5a964f20 | 470 | @fields = split(/\b(?:a|b|c)\b/) |
a0d0e21e LW |
471 | |
472 | is like | |
473 | ||
5a964f20 | 474 | @fields = split(/\b(a|b|c)\b/) |
a0d0e21e | 475 | |
19799a22 GS |
476 | but doesn't spit out extra fields. It's also cheaper not to capture |
477 | characters if you don't need to. | |
a0d0e21e | 478 | |
19799a22 GS |
479 | Any letters between C<?> and C<:> act as flags modifiers as with |
480 | C<(?imsx-imsx)>. For example, | |
ca9dfc88 IZ |
481 | |
482 | /(?s-i:more.*than).*million/i | |
483 | ||
14218588 | 484 | is equivalent to the more verbose |
ca9dfc88 IZ |
485 | |
486 | /(?:(?s-i)more.*than).*million/i | |
487 | ||
5a964f20 | 488 | =item C<(?=pattern)> |
a0d0e21e | 489 | |
19799a22 | 490 | A zero-width positive look-ahead assertion. For example, C</\w+(?=\t)/> |
a0d0e21e LW |
491 | matches a word followed by a tab, without including the tab in C<$&>. |
492 | ||
5a964f20 | 493 | =item C<(?!pattern)> |
a0d0e21e | 494 | |
19799a22 | 495 | A zero-width negative look-ahead assertion. For example C</foo(?!bar)/> |
a0d0e21e | 496 | matches any occurrence of "foo" that isn't followed by "bar". Note |
19799a22 GS |
497 | however that look-ahead and look-behind are NOT the same thing. You cannot |
498 | use this for look-behind. | |
7b8d334a | 499 | |
5a964f20 | 500 | If you are looking for a "bar" that isn't preceded by a "foo", C</(?!foo)bar/> |
7b8d334a GS |
501 | will not do what you want. That's because the C<(?!foo)> is just saying that |
502 | the next thing cannot be "foo"--and it's not, it's a "bar", so "foobar" will | |
503 | match. You would have to do something like C</(?!foo)...bar/> for that. We | |
504 | say "like" because there's the case of your "bar" not having three characters | |
505 | before it. You could cover that this way: C</(?:(?!foo)...|^.{0,2})bar/>. | |
506 | Sometimes it's still easier just to say: | |
a0d0e21e | 507 | |
a3cb178b | 508 | if (/bar/ && $` !~ /foo$/) |
a0d0e21e | 509 | |
19799a22 | 510 | For look-behind see below. |
c277df42 | 511 | |
c47ff5f1 | 512 | =item C<(?<=pattern)> |
c277df42 | 513 | |
c47ff5f1 | 514 | A zero-width positive look-behind assertion. For example, C</(?<=\t)\w+/> |
19799a22 GS |
515 | matches a word that follows a tab, without including the tab in C<$&>. |
516 | Works only for fixed-width look-behind. | |
c277df42 | 517 | |
5a964f20 | 518 | =item C<(?<!pattern)> |
c277df42 | 519 | |
19799a22 GS |
520 | A zero-width negative look-behind assertion. For example C</(?<!bar)foo/> |
521 | matches any occurrence of "foo" that does not follow "bar". Works | |
522 | only for fixed-width look-behind. | |
c277df42 | 523 | |
cc6b7395 | 524 | =item C<(?{ code })> |
c277df42 | 525 | |
19799a22 GS |
526 | B<WARNING>: This extended regular expression feature is considered |
527 | highly experimental, and may be changed or deleted without notice. | |
c277df42 | 528 | |
19799a22 GS |
529 | This zero-width assertion evaluate any embedded Perl code. It |
530 | always succeeds, and its C<code> is not interpolated. Currently, | |
531 | the rules to determine where the C<code> ends are somewhat convoluted. | |
532 | ||
533 | The C<code> is properly scoped in the following sense: If the assertion | |
534 | is backtracked (compare L<"Backtracking">), all changes introduced after | |
535 | C<local>ization are undone, so that | |
b9ac3b5b GS |
536 | |
537 | $_ = 'a' x 8; | |
538 | m< | |
539 | (?{ $cnt = 0 }) # Initialize $cnt. | |
540 | ( | |
541 | a | |
542 | (?{ | |
543 | local $cnt = $cnt + 1; # Update $cnt, backtracking-safe. | |
544 | }) | |
545 | )* | |
546 | aaaa | |
547 | (?{ $res = $cnt }) # On success copy to non-localized | |
548 | # location. | |
549 | >x; | |
550 | ||
19799a22 | 551 | will set C<$res = 4>. Note that after the match, $cnt returns to the globally |
14218588 | 552 | introduced value, because the scopes that restrict C<local> operators |
b9ac3b5b GS |
553 | are unwound. |
554 | ||
19799a22 GS |
555 | This assertion may be used as a C<(?(condition)yes-pattern|no-pattern)> |
556 | switch. If I<not> used in this way, the result of evaluation of | |
557 | C<code> is put into the special variable C<$^R>. This happens | |
558 | immediately, so C<$^R> can be used from other C<(?{ code })> assertions | |
559 | inside the same regular expression. | |
b9ac3b5b | 560 | |
19799a22 GS |
561 | The assignment to C<$^R> above is properly localized, so the old |
562 | value of C<$^R> is restored if the assertion is backtracked; compare | |
563 | L<"Backtracking">. | |
b9ac3b5b | 564 | |
19799a22 GS |
565 | For reasons of security, this construct is forbidden if the regular |
566 | expression involves run-time interpolation of variables, unless the | |
567 | perilous C<use re 'eval'> pragma has been used (see L<re>), or the | |
568 | variables contain results of C<qr//> operator (see | |
569 | L<perlop/"qr/STRING/imosx">). | |
871b0233 | 570 | |
14218588 | 571 | This restriction is because of the wide-spread and remarkably convenient |
19799a22 | 572 | custom of using run-time determined strings as patterns. For example: |
871b0233 IZ |
573 | |
574 | $re = <>; | |
575 | chomp $re; | |
