3 perlretut - Perl regular expressions tutorial
7 This page provides a basic tutorial on understanding, creating and
8 using regular expressions in Perl. It serves as a complement to the
9 reference page on regular expressions L<perlre>. Regular expressions
10 are an integral part of the C<m//>, C<s///>, C<qr//> and C<split>
11 operators and so this tutorial also overlaps with
12 L<perlop/"Regexp Quote-Like Operators"> and L<perlfunc/split>.
14 Perl is widely renowned for excellence in text processing, and regular
15 expressions are one of the big factors behind this fame. Perl regular
16 expressions display an efficiency and flexibility unknown in most
17 other computer languages. Mastering even the basics of regular
18 expressions will allow you to manipulate text with surprising ease.
20 What is a regular expression? A regular expression is simply a string
21 that describes a pattern. Patterns are in common use these days;
22 examples are the patterns typed into a search engine to find web pages
23 and the patterns used to list files in a directory, e.g., C<ls *.txt>
24 or C<dir *.*>. In Perl, the patterns described by regular expressions
25 are used to search strings, extract desired parts of strings, and to
26 do search and replace operations.
28 Regular expressions have the undeserved reputation of being abstract
29 and difficult to understand. Regular expressions are constructed using
30 simple concepts like conditionals and loops and are no more difficult
31 to understand than the corresponding C<if> conditionals and C<while>
32 loops in the Perl language itself. In fact, the main challenge in
33 learning regular expressions is just getting used to the terse
34 notation used to express these concepts.
36 This tutorial flattens the learning curve by discussing regular
37 expression concepts, along with their notation, one at a time and with
38 many examples. The first part of the tutorial will progress from the
39 simplest word searches to the basic regular expression concepts. If
40 you master the first part, you will have all the tools needed to solve
41 about 98% of your needs. The second part of the tutorial is for those
42 comfortable with the basics and hungry for more power tools. It
43 discusses the more advanced regular expression operators and
44 introduces the latest cutting-edge innovations.
46 A note: to save time, 'regular expression' is often abbreviated as
47 regexp or regex. Regexp is a more natural abbreviation than regex, but
48 is harder to pronounce. The Perl pod documentation is evenly split on
49 regexp vs regex; in Perl, there is more than one way to abbreviate it.
50 We'll use regexp in this tutorial.
52 =head1 Part 1: The basics
54 =head2 Simple word matching
56 The simplest regexp is simply a word, or more generally, a string of
57 characters. A regexp consisting of a word matches any string that
60 "Hello World" =~ /World/; # matches
62 What is this Perl statement all about? C<"Hello World"> is a simple
63 double-quoted string. C<World> is the regular expression and the
64 C<//> enclosing C</World/> tells Perl to search a string for a match.
65 The operator C<=~> associates the string with the regexp match and
66 produces a true value if the regexp matched, or false if the regexp
67 did not match. In our case, C<World> matches the second word in
68 C<"Hello World">, so the expression is true. Expressions like this
69 are useful in conditionals:
71 if ("Hello World" =~ /World/) {
75 print "It doesn't match\n";
78 There are useful variations on this theme. The sense of the match can
79 be reversed by using the C<!~> operator:
81 if ("Hello World" !~ /World/) {
82 print "It doesn't match\n";
88 The literal string in the regexp can be replaced by a variable:
91 if ("Hello World" =~ /$greeting/) {
95 print "It doesn't match\n";
98 If you're matching against the special default variable C<$_>, the
99 C<$_ =~> part can be omitted:
103 print "It matches\n";
106 print "It doesn't match\n";
109 And finally, the C<//> default delimiters for a match can be changed
110 to arbitrary delimiters by putting an C<'m'> out front:
112 "Hello World" =~ m!World!; # matches, delimited by '!'
113 "Hello World" =~ m{World}; # matches, note the matching '{}'
114 "/usr/bin/perl" =~ m"/perl"; # matches after '/usr/bin',
115 # '/' becomes an ordinary char
117 C</World/>, C<m!World!>, and C<m{World}> all represent the
118 same thing. When, e.g., the quote (C<">) is used as a delimiter, the forward
119 slash C<'/'> becomes an ordinary character and can be used in this regexp
122 Let's consider how different regexps would match C<"Hello World">:
124 "Hello World" =~ /world/; # doesn't match
125 "Hello World" =~ /o W/; # matches
126 "Hello World" =~ /oW/; # doesn't match
127 "Hello World" =~ /World /; # doesn't match
129 The first regexp C<world> doesn't match because regexps are
130 case-sensitive. The second regexp matches because the substring
131 S<C<'o W'>> occurs in the string S<C<"Hello World">>. The space
132 character ' ' is treated like any other character in a regexp and is
133 needed to match in this case. The lack of a space character is the
134 reason the third regexp C<'oW'> doesn't match. The fourth regexp
135 C<'World '> doesn't match because there is a space at the end of the
136 regexp, but not at the end of the string. The lesson here is that
137 regexps must match a part of the string I<exactly> in order for the
138 statement to be true.
140 If a regexp matches in more than one place in the string, Perl will
141 always match at the earliest possible point in the string:
143 "Hello World" =~ /o/; # matches 'o' in 'Hello'
144 "That hat is red" =~ /hat/; # matches 'hat' in 'That'
146 With respect to character matching, there are a few more points you
147 need to know about. First of all, not all characters can be used 'as
148 is' in a match. Some characters, called I<metacharacters>, are reserved
149 for use in regexp notation. The metacharacters are
153 The significance of each of these will be explained
154 in the rest of the tutorial, but for now, it is important only to know
155 that a metacharacter can be matched by putting a backslash before it:
157 "2+2=4" =~ /2+2/; # doesn't match, + is a metacharacter
158 "2+2=4" =~ /2\+2/; # matches, \+ is treated like an ordinary +
159 "The interval is [0,1)." =~ /[0,1)./ # is a syntax error!
160 "The interval is [0,1)." =~ /\[0,1\)\./ # matches
161 "#!/usr/bin/perl" =~ /#!\/usr\/bin\/perl/; # matches
163 In the last regexp, the forward slash C<'/'> is also backslashed,
164 because it is used to delimit the regexp. This can lead to LTS
165 (leaning toothpick syndrome), however, and it is often more readable
166 to change delimiters.
168 "#!/usr/bin/perl" =~ m!#\!/usr/bin/perl!; # easier to read
170 The backslash character C<'\'> is a metacharacter itself and needs to
173 'C:\WIN32' =~ /C:\\WIN/; # matches
175 In addition to the metacharacters, there are some ASCII characters
176 which don't have printable character equivalents and are instead
177 represented by I<escape sequences>. Common examples are C<\t> for a
178 tab, C<\n> for a newline, C<\r> for a carriage return and C<\a> for a
179 bell (or alert). If your string is better thought of as a sequence of arbitrary
180 bytes, the octal escape sequence, e.g., C<\033>, or hexadecimal escape
181 sequence, e.g., C<\x1B> may be a more natural representation for your
182 bytes. Here are some examples of escapes:
184 "1000\t2000" =~ m(0\t2) # matches
185 "1000\n2000" =~ /0\n20/ # matches
186 "1000\t2000" =~ /\000\t2/ # doesn't match, "0" ne "\000"
187 "cat" =~ /\o{143}\x61\x74/ # matches in ASCII, but a weird way
190 If you've been around Perl a while, all this talk of escape sequences
191 may seem familiar. Similar escape sequences are used in double-quoted
192 strings and in fact the regexps in Perl are mostly treated as
193 double-quoted strings. This means that variables can be used in
194 regexps as well. Just like double-quoted strings, the values of the
195 variables in the regexp will be substituted in before the regexp is
196 evaluated for matching purposes. So we have:
199 'housecat' =~ /$foo/; # matches
200 'cathouse' =~ /cat$foo/; # matches
201 'housecat' =~ /${foo}cat/; # matches
203 So far, so good. With the knowledge above you can already perform
204 searches with just about any literal string regexp you can dream up.
205 Here is a I<very simple> emulation of the Unix grep program:
215 % chmod +x simple_grep
217 % simple_grep abba /usr/dict/words
228 This program is easy to understand. C<#!/usr/bin/perl> is the standard
229 way to invoke a perl program from the shell.
230 S<C<$regexp = shift;>> saves the first command line argument as the
231 regexp to be used, leaving the rest of the command line arguments to
232 be treated as files. S<C<< while (<>) >>> loops over all the lines in
233 all the files. For each line, S<C<print if /$regexp/;>> prints the
234 line if the regexp matches the line. In this line, both C<print> and
235 C</$regexp/> use the default variable C<$_> implicitly.
237 With all of the regexps above, if the regexp matched anywhere in the
238 string, it was considered a match. Sometimes, however, we'd like to
239 specify I<where> in the string the regexp should try to match. To do
240 this, we would use the I<anchor> metacharacters C<^> and C<$>. The
241 anchor C<^> means match at the beginning of the string and the anchor
242 C<$> means match at the end of the string, or before a newline at the
243 end of the string. Here is how they are used:
245 "housekeeper" =~ /keeper/; # matches
246 "housekeeper" =~ /^keeper/; # doesn't match
247 "housekeeper" =~ /keeper$/; # matches
248 "housekeeper\n" =~ /keeper$/; # matches
250 The second regexp doesn't match because C<^> constrains C<keeper> to
251 match only at the beginning of the string, but C<"housekeeper"> has
252 keeper starting in the middle. The third regexp does match, since the
253 C<$> constrains C<keeper> to match only at the end of the string.
255 When both C<^> and C<$> are used at the same time, the regexp has to
256 match both the beginning and the end of the string, i.e., the regexp
257 matches the whole string. Consider
259 "keeper" =~ /^keep$/; # doesn't match
260 "keeper" =~ /^keeper$/; # matches
261 "" =~ /^$/; # ^$ matches an empty string
263 The first regexp doesn't match because the string has more to it than
264 C<keep>. Since the second regexp is exactly the string, it
265 matches. Using both C<^> and C<$> in a regexp forces the complete
266 string to match, so it gives you complete control over which strings
267 match and which don't. Suppose you are looking for a fellow named
268 bert, off in a string by himself:
270 "dogbert" =~ /bert/; # matches, but not what you want
272 "dilbert" =~ /^bert/; # doesn't match, but ..
273 "bertram" =~ /^bert/; # matches, so still not good enough
275 "bertram" =~ /^bert$/; # doesn't match, good
276 "dilbert" =~ /^bert$/; # doesn't match, good
277 "bert" =~ /^bert$/; # matches, perfect
279 Of course, in the case of a literal string, one could just as easily
280 use the string comparison S<C<$string eq 'bert'>> and it would be
281 more efficient. The C<^...$> regexp really becomes useful when we
282 add in the more powerful regexp tools below.
284 =head2 Using character classes
286 Although one can already do quite a lot with the literal string
287 regexps above, we've only scratched the surface of regular expression
288 technology. In this and subsequent sections we will introduce regexp
289 concepts (and associated metacharacter notations) that will allow a
290 regexp to represent not just a single character sequence, but a I<whole
293 One such concept is that of a I<character class>. A character class
294 allows a set of possible characters, rather than just a single
295 character, to match at a particular point in a regexp. You can define
296 your own custom character classes. These
297 are denoted by brackets C<[...]>, with the set of characters
298 to be possibly matched inside. Here are some examples:
300 /cat/; # matches 'cat'
301 /[bcr]at/; # matches 'bat, 'cat', or 'rat'
302 /item[0123456789]/; # matches 'item0' or ... or 'item9'
303 "abc" =~ /[cab]/; # matches 'a'
305 In the last statement, even though C<'c'> is the first character in
306 the class, C<'a'> matches because the first character position in the
307 string is the earliest point at which the regexp can match.
309 /[yY][eE][sS]/; # match 'yes' in a case-insensitive way
310 # 'yes', 'Yes', 'YES', etc.
312 This regexp displays a common task: perform a case-insensitive
313 match. Perl provides a way of avoiding all those brackets by simply
314 appending an C<'i'> to the end of the match. Then C</[yY][eE][sS]/;>
315 can be rewritten as C</yes/i;>. The C<'i'> stands for
316 case-insensitive and is an example of a I<modifier> of the matching
317 operation. We will meet other modifiers later in the tutorial.
319 We saw in the section above that there were ordinary characters, which
320 represented themselves, and special characters, which needed a
321 backslash C<\> to represent themselves. The same is true in a
322 character class, but the sets of ordinary and special characters
323 inside a character class are different than those outside a character
324 class. The special characters for a character class are C<-]\^$> (and
325 the pattern delimiter, whatever it is).
326 C<]> is special because it denotes the end of a character class. C<$> is
327 special because it denotes a scalar variable. C<\> is special because
328 it is used in escape sequences, just like above. Here is how the
329 special characters C<]$\> are handled:
331 /[\]c]def/; # matches ']def' or 'cdef'
333 /[$x]at/; # matches 'bat', 'cat', or 'rat'
334 /[\$x]at/; # matches '$at' or 'xat'
335 /[\\$x]at/; # matches '\at', 'bat, 'cat', or 'rat'
337 The last two are a little tricky. In C<[\$x]>, the backslash protects
338 the dollar sign, so the character class has two members C<$> and C<x>.
339 In C<[\\$x]>, the backslash is protected, so C<$x> is treated as a
340 variable and substituted in double quote fashion.
342 The special character C<'-'> acts as a range operator within character
343 classes, so that a contiguous set of characters can be written as a
344 range. With ranges, the unwieldy C<[0123456789]> and C<[abc...xyz]>
345 become the svelte C<[0-9]> and C<[a-z]>. Some examples are
347 /item[0-9]/; # matches 'item0' or ... or 'item9'
348 /[0-9bx-z]aa/; # matches '0aa', ..., '9aa',
349 # 'baa', 'xaa', 'yaa', or 'zaa'
350 /[0-9a-fA-F]/; # matches a hexadecimal digit
351 /[0-9a-zA-Z_]/; # matches a "word" character,
352 # like those in a Perl variable name
354 If C<'-'> is the first or last character in a character class, it is
355 treated as an ordinary character; C<[-ab]>, C<[ab-]> and C<[a\-b]> are
358 The special character C<^> in the first position of a character class
359 denotes a I<negated character class>, which matches any character but
360 those in the brackets. Both C<[...]> and C<[^...]> must match a
361 character, or the match fails. Then
363 /[^a]at/; # doesn't match 'aat' or 'at', but matches
364 # all other 'bat', 'cat, '0at', '%at', etc.