576 | $string =~ /$re/; | |
577 | ||
14218588 GS |
578 | Before Perl knew how to execute interpolated code within a pattern, |
579 | this operation was completely safe from a security point of view, | |
580 | although it could raise an exception from an illegal pattern. If | |
581 | you turn on the C<use re 'eval'>, though, it is no longer secure, | |
582 | so you should only do so if you are also using taint checking. | |
583 | Better yet, use the carefully constrained evaluation within a Safe | |
584 | module. See L<perlsec> for details about both these mechanisms. | |
871b0233 | 585 | |
14455d6c | 586 | =item C<(??{ code })> |
0f5d15d6 | 587 | |
19799a22 GS |
588 | B<WARNING>: This extended regular expression feature is considered |
589 | highly experimental, and may be changed or deleted without notice. | |
9da458fc IZ |
590 | A simplified version of the syntax may be introduced for commonly |
591 | used idioms. | |
0f5d15d6 | 592 | |
19799a22 GS |
593 | This is a "postponed" regular subexpression. The C<code> is evaluated |
594 | at run time, at the moment this subexpression may match. The result | |
595 | of evaluation is considered as a regular expression and matched as | |
596 | if it were inserted instead of this construct. | |
0f5d15d6 | 597 | |
428594d9 | 598 | The C<code> is not interpolated. As before, the rules to determine |
19799a22 GS |
599 | where the C<code> ends are currently somewhat convoluted. |
600 | ||
601 | The following pattern matches a parenthesized group: | |
0f5d15d6 IZ |
602 | |
603 | $re = qr{ | |
604 | \( | |
605 | (?: | |
606 | (?> [^()]+ ) # Non-parens without backtracking | |
607 | | | |
14455d6c | 608 | (??{ $re }) # Group with matching parens |
0f5d15d6 IZ |
609 | )* |
610 | \) | |
611 | }x; | |
612 | ||
c47ff5f1 | 613 | =item C<< (?>pattern) >> |
5a964f20 | 614 | |
19799a22 GS |
615 | B<WARNING>: This extended regular expression feature is considered |
616 | highly experimental, and may be changed or deleted without notice. | |
617 | ||
618 | An "independent" subexpression, one which matches the substring | |
619 | that a I<standalone> C<pattern> would match if anchored at the given | |
9da458fc | 620 | position, and it matches I<nothing other than this substring>. This |
19799a22 GS |
621 | construct is useful for optimizations of what would otherwise be |
622 | "eternal" matches, because it will not backtrack (see L<"Backtracking">). | |
9da458fc IZ |
623 | It may also be useful in places where the "grab all you can, and do not |
624 | give anything back" semantic is desirable. | |
19799a22 | 625 | |
c47ff5f1 | 626 | For example: C<< ^(?>a*)ab >> will never match, since C<< (?>a*) >> |
19799a22 GS |
627 | (anchored at the beginning of string, as above) will match I<all> |
628 | characters C<a> at the beginning of string, leaving no C<a> for | |
629 | C<ab> to match. In contrast, C<a*ab> will match the same as C<a+b>, | |
630 | since the match of the subgroup C<a*> is influenced by the following | |
631 | group C<ab> (see L<"Backtracking">). In particular, C<a*> inside | |
632 | C<a*ab> will match fewer characters than a standalone C<a*>, since | |
633 | this makes the tail match. | |
634 | ||
c47ff5f1 | 635 | An effect similar to C<< (?>pattern) >> may be achieved by writing |
19799a22 GS |
636 | C<(?=(pattern))\1>. This matches the same substring as a standalone |
637 | C<a+>, and the following C<\1> eats the matched string; it therefore | |
c47ff5f1 | 638 | makes a zero-length assertion into an analogue of C<< (?>...) >>. |
19799a22 GS |
639 | (The difference between these two constructs is that the second one |
640 | uses a capturing group, thus shifting ordinals of backreferences | |
641 | in the rest of a regular expression.) | |
642 | ||
643 | Consider this pattern: | |
c277df42 | 644 | |
871b0233 IZ |
645 | m{ \( |
646 | ( | |
9da458fc | 647 | [^()]+ # x+ |
871b0233 IZ |
648 | | |
649 | \( [^()]* \) | |
650 | )+ | |
651 | \) | |
652 | }x | |
5a964f20 | 653 | |
19799a22 GS |
654 | That will efficiently match a nonempty group with matching parentheses |
655 | two levels deep or less. However, if there is no such group, it | |
656 | will take virtually forever on a long string. That's because there | |
657 | are so many different ways to split a long string into several | |
658 | substrings. This is what C<(.+)+> is doing, and C<(.+)+> is similar | |
659 | to a subpattern of the above pattern. Consider how the pattern | |
660 | above detects no-match on C<((()aaaaaaaaaaaaaaaaaa> in several | |
661 | seconds, but that each extra letter doubles this time. This | |
662 | exponential performance will make it appear that your program has | |
14218588 | 663 | hung. However, a tiny change to this pattern |
5a964f20 | 664 | |
871b0233 IZ |
665 | m{ \( |
666 | ( | |
9da458fc | 667 | (?> [^()]+ ) # change x+ above to (?> x+ ) |
871b0233 IZ |
668 | | |
669 | \( [^()]* \) | |
670 | )+ | |
671 | \) | |
672 | }x | |
c277df42 | 673 | |
c47ff5f1 | 674 | which uses C<< (?>...) >> matches exactly when the one above does (verifying |
5a964f20 TC |
675 | this yourself would be a productive exercise), but finishes in a fourth |