365 /[^0-9]/; # matches a non-numeric character
366 /[a^]at/; # matches 'aat' or '^at'; here '^' is ordinary
368 Now, even C<[0-9]> can be a bother to write multiple times, so in the
369 interest of saving keystrokes and making regexps more readable, Perl
370 has several abbreviations for common character classes, as shown below.
371 Since the introduction of Unicode, unless the C<//a> modifier is in
372 effect, these character classes match more than just a few characters in
379 \d matches a digit, not just [0-9] but also digits from non-roman scripts
383 \s matches a whitespace character, the set [\ \t\r\n\f] and others
387 \w matches a word character (alphanumeric or _), not just [0-9a-zA-Z_]
388 but also digits and characters from non-roman scripts
392 \D is a negated \d; it represents any other character than a digit, or [^\d]
396 \S is a negated \s; it represents any non-whitespace character [^\s]
400 \W is a negated \w; it represents any non-word character [^\w]
404 The period '.' matches any character but "\n" (unless the modifier C<//s> is
405 in effect, as explained below).
409 \N, like the period, matches any character but "\n", but it does so
410 regardless of whether the modifier C<//s> is in effect.
414 The C<//a> modifier, available starting in Perl 5.14, is used to
415 restrict the matches of \d, \s, and \w to just those in the ASCII range.
416 It is useful to keep your program from being needlessly exposed to full
417 Unicode (and its accompanying security considerations) when all you want
418 is to process English-like text. (The "a" may be doubled, C<//aa>, to
419 provide even more restrictions, preventing case-insensitive matching of
420 ASCII with non-ASCII characters; otherwise a Unicode "Kelvin Sign"
421 would caselessly match a "k" or "K".)
423 The C<\d\s\w\D\S\W> abbreviations can be used both inside and outside
424 of bracketed character classes. Here are some in use:
426 /\d\d:\d\d:\d\d/; # matches a hh:mm:ss time format
427 /[\d\s]/; # matches any digit or whitespace character
428 /\w\W\w/; # matches a word char, followed by a
429 # non-word char, followed by a word char
430 /..rt/; # matches any two chars, followed by 'rt'
431 /end\./; # matches 'end.'
432 /end[.]/; # same thing, matches 'end.'
434 Because a period is a metacharacter, it needs to be escaped to match
435 as an ordinary period. Because, for example, C<\d> and C<\w> are sets
436 of characters, it is incorrect to think of C<[^\d\w]> as C<[\D\W]>; in
437 fact C<[^\d\w]> is the same as C<[^\w]>, which is the same as
438 C<[\W]>. Think DeMorgan's laws.
440 In actuality, the period and C<\d\s\w\D\S\W> abbreviations are
441 themselves types of character classes, so the ones surrounded by
442 brackets are just one type of character class. When we need to make a
443 distinction, we refer to them as "bracketed character classes."
445 An anchor useful in basic regexps is the I<word anchor>
446 C<\b>. This matches a boundary between a word character and a non-word
447 character C<\w\W> or C<\W\w>:
449 $x = "Housecat catenates house and cat";
450 $x =~ /cat/; # matches cat in 'housecat'
451 $x =~ /\bcat/; # matches cat in 'catenates'
452 $x =~ /cat\b/; # matches cat in 'housecat'
453 $x =~ /\bcat\b/; # matches 'cat' at end of string
455 Note in the last example, the end of the string is considered a word
458 You might wonder why C<'.'> matches everything but C<"\n"> - why not
459 every character? The reason is that often one is matching against
460 lines and would like to ignore the newline characters. For instance,
461 while the string C<"\n"> represents one line, we would like to think
464 "" =~ /^$/; # matches
465 "\n" =~ /^$/; # matches, $ anchors before "\n"
467 "" =~ /./; # doesn't match; it needs a char
468 "" =~ /^.$/; # doesn't match; it needs a char
469 "\n" =~ /^.$/; # doesn't match; it needs a char other than "\n"
470 "a" =~ /^.$/; # matches
471 "a\n" =~ /^.$/; # matches, $ anchors before "\n"
473 This behavior is convenient, because we usually want to ignore
474 newlines when we count and match characters in a line. Sometimes,
475 however, we want to keep track of newlines. We might even want C<^>
476 and C<$> to anchor at the beginning and end of lines within the
477 string, rather than just the beginning and end of the string. Perl
478 allows us to choose between ignoring and paying attention to newlines
479 by using the C<//s> and C<//m> modifiers. C<//s> and C<//m> stand for
480 single line and multi-line and they determine whether a string is to
481 be treated as one continuous string, or as a set of lines. The two
482 modifiers affect two aspects of how the regexp is interpreted: 1) how
483 the C<'.'> character class is defined, and 2) where the anchors C<^>
484 and C<$> are able to match. Here are the four possible combinations:
490 no modifiers (//): Default behavior. C<'.'> matches any character
491 except C<"\n">. C<^> matches only at the beginning of the string and
492 C<$> matches only at the end or before a newline at the end.
496 s modifier (//s): Treat string as a single long line. C<'.'> matches
497 any character, even C<"\n">. C<^> matches only at the beginning of
498 the string and C<$> matches only at the end or before a newline at the
503 m modifier (//m): Treat string as a set of multiple lines. C<'.'>
504 matches any character except C<"\n">. C<^> and C<$> are able to match
505 at the start or end of I<any> line within the string.
509 both s and m modifiers (//sm): Treat string as a single long line, but
510 detect multiple lines. C<'.'> matches any character, even
511 C<"\n">. C<^> and C<$>, however, are able to match at the start or end
512 of I<any> line within the string.
516 Here are examples of C<//s> and C<//m> in action:
518 $x = "There once was a girl\nWho programmed in Perl\n";
520 $x =~ /^Who/; # doesn't match, "Who" not at start of string
521 $x =~ /^Who/s; # doesn't match, "Who" not at start of string
522 $x =~ /^Who/m; # matches, "Who" at start of second line
523 $x =~ /^Who/sm; # matches, "Who" at start of second line
525 $x =~ /girl.Who/; # doesn't match, "." doesn't match "\n"
526 $x =~ /girl.Who/s; # matches, "." matches "\n"
527 $x =~ /girl.Who/m; # doesn't match, "." doesn't match "\n"
528 $x =~ /girl.Who/sm; # matches, "." matches "\n"
530 Most of the time, the default behavior is what is wanted, but C<//s> and
531 C<//m> are occasionally very useful. If C<//m> is being used, the start
532 of the string can still be matched with C<\A> and the end of the string
533 can still be matched with the anchors C<\Z> (matches both the end and
534 the newline before, like C<$>), and C<\z> (matches only the end):
536 $x =~ /^Who/m; # matches, "Who" at start of second line
537 $x =~ /\AWho/m; # doesn't match, "Who" is not at start of string
539 $x =~ /girl$/m; # matches, "girl" at end of first line
540 $x =~ /girl\Z/m; # doesn't match, "girl" is not at end of string
542 $x =~ /Perl\Z/m; # matches, "Perl" is at newline before end
543 $x =~ /Perl\z/m; # doesn't match, "Perl" is not at end of string
545 We now know how to create choices among classes of characters in a
546 regexp. What about choices among words or character strings? Such
547 choices are described in the next section.
549 =head2 Matching this or that
551 Sometimes we would like our regexp to be able to match different
552 possible words or character strings. This is accomplished by using
553 the I<alternation> metacharacter C<|>. To match C<dog> or C<cat>, we
554 form the regexp C<dog|cat>. As before, Perl will try to match the
555 regexp at the earliest possible point in the string. At each
556 character position, Perl will first try to match the first
557 alternative, C<dog>. If C<dog> doesn't match, Perl will then try the
558 next alternative, C<cat>. If C<cat> doesn't match either, then the
559 match fails and Perl moves to the next position in the string. Some
562 "cats and dogs" =~ /cat|dog|bird/; # matches "cat"
563 "cats and dogs" =~ /dog|cat|bird/; # matches "cat"
565 Even though C<dog> is the first alternative in the second regexp,
566 C<cat> is able to match earlier in the string.
568 "cats" =~ /c|ca|cat|cats/; # matches "c"
569 "cats" =~ /cats|cat|ca|c/; # matches "cats"
571 Here, all the alternatives match at the first string position, so the
572 first alternative is the one that matches. If some of the
573 alternatives are truncations of the others, put the longest ones first
574 to give them a chance to match.
576 "cab" =~ /a|b|c/ # matches "c"
579 The last example points out that character classes are like
580 alternations of characters. At a given character position, the first
581 alternative that allows the regexp match to succeed will be the one
584 =head2 Grouping things and hierarchical matching
586 Alternation allows a regexp to choose among alternatives, but by
587 itself it is unsatisfying. The reason is that each alternative is a whole
588 regexp, but sometime we want alternatives for just part of a
589 regexp. For instance, suppose we want to search for housecats or
590 housekeepers. The regexp C<housecat|housekeeper> fits the bill, but is
591 inefficient because we had to type C<house> twice. It would be nice to
592 have parts of the regexp be constant, like C<house>, and some
593 parts have alternatives, like C<cat|keeper>.
595 The I<grouping> metacharacters C<()> solve this problem. Grouping
596 allows parts of a regexp to be treated as a single unit. Parts of a
597 regexp are grouped by enclosing them in parentheses. Thus we could solve
598 the C<housecat|housekeeper> by forming the regexp as
599 C<house(cat|keeper)>. The regexp C<house(cat|keeper)> means match
600 C<house> followed by either C<cat> or C<keeper>. Some more examples
603 /(a|b)b/; # matches 'ab' or 'bb'
604 /(ac|b)b/; # matches 'acb' or 'bb'
605 /(^a|b)c/; # matches 'ac' at start of string or 'bc' anywhere
606 /(a|[bc])d/; # matches 'ad', 'bd', or 'cd'
608 /house(cat|)/; # matches either 'housecat' or 'house'
609 /house(cat(s|)|)/; # matches either 'housecats' or 'housecat' or
610 # 'house'. Note groups can be nested.
612 /(19|20|)\d\d/; # match years 19xx, 20xx, or the Y2K problem, xx
613 "20" =~ /(19|20|)\d\d/; # matches the null alternative '()\d\d',
614 # because '20\d\d' can't match
616 Alternations behave the same way in groups as out of them: at a given
617 string position, the leftmost alternative that allows the regexp to
618 match is taken. So in the last example at the first string position,
619 C<"20"> matches the second alternative, but there is nothing left over
620 to match the next two digits C<\d\d>. So Perl moves on to the next
621 alternative, which is the null alternative and that works, since
622 C<"20"> is two digits.
624 The process of trying one alternative, seeing if it matches, and
625 moving on to the next alternative, while going back in the string
626 from where the previous alternative was tried, if it doesn't, is called
627 I<backtracking>. The term 'backtracking' comes from the idea that
628 matching a regexp is like a walk in the woods. Successfully matching
629 a regexp is like arriving at a destination. There are many possible
630 trailheads, one for each string position, and each one is tried in
631 order, left to right. From each trailhead there may be many paths,
632 some of which get you there, and some which are dead ends. When you
633 walk along a trail and hit a dead end, you have to backtrack along the
634 trail to an earlier point to try another trail. If you hit your
635 destination, you stop immediately and forget about trying all the
636 other trails. You are persistent, and only if you have tried all the
637 trails from all the trailheads and not arrived at your destination, do
638 you declare failure. To be concrete, here is a step-by-step analysis
639 of what Perl does when it tries to match the regexp
641 "abcde" =~ /(abd|abc)(df|d|de)/;
647 Start with the first letter in the string 'a'.
651 Try the first alternative in the first group 'abd'.
655 Match 'a' followed by 'b'. So far so good.
659 'd' in the regexp doesn't match 'c' in the string - a dead
660 end. So backtrack two characters and pick the second alternative in
661 the first group 'abc'.
665 Match 'a' followed by 'b' followed by 'c'. We are on a roll
666 and have satisfied the first group. Set $1 to 'abc'.
670 Move on to the second group and pick the first alternative
679 'f' in the regexp doesn't match 'e' in the string, so a dead
680 end. Backtrack one character and pick the second alternative in the
685 'd' matches. The second grouping is satisfied, so set $2 to
690 We are at the end of the regexp, so we are done! We have
691 matched 'abcd' out of the string "abcde".
695 There are a couple of things to note about this analysis. First, the
696 third alternative in the second group 'de' also allows a match, but we
697 stopped before we got to it - at a given character position, leftmost
698 wins. Second, we were able to get a match at the first character
699 position of the string 'a'. If there were no matches at the first
700 position, Perl would move to the second character position 'b' and
701 attempt the match all over again. Only when all possible paths at all
702 possible character positions have been exhausted does Perl give
703 up and declare S<C<$string =~ /(abd|abc)(df|d|de)/;>> to be false.
705 Even with all this work, regexp matching happens remarkably fast. To
706 speed things up, Perl compiles the regexp into a compact sequence of
707 opcodes that can often fit inside a processor cache. When the code is
708 executed, these opcodes can then run at full throttle and search very
711 =head2 Extracting matches
713 The grouping metacharacters C<()> also serve another completely
714 different function: they allow the extraction of the parts of a string
715 that matched. This is very useful to find out what matched and for
716 text processing in general. For each grouping, the part that matched
717 inside goes into the special variables C<$1>, C<$2>, etc. They can be
718 used just as ordinary variables:
720 # extract hours, minutes, seconds
721 if ($time =~ /(\d\d):(\d\d):(\d\d)/) { # match hh:mm:ss format
727 Now, we know that in scalar context,
728 S<C<$time =~ /(\d\d):(\d\d):(\d\d)/>> returns a true or false
729 value. In list context, however, it returns the list of matched values
730 C<($1,$2,$3)>. So we could write the code more compactly as
732 # extract hours, minutes, seconds
733 ($hours, $minutes, $second) = ($time =~ /(\d\d):(\d\d):(\d\d)/);
735 If the groupings in a regexp are nested, C<$1> gets the group with the
736 leftmost opening parenthesis, C<$2> the next opening parenthesis,
737 etc. Here is a regexp with nested groups:
739 /(ab(cd|ef)((gi)|j))/;
742 If this regexp matches, C<$1> contains a string starting with
743 C<'ab'>, C<$2> is either set to C<'cd'> or C<'ef'>, C<$3> equals either
744 C<'gi'> or C<'j'>, and C<$4> is either set to C<'gi'>, just like C<$3>,
745 or it remains undefined.