676 | the time when used on a similar string with 1000000 C<a>s. Be aware, | |
677 | however, that this pattern currently triggers a warning message under | |
9f1b1f2d GS |
678 | the C<use warnings> pragma or B<-w> switch saying it |
679 | C<"matches the null string many times">): | |
c277df42 | 680 | |
c47ff5f1 | 681 | On simple groups, such as the pattern C<< (?> [^()]+ ) >>, a comparable |
19799a22 | 682 | effect may be achieved by negative look-ahead, as in C<[^()]+ (?! [^()] )>. |
c277df42 IZ |
683 | This was only 4 times slower on a string with 1000000 C<a>s. |
684 | ||
9da458fc IZ |
685 | The "grab all you can, and do not give anything back" semantic is desirable |
686 | in many situations where on the first sight a simple C<()*> looks like | |
687 | the correct solution. Suppose we parse text with comments being delimited | |
688 | by C<#> followed by some optional (horizontal) whitespace. Contrary to | |
4375e838 | 689 | its appearance, C<#[ \t]*> I<is not> the correct subexpression to match |
9da458fc IZ |
690 | the comment delimiter, because it may "give up" some whitespace if |
691 | the remainder of the pattern can be made to match that way. The correct | |
692 | answer is either one of these: | |
693 | ||
694 | (?>#[ \t]*) | |
695 | #[ \t]*(?![ \t]) | |
696 | ||
697 | For example, to grab non-empty comments into $1, one should use either | |
698 | one of these: | |
699 | ||
700 | / (?> \# [ \t]* ) ( .+ ) /x; | |
701 | / \# [ \t]* ( [^ \t] .* ) /x; | |
702 | ||
703 | Which one you pick depends on which of these expressions better reflects | |
704 | the above specification of comments. | |
705 | ||
5a964f20 | 706 | =item C<(?(condition)yes-pattern|no-pattern)> |
c277df42 | 707 | |
5a964f20 | 708 | =item C<(?(condition)yes-pattern)> |
c277df42 | 709 | |
19799a22 GS |
710 | B<WARNING>: This extended regular expression feature is considered |
711 | highly experimental, and may be changed or deleted without notice. | |
712 | ||
c277df42 IZ |
713 | Conditional expression. C<(condition)> should be either an integer in |
714 | parentheses (which is valid if the corresponding pair of parentheses | |
19799a22 | 715 | matched), or look-ahead/look-behind/evaluate zero-width assertion. |
c277df42 | 716 | |
19799a22 | 717 | For example: |
c277df42 | 718 | |
5a964f20 | 719 | m{ ( \( )? |
871b0233 | 720 | [^()]+ |
5a964f20 | 721 | (?(1) \) ) |
871b0233 | 722 | }x |
c277df42 IZ |
723 | |
724 | matches a chunk of non-parentheses, possibly included in parentheses | |
725 | themselves. | |
a0d0e21e | 726 | |
a0d0e21e LW |
727 | =back |
728 | ||
c07a80fd | 729 | =head2 Backtracking |
730 | ||
35a734be IZ |
731 | NOTE: This section presents an abstract approximation of regular |
732 | expression behavior. For a more rigorous (and complicated) view of | |
733 | the rules involved in selecting a match among possible alternatives, | |
734 | see L<Combining pieces together>. | |
735 | ||
c277df42 | 736 | A fundamental feature of regular expression matching involves the |
5a964f20 | 737 | notion called I<backtracking>, which is currently used (when needed) |
c277df42 | 738 | by all regular expression quantifiers, namely C<*>, C<*?>, C<+>, |
9da458fc IZ |
739 | C<+?>, C<{n,m}>, and C<{n,m}?>. Backtracking is often optimized |
740 | internally, but the general principle outlined here is valid. | |
c07a80fd | 741 | |
742 | For a regular expression to match, the I<entire> regular expression must | |
743 | match, not just part of it. So if the beginning of a pattern containing a | |
744 | quantifier succeeds in a way that causes later parts in the pattern to | |
745 | fail, the matching engine backs up and recalculates the beginning | |
746 | part--that's why it's called backtracking. | |
747 | ||
748 | Here is an example of backtracking: Let's say you want to find the | |
749 | word following "foo" in the string "Food is on the foo table.": | |
750 | ||
751 | $_ = "Food is on the foo table."; | |
752 | if ( /\b(foo)\s+(\w+)/i ) { | |
753 | print "$2 follows $1.\n"; | |
754 | } | |
755 | ||
756 | When the match runs, the first part of the regular expression (C<\b(foo)>) | |
757 | finds a possible match right at the beginning of the string, and loads up | |
758 | $1 with "Foo". However, as soon as the matching engine sees that there's | |
759 | no whitespace following the "Foo" that it had saved in $1, it realizes its | |
68dc0745 | 760 | mistake and starts over again one character after where it had the |
c07a80fd | 761 | tentative match. This time it goes all the way until the next occurrence |
762 | of "foo". The complete regular expression matches this time, and you get | |
763 | the expected output of "table follows foo." | |
764 | ||
765 | Sometimes minimal matching can help a lot. Imagine you'd like to match | |
766 | everything between "foo" and "bar". Initially, you write something | |
767 | like this: | |
768 | ||
769 | $_ = "The food is under the bar in the barn."; | |
770 | if ( /foo(.*)bar/ ) { | |
771 | print "got <$1>\n"; | |
772 | } | |
773 | ||
774 | Which perhaps unexpectedly yields: | |
775 | ||
776 | got <d is under the bar in the > | |
777 | ||
778 | That's because C<.*> was greedy, so you get everything between the | |
14218588 | 779 | I<first> "foo" and the I<last> "bar". Here it's more effective |
c07a80fd | 780 | to use minimal matching to make sure you get the text between a "foo" |
781 | and the first "bar" thereafter. | |
782 | ||
783 | if ( /foo(.*?)bar/ ) { print "got <$1>\n" } | |
784 | got <d is under the > | |
785 | ||
786 | Here's another example: let's say you'd like to match a number at the end | |
9b9391b2 | 787 | of a string, and you also want to keep the preceding of part the match. |
c07a80fd | 788 | So you write this: |
789 | ||
790 | $_ = "I have 2 numbers: 53147"; | |
791 | if ( /(.*)(\d*)/ ) { # Wrong! | |
792 | print "Beginning is <$1>, number is <$2>.\n"; | |
793 | } | |
794 | ||
795 | That won't work at all, because C<.*> was greedy and gobbled up the | |
796 | whole string. As C<\d*> can match on an empty string the complete | |
797 | regular expression matched successfully. | |
798 | ||
8e1088bc | 799 | Beginning is <I have 2 numbers: 53147>, number is <>. |
c07a80fd | 800 | |
801 | Here are some variants, most of which don't work: | |
802 | ||
803 | $_ = "I have 2 numbers: 53147"; | |
804 | @pats = qw{ | |
805 | (.*)(\d*) | |
806 | (.*)(\d+) | |
807 | (.*?)(\d*) | |
808 | (.*?)(\d+) | |
809 | (.*)(\d+)$ | |
810 | (.*?)(\d+)$ | |
811 | (.*)\b(\d+)$ | |
812 | (.*\D)(\d+)$ | |
813 | }; | |
814 | ||
815 | for $pat (@pats) { | |
816 | printf "%-12s ", $pat; | |
817 | if ( /$pat/ ) { | |
818 | print "<$1> <$2>\n"; | |
819 | } else { | |
820 | print "FAIL\n"; | |
821 | } | |
822 | } | |
823 | ||
824 | That will print out: | |
825 | ||
826 | (.*)(\d*) <I have 2 numbers: 53147> <> | |
827 | (.*)(\d+) <I have 2 numbers: 5314> <7> | |
828 | (.*?)(\d*) <> <> | |
829 | (.*?)(\d+) <I have > <2> | |
830 | (.*)(\d+)$ <I have 2 numbers: 5314> <7> | |
831 | (.*?)(\d+)$ <I have 2 numbers: > <53147> | |
832 | (.*)\b(\d+)$ <I have 2 numbers: > <53147> | |
833 | (.*\D)(\d+)$ <I have 2 numbers: > <53147> | |
834 | ||
835 | As you see, this can be a bit tricky. It's important to realize that a | |
836 | regular expression is merely a set of assertions that gives a definition | |
837 | of success. There may be 0, 1, or several different ways that the | |
838 | definition might succeed against a particular string. And if there are | |
5a964f20 TC |
839 | multiple ways it might succeed, you need to understand backtracking to |
840 | know which variety of success you will achieve. | |
c07a80fd | 841 | |
19799a22 | 842 | When using look-ahead assertions and negations, this can all get even |
54310121 | 843 | tricker. Imagine you'd like to find a sequence of non-digits not |
c07a80fd | 844 | followed by "123". You might try to write that as |
845 | ||
871b0233 IZ |
846 | $_ = "ABC123"; |
847 | if ( /^\D*(?!123)/ ) { # Wrong! | |
848 | print "Yup, no 123 in $_\n"; | |
849 | } | |
c07a80fd | 850 | |
851 | But that isn't going to match; at least, not the way you're hoping. It | |
852 | claims that there is no 123 in the string. Here's a clearer picture of | |
9b9391b2 | 853 | why that pattern matches, contrary to popular expectations: |
c07a80fd | 854 | |
855 | $x = 'ABC123' ; | |
856 | $y = 'ABC445' ; | |
857 | ||
858 | print "1: got $1\n" if $x =~ /^(ABC)(?!123)/ ; | |
859 | print "2: got $1\n" if $y =~ /^(ABC)(?!123)/ ; | |
860 | ||
861 | print "3: got $1\n" if $x =~ /^(\D*)(?!123)/ ; | |
862 | print "4: got $1\n" if $y =~ /^(\D*)(?!123)/ ; | |
863 | ||
864 | This prints | |
865 | ||
866 | 2: got ABC | |
867 | 3: got AB | |
868 | 4: got ABC | |
869 | ||
5f05dabc | 870 | You might have expected test 3 to fail because it seems to a more |
c07a80fd | 871 | general purpose version of test 1. The important difference between |
872 | them is that test 3 contains a quantifier (C<\D*>) and so can use | |
873 | backtracking, whereas test 1 will not. What's happening is | |
874 | that you've asked "Is it true that at the start of $x, following 0 or more | |
5f05dabc | 875 | non-digits, you have something that's not 123?" If the pattern matcher had |
c07a80fd | 876 | let C<\D*> expand to "ABC", this would have caused the whole pattern to |
54310121 | 877 | fail. |
14218588 | 878 | |
c07a80fd | 879 | The search engine will initially match C<\D*> with "ABC". Then it will |
14218588 | 880 | try to match C<(?!123> with "123", which fails. But because |
c07a80fd | 881 | a quantifier (C<\D*>) has been used in the regular expression, the |
882 | search engine can backtrack and retry the match differently | |
54310121 | 883 | in the hope of matching the complete regular expression. |
c07a80fd | 884 | |
5a964f20 TC |
885 | The pattern really, I<really> wants to succeed, so it uses the |
886 | standard pattern back-off-and-retry and lets C<\D*> expand to just "AB" this | |
c07a80fd | 887 | time. Now there's indeed something following "AB" that is not |
14218588 | 888 | "123". It's "C123", which suffices. |
c07a80fd | 889 | |
14218588 GS |
890 | We can deal with this by using both an assertion and a negation. |
891 | We'll say that the first part in $1 must be followed both by a digit | |