747 For convenience, Perl sets C<$+> to the string held by the highest numbered
748 C<$1>, C<$2>,... that got assigned (and, somewhat related, C<$^N> to the
749 value of the C<$1>, C<$2>,... most-recently assigned; i.e. the C<$1>,
750 C<$2>,... associated with the rightmost closing parenthesis used in the
754 =head2 Backreferences
756 Closely associated with the matching variables C<$1>, C<$2>, ... are
757 the I<backreferences> C<\g1>, C<\g2>,... Backreferences are simply
758 matching variables that can be used I<inside> a regexp. This is a
759 really nice feature; what matches later in a regexp is made to depend on
760 what matched earlier in the regexp. Suppose we wanted to look
761 for doubled words in a text, like 'the the'. The following regexp finds
762 all 3-letter doubles with a space in between:
766 The grouping assigns a value to \g1, so that the same 3-letter sequence
767 is used for both parts.
769 A similar task is to find words consisting of two identical parts:
771 % simple_grep '^(\w\w\w\w|\w\w\w|\w\w|\w)\g1$' /usr/dict/words
779 The regexp has a single grouping which considers 4-letter
780 combinations, then 3-letter combinations, etc., and uses C<\g1> to look for
781 a repeat. Although C<$1> and C<\g1> represent the same thing, care should be
782 taken to use matched variables C<$1>, C<$2>,... only I<outside> a regexp
783 and backreferences C<\g1>, C<\g2>,... only I<inside> a regexp; not doing
784 so may lead to surprising and unsatisfactory results.
787 =head2 Relative backreferences
789 Counting the opening parentheses to get the correct number for a
790 backreference is error-prone as soon as there is more than one
791 capturing group. A more convenient technique became available
792 with Perl 5.10: relative backreferences. To refer to the immediately
793 preceding capture group one now may write C<\g{-1}>, the next but
794 last is available via C<\g{-2}>, and so on.
796 Another good reason in addition to readability and maintainability
797 for using relative backreferences is illustrated by the following example,
798 where a simple pattern for matching peculiar strings is used:
800 $a99a = '([a-z])(\d)\g2\g1'; # matches a11a, g22g, x33x, etc.
802 Now that we have this pattern stored as a handy string, we might feel
803 tempted to use it as a part of some other pattern:
806 if ($line =~ /^(\w+)=$a99a$/){ # unexpected behavior!
807 print "$1 is valid\n";
809 print "bad line: '$line'\n";
812 But this doesn't match, at least not the way one might expect. Only
813 after inserting the interpolated C<$a99a> and looking at the resulting
814 full text of the regexp is it obvious that the backreferences have
815 backfired. The subexpression C<(\w+)> has snatched number 1 and
816 demoted the groups in C<$a99a> by one rank. This can be avoided by
817 using relative backreferences:
819 $a99a = '([a-z])(\d)\g{-1}\g{-2}'; # safe for being interpolated
822 =head2 Named backreferences
824 Perl 5.10 also introduced named capture groups and named backreferences.
825 To attach a name to a capturing group, you write either
826 C<< (?<name>...) >> or C<< (?'name'...) >>. The backreference may
827 then be written as C<\g{name}>. It is permissible to attach the
828 same name to more than one group, but then only the leftmost one of the
829 eponymous set can be referenced. Outside of the pattern a named
830 capture group is accessible through the C<%+> hash.
832 Assuming that we have to match calendar dates which may be given in one
833 of the three formats yyyy-mm-dd, mm/dd/yyyy or dd.mm.yyyy, we can write
834 three suitable patterns where we use 'd', 'm' and 'y' respectively as the
835 names of the groups capturing the pertaining components of a date. The
836 matching operation combines the three patterns as alternatives:
838 $fmt1 = '(?<y>\d\d\d\d)-(?<m>\d\d)-(?<d>\d\d)';
839 $fmt2 = '(?<m>\d\d)/(?<d>\d\d)/(?<y>\d\d\d\d)';
840 $fmt3 = '(?<d>\d\d)\.(?<m>\d\d)\.(?<y>\d\d\d\d)';
841 for my $d qw( 2006-10-21 15.01.2007 10/31/2005 ){
842 if ( $d =~ m{$fmt1|$fmt2|$fmt3} ){
843 print "day=$+{d} month=$+{m} year=$+{y}\n";
847 If any of the alternatives matches, the hash C<%+> is bound to contain the
848 three key-value pairs.
851 =head2 Alternative capture group numbering
853 Yet another capturing group numbering technique (also as from Perl 5.10)
854 deals with the problem of referring to groups within a set of alternatives.
855 Consider a pattern for matching a time of the day, civil or military style:
857 if ( $time =~ /(\d\d|\d):(\d\d)|(\d\d)(\d\d)/ ){
858 # process hour and minute
861 Processing the results requires an additional if statement to determine
862 whether C<$1> and C<$2> or C<$3> and C<$4> contain the goodies. It would
863 be easier if we could use group numbers 1 and 2 in second alternative as
864 well, and this is exactly what the parenthesized construct C<(?|...)>,
865 set around an alternative achieves. Here is an extended version of the
868 if($time =~ /(?|(\d\d|\d):(\d\d)|(\d\d)(\d\d))\s+([A-Z][A-Z][A-Z])/){
869 print "hour=$1 minute=$2 zone=$3\n";
872 Within the alternative numbering group, group numbers start at the same
873 position for each alternative. After the group, numbering continues
874 with one higher than the maximum reached across all the alternatives.
876 =head2 Position information
878 In addition to what was matched, Perl also provides the
879 positions of what was matched as contents of the C<@-> and C<@+>
880 arrays. C<$-[0]> is the position of the start of the entire match and
881 C<$+[0]> is the position of the end. Similarly, C<$-[n]> is the
882 position of the start of the C<$n> match and C<$+[n]> is the position
883 of the end. If C<$n> is undefined, so are C<$-[n]> and C<$+[n]>. Then
886 $x = "Mmm...donut, thought Homer";
887 $x =~ /^(Mmm|Yech)\.\.\.(donut|peas)/; # matches
888 foreach $exp (1..$#-) {
889 print "Match $exp: '${$exp}' at position ($-[$exp],$+[$exp])\n";
894 Match 1: 'Mmm' at position (0,3)
895 Match 2: 'donut' at position (6,11)
897 Even if there are no groupings in a regexp, it is still possible to
898 find out what exactly matched in a string. If you use them, Perl
899 will set C<$`> to the part of the string before the match, will set C<$&>
900 to the part of the string that matched, and will set C<$'> to the part
901 of the string after the match. An example:
903 $x = "the cat caught the mouse";
904 $x =~ /cat/; # $` = 'the ', $& = 'cat', $' = ' caught the mouse'
905 $x =~ /the/; # $` = '', $& = 'the', $' = ' cat caught the mouse'
907 In the second match, C<$`> equals C<''> because the regexp matched at the
908 first character position in the string and stopped; it never saw the
911 If your code is to run on Perl versions earlier than
912 5.20, it is worthwhile to note that using C<$`> and C<$'>
913 slows down regexp matching quite a bit, while C<$&> slows it down to a
914 lesser extent, because if they are used in one regexp in a program,
915 they are generated for I<all> regexps in the program. So if raw
916 performance is a goal of your application, they should be avoided.
917 If you need to extract the corresponding substrings, use C<@-> and
920 $` is the same as substr( $x, 0, $-[0] )
921 $& is the same as substr( $x, $-[0], $+[0]-$-[0] )
922 $' is the same as substr( $x, $+[0] )
924 As of Perl 5.10, the C<${^PREMATCH}>, C<${^MATCH}> and C<${^POSTMATCH}>
925 variables may be used. These are only set if the C</p> modifier is
926 present. Consequently they do not penalize the rest of the program. In
927 Perl 5.20, C<${^PREMATCH}>, C<${^MATCH}> and C<${^POSTMATCH}> are available
928 whether the C</p> has been used or not (the modifier is ignored), and
929 C<$`>, C<$'> and C<$&> do not cause any speed difference.
931 =head2 Non-capturing groupings
933 A group that is required to bundle a set of alternatives may or may not be
934 useful as a capturing group. If it isn't, it just creates a superfluous
935 addition to the set of available capture group values, inside as well as
936 outside the regexp. Non-capturing groupings, denoted by C<(?:regexp)>,
937 still allow the regexp to be treated as a single unit, but don't establish
938 a capturing group at the same time. Both capturing and non-capturing
939 groupings are allowed to co-exist in the same regexp. Because there is
940 no extraction, non-capturing groupings are faster than capturing
941 groupings. Non-capturing groupings are also handy for choosing exactly
942 which parts of a regexp are to be extracted to matching variables:
944 # match a number, $1-$4 are set, but we only want $1
945 /([+-]?\ *(\d+(\.\d*)?|\.\d+)([eE][+-]?\d+)?)/;
947 # match a number faster , only $1 is set
948 /([+-]?\ *(?:\d+(?:\.\d*)?|\.\d+)(?:[eE][+-]?\d+)?)/;
950 # match a number, get $1 = whole number, $2 = exponent
951 /([+-]?\ *(?:\d+(?:\.\d*)?|\.\d+)(?:[eE]([+-]?\d+))?)/;
953 Non-capturing groupings are also useful for removing nuisance
954 elements gathered from a split operation where parentheses are
955 required for some reason:
958 @num = split /(a|b)+/, $x; # @num = ('12','a','34','a','5')
959 @num = split /(?:a|b)+/, $x; # @num = ('12','34','5')
962 =head2 Matching repetitions
964 The examples in the previous section display an annoying weakness. We
965 were only matching 3-letter words, or chunks of words of 4 letters or
966 less. We'd like to be able to match words or, more generally, strings
967 of any length, without writing out tedious alternatives like
968 C<\w\w\w\w|\w\w\w|\w\w|\w>.
970 This is exactly the problem the I<quantifier> metacharacters C<?>,
971 C<*>, C<+>, and C<{}> were created for. They allow us to delimit the
972 number of repeats for a portion of a regexp we consider to be a
973 match. Quantifiers are put immediately after the character, character
974 class, or grouping that we want to specify. They have the following
981 C<a?> means: match 'a' 1 or 0 times
985 C<a*> means: match 'a' 0 or more times, i.e., any number of times
989 C<a+> means: match 'a' 1 or more times, i.e., at least once
993 C<a{n,m}> means: match at least C<n> times, but not more than C<m>
998 C<a{n,}> means: match at least C<n> or more times
1002 C<a{n}> means: match exactly C<n> times
1006 Here are some examples:
1008 /[a-z]+\s+\d*/; # match a lowercase word, at least one space, and
1009 # any number of digits
1010 /(\w+)\s+\g1/; # match doubled words of arbitrary length
1011 /y(es)?/i; # matches 'y', 'Y', or a case-insensitive 'yes'
1012 $year =~ /^\d{2,4}$/; # make sure year is at least 2 but not more
1014 $year =~ /^\d{4}$|^\d{2}$/; # better match; throw out 3-digit dates
1015 $year =~ /^\d{2}(\d{2})?$/; # same thing written differently.
1016 # However, this captures the last two
1017 # digits in $1 and the other does not.
1019 % simple_grep '^(\w+)\g1$' /usr/dict/words # isn't this easier?
1027 For all of these quantifiers, Perl will try to match as much of the
1028 string as possible, while still allowing the regexp to succeed. Thus
1029 with C</a?.../>, Perl will first try to match the regexp with the C<a>
1030 present; if that fails, Perl will try to match the regexp without the
1031 C<a> present. For the quantifier C<*>, we get the following:
1033 $x = "the cat in the hat";
1034 $x =~ /^(.*)(cat)(.*)$/; # matches,
1037 # $3 = ' in the hat'
1039 Which is what we might expect, the match finds the only C<cat> in the
1040 string and locks onto it. Consider, however, this regexp:
1042 $x =~ /^(.*)(at)(.*)$/; # matches,
1043 # $1 = 'the cat in the h'
1045 # $3 = '' (0 characters match)
1047 One might initially guess that Perl would find the C<at> in C<cat> and
1048 stop there, but that wouldn't give the longest possible string to the
1049 first quantifier C<.*>. Instead, the first quantifier C<.*> grabs as
1050 much of the string as possible while still having the regexp match. In
1051 this example, that means having the C<at> sequence with the final C<at>
1052 in the string. The other important principle illustrated here is that,
1053 when there are two or more elements in a regexp, the I<leftmost>
1054 quantifier, if there is one, gets to grab as much of the string as
1055 possible, leaving the rest of the regexp to fight over scraps. Thus in
1056 our example, the first quantifier C<.*> grabs most of the string, while
1057 the second quantifier C<.*> gets the empty string. Quantifiers that
1058 grab as much of the string as possible are called I<maximal match> or
1059 I<greedy> quantifiers.
1061 When a regexp can match a string in several different ways, we can use
1062 the principles above to predict which way the regexp will match:
1068 Principle 0: Taken as a whole, any regexp will be matched at the
1069 earliest possible position in the string.
1073 Principle 1: In an alternation C<a|b|c...>, the leftmost alternative
1074 that allows a match for the whole regexp will be the one used.
1078 Principle 2: The maximal matching quantifiers C<?>, C<*>, C<+> and
1079 C<{n,m}> will in general match as much of the string as possible while
1080 still allowing the whole regexp to match.