892 | and by something that's not "123". Remember that the look-aheads | |
893 | are zero-width expressions--they only look, but don't consume any | |
894 | of the string in their match. So rewriting this way produces what | |
c07a80fd | 895 | you'd expect; that is, case 5 will fail, but case 6 succeeds: |
896 | ||
897 | print "5: got $1\n" if $x =~ /^(\D*)(?=\d)(?!123)/ ; | |
898 | print "6: got $1\n" if $y =~ /^(\D*)(?=\d)(?!123)/ ; | |
899 | ||
900 | 6: got ABC | |
901 | ||
5a964f20 | 902 | In other words, the two zero-width assertions next to each other work as though |
19799a22 | 903 | they're ANDed together, just as you'd use any built-in assertions: C</^$/> |
c07a80fd | 904 | matches only if you're at the beginning of the line AND the end of the |
905 | line simultaneously. The deeper underlying truth is that juxtaposition in | |
906 | regular expressions always means AND, except when you write an explicit OR | |
907 | using the vertical bar. C</ab/> means match "a" AND (then) match "b", | |
908 | although the attempted matches are made at different positions because "a" | |
909 | is not a zero-width assertion, but a one-width assertion. | |
910 | ||
19799a22 | 911 | B<WARNING>: particularly complicated regular expressions can take |
14218588 | 912 | exponential time to solve because of the immense number of possible |
9da458fc IZ |
913 | ways they can use backtracking to try match. For example, without |
914 | internal optimizations done by the regular expression engine, this will | |
915 | take a painfully long time to run: | |
c07a80fd | 916 | |
e1901655 IZ |
917 | 'aaaaaaaaaaaa' =~ /((a{0,5}){0,5})*[c]/ |
918 | ||
919 | And if you used C<*>'s in the internal groups instead of limiting them | |
920 | to 0 through 5 matches, then it would take forever--or until you ran | |
921 | out of stack space. Moreover, these internal optimizations are not | |
922 | always applicable. For example, if you put C<{0,5}> instead of C<*> | |
923 | on the external group, no current optimization is applicable, and the | |
924 | match takes a long time to finish. | |
c07a80fd | 925 | |
9da458fc IZ |
926 | A powerful tool for optimizing such beasts is what is known as an |
927 | "independent group", | |
c47ff5f1 | 928 | which does not backtrack (see L<C<< (?>pattern) >>>). Note also that |
9da458fc | 929 | zero-length look-ahead/look-behind assertions will not backtrack to make |
14218588 GS |
930 | the tail match, since they are in "logical" context: only |
931 | whether they match is considered relevant. For an example | |
9da458fc | 932 | where side-effects of look-ahead I<might> have influenced the |
c47ff5f1 | 933 | following match, see L<C<< (?>pattern) >>>. |
c277df42 | 934 | |
a0d0e21e LW |
935 | =head2 Version 8 Regular Expressions |
936 | ||
5a964f20 | 937 | In case you're not familiar with the "regular" Version 8 regex |
a0d0e21e LW |
938 | routines, here are the pattern-matching rules not described above. |
939 | ||
54310121 | 940 | Any single character matches itself, unless it is a I<metacharacter> |
a0d0e21e | 941 | with a special meaning described here or above. You can cause |
5a964f20 | 942 | characters that normally function as metacharacters to be interpreted |
5f05dabc | 943 | literally by prefixing them with a "\" (e.g., "\." matches a ".", not any |
a0d0e21e LW |
944 | character; "\\" matches a "\"). A series of characters matches that |
945 | series of characters in the target string, so the pattern C<blurfl> | |
946 | would match "blurfl" in the target string. | |
947 | ||
948 | You can specify a character class, by enclosing a list of characters | |
5a964f20 | 949 | in C<[]>, which will match any one character from the list. If the |
a0d0e21e | 950 | first character after the "[" is "^", the class matches any character not |
14218588 | 951 | in the list. Within a list, the "-" character specifies a |
5a964f20 | 952 | range, so that C<a-z> represents all characters between "a" and "z", |
8a4f6ac2 GS |
953 | inclusive. If you want either "-" or "]" itself to be a member of a |
954 | class, put it at the start of the list (possibly after a "^"), or | |
955 | escape it with a backslash. "-" is also taken literally when it is | |
956 | at the end of the list, just before the closing "]". (The | |
84850974 DD |
957 | following all specify the same class of three characters: C<[-az]>, |
958 | C<[az-]>, and C<[a\-z]>. All are different from C<[a-z]>, which | |
fb55449c JH |
959 | specifies a class containing twenty-six characters, even on EBCDIC |
960 | based coded character sets.) Also, if you try to use the character | |
961 | classes C<\w>, C<\W>, C<\s>, C<\S>, C<\d>, or C<\D> as endpoints of | |
962 | a range, that's not a range, the "-" is understood literally. | |
a0d0e21e | 963 | |
8ada0baa JH |
964 | Note also that the whole range idea is rather unportable between |
965 | character sets--and even within character sets they may cause results | |
966 | you probably didn't expect. A sound principle is to use only ranges | |
967 | that begin from and end at either alphabets of equal case ([a-e], | |
968 | [A-E]), or digits ([0-9]). Anything else is unsafe. If in doubt, | |