1084 Principle 3: If there are two or more elements in a regexp, the
1085 leftmost greedy quantifier, if any, will match as much of the string
1086 as possible while still allowing the whole regexp to match. The next
1087 leftmost greedy quantifier, if any, will try to match as much of the
1088 string remaining available to it as possible, while still allowing the
1089 whole regexp to match. And so on, until all the regexp elements are
1094 As we have seen above, Principle 0 overrides the others. The regexp
1095 will be matched as early as possible, with the other principles
1096 determining how the regexp matches at that earliest character
1099 Here is an example of these principles in action:
1101 $x = "The programming republic of Perl";
1102 $x =~ /^(.+)(e|r)(.*)$/; # matches,
1103 # $1 = 'The programming republic of Pe'
1107 This regexp matches at the earliest string position, C<'T'>. One
1108 might think that C<e>, being leftmost in the alternation, would be
1109 matched, but C<r> produces the longest string in the first quantifier.
1111 $x =~ /(m{1,2})(.*)$/; # matches,
1113 # $2 = 'ing republic of Perl'
1115 Here, The earliest possible match is at the first C<'m'> in
1116 C<programming>. C<m{1,2}> is the first quantifier, so it gets to match
1119 $x =~ /.*(m{1,2})(.*)$/; # matches,
1121 # $2 = 'ing republic of Perl'
1123 Here, the regexp matches at the start of the string. The first
1124 quantifier C<.*> grabs as much as possible, leaving just a single
1125 C<'m'> for the second quantifier C<m{1,2}>.
1127 $x =~ /(.?)(m{1,2})(.*)$/; # matches,
1130 # $3 = 'ing republic of Perl'
1132 Here, C<.?> eats its maximal one character at the earliest possible
1133 position in the string, C<'a'> in C<programming>, leaving C<m{1,2}>
1134 the opportunity to match both C<m>'s. Finally,
1136 "aXXXb" =~ /(X*)/; # matches with $1 = ''
1138 because it can match zero copies of C<'X'> at the beginning of the
1139 string. If you definitely want to match at least one C<'X'>, use
1142 Sometimes greed is not good. At times, we would like quantifiers to
1143 match a I<minimal> piece of string, rather than a maximal piece. For
1144 this purpose, Larry Wall created the I<minimal match> or
1145 I<non-greedy> quantifiers C<??>, C<*?>, C<+?>, and C<{}?>. These are
1146 the usual quantifiers with a C<?> appended to them. They have the
1153 C<a??> means: match 'a' 0 or 1 times. Try 0 first, then 1.
1157 C<a*?> means: match 'a' 0 or more times, i.e., any number of times,
1158 but as few times as possible
1162 C<a+?> means: match 'a' 1 or more times, i.e., at least once, but
1163 as few times as possible
1167 C<a{n,m}?> means: match at least C<n> times, not more than C<m>
1168 times, as few times as possible
1172 C<a{n,}?> means: match at least C<n> times, but as few times as
1177 C<a{n}?> means: match exactly C<n> times. Because we match exactly
1178 C<n> times, C<a{n}?> is equivalent to C<a{n}> and is just there for
1179 notational consistency.
1183 Let's look at the example above, but with minimal quantifiers:
1185 $x = "The programming republic of Perl";
1186 $x =~ /^(.+?)(e|r)(.*)$/; # matches,
1189 # $3 = ' programming republic of Perl'
1191 The minimal string that will allow both the start of the string C<^>
1192 and the alternation to match is C<Th>, with the alternation C<e|r>
1193 matching C<e>. The second quantifier C<.*> is free to gobble up the
1196 $x =~ /(m{1,2}?)(.*?)$/; # matches,
1198 # $2 = 'ming republic of Perl'
1200 The first string position that this regexp can match is at the first
1201 C<'m'> in C<programming>. At this position, the minimal C<m{1,2}?>
1202 matches just one C<'m'>. Although the second quantifier C<.*?> would
1203 prefer to match no characters, it is constrained by the end-of-string
1204 anchor C<$> to match the rest of the string.
1206 $x =~ /(.*?)(m{1,2}?)(.*)$/; # matches,
1209 # $3 = 'ming republic of Perl'
1211 In this regexp, you might expect the first minimal quantifier C<.*?>
1212 to match the empty string, because it is not constrained by a C<^>
1213 anchor to match the beginning of the word. Principle 0 applies here,
1214 however. Because it is possible for the whole regexp to match at the
1215 start of the string, it I<will> match at the start of the string. Thus
1216 the first quantifier has to match everything up to the first C<m>. The
1217 second minimal quantifier matches just one C<m> and the third
1218 quantifier matches the rest of the string.
1220 $x =~ /(.??)(m{1,2})(.*)$/; # matches,
1223 # $3 = 'ing republic of Perl'
1225 Just as in the previous regexp, the first quantifier C<.??> can match
1226 earliest at position C<'a'>, so it does. The second quantifier is
1227 greedy, so it matches C<mm>, and the third matches the rest of the
1230 We can modify principle 3 above to take into account non-greedy
1237 Principle 3: If there are two or more elements in a regexp, the
1238 leftmost greedy (non-greedy) quantifier, if any, will match as much
1239 (little) of the string as possible while still allowing the whole
1240 regexp to match. The next leftmost greedy (non-greedy) quantifier, if
1241 any, will try to match as much (little) of the string remaining
1242 available to it as possible, while still allowing the whole regexp to
1243 match. And so on, until all the regexp elements are satisfied.
1247 Just like alternation, quantifiers are also susceptible to
1248 backtracking. Here is a step-by-step analysis of the example
1250 $x = "the cat in the hat";
1251 $x =~ /^(.*)(at)(.*)$/; # matches,
1252 # $1 = 'the cat in the h'
1254 # $3 = '' (0 matches)
1260 Start with the first letter in the string 't'.
1264 The first quantifier '.*' starts out by matching the whole
1265 string 'the cat in the hat'.
1269 'a' in the regexp element 'at' doesn't match the end of the
1270 string. Backtrack one character.
1274 'a' in the regexp element 'at' still doesn't match the last
1275 letter of the string 't', so backtrack one more character.
1279 Now we can match the 'a' and the 't'.
1283 Move on to the third element '.*'. Since we are at the end of
1284 the string and '.*' can match 0 times, assign it the empty string.
1292 Most of the time, all this moving forward and backtracking happens
1293 quickly and searching is fast. There are some pathological regexps,
1294 however, whose execution time exponentially grows with the size of the
1295 string. A typical structure that blows up in your face is of the form
1299 The problem is the nested indeterminate quantifiers. There are many
1300 different ways of partitioning a string of length n between the C<+>
1301 and C<*>: one repetition with C<b+> of length n, two repetitions with
1302 the first C<b+> length k and the second with length n-k, m repetitions
1303 whose bits add up to length n, etc. In fact there are an exponential
1304 number of ways to partition a string as a function of its length. A
1305 regexp may get lucky and match early in the process, but if there is
1306 no match, Perl will try I<every> possibility before giving up. So be
1307 careful with nested C<*>'s, C<{n,m}>'s, and C<+>'s. The book
1308 I<Mastering Regular Expressions> by Jeffrey Friedl gives a wonderful
1309 discussion of this and other efficiency issues.
1312 =head2 Possessive quantifiers
1314 Backtracking during the relentless search for a match may be a waste
1315 of time, particularly when the match is bound to fail. Consider
1318 /^\w+\s+\w+$/; # a word, spaces, a word
1320 Whenever this is applied to a string which doesn't quite meet the
1321 pattern's expectations such as S<C<"abc ">> or S<C<"abc def ">>,
1322 the regex engine will backtrack, approximately once for each character
1323 in the string. But we know that there is no way around taking I<all>
1324 of the initial word characters to match the first repetition, that I<all>
1325 spaces must be eaten by the middle part, and the same goes for the second
1328 With the introduction of the I<possessive quantifiers> in Perl 5.10, we
1329 have a way of instructing the regex engine not to backtrack, with the
1330 usual quantifiers with a C<+> appended to them. This makes them greedy as
1331 well as stingy; once they succeed they won't give anything back to permit
1332 another solution. They have the following meanings:
1338 C<a{n,m}+> means: match at least C<n> times, not more than C<m> times,
1339 as many times as possible, and don't give anything up. C<a?+> is short
1344 C<a{n,}+> means: match at least C<n> times, but as many times as possible,
1345 and don't give anything up. C<a*+> is short for C<a{0,}+> and C<a++> is
1346 short for C<a{1,}+>.
1350 C<a{n}+> means: match exactly C<n> times. It is just there for
1351 notational consistency.
1355 These possessive quantifiers represent a special case of a more general
1356 concept, the I<independent subexpression>, see below.
1358 As an example where a possessive quantifier is suitable we consider
1359 matching a quoted string, as it appears in several programming languages.
1360 The backslash is used as an escape character that indicates that the
1361 next character is to be taken literally, as another character for the
1362 string. Therefore, after the opening quote, we expect a (possibly
1363 empty) sequence of alternatives: either some character except an
1364 unescaped quote or backslash or an escaped character.
1366 /"(?:[^"\\]++|\\.)*+"/;
1369 =head2 Building a regexp
1371 At this point, we have all the basic regexp concepts covered, so let's
1372 give a more involved example of a regular expression. We will build a
1373 regexp that matches numbers.
1375 The first task in building a regexp is to decide what we want to match
1376 and what we want to exclude. In our case, we want to match both
1377 integers and floating point numbers and we want to reject any string
1378 that isn't a number.
1380 The next task is to break the problem down into smaller problems that
1381 are easily converted into a regexp.
1383 The simplest case is integers. These consist of a sequence of digits,
1384 with an optional sign in front. The digits we can represent with
1385 C<\d+> and the sign can be matched with C<[+-]>. Thus the integer
1388 /[+-]?\d+/; # matches integers
1390 A floating point number potentially has a sign, an integral part, a
1391 decimal point, a fractional part, and an exponent. One or more of these
1392 parts is optional, so we need to check out the different
1393 possibilities. Floating point numbers which are in proper form include
1394 123., 0.345, .34, -1e6, and 25.4E-72. As with integers, the sign out
1395 front is completely optional and can be matched by C<[+-]?>. We can
1396 see that if there is no exponent, floating point numbers must have a
1397 decimal point, otherwise they are integers. We might be tempted to
1398 model these with C<\d*\.\d*>, but this would also match just a single
1399 decimal point, which is not a number. So the three cases of floating
1400 point number without exponent are
1402 /[+-]?\d+\./; # 1., 321., etc.
1403 /[+-]?\.\d+/; # .1, .234, etc.
1404 /[+-]?\d+\.\d+/; # 1.0, 30.56, etc.
1406 These can be combined into a single regexp with a three-way alternation:
1408 /[+-]?(\d+\.\d+|\d+\.|\.\d+)/; # floating point, no exponent
1410 In this alternation, it is important to put C<'\d+\.\d+'> before
1411 C<'\d+\.'>. If C<'\d+\.'> were first, the regexp would happily match that
1412 and ignore the fractional part of the number.
1414 Now consider floating point numbers with exponents. The key
1415 observation here is that I<both> integers and numbers with decimal
1416 points are allowed in front of an exponent. Then exponents, like the
1417 overall sign, are independent of whether we are matching numbers with
1418 or without decimal points, and can be 'decoupled' from the
1419 mantissa. The overall form of the regexp now becomes clear:
1421 /^(optional sign)(integer | f.p. mantissa)(optional exponent)$/;
1423 The exponent is an C<e> or C<E>, followed by an integer. So the
1426 /[eE][+-]?\d+/; # exponent
1428 Putting all the parts together, we get a regexp that matches numbers:
1430 /^[+-]?(\d+\.\d+|\d+\.|\.\d+|\d+)([eE][+-]?\d+)?$/; # Ta da!
1432 Long regexps like this may impress your friends, but can be hard to
1433 decipher. In complex situations like this, the C<//x> modifier for a
1434 match is invaluable. It allows one to put nearly arbitrary whitespace
1435 and comments into a regexp without affecting their meaning. Using it,
1436 we can rewrite our 'extended' regexp in the more pleasing form
1439 [+-]? # first, match an optional sign
1440 ( # then match integers or f.p. mantissas:
1441 \d+\.\d+ # mantissa of the form a.b
1442 |\d+\. # mantissa of the form a.
1443 |\.\d+ # mantissa of the form .b
1444 |\d+ # integer of the form a
1446 ([eE][+-]?\d+)? # finally, optionally match an exponent
1449 If whitespace is mostly irrelevant, how does one include space
1450 characters in an extended regexp? The answer is to backslash it
1451 S<C<'\ '>> or put it in a character class S<C<[ ]>>. The same thing
1452 goes for pound signs: use C<\#> or C<[#]>. For instance, Perl allows
1453 a space between the sign and the mantissa or integer, and we could add
1454 this to our regexp as follows:
1457 [+-]?\ * # first, match an optional sign *and space*
1458 ( # then match integers or f.p. mantissas:
1459 \d+\.\d+ # mantissa of the form a.b
1460 |\d+\. # mantissa of the form a.
1461 |\.\d+ # mantissa of the form .b
1462 |\d+ # integer of the form a
1464 ([eE][+-]?\d+)? # finally, optionally match an exponent
1467 In this form, it is easier to see a way to simplify the
1468 alternation. Alternatives 1, 2, and 4 all start with C<\d+>, so it
1469 could be factored out:
1472 [+-]?\ * # first, match an optional sign
1473 ( # then match integers or f.p. mantissas:
1474 \d+ # start out with a ...
1476 \.\d* # mantissa of the form a.b or a.
1477 )? # ? takes care of integers of the form a
1478 |\.\d+ # mantissa of the form .b
1480 ([eE][+-]?\d+)? # finally, optionally match an exponent
1483 or written in the compact form,
1485 /^[+-]?\ *(\d+(\.\d*)?|\.\d+)([eE][+-]?\d+)?$/;
1487 This is our final regexp. To recap, we built a regexp by
1493 specifying the task in detail,
1497 breaking down the problem into smaller parts,
1501 translating the small parts into regexps,
1505 combining the regexps,
1509 and optimizing the final combined regexp.
1513 These are also the typical steps involved in writing a computer
1514 program. This makes perfect sense, because regular expressions are
1515 essentially programs written in a little computer language that specifies
1518 =head2 Using regular expressions in Perl
1520 The last topic of Part 1 briefly covers how regexps are used in Perl
1521 programs. Where do they fit into Perl syntax?
1523 We have already introduced the matching operator in its default
1524 C</regexp/> and arbitrary delimiter C<m!regexp!> forms. We have used
1525 the binding operator C<=~> and its negation C<!~> to test for string
1526 matches. Associated with the matching operator, we have discussed the
1527 single line C<//s>, multi-line C<//m>, case-insensitive C<//i> and
1528 extended C<//x> modifiers. There are a few more things you might
1529 want to know about matching operators.
1531 =head3 Prohibiting substitution
1533 If you change C<$pattern> after the first substitution happens, Perl
1534 will ignore it. If you don't want any substitutions at all, use the
1535 special delimiter C<m''>:
1537 @pattern = ('Seuss');
1539 print if m'@pattern'; # matches literal '@pattern', not 'Seuss'
1542 Similar to strings, C<m''> acts like apostrophes on a regexp; all other
1543 C<m> delimiters act like quotes. If the regexp evaluates to the empty string,
1544 the regexp in the I<last successful match> is used instead. So we have
1546 "dog" =~ /d/; # 'd' matches
1547 "dogbert =~ //; # this matches the 'd' regexp used before
1550 =head3 Global matching
1552 The final two modifiers we will discuss here,
1553 C<//g> and C<//c>, concern multiple matches.
1554 The modifier C<//g> stands for global matching and allows the
1555 matching operator to match within a string as many times as possible.
1556 In scalar context, successive invocations against a string will have
1557 C<//g> jump from match to match, keeping track of position in the
1558 string as it goes along. You can get or set the position with the
1561 The use of C<//g> is shown in the following example. Suppose we have
1562 a string that consists of words separated by spaces. If we know how
1563 many words there are in advance, we could extract the words using
1566 $x = "cat dog house"; # 3 words
1567 $x =~ /^\s*(\w+)\s+(\w+)\s+(\w+)\s*$/; # matches,
1572 But what if we had an indeterminate number of words? This is the sort
1573 of task C<//g> was made for. To extract all words, form the simple
1574 regexp C<(\w+)> and loop over all matches with C</(\w+)/g>:
1576 while ($x =~ /(\w+)/g) {
1577 print "Word is $1, ends at position ", pos $x, "\n";
1582 Word is cat, ends at position 3
1583 Word is dog, ends at position 7
1584 Word is house, ends at position 13
1586 A failed match or changing the target string resets the position. If
1587 you don't want the position reset after failure to match, add the
1588 C<//c>, as in C</regexp/gc>. The current position in the string is
1589 associated with the string, not the regexp. This means that different
1590 strings have different positions and their respective positions can be
1591 set or read independently.
1593 In list context, C<//g> returns a list of matched groupings, or if
1594 there are no groupings, a list of matches to the whole regexp. So if
1595 we wanted just the words, we could use
1597 @words = ($x =~ /(\w+)/g); # matches,
1600 # $words[2] = 'house'
1602 Closely associated with the C<//g> modifier is the C<\G> anchor. The
1603 C<\G> anchor matches at the point where the previous C<//g> match left
1604 off. C<\G> allows us to easily do context-sensitive matching:
1606 $metric = 1; # use metric units
1608 $x = <FILE>; # read in measurement
1609 $x =~ /^([+-]?\d+)\s*/g; # get magnitude
1611 if ($metric) { # error checking
1612 print "Units error!" unless $x =~ /\Gkg\./g;
1615 print "Units error!" unless $x =~ /\Glbs\./g;
1617 $x =~ /\G\s+(widget|sprocket)/g; # continue processing
1619 The combination of C<//g> and C<\G> allows us to process the string a
1620 bit at a time and use arbitrary Perl logic to decide what to do next.
1621 Currently, the C<\G> anchor is only fully supported when used to anchor
1622 to the start of the pattern.
1624 C<\G> is also invaluable in processing fixed-length records with
1625 regexps. Suppose we have a snippet of coding region DNA, encoded as
1626 base pair letters C<ATCGTTGAAT...> and we want to find all the stop
1627 codons C<TGA>. In a coding region, codons are 3-letter sequences, so
1628 we can think of the DNA snippet as a sequence of 3-letter records. The
1631 # expanded, this is "ATC GTT GAA TGC AAA TGA CAT GAC"
1632 $dna = "ATCGTTGAATGCAAATGACATGAC";
1635 doesn't work; it may match a C<TGA>, but there is no guarantee that
1636 the match is aligned with codon boundaries, e.g., the substring
1637 S<C<GTT GAA>> gives a match. A better solution is
1639 while ($dna =~ /(\w\w\w)*?TGA/g) { # note the minimal *?
1640 print "Got a TGA stop codon at position ", pos $dna, "\n";
1645 Got a TGA stop codon at position 18
1646 Got a TGA stop codon at position 23
1648 Position 18 is good, but position 23 is bogus. What happened?
1650 The answer is that our regexp works well until we get past the last
1651 real match. Then the regexp will fail to match a synchronized C<TGA>
1652 and start stepping ahead one character position at a time, not what we
1653 want. The solution is to use C<\G> to anchor the match to the codon
1656 while ($dna =~ /\G(\w\w\w)*?TGA/g) {
1657 print "Got a TGA stop codon at position ", pos $dna, "\n";
1662 Got a TGA stop codon at position 18
1664 which is the correct answer. This example illustrates that it is
1665 important not only to match what is desired, but to reject what is not
1668 (There are other regexp modifiers that are available, such as
1669 C<//o>, but their specialized uses are beyond the
1670 scope of this introduction. )
1672 =head3 Search and replace
1674 Regular expressions also play a big role in I<search and replace>
1675 operations in Perl. Search and replace is accomplished with the
1676 C<s///> operator. The general form is
1677 C<s/regexp/replacement/modifiers>, with everything we know about
1678 regexps and modifiers applying in this case as well. The
1679 C<replacement> is a Perl double-quoted string that replaces in the
1680 string whatever is matched with the C<regexp>. The operator C<=~> is
1681 also used here to associate a string with C<s///>. If matching
1682 against C<$_>, the S<C<$_ =~>> can be dropped. If there is a match,
1683 C<s///> returns the number of substitutions made; otherwise it returns
1684 false. Here are a few examples:
1686 $x = "Time to feed the cat!";
1687 $x =~ s/cat/hacker/; # $x contains "Time to feed the hacker!"
1688 if ($x =~ s/^(Time.*hacker)!$/$1 now!/) {
1689 $more_insistent = 1;
1691 $y = "'quoted words'";
1692 $y =~ s/^'(.*)'$/$1/; # strip single quotes,
1693 # $y contains "quoted words"
1695 In the last example, the whole string was matched, but only the part
1696 inside the single quotes was grouped. With the C<s///> operator, the
1697 matched variables C<$1>, C<$2>, etc. are immediately available for use
1698 in the replacement expression, so we use C<$1> to replace the quoted
1699 string with just what was quoted. With the global modifier, C<s///g>
1700 will search and replace all occurrences of the regexp in the string:
1702 $x = "I batted 4 for 4";
1703 $x =~ s/4/four/; # doesn't do it all:
1704 # $x contains "I batted four for 4"
1705 $x = "I batted 4 for 4";
1706 $x =~ s/4/four/g; # does it all:
1707 # $x contains "I batted four for four"
1709 If you prefer 'regex' over 'regexp' in this tutorial, you could use
1710 the following program to replace it:
1712 % cat > simple_replace
1715 $replacement = shift;
1717 s/$regexp/$replacement/g;
1722 % simple_replace regexp regex perlretut.pod
1724 In C<simple_replace> we used the C<s///g> modifier to replace all
1725 occurrences of the regexp on each line. (Even though the regular
1726 expression appears in a loop, Perl is smart enough to compile it
1727 only once.) As with C<simple_grep>, both the
1728 C<print> and the C<s/$regexp/$replacement/g> use C<$_> implicitly.
1730 If you don't want C<s///> to change your original variable you can use
1731 the non-destructive substitute modifier, C<s///r>. This changes the
1732 behavior so that C<s///r> returns the final substituted string
1733 (instead of the number of substitutions):
1735 $x = "I like dogs.";
1736 $y = $x =~ s/dogs/cats/r;
1739 That example will print "I like dogs. I like cats". Notice the original
1740 C<$x> variable has not been affected. The overall
1741 result of the substitution is instead stored in C<$y>. If the
1742 substitution doesn't affect anything then the original string is
1745 $x = "I like dogs.";
1746 $y = $x =~ s/elephants/cougars/r;
1747 print "$x $y\n"; # prints "I like dogs. I like dogs."
1749 One other interesting thing that the C<s///r> flag allows is chaining
1752 $x = "Cats are great.";
1753 print $x =~ s/Cats/Dogs/r =~ s/Dogs/Frogs/r =~
1754 s/Frogs/Hedgehogs/r, "\n";
1755 # prints "Hedgehogs are great."
1757 A modifier available specifically to search and replace is the
1758 C<s///e> evaluation modifier. C<s///e> treats the
1759 replacement text as Perl code, rather than a double-quoted
1760 string. The value that the code returns is substituted for the
1761 matched substring. C<s///e> is useful if you need to do a bit of
1762 computation in the process of replacing text. This example counts
1763 character frequencies in a line:
1765 $x = "Bill the cat";
1766 $x =~ s/(.)/$chars{$1}++;$1/eg; # final $1 replaces char with itself
1767 print "frequency of '$_' is $chars{$_}\n"
1768 foreach (sort {$chars{$b} <=> $chars{$a}} keys %chars);
1772 frequency of ' ' is 2
1773 frequency of 't' is 2
1774 frequency of 'l' is 2
1775 frequency of 'B' is 1
1776 frequency of 'c' is 1
1777 frequency of 'e' is 1
1778 frequency of 'h' is 1
1779 frequency of 'i' is 1
1780 frequency of 'a' is 1
1782 As with the match C<m//> operator, C<s///> can use other delimiters,
1783 such as C<s!!!> and C<s{}{}>, and even C<s{}//>. If single quotes are
1784 used C<s'''>, then the regexp and replacement are
1785 treated as single-quoted strings and there are no
1786 variable substitutions. C<s///> in list context
1787 returns the same thing as in scalar context, i.e., the number of
1790 =head3 The split function
1792 The C<split()> function is another place where a regexp is used.
1793 C<split /regexp/, string, limit> separates the C<string> operand into
1794 a list of substrings and returns that list. The regexp must be designed
1795 to match whatever constitutes the separators for the desired substrings.
1796 The C<limit>, if present, constrains splitting into no more than C<limit>
1797 number of strings. For example, to split a string into words, use
1799 $x = "Calvin and Hobbes";
1800 @words = split /\s+/, $x; # $word[0] = 'Calvin'
1802 # $word[2] = 'Hobbes'
1804 If the empty regexp C<//> is used, the regexp always matches and
1805 the string is split into individual characters. If the regexp has
1806 groupings, then the resulting list contains the matched substrings from the
1807 groupings as well. For instance,
1809 $x = "/usr/bin/perl";
1810 @dirs = split m!/!, $x; # $dirs[0] = ''
1814 @parts = split m!(/)!, $x; # $parts[0] = ''
1820 # $parts[6] = 'perl'
1822 Since the first character of $x matched the regexp, C<split> prepended
1823 an empty initial element to the list.
1825 If you have read this far, congratulations! You now have all the basic
1826 tools needed to use regular expressions to solve a wide range of text
1827 processing problems. If this is your first time through the tutorial,
1828 why not stop here and play around with regexps a while.... S<Part 2>
1829 concerns the more esoteric aspects of regular expressions and those
1830 concepts certainly aren't needed right at the start.
1832 =head1 Part 2: Power tools
1834 OK, you know the basics of regexps and you want to know more. If
1835 matching regular expressions is analogous to a walk in the woods, then
1836 the tools discussed in Part 1 are analogous to topo maps and a
1837 compass, basic tools we use all the time. Most of the tools in part 2
1838 are analogous to flare guns and satellite phones. They aren't used
1839 too often on a hike, but when we are stuck, they can be invaluable.
1841 What follows are the more advanced, less used, or sometimes esoteric
1842 capabilities of Perl regexps. In Part 2, we will assume you are
1843 comfortable with the basics and concentrate on the advanced features.
1845 =head2 More on characters, strings, and character classes
1847 There are a number of escape sequences and character classes that we
1848 haven't covered yet.
1850 There are several escape sequences that convert characters or strings
1851 between upper and lower case, and they are also available within
1852 patterns. C<\l> and C<\u> convert the next character to lower or
1853 upper case, respectively:
1856 $string =~ /\u$x/; # matches 'Perl' in $string
1857 $x = "M(rs?|s)\\."; # note the double backslash
1858 $string =~ /\l$x/; # matches 'mr.', 'mrs.', and 'ms.',
1860 A C<\L> or C<\U> indicates a lasting conversion of case, until
1861 terminated by C<\E> or thrown over by another C<\U> or C<\L>:
1863 $x = "This word is in lower case:\L SHOUT\E";
1864 $x =~ /shout/; # matches
1865 $x = "I STILL KEYPUNCH CARDS FOR MY 360"
1866 $x =~ /\Ukeypunch/; # matches punch card string
1868 If there is no C<\E>, case is converted until the end of the
1869 string. The regexps C<\L\u$word> or C<\u\L$word> convert the first
1870 character of C<$word> to uppercase and the rest of the characters to
1873 Control characters can be escaped with C<\c>, so that a control-Z
1874 character would be matched with C<\cZ>. The escape sequence
1875 C<\Q>...C<\E> quotes, or protects most non-alphabetic characters. For
1878 $x = "\QThat !^*&%~& cat!";
1879 $x =~ /\Q!^*&%~&\E/; # check for rough language
1881 It does not protect C<$> or C<@>, so that variables can still be
1884 C<\Q>, C<\L>, C<\l>, C<\U>, C<\u> and C<\E> are actually part of
1885 double-quotish syntax, and not part of regexp syntax proper. They will
1886 work if they appear in a regular expression embedded directly in a
1887 program, but not when contained in a string that is interpolated in a
1890 Perl regexps can handle more than just the
1891 standard ASCII character set. Perl supports I<Unicode>, a standard
1892 for representing the alphabets from virtually all of the world's written
1893 languages, and a host of symbols. Perl's text strings are Unicode strings, so
1894 they can contain characters with a value (codepoint or character number) higher
1897 What does this mean for regexps? Well, regexp users don't need to know
1898 much about Perl's internal representation of strings. But they do need
1899 to know 1) how to represent Unicode characters in a regexp and 2) that
1900 a matching operation will treat the string to be searched as a sequence
1901 of characters, not bytes. The answer to 1) is that Unicode characters
1902 greater than C<chr(255)> are represented using the C<\x{hex}> notation, because
1903 \x hex (without curly braces) doesn't go further than 255. (Starting in Perl
1904 5.14, if you're an octal fan, you can also use C<\o{oct}>.)