969 | spell out the character sets in full. | |
970 | ||
54310121 | 971 | Characters may be specified using a metacharacter syntax much like that |
a0d0e21e LW |
972 | used in C: "\n" matches a newline, "\t" a tab, "\r" a carriage return, |
973 | "\f" a form feed, etc. More generally, \I<nnn>, where I<nnn> is a string | |
fb55449c JH |
974 | of octal digits, matches the character whose coded character set value |
975 | is I<nnn>. Similarly, \xI<nn>, where I<nn> are hexadecimal digits, | |
976 | matches the character whose numeric value is I<nn>. The expression \cI<x> | |
977 | matches the character control-I<x>. Finally, the "." metacharacter | |
978 | matches any character except "\n" (unless you use C</s>). | |
a0d0e21e LW |
979 | |
980 | You can specify a series of alternatives for a pattern using "|" to | |
981 | separate them, so that C<fee|fie|foe> will match any of "fee", "fie", | |
5a964f20 | 982 | or "foe" in the target string (as would C<f(e|i|o)e>). The |
a0d0e21e LW |
983 | first alternative includes everything from the last pattern delimiter |
984 | ("(", "[", or the beginning of the pattern) up to the first "|", and | |
985 | the last alternative contains everything from the last "|" to the next | |
14218588 GS |
986 | pattern delimiter. That's why it's common practice to include |
987 | alternatives in parentheses: to minimize confusion about where they | |
a3cb178b GS |
988 | start and end. |
989 | ||
5a964f20 | 990 | Alternatives are tried from left to right, so the first |
a3cb178b GS |
991 | alternative found for which the entire expression matches, is the one that |
992 | is chosen. This means that alternatives are not necessarily greedy. For | |
628afcb5 | 993 | example: when matching C<foo|foot> against "barefoot", only the "foo" |
a3cb178b GS |
994 | part will match, as that is the first alternative tried, and it successfully |
995 | matches the target string. (This might not seem important, but it is | |
996 | important when you are capturing matched text using parentheses.) | |
997 | ||
5a964f20 | 998 | Also remember that "|" is interpreted as a literal within square brackets, |
a3cb178b | 999 | so if you write C<[fee|fie|foe]> you're really only matching C<[feio|]>. |
a0d0e21e | 1000 | |
14218588 GS |
1001 | Within a pattern, you may designate subpatterns for later reference |
1002 | by enclosing them in parentheses, and you may refer back to the | |
1003 | I<n>th subpattern later in the pattern using the metacharacter | |
1004 | \I<n>. Subpatterns are numbered based on the left to right order | |
1005 | of their opening parenthesis. A backreference matches whatever | |
1006 | actually matched the subpattern in the string being examined, not | |
1007 | the rules for that subpattern. Therefore, C<(0|0x)\d*\s\1\d*> will | |
1008 | match "0x1234 0x4321", but not "0x1234 01234", because subpattern | |
1009 | 1 matched "0x", even though the rule C<0|0x> could potentially match | |
1010 | the leading 0 in the second number. | |
cb1a09d0 | 1011 | |
19799a22 | 1012 | =head2 Warning on \1 vs $1 |
cb1a09d0 | 1013 | |
5a964f20 | 1014 | Some people get too used to writing things like: |
cb1a09d0 AD |
1015 | |
1016 | $pattern =~ s/(\W)/\\\1/g; | |
1017 | ||
1018 | This is grandfathered for the RHS of a substitute to avoid shocking the | |
1019 | B<sed> addicts, but it's a dirty habit to get into. That's because in | |
5f05dabc | 1020 | PerlThink, the righthand side of a C<s///> is a double-quoted string. C<\1> in |
cb1a09d0 AD |
1021 | the usual double-quoted string means a control-A. The customary Unix |
1022 | meaning of C<\1> is kludged in for C<s///>. However, if you get into the habit | |
1023 | of doing that, you get yourself into trouble if you then add an C</e> | |
1024 | modifier. | |
1025 | ||
5a964f20 | 1026 | s/(\d+)/ \1 + 1 /eg; # causes warning under -w |
cb1a09d0 AD |
1027 | |
1028 | Or if you try to do | |
1029 | ||
1030 | s/(\d+)/\1000/; | |
1031 | ||
1032 | You can't disambiguate that by saying C<\{1}000>, whereas you can fix it with | |
14218588 | 1033 | C<${1}000>. The operation of interpolation should not be confused |
cb1a09d0 AD |
1034 | with the operation of matching a backreference. Certainly they mean two |
1035 | different things on the I<left> side of the C<s///>. | |
9fa51da4 | 1036 | |
c84d73f1 IZ |
1037 | =head2 Repeated patterns matching zero-length substring |
1038 | ||
19799a22 | 1039 | B<WARNING>: Difficult material (and prose) ahead. This section needs a rewrite. |
c84d73f1 IZ |
1040 | |
1041 | Regular expressions provide a terse and powerful programming language. As | |
1042 | with most other power tools, power comes together with the ability | |
1043 | to wreak havoc. | |
1044 | ||
1045 | A common abuse of this power stems from the ability to make infinite | |
628afcb5 | 1046 | loops using regular expressions, with something as innocuous as: |
c84d73f1 IZ |
1047 | |
1048 | 'foo' =~ m{ ( o? )* }x; | |
1049 | ||
1050 | The C<o?> can match at the beginning of C<'foo'>, and since the position | |
1051 | in the string is not moved by the match, C<o?> would match again and again | |
14218588 | 1052 | because of the C<*> modifier. Another common way to create a similar cycle |
c84d73f1 IZ |
1053 | is with the looping modifier C<//g>: |
1054 | ||
1055 | @matches = ( 'foo' =~ m{ o? }xg ); | |
1056 | ||
1057 | or | |
1058 | ||
1059 | print "match: <$&>\n" while 'foo' =~ m{ o? }xg; | |
1060 | ||
1061 | or the loop implied by split(). | |
1062 | ||
1063 | However, long experience has shown that many programming tasks may | |
14218588 GS |
1064 | be significantly simplified by using repeated subexpressions that |
1065 | may match zero-length substrings. Here's a simple example being: | |
c84d73f1 IZ |
1066 | |
1067 | @chars = split //, $string; # // is not magic in split | |
1068 | ($whitewashed = $string) =~ s/()/ /g; # parens avoid magic s// / | |
1069 | ||
9da458fc | 1070 | Thus Perl allows such constructs, by I<forcefully breaking |
c84d73f1 IZ |
1071 | the infinite loop>. The rules for this are different for lower-level |
1072 | loops given by the greedy modifiers C<*+{}>, and for higher-level | |
1073 | ones like the C</g> modifier or split() operator. | |
1074 | ||
19799a22 GS |
1075 | The lower-level loops are I<interrupted> (that is, the loop is |
1076 | broken) when Perl detects that a repeated expression matched a | |
1077 | zero-length substring. Thus | |
c84d73f1 IZ |
1078 | |
1079 | m{ (?: NON_ZERO_LENGTH | ZERO_LENGTH )* }x; | |
1080 | ||
1081 | is made equivalent to | |
1082 | ||
1083 | m{ (?: NON_ZERO_LENGTH )* | |
1084 | | | |
1085 | (?: ZERO_LENGTH )? | |
1086 | }x; | |
1087 | ||
1088 | The higher level-loops preserve an additional state between iterations: | |
1089 | whether the last match was zero-length. To break the loop, the following | |
1090 | match after a zero-length match is prohibited to have a length of zero. | |
1091 | This prohibition interacts with backtracking (see L<"Backtracking">), | |
1092 | and so the I<second best> match is chosen if the I<best> match is of | |
1093 | zero length. | |
1094 | ||
19799a22 | 1095 | For example: |
c84d73f1 IZ |
1096 | |
1097 | $_ = 'bar'; | |
1098 | s/\w??/<$&>/g; | |
1099 | ||
20fb949f | 1100 | results in C<< <><b><><a><><r><> >>. At each position of the string the best |
c84d73f1 IZ |
1101 | match given by non-greedy C<??> is the zero-length match, and the I<second |
1102 | best> match is what is matched by C<\w>. Thus zero-length matches | |
1103 | alternate with one-character-long matches. | |
1104 | ||
1105 | Similarly, for repeated C<m/()/g> the second-best match is the match at the | |
1106 | position one notch further in the string. | |
1107 | ||
19799a22 | 1108 | The additional state of being I<matched with zero-length> is associated with |
c84d73f1 | 1109 | the matched string, and is reset by each assignment to pos(). |
9da458fc IZ |
1110 | Zero-length matches at the end of the previous match are ignored |
1111 | during C<split>. | |
c84d73f1 | 1112 | |
35a734be IZ |
1113 | =head2 Combining pieces together |
1114 | ||
1115 | Each of the elementary pieces of regular expressions which were described | |
1116 | before (such as C<ab> or C<\Z>) could match at most one substring | |
1117 | at the given position of the input string. However, in a typical regular | |
1118 | expression these elementary pieces are combined into more complicated | |
1119 | patterns using combining operators C<ST>, C<S|T>, C<S*> etc | |
1120 | (in these examples C<S> and C<T> are regular subexpressions). | |
1121 | ||
1122 | Such combinations can include alternatives, leading to a problem of choice: | |
1123 | if we match a regular expression C<a|ab> against C<"abc">, will it match | |
1124 | substring C<"a"> or C<"ab">? One way to describe which substring is | |
1125 | actually matched is the concept of backtracking (see L<"Backtracking">). | |
1126 | However, this description is too low-level and makes you think | |
1127 | in terms of a particular implementation. | |
1128 | ||
1129 | Another description starts with notions of "better"/"worse". All the | |
1130 | substrings which may be matched by the given regular expression can be | |
1131 | sorted from the "best" match to the "worst" match, and it is the "best" | |
1132 | match which is chosen. This substitutes the question of "what is chosen?" | |
1133 | by the question of "which matches are better, and which are worse?". | |
1134 | ||
1135 | Again, for elementary pieces there is no such question, since at most | |
1136 | one match at a given position is possible. This section describes the | |
1137 | notion of better/worse for combining operators. In the description | |
1138 | below C<S> and C<T> are regular subexpressions. | |
1139 | ||
13a2d996 | 1140 | =over 4 |
35a734be IZ |
1141 | |
1142 | =item C<ST> | |
1143 | ||
1144 | Consider two possible matches, C<AB> and C<A'B'>, C<A> and C<A'> are | |
1145 | substrings which can be matched by C<S>, C<B> and C<B'> are substrings | |
1146 | which can be matched by C<T>. | |
1147 | ||
1148 | If C<A> is better match for C<S> than C<A'>, C<AB> is a better | |
1149 | match than C<A'B'>. | |
1150 | ||
1151 | If C<A> and C<A'> coincide: C<AB> is a better match than C<AB'> if | |