1906 /\x{263a}/; # match a Unicode smiley face :)
1908 B<NOTE>: In Perl 5.6.0 it used to be that one needed to say C<use
1909 utf8> to use any Unicode features. This is no more the case: for
1910 almost all Unicode processing, the explicit C<utf8> pragma is not
1911 needed. (The only case where it matters is if your Perl script is in
1912 Unicode and encoded in UTF-8, then an explicit C<use utf8> is needed.)
1914 Figuring out the hexadecimal sequence of a Unicode character you want
1915 or deciphering someone else's hexadecimal Unicode regexp is about as
1916 much fun as programming in machine code. So another way to specify
1917 Unicode characters is to use the I<named character> escape
1918 sequence C<\N{I<name>}>. I<name> is a name for the Unicode character, as
1919 specified in the Unicode standard. For instance, if we wanted to
1920 represent or match the astrological sign for the planet Mercury, we
1923 $x = "abc\N{MERCURY}def";
1924 $x =~ /\N{MERCURY}/; # matches
1926 One can also use "short" names:
1928 print "\N{GREEK SMALL LETTER SIGMA} is called sigma.\n";
1929 print "\N{greek:Sigma} is an upper-case sigma.\n";
1931 You can also restrict names to a certain alphabet by specifying the
1932 L<charnames> pragma:
1934 use charnames qw(greek);
1935 print "\N{sigma} is Greek sigma\n";
1937 An index of character names is available on-line from the Unicode
1938 Consortium, L<http://www.unicode.org/charts/charindex.html>; explanatory
1939 material with links to other resources at
1940 L<http://www.unicode.org/standard/where>.
1942 The answer to requirement 2) is that a regexp (mostly)
1943 uses Unicode characters. The "mostly" is for messy backward
1944 compatibility reasons, but starting in Perl 5.14, any regex compiled in
1945 the scope of a C<use feature 'unicode_strings'> (which is automatically
1946 turned on within the scope of a C<use 5.012> or higher) will turn that
1947 "mostly" into "always". If you want to handle Unicode properly, you
1948 should ensure that C<'unicode_strings'> is turned on.
1949 Internally, this is encoded to bytes using either UTF-8 or a native 8
1950 bit encoding, depending on the history of the string, but conceptually
1951 it is a sequence of characters, not bytes. See L<perlunitut> for a
1952 tutorial about that.
1954 Let us now discuss Unicode character classes, most usually called
1955 "character properties". These are represented by the
1956 C<\p{name}> escape sequence. Closely associated is the C<\P{name}>
1957 property, which is the negation of the C<\p{name}> one. For
1958 example, to match lower and uppercase characters,
1961 $x =~ /^\p{IsUpper}/; # matches, uppercase char class
1962 $x =~ /^\P{IsUpper}/; # doesn't match, char class sans uppercase
1963 $x =~ /^\p{IsLower}/; # doesn't match, lowercase char class
1964 $x =~ /^\P{IsLower}/; # matches, char class sans lowercase
1966 (The "Is" is optional.)
1968 There are many, many Unicode character properties. For the full list
1969 see L<perluniprops>. Most of them have synonyms with shorter names,
1970 also listed there. Some synonyms are a single character. For these,
1971 you can drop the braces. For instance, C<\pM> is the same thing as
1972 C<\p{Mark}>, meaning things like accent marks.
1974 The Unicode C<\p{Script}> property is used to categorize every Unicode
1975 character into the language script it is written in. For example,
1976 English, French, and a bunch of other European languages are written in
1977 the Latin script. But there is also the Greek script, the Thai script,
1978 the Katakana script, etc. You can test whether a character is in a
1979 particular script with, for example C<\p{Latin}>, C<\p{Greek}>,
1980 or C<\p{Katakana}>. To test if it isn't in the Balinese script, you
1981 would use C<\P{Balinese}>.
1983 What we have described so far is the single form of the C<\p{...}> character
1984 classes. There is also a compound form which you may run into. These
1985 look like C<\p{name=value}> or C<\p{name:value}> (the equals sign and colon
1986 can be used interchangeably). These are more general than the single form,
1987 and in fact most of the single forms are just Perl-defined shortcuts for common
1988 compound forms. For example, the script examples in the previous paragraph
1989 could be written equivalently as C<\p{Script=Latin}>, C<\p{Script:Greek}>,
1990 C<\p{script=katakana}>, and C<\P{script=balinese}> (case is irrelevant
1991 between the C<{}> braces). You may
1992 never have to use the compound forms, but sometimes it is necessary, and their
1993 use can make your code easier to understand.
1995 C<\X> is an abbreviation for a character class that comprises
1996 a Unicode I<extended grapheme cluster>. This represents a "logical character":
1997 what appears to be a single character, but may be represented internally by more
1998 than one. As an example, using the Unicode full names, e.g., S<C<A + COMBINING
1999 RING>> is a grapheme cluster with base character C<A> and combining character
2000 S<C<COMBINING RING>>, which translates in Danish to A with the circle atop it,
2001 as in the word E<Aring>ngstrom.
2003 For the full and latest information about Unicode see the latest
2004 Unicode standard, or the Unicode Consortium's website L<http://www.unicode.org>
2006 As if all those classes weren't enough, Perl also defines POSIX-style
2007 character classes. These have the form C<[:name:]>, with C<name> the
2008 name of the POSIX class. The POSIX classes are C<alpha>, C<alnum>,
2009 C<ascii>, C<cntrl>, C<digit>, C<graph>, C<lower>, C<print>, C<punct>,
2010 C<space>, C<upper>, and C<xdigit>, and two extensions, C<word> (a Perl
2011 extension to match C<\w>), and C<blank> (a GNU extension). The C<//a>
2012 modifier restricts these to matching just in the ASCII range; otherwise
2013 they can match the same as their corresponding Perl Unicode classes:
2014 C<[:upper:]> is the same as C<\p{IsUpper}>, etc. (There are some
2015 exceptions and gotchas with this; see L<perlrecharclass> for a full
2016 discussion.) The C<[:digit:]>, C<[:word:]>, and
2017 C<[:space:]> correspond to the familiar C<\d>, C<\w>, and C<\s>
2018 character classes. To negate a POSIX class, put a C<^> in front of
2019 the name, so that, e.g., C<[:^digit:]> corresponds to C<\D> and, under
2020 Unicode, C<\P{IsDigit}>. The Unicode and POSIX character classes can
2021 be used just like C<\d>, with the exception that POSIX character
2022 classes can only be used inside of a character class:
2024 /\s+[abc[:digit:]xyz]\s*/; # match a,b,c,x,y,z, or a digit
2025 /^=item\s[[:digit:]]/; # match '=item',
2026 # followed by a space and a digit
2027 /\s+[abc\p{IsDigit}xyz]\s+/; # match a,b,c,x,y,z, or a digit
2028 /^=item\s\p{IsDigit}/; # match '=item',
2029 # followed by a space and a digit
2031 Whew! That is all the rest of the characters and character classes.
2033 =head2 Compiling and saving regular expressions
2035 In Part 1 we mentioned that Perl compiles a regexp into a compact
2036 sequence of opcodes. Thus, a compiled regexp is a data structure
2037 that can be stored once and used again and again. The regexp quote
2038 C<qr//> does exactly that: C<qr/string/> compiles the C<string> as a
2039 regexp and transforms the result into a form that can be assigned to a
2042 $reg = qr/foo+bar?/; # reg contains a compiled regexp
2044 Then C<$reg> can be used as a regexp:
2047 $x =~ $reg; # matches, just like /foo+bar?/
2048 $x =~ /$reg/; # same thing, alternate form
2050 C<$reg> can also be interpolated into a larger regexp:
2052 $x =~ /(abc)?$reg/; # still matches
2054 As with the matching operator, the regexp quote can use different
2055 delimiters, e.g., C<qr!!>, C<qr{}> or C<qr~~>. Apostrophes
2056 as delimiters (C<qr''>) inhibit any interpolation.
2058 Pre-compiled regexps are useful for creating dynamic matches that
2059 don't need to be recompiled each time they are encountered. Using
2060 pre-compiled regexps, we write a C<grep_step> program which greps
2061 for a sequence of patterns, advancing to the next pattern as soon
2062 as one has been satisfied.
2066 # grep_step - match <number> regexps, one after the other
2067 # usage: multi_grep <number> regexp1 regexp2 ... file1 file2 ...
2070 $regexp[$_] = shift foreach (0..$number-1);
2071 @compiled = map qr/$_/, @regexp;
2072 while ($line = <>) {
2073 if ($line =~ /$compiled[0]/) {
2076 last unless @compiled;
2081 % grep_step 3 shift print last grep_step
2084 last unless @compiled;
2086 Storing pre-compiled regexps in an array C<@compiled> allows us to
2087 simply loop through the regexps without any recompilation, thus gaining
2088 flexibility without sacrificing speed.
2091 =head2 Composing regular expressions at runtime
2093 Backtracking is more efficient than repeated tries with different regular
2094 expressions. If there are several regular expressions and a match with
2095 any of them is acceptable, then it is possible to combine them into a set
2096 of alternatives. If the individual expressions are input data, this
2097 can be done by programming a join operation. We'll exploit this idea in
2098 an improved version of the C<simple_grep> program: a program that matches
2103 # multi_grep - match any of <number> regexps
2104 # usage: multi_grep <number> regexp1 regexp2 ... file1 file2 ...
2107 $regexp[$_] = shift foreach (0..$number-1);
2108 $pattern = join '|', @regexp;
2110 while ($line = <>) {
2111 print $line if $line =~ /$pattern/;
2115 % multi_grep 2 shift for multi_grep
2117 $regexp[$_] = shift foreach (0..$number-1);
2119 Sometimes it is advantageous to construct a pattern from the I<input>
2120 that is to be analyzed and use the permissible values on the left
2121 hand side of the matching operations. As an example for this somewhat
2122 paradoxical situation, let's assume that our input contains a command
2123 verb which should match one out of a set of available command verbs,
2124 with the additional twist that commands may be abbreviated as long as
2125 the given string is unique. The program below demonstrates the basic
2130 $kwds = 'copy compare list print';
2132 $cmd =~ s/^\s+|\s+$//g; # trim leading and trailing spaces
2133 if( ( @matches = $kwds =~ /\b$cmd\w*/g ) == 1 ){
2134 print "command: '@matches'\n";
2135 } elsif( @matches == 0 ){
2136 print "no such command: '$cmd'\n";
2138 print "not unique: '$cmd' (could be one of: @matches)\n";
2147 not unique: 'co' (could be one of: copy compare)
2149 no such command: 'printer'
2151 Rather than trying to match the input against the keywords, we match the
2152 combined set of keywords against the input. The pattern matching
2153 operation S<C<$kwds =~ /\b($cmd\w*)/g>> does several things at the
2154 same time. It makes sure that the given command begins where a keyword
2155 begins (C<\b>). It tolerates abbreviations due to the added C<\w*>. It
2156 tells us the number of matches (C<scalar @matches>) and all the keywords
2157 that were actually matched. You could hardly ask for more.
2159 =head2 Embedding comments and modifiers in a regular expression
2161 Starting with this section, we will be discussing Perl's set of
2162 I<extended patterns>. These are extensions to the traditional regular
2163 expression syntax that provide powerful new tools for pattern
2164 matching. We have already seen extensions in the form of the minimal
2165 matching constructs C<??>, C<*?>, C<+?>, C<{n,m}?>, and C<{n,}?>. Most
2166 of the extensions below have the form C<(?char...)>, where the
2167 C<char> is a character that determines the type of extension.
2169 The first extension is an embedded comment C<(?#text)>. This embeds a
2170 comment into the regular expression without affecting its meaning. The
2171 comment should not have any closing parentheses in the text. An
2174 /(?# Match an integer:)[+-]?\d+/;
2176 This style of commenting has been largely superseded by the raw,
2177 freeform commenting that is allowed with the C<//x> modifier.
2179 Most modifiers, such as C<//i>, C<//m>, C<//s> and C<//x> (or any
2180 combination thereof) can also be embedded in
2181 a regexp using C<(?i)>, C<(?m)>, C<(?s)>, and C<(?x)>. For instance,
2183 /(?i)yes/; # match 'yes' case insensitively
2184 /yes/i; # same thing
2185 /(?x)( # freeform version of an integer regexp
2186 [+-]? # match an optional sign
2187 \d+ # match a sequence of digits
2191 Embedded modifiers can have two important advantages over the usual
2192 modifiers. Embedded modifiers allow a custom set of modifiers to
2193 I<each> regexp pattern. This is great for matching an array of regexps
2194 that must have different modifiers:
2196 $pattern[0] = '(?i)doctor';
2197 $pattern[1] = 'Johnson';
2200 foreach $patt (@pattern) {
2205 The second advantage is that embedded modifiers (except C<//p>, which
2206 modifies the entire regexp) only affect the regexp
2207 inside the group the embedded modifier is contained in. So grouping
2208 can be used to localize the modifier's effects:
2210 /Answer: ((?i)yes)/; # matches 'Answer: yes', 'Answer: YES', etc.
2212 Embedded modifiers can also turn off any modifiers already present
2213 by using, e.g., C<(?-i)>. Modifiers can also be combined into
2214 a single expression, e.g., C<(?s-i)> turns on single line mode and
2215 turns off case insensitivity.
2217 Embedded modifiers may also be added to a non-capturing grouping.
2218 C<(?i-m:regexp)> is a non-capturing grouping that matches C<regexp>
2219 case insensitively and turns off multi-line mode.
2222 =head2 Looking ahead and looking behind
2224 This section concerns the lookahead and lookbehind assertions. First,
2225 a little background.
2227 In Perl regular expressions, most regexp elements 'eat up' a certain
2228 amount of string when they match. For instance, the regexp element
2229 C<[abc}]> eats up one character of the string when it matches, in the
2230 sense that Perl moves to the next character position in the string
2231 after the match. There are some elements, however, that don't eat up
2232 characters (advance the character position) if they match. The examples
2233 we have seen so far are the anchors. The anchor C<^> matches the
2234 beginning of the line, but doesn't eat any characters. Similarly, the
2235 word boundary anchor C<\b> matches wherever a character matching C<\w>
2236 is next to a character that doesn't, but it doesn't eat up any
2237 characters itself. Anchors are examples of I<zero-width assertions>:
2238 zero-width, because they consume
2239 no characters, and assertions, because they test some property of the
2240 string. In the context of our walk in the woods analogy to regexp
2241 matching, most regexp elements move us along a trail, but anchors have
2242 us stop a moment and check our surroundings. If the local environment
2243 checks out, we can proceed forward. But if the local environment
2244 doesn't satisfy us, we must backtrack.
2246 Checking the environment entails either looking ahead on the trail,
2247 looking behind, or both. C<^> looks behind, to see that there are no
2248 characters before. C<$> looks ahead, to see that there are no
2249 characters after. C<\b> looks both ahead and behind, to see if the
2250 characters on either side differ in their "word-ness".
2252 The lookahead and lookbehind assertions are generalizations of the
2253 anchor concept. Lookahead and lookbehind are zero-width assertions
2254 that let us specify which characters we want to test for. The
2255 lookahead assertion is denoted by C<(?=regexp)> and the lookbehind
2256 assertion is denoted by C<< (?<=fixed-regexp) >>. Some examples are
2258 $x = "I catch the housecat 'Tom-cat' with catnip";
2259 $x =~ /cat(?=\s)/; # matches 'cat' in 'housecat'
2260 @catwords = ($x =~ /(?<=\s)cat\w+/g); # matches,
2261 # $catwords[0] = 'catch'
2262 # $catwords[1] = 'catnip'
2263 $x =~ /\bcat\b/; # matches 'cat' in 'Tom-cat'
2264 $x =~ /(?<=\s)cat(?=\s)/; # doesn't match; no isolated 'cat' in
2267 Note that the parentheses in C<(?=regexp)> and C<< (?<=regexp) >> are
2268 non-capturing, since these are zero-width assertions. Thus in the
2269 second regexp, the substrings captured are those of the whole regexp
2270 itself. Lookahead C<(?=regexp)> can match arbitrary regexps, but
2271 lookbehind C<< (?<=fixed-regexp) >> only works for regexps of fixed
2272 width, i.e., a fixed number of characters long. Thus
2273 C<< (?<=(ab|bc)) >> is fine, but C<< (?<=(ab)*) >> is not. The
2274 negated versions of the lookahead and lookbehind assertions are
2275 denoted by C<(?!regexp)> and C<< (?<!fixed-regexp) >> respectively.
2276 They evaluate true if the regexps do I<not> match:
2279 $x =~ /foo(?!bar)/; # doesn't match, 'bar' follows 'foo'
2280 $x =~ /foo(?!baz)/; # matches, 'baz' doesn't follow 'foo'
2281 $x =~ /(?<!\s)foo/; # matches, there is no \s before 'foo'
2283 The C<\C> is unsupported in lookbehind, because the already
2284 treacherous definition of C<\C> would become even more so
2285 when going backwards.
2287 Here is an example where a string containing blank-separated words,
2288 numbers and single dashes is to be split into its components.
2289 Using C</\s+/> alone won't work, because spaces are not required between
2290 dashes, or a word or a dash. Additional places for a split are established
2291 by looking ahead and behind:
2293 $str = "one two - --6-8";
2294 @toks = split / \s+ # a run of spaces
2295 | (?<=\S) (?=-) # any non-space followed by '-'
2296 | (?<=-) (?=\S) # a '-' followed by any non-space
2297 /x, $str; # @toks = qw(one two - - - 6 - 8)
2300 =head2 Using independent subexpressions to prevent backtracking
2302 I<Independent subexpressions> are regular expressions, in the
2303 context of a larger regular expression, that function independently of
2304 the larger regular expression. That is, they consume as much or as
2305 little of the string as they wish without regard for the ability of
2306 the larger regexp to match. Independent subexpressions are represented
2307 by C<< (?>regexp) >>. We can illustrate their behavior by first
2308 considering an ordinary regexp:
2311 $x =~ /a*ab/; # matches
2313 This obviously matches, but in the process of matching, the
2314 subexpression C<a*> first grabbed the C<a>. Doing so, however,
2315 wouldn't allow the whole regexp to match, so after backtracking, C<a*>
2316 eventually gave back the C<a> and matched the empty string. Here, what
2317 C<a*> matched was I<dependent> on what the rest of the regexp matched.
2319 Contrast that with an independent subexpression:
2321 $x =~ /(?>a*)ab/; # doesn't match!
2323 The independent subexpression C<< (?>a*) >> doesn't care about the rest
2324 of the regexp, so it sees an C<a> and grabs it. Then the rest of the
2325 regexp C<ab> cannot match. Because C<< (?>a*) >> is independent, there
2326 is no backtracking and the independent subexpression does not give
2327 up its C<a>. Thus the match of the regexp as a whole fails. A similar
2328 behavior occurs with completely independent regexps:
2331 $x =~ /a*/g; # matches, eats an 'a'
2332 $x =~ /\Gab/g; # doesn't match, no 'a' available
2334 Here C<//g> and C<\G> create a 'tag team' handoff of the string from
2335 one regexp to the other. Regexps with an independent subexpression are
2336 much like this, with a handoff of the string to the independent
2337 subexpression, and a handoff of the string back to the enclosing
2340 The ability of an independent subexpression to prevent backtracking
2341 can be quite useful. Suppose we want to match a non-empty string
2342 enclosed in parentheses up to two levels deep. Then the following
2345 $x = "abc(de(fg)h"; # unbalanced parentheses
2346 $x =~ /\( ( [^()]+ | \([^()]*\) )+ \)/x;
2348 The regexp matches an open parenthesis, one or more copies of an
2349 alternation, and a close parenthesis. The alternation is two-way, with
2350 the first alternative C<[^()]+> matching a substring with no
2351 parentheses and the second alternative C<\([^()]*\)> matching a
2352 substring delimited by parentheses. The problem with this regexp is
2353 that it is pathological: it has nested indeterminate quantifiers
2354 of the form C<(a+|b)+>. We discussed in Part 1 how nested quantifiers
2355 like this could take an exponentially long time to execute if there
2356 was no match possible. To prevent the exponential blowup, we need to
2357 prevent useless backtracking at some point. This can be done by
2358 enclosing the inner quantifier as an independent subexpression:
2360 $x =~ /\( ( (?>[^()]+) | \([^()]*\) )+ \)/x;
2362 Here, C<< (?>[^()]+) >> breaks the degeneracy of string partitioning
2363 by gobbling up as much of the string as possible and keeping it. Then
2364 match failures fail much more quickly.
2367 =head2 Conditional expressions
2369 A I<conditional expression> is a form of if-then-else statement
2370 that allows one to choose which patterns are to be matched, based on
2371 some condition. There are two types of conditional expression:
2372 C<(?(condition)yes-regexp)> and
2373 C<(?(condition)yes-regexp|no-regexp)>. C<(?(condition)yes-regexp)> is
2374 like an S<C<'if () {}'>> statement in Perl. If the C<condition> is true,
2375 the C<yes-regexp> will be matched. If the C<condition> is false, the
2376 C<yes-regexp> will be skipped and Perl will move onto the next regexp
2377 element. The second form is like an S<C<'if () {} else {}'>> statement
2378 in Perl. If the C<condition> is true, the C<yes-regexp> will be
2379 matched, otherwise the C<no-regexp> will be matched.
2381 The C<condition> can have several forms. The first form is simply an
2382 integer in parentheses C<(integer)>. It is true if the corresponding
2383 backreference C<\integer> matched earlier in the regexp. The same
2384 thing can be done with a name associated with a capture group, written
2385 as C<< (<name>) >> or C<< ('name') >>. The second form is a bare
2386 zero-width assertion C<(?...)>, either a lookahead, a lookbehind, or a
2387 code assertion (discussed in the next section). The third set of forms
2388 provides tests that return true if the expression is executed within
2389 a recursion (C<(R)>) or is being called from some capturing group,
2390 referenced either by number (C<(R1)>, C<(R2)>,...) or by name
2393 The integer or name form of the C<condition> allows us to choose,
2394 with more flexibility, what to match based on what matched earlier in the
2395 regexp. This searches for words of the form C<"$x$x"> or C<"$x$y$y$x">:
2397 % simple_grep '^(\w+)(\w+)?(?(2)\g2\g1|\g1)$' /usr/dict/words
2407 The lookbehind C<condition> allows, along with backreferences,
2408 an earlier part of the match to influence a later part of the
2409 match. For instance,
2411 /[ATGC]+(?(?<=AA)G|C)$/;
2413 matches a DNA sequence such that it either ends in C<AAG>, or some
2414 other base pair combination and C<C>. Note that the form is
2415 C<< (?(?<=AA)G|C) >> and not C<< (?((?<=AA))G|C) >>; for the
2416 lookahead, lookbehind or code assertions, the parentheses around the
2417 conditional are not needed.
2420 =head2 Defining named patterns
2422 Some regular expressions use identical subpatterns in several places.
2423 Starting with Perl 5.10, it is possible to define named subpatterns in
2424 a section of the pattern so that they can be called up by name
2425 anywhere in the pattern. This syntactic pattern for this definition
2426 group is C<< (?(DEFINE)(?<name>pattern)...) >>. An insertion
2427 of a named pattern is written as C<(?&name)>.
2429 The example below illustrates this feature using the pattern for
2430 floating point numbers that was presented earlier on. The three
2431 subpatterns that are used more than once are the optional sign, the
2432 digit sequence for an integer and the decimal fraction. The DEFINE
2433 group at the end of the pattern contains their definition. Notice
2434 that the decimal fraction pattern is the first place where we can
2435 reuse the integer pattern.
2437 /^ (?&osg)\ * ( (?&int)(?&dec)? | (?&dec) )
2438 (?: [eE](?&osg)(?&int) )?
2441 (?<osg>[-+]?) # optional sign
2442 (?<int>\d++) # integer
2443 (?<dec>\.(?&int)) # decimal fraction
2447 =head2 Recursive patterns
2449 This feature (introduced in Perl 5.10) significantly extends the
2450 power of Perl's pattern matching. By referring to some other
2451 capture group anywhere in the pattern with the construct
2452 C<(?group-ref)>, the I<pattern> within the referenced group is used
2453 as an independent subpattern in place of the group reference itself.
2454 Because the group reference may be contained I<within> the group it
2455 refers to, it is now possible to apply pattern matching to tasks that
2456 hitherto required a recursive parser.
2458 To illustrate this feature, we'll design a pattern that matches if
2459 a string contains a palindrome. (This is a word or a sentence that,
2460 while ignoring spaces, interpunctuation and case, reads the same backwards
2461 as forwards. We begin by observing that the empty string or a string
2462 containing just one word character is a palindrome. Otherwise it must
2463 have a word character up front and the same at its end, with another
2464 palindrome in between.
2466 /(?: (\w) (?...Here be a palindrome...) \g{-1} | \w? )/x
2468 Adding C<\W*> at either end to eliminate what is to be ignored, we already
2469 have the full pattern:
2471 my $pp = qr/^(\W* (?: (\w) (?1) \g{-1} | \w? ) \W*)$/ix;
2472 for $s ( "saippuakauppias", "A man, a plan, a canal: Panama!" ){
2473 print "'$s' is a palindrome\n" if $s =~ /$pp/;
2476 In C<(?...)> both absolute and relative backreferences may be used.
2477 The entire pattern can be reinserted with C<(?R)> or C<(?0)>.
2478 If you prefer to name your groups, you can use C<(?&name)> to
2479 recurse into that group.
2482 =head2 A bit of magic: executing Perl code in a regular expression
2484 Normally, regexps are a part of Perl expressions.
2485 I<Code evaluation> expressions turn that around by allowing
2486 arbitrary Perl code to be a part of a regexp. A code evaluation
2487 expression is denoted C<(?{code})>, with I<code> a string of Perl
2490 Be warned that this feature is considered experimental, and may be
2491 changed without notice.
2493 Code expressions are zero-width assertions, and the value they return
2494 depends on their environment. There are two possibilities: either the
2495 code expression is used as a conditional in a conditional expression
2496 C<(?(condition)...)>, or it is not. If the code expression is a
2497 conditional, the code is evaluated and the result (i.e., the result of
2498 the last statement) is used to determine truth or falsehood. If the
2499 code expression is not used as a conditional, the assertion always
2500 evaluates true and the result is put into the special variable
2501 C<$^R>. The variable C<$^R> can then be used in code expressions later
2502 in the regexp. Here are some silly examples:
2505 $x =~ /abc(?{print "Hi Mom!";})def/; # matches,
2507 $x =~ /aaa(?{print "Hi Mom!";})def/; # doesn't match,
2510 Pay careful attention to the next example:
2512 $x =~ /abc(?{print "Hi Mom!";})ddd/; # doesn't match,
2516 At first glance, you'd think that it shouldn't print, because obviously
2517 the C<ddd> isn't going to match the target string. But look at this
2520 $x =~ /abc(?{print "Hi Mom!";})[dD]dd/; # doesn't match,
2523 Hmm. What happened here? If you've been following along, you know that
2524 the above pattern should be effectively (almost) the same as the last one;
2525 enclosing the C<d> in a character class isn't going to change what it
2526 matches. So why does the first not print while the second one does?