1152 | C<B> is better match for C<T> than C<B'>. | |
1153 | ||
1154 | =item C<S|T> | |
1155 | ||
1156 | When C<S> can match, it is a better match than when only C<T> can match. | |
1157 | ||
1158 | Ordering of two matches for C<S> is the same as for C<S>. Similar for | |
1159 | two matches for C<T>. | |
1160 | ||
1161 | =item C<S{REPEAT_COUNT}> | |
1162 | ||
1163 | Matches as C<SSS...S> (repeated as many times as necessary). | |
1164 | ||
1165 | =item C<S{min,max}> | |
1166 | ||
1167 | Matches as C<S{max}|S{max-1}|...|S{min+1}|S{min}>. | |
1168 | ||
1169 | =item C<S{min,max}?> | |
1170 | ||
1171 | Matches as C<S{min}|S{min+1}|...|S{max-1}|S{max}>. | |
1172 | ||
1173 | =item C<S?>, C<S*>, C<S+> | |
1174 | ||
1175 | Same as C<S{0,1}>, C<S{0,BIG_NUMBER}>, C<S{1,BIG_NUMBER}> respectively. | |
1176 | ||
1177 | =item C<S??>, C<S*?>, C<S+?> | |
1178 | ||
1179 | Same as C<S{0,1}?>, C<S{0,BIG_NUMBER}?>, C<S{1,BIG_NUMBER}?> respectively. | |
1180 | ||
c47ff5f1 | 1181 | =item C<< (?>S) >> |
35a734be IZ |
1182 | |
1183 | Matches the best match for C<S> and only that. | |
1184 | ||
1185 | =item C<(?=S)>, C<(?<=S)> | |
1186 | ||
1187 | Only the best match for C<S> is considered. (This is important only if | |
1188 | C<S> has capturing parentheses, and backreferences are used somewhere | |
1189 | else in the whole regular expression.) | |
1190 | ||
1191 | =item C<(?!S)>, C<(?<!S)> | |
1192 | ||
1193 | For this grouping operator there is no need to describe the ordering, since | |
1194 | only whether or not C<S> can match is important. | |
1195 | ||
14455d6c | 1196 | =item C<(??{ EXPR })> |
35a734be IZ |
1197 | |
1198 | The ordering is the same as for the regular expression which is | |
1199 | the result of EXPR. | |
1200 | ||
1201 | =item C<(?(condition)yes-pattern|no-pattern)> | |
1202 | ||
1203 | Recall that which of C<yes-pattern> or C<no-pattern> actually matches is | |
1204 | already determined. The ordering of the matches is the same as for the | |
1205 | chosen subexpression. | |
1206 | ||
1207 | =back | |
1208 | ||
1209 | The above recipes describe the ordering of matches I<at a given position>. | |
1210 | One more rule is needed to understand how a match is determined for the | |
1211 | whole regular expression: a match at an earlier position is always better | |
1212 | than a match at a later position. | |
1213 | ||
c84d73f1 IZ |
1214 | =head2 Creating custom RE engines |
1215 | ||
1216 | Overloaded constants (see L<overload>) provide a simple way to extend | |
1217 | the functionality of the RE engine. | |
1218 | ||
1219 | Suppose that we want to enable a new RE escape-sequence C<\Y|> which | |
1220 | matches at boundary between white-space characters and non-whitespace | |
1221 | characters. Note that C<(?=\S)(?<!\S)|(?!\S)(?<=\S)> matches exactly | |
1222 | at these positions, so we want to have each C<\Y|> in the place of the | |
1223 | more complicated version. We can create a module C<customre> to do | |
1224 | this: | |
1225 | ||
1226 | package customre; | |
1227 | use overload; | |
1228 | ||
1229 | sub import { | |
1230 | shift; | |
1231 | die "No argument to customre::import allowed" if @_; | |
1232 | overload::constant 'qr' => \&convert; | |
1233 | } | |
1234 | ||
1235 | sub invalid { die "/$_[0]/: invalid escape '\\$_[1]'"} | |
1236 | ||
1237 | my %rules = ( '\\' => '\\', | |
1238 | 'Y|' => qr/(?=\S)(?<!\S)|(?!\S)(?<=\S)/ ); | |
1239 | sub convert { | |
1240 | my $re = shift; | |
1241 | $re =~ s{ | |
1242 | \\ ( \\ | Y . ) | |
1243 | } | |
1244 | { $rules{$1} or invalid($re,$1) }sgex; | |
1245 | return $re; | |
1246 | } | |
1247 | ||
1248 | Now C<use customre> enables the new escape in constant regular | |
1249 | expressions, i.e., those without any runtime variable interpolations. | |
1250 | As documented in L<overload>, this conversion will work only over | |
1251 | literal parts of regular expressions. For C<\Y|$re\Y|> the variable | |
1252 | part of this regular expression needs to be converted explicitly | |
1253 | (but only if the special meaning of C<\Y|> should be enabled inside $re): | |
1254 | ||
1255 | use customre; | |
1256 | $re = <>; | |
1257 | chomp $re; | |
1258 | $re = customre::convert $re; | |
1259 | /\Y|$re\Y|/; | |
1260 | ||
19799a22 GS |
1261 | =head1 BUGS |
1262 | ||
9da458fc IZ |
1263 | This document varies from difficult to understand to completely |
1264 | and utterly opaque. The wandering prose riddled with jargon is | |
1265 | hard to fathom in several places. | |
1266 | ||
1267 | This document needs a rewrite that separates the tutorial content | |
1268 | from the reference content. | |
19799a22 GS |
1269 | |
1270 | =head1 SEE ALSO | |
9fa51da4 | 1271 | |
9b599b2a GS |
1272 | L<perlop/"Regexp Quote-Like Operators">. |
1273 | ||
1e66bd83 PP |
1274 | L<perlop/"Gory details of parsing quoted constructs">. |
1275 | ||
14218588 GS |
1276 | L<perlfaq6>. |
1277 | ||
9b599b2a GS |
1278 | L<perlfunc/pos>. |
1279 | ||
1280 | L<perllocale>. | |
1281 | ||
fb55449c JH |
1282 | L<perlebcdic>. |
1283 | ||
14218588 GS |
1284 | I<Mastering Regular Expressions> by Jeffrey Friedl, published |
1285 | by O'Reilly and Associates. |