2528 The answer lies in the optimizations the regex engine makes. In the first
2529 case, all the engine sees are plain old characters (aside from the
2530 C<?{}> construct). It's smart enough to realize that the string 'ddd'
2531 doesn't occur in our target string before actually running the pattern
2532 through. But in the second case, we've tricked it into thinking that our
2533 pattern is more complicated. It takes a look, sees our
2534 character class, and decides that it will have to actually run the
2535 pattern to determine whether or not it matches, and in the process of
2536 running it hits the print statement before it discovers that we don't
2539 To take a closer look at how the engine does optimizations, see the
2540 section L<"Pragmas and debugging"> below.
2542 More fun with C<?{}>:
2544 $x =~ /(?{print "Hi Mom!";})/; # matches,
2546 $x =~ /(?{$c = 1;})(?{print "$c";})/; # matches,
2548 $x =~ /(?{$c = 1;})(?{print "$^R";})/; # matches,
2551 The bit of magic mentioned in the section title occurs when the regexp
2552 backtracks in the process of searching for a match. If the regexp
2553 backtracks over a code expression and if the variables used within are
2554 localized using C<local>, the changes in the variables produced by the
2555 code expression are undone! Thus, if we wanted to count how many times
2556 a character got matched inside a group, we could use, e.g.,
2559 $count = 0; # initialize 'a' count
2560 $c = "bob"; # test if $c gets clobbered
2561 $x =~ /(?{local $c = 0;}) # initialize count
2563 (?{local $c = $c + 1;}) # increment count
2564 )* # do this any number of times,
2565 aa # but match 'aa' at the end
2566 (?{$count = $c;}) # copy local $c var into $count
2568 print "'a' count is $count, \$c variable is '$c'\n";
2572 'a' count is 2, $c variable is 'bob'
2574 If we replace the S<C< (?{local $c = $c + 1;})>> with
2575 S<C< (?{$c = $c + 1;})>>, the variable changes are I<not> undone
2576 during backtracking, and we get
2578 'a' count is 4, $c variable is 'bob'
2580 Note that only localized variable changes are undone. Other side
2581 effects of code expression execution are permanent. Thus
2584 $x =~ /(a(?{print "Yow\n";}))*aa/;
2593 The result C<$^R> is automatically localized, so that it will behave
2594 properly in the presence of backtracking.
2596 This example uses a code expression in a conditional to match a
2597 definite article, either 'the' in English or 'der|die|das' in German:
2599 $lang = 'DE'; # use German
2604 $lang eq 'EN'; # is the language English?
2606 the | # if so, then match 'the'
2607 (der|die|das) # else, match 'der|die|das'
2611 Note that the syntax here is C<(?(?{...})yes-regexp|no-regexp)>, not
2612 C<(?((?{...}))yes-regexp|no-regexp)>. In other words, in the case of a
2613 code expression, we don't need the extra parentheses around the
2616 If you try to use code expressions where the code text is contained within
2617 an interpolated variable, rather than appearing literally in the pattern,
2618 Perl may surprise you:
2622 /foo(?{ $bar })bar/; # compiles ok, $bar not interpolated
2623 /foo(?{ 1 })$bar/; # compiles ok, $bar interpolated
2624 /foo${pat}bar/; # compile error!
2626 $pat = qr/(?{ $foo = 1 })/; # precompile code regexp
2627 /foo${pat}bar/; # compiles ok
2629 If a regexp has a variable that interpolates a code expression, Perl
2630 treats the regexp as an error. If the code expression is precompiled into
2631 a variable, however, interpolating is ok. The question is, why is this an
2634 The reason is that variable interpolation and code expressions
2635 together pose a security risk. The combination is dangerous because
2636 many programmers who write search engines often take user input and
2637 plug it directly into a regexp:
2639 $regexp = <>; # read user-supplied regexp
2640 $chomp $regexp; # get rid of possible newline
2641 $text =~ /$regexp/; # search $text for the $regexp
2643 If the C<$regexp> variable contains a code expression, the user could
2644 then execute arbitrary Perl code. For instance, some joker could
2645 search for S<C<system('rm -rf *');>> to erase your files. In this
2646 sense, the combination of interpolation and code expressions I<taints>
2647 your regexp. So by default, using both interpolation and code
2648 expressions in the same regexp is not allowed. If you're not
2649 concerned about malicious users, it is possible to bypass this
2650 security check by invoking S<C<use re 'eval'>>:
2652 use re 'eval'; # throw caution out the door
2655 /foo${pat}bar/; # compiles ok
2657 Another form of code expression is the I<pattern code expression>.
2658 The pattern code expression is like a regular code expression, except
2659 that the result of the code evaluation is treated as a regular
2660 expression and matched immediately. A simple example is
2665 $x =~ /(??{$char x $length})/x; # matches, there are 5 of 'a'
2668 This final example contains both ordinary and pattern code
2669 expressions. It detects whether a binary string C<1101010010001...> has a
2670 Fibonacci spacing 0,1,1,2,3,5,... of the C<1>'s:
2672 $x = "1101010010001000001";
2673 $z0 = ''; $z1 = '0'; # initial conditions
2674 print "It is a Fibonacci sequence\n"
2675 if $x =~ /^1 # match an initial '1'
2677 ((??{ $z0 })) # match some '0'
2679 (?{ $z0 = $z1; $z1 .= $^N; })
2680 )+ # repeat as needed
2681 $ # that is all there is
2683 printf "Largest sequence matched was %d\n", length($z1)-length($z0);
2685 Remember that C<$^N> is set to whatever was matched by the last
2686 completed capture group. This prints
2688 It is a Fibonacci sequence
2689 Largest sequence matched was 5
2691 Ha! Try that with your garden variety regexp package...
2693 Note that the variables C<$z0> and C<$z1> are not substituted when the
2694 regexp is compiled, as happens for ordinary variables outside a code
2695 expression. Rather, the whole code block is parsed as perl code at the
2696 same time as perl is compiling the code containing the literal regexp
2699 The regexp without the C<//x> modifier is
2701 /^1(?:((??{ $z0 }))1(?{ $z0 = $z1; $z1 .= $^N; }))+$/
2703 which shows that spaces are still possible in the code parts. Nevertheless,
2704 when working with code and conditional expressions, the extended form of
2705 regexps is almost necessary in creating and debugging regexps.
2708 =head2 Backtracking control verbs
2710 Perl 5.10 introduced a number of control verbs intended to provide
2711 detailed control over the backtracking process, by directly influencing
2712 the regexp engine and by providing monitoring techniques. As all
2713 the features in this group are experimental and subject to change or
2714 removal in a future version of Perl, the interested reader is
2715 referred to L<perlre/"Special Backtracking Control Verbs"> for a
2716 detailed description.
2718 Below is just one example, illustrating the control verb C<(*FAIL)>,
2719 which may be abbreviated as C<(*F)>. If this is inserted in a regexp
2720 it will cause it to fail, just as it would at some
2721 mismatch between the pattern and the string. Processing
2722 of the regexp continues as it would after any "normal"
2723 failure, so that, for instance, the next position in the string or another
2724 alternative will be tried. As failing to match doesn't preserve capture
2725 groups or produce results, it may be necessary to use this in
2726 combination with embedded code.
2729 "supercalifragilisticexpialidocious" =~
2730 /([aeiou])(?{ $count{$1}++; })(*FAIL)/i;
2731 printf "%3d '%s'\n", $count{$_}, $_ for (sort keys %count);
2733 The pattern begins with a class matching a subset of letters. Whenever
2734 this matches, a statement like C<$count{'a'}++;> is executed, incrementing
2735 the letter's counter. Then C<(*FAIL)> does what it says, and
2736 the regexp engine proceeds according to the book: as long as the end of
2737 the string hasn't been reached, the position is advanced before looking
2738 for another vowel. Thus, match or no match makes no difference, and the
2739 regexp engine proceeds until the entire string has been inspected.
2740 (It's remarkable that an alternative solution using something like
2742 $count{lc($_)}++ for split('', "supercalifragilisticexpialidocious");
2743 printf "%3d '%s'\n", $count2{$_}, $_ for ( qw{ a e i o u } );
2745 is considerably slower.)
2748 =head2 Pragmas and debugging
2750 Speaking of debugging, there are several pragmas available to control
2751 and debug regexps in Perl. We have already encountered one pragma in
2752 the previous section, S<C<use re 'eval';>>, that allows variable
2753 interpolation and code expressions to coexist in a regexp. The other
2758 @parts = ($tainted =~ /(\w+)\s+(\w+)/; # @parts is now tainted
2760 The C<taint> pragma causes any substrings from a match with a tainted
2761 variable to be tainted as well. This is not normally the case, as
2762 regexps are often used to extract the safe bits from a tainted
2763 variable. Use C<taint> when you are not extracting safe bits, but are
2764 performing some other processing. Both C<taint> and C<eval> pragmas
2765 are lexically scoped, which means they are in effect only until
2766 the end of the block enclosing the pragmas.
2768 use re '/m'; # or any other flags
2769 $multiline_string =~ /^foo/; # /m is implied
2771 The C<re '/flags'> pragma (introduced in Perl
2772 5.14) turns on the given regular expression flags
2773 until the end of the lexical scope. See
2774 L<re/"'E<sol>flags' mode"> for more
2778 /^(.*)$/s; # output debugging info
2780 use re 'debugcolor';
2781 /^(.*)$/s; # output debugging info in living color
2783 The global C<debug> and C<debugcolor> pragmas allow one to get
2784 detailed debugging info about regexp compilation and
2785 execution. C<debugcolor> is the same as debug, except the debugging
2786 information is displayed in color on terminals that can display
2787 termcap color sequences. Here is example output:
2789 % perl -e 'use re "debug"; "abc" =~ /a*b+c/;'
2790 Compiling REx 'a*b+c'
2798 floating 'bc' at 0..2147483647 (checking floating) minlen 2
2799 Guessing start of match, REx 'a*b+c' against 'abc'...
2800 Found floating substr 'bc' at offset 1...
2801 Guessed: match at offset 0
2802 Matching REx 'a*b+c' against 'abc'
2803 Setting an EVAL scope, savestack=3
2804 0 <> <abc> | 1: STAR
2805 EXACT <a> can match 1 times out of 32767...
2806 Setting an EVAL scope, savestack=3
2807 1 <a> <bc> | 4: PLUS
2808 EXACT <b> can match 1 times out of 32767...
2809 Setting an EVAL scope, savestack=3
2810 2 <ab> <c> | 7: EXACT <c>
2813 Freeing REx: 'a*b+c'
2815 If you have gotten this far into the tutorial, you can probably guess
2816 what the different parts of the debugging output tell you. The first
2819 Compiling REx 'a*b+c'
2828 describes the compilation stage. C<STAR(4)> means that there is a
2829 starred object, in this case C<'a'>, and if it matches, goto line 4,
2830 i.e., C<PLUS(7)>. The middle lines describe some heuristics and
2831 optimizations performed before a match:
2833 floating 'bc' at 0..2147483647 (checking floating) minlen 2
2834 Guessing start of match, REx 'a*b+c' against 'abc'...
2835 Found floating substr 'bc' at offset 1...
2836 Guessed: match at offset 0
2838 Then the match is executed and the remaining lines describe the
2841 Matching REx 'a*b+c' against 'abc'
2842 Setting an EVAL scope, savestack=3
2843 0 <> <abc> | 1: STAR
2844 EXACT <a> can match 1 times out of 32767...
2845 Setting an EVAL scope, savestack=3
2846 1 <a> <bc> | 4: PLUS
2847 EXACT <b> can match 1 times out of 32767...
2848 Setting an EVAL scope, savestack=3
2849 2 <ab> <c> | 7: EXACT <c>
2852 Freeing REx: 'a*b+c'
2854 Each step is of the form S<C<< n <x> <y> >>>, with C<< <x> >> the
2855 part of the string matched and C<< <y> >> the part not yet
2856 matched. The S<C<< | 1: STAR >>> says that Perl is at line number 1
2857 in the compilation list above. See
2858 L<perldebguts/"Debugging Regular Expressions"> for much more detail.
2860 An alternative method of debugging regexps is to embed C<print>
2861 statements within the regexp. This provides a blow-by-blow account of
2862 the backtracking in an alternation:
2864 "that this" =~ m@(?{print "Start at position ", pos, "\n";})
2874 (?{print "Done at position ", pos, "\n";})
2890 Code expressions, conditional expressions, and independent expressions
2891 are I<experimental>. Don't use them in production code. Yet.
2895 This is just a tutorial. For the full story on Perl regular
2896 expressions, see the L<perlre> regular expressions reference page.
2898 For more information on the matching C<m//> and substitution C<s///>
2899 operators, see L<perlop/"Regexp Quote-Like Operators">. For
2900 information on the C<split> operation, see L<perlfunc/split>.
2902 For an excellent all-around resource on the care and feeding of
2903 regular expressions, see the book I<Mastering Regular Expressions> by
2904 Jeffrey Friedl (published by O'Reilly, ISBN 1556592-257-3).
2906 =head1 AUTHOR AND COPYRIGHT
2908 Copyright (c) 2000 Mark Kvale
2909 All rights reserved.
2911 This document may be distributed under the same terms as Perl itself.
2913 =head2 Acknowledgments
2915 The inspiration for the stop codon DNA example came from the ZIP
2916 code example in chapter 7 of I<Mastering Regular Expressions>.
2918 The author would like to thank Jeff Pinyan, Andrew Johnson, Peter
2919 Haworth, Ronald J Kimball, and Joe Smith for all their helpful