1PERLREQUICK(1) Perl Programmers Reference Guide PERLREQUICK(1)
2
6 perlrequick - Perl regular expressions quick start
7
9 This page covers the very basics of understanding, creating and using
10 regular expressions ('regexes') in Perl.
11
13 This page assumes you already know things, like what a "pattern" is,
14 and the basic syntax of using them. If you don't, see perlretut.
15 Simple word matching
17 The simplest regex is simply a word, or more generally, a string of
18 characters. A regex consisting of a word matches any string that
19 contains that word:
20 "Hello World" =~ /World/; # matches
22 In this statement, "World" is a regex and the "//" enclosing "/World/"
24 tells Perl to search a string for a match. The operator "=~"
25 associates the string with the regex match and produces a true value if
26 the regex matched, or false if the regex did not match. In our case,
27 "World" matches the second word in "Hello World", so the expression is
28 true. This idea has several variations.
29 Expressions like this are useful in conditionals:
31 print "It matches\n" if "Hello World" =~ /World/;
33 The sense of the match can be reversed by using "!~" operator:
35 print "It doesn't match\n" if "Hello World" !~ /World/;
37 The literal string in the regex can be replaced by a variable:
39 $greeting = "World";
41 print "It matches\n" if "Hello World" =~ /$greeting/;
42 If you're matching against $_, the "$_ =~" part can be omitted:
44 $_ = "Hello World";
46 print "It matches\n" if /World/;
47 Finally, the "//" default delimiters for a match can be changed to
49 arbitrary delimiters by putting an 'm' out front:
50 "Hello World" =~ m!World!; # matches, delimited by '!'
52 "Hello World" =~ m{World}; # matches, note the matching '{}'
53 "/usr/bin/perl" =~ m"/perl"; # matches after '/usr/bin',
54 # '/' becomes an ordinary char
55 Regexes must match a part of the string exactly in order for the
57 statement to be true:
58 "Hello World" =~ /world/; # doesn't match, case sensitive
60 "Hello World" =~ /o W/; # matches, ' ' is an ordinary char
61 "Hello World" =~ /World /; # doesn't match, no ' ' at end
62 Perl will always match at the earliest possible point in the string:
64 "Hello World" =~ /o/; # matches 'o' in 'Hello'
66 "That hat is red" =~ /hat/; # matches 'hat' in 'That'
67 Not all characters can be used 'as is' in a match. Some characters,
69 called metacharacters, are considered special, and reserved for use in
70 regex notation. The metacharacters are
71 {}[]()^$.|*+?\
73 A metacharacter can be matched literally by putting a backslash before
75 it:
76 "2+2=4" =~ /2+2/; # doesn't match, + is a metacharacter
78 "2+2=4" =~ /2\+2/; # matches, \+ is treated like an ordinary +
79 'C:\WIN32' =~ /C:\\WIN/; # matches
80 "/usr/bin/perl" =~ /\/usr\/bin\/perl/; # matches
81 In the last regex, the forward slash '/' is also backslashed, because
83 it is used to delimit the regex.
84 Most of the metacharacters aren't always special, and other characters
86 (such as the ones delimitting the pattern) become special under various
87 circumstances. This can be confusing and lead to unexpected results.
88 "use re 'strict'" can notify you of potential pitfalls.
89 Non-printable ASCII characters are represented by escape sequences.
91 Common examples are "\t" for a tab, "\n" for a newline, and "\r" for a
92 carriage return. Arbitrary bytes are represented by octal escape
93 sequences, e.g., "\033", or hexadecimal escape sequences, e.g., "\x1B":
94 "1000\t2000" =~ m(0\t2) # matches
96 "cat" =~ /\143\x61\x74/ # matches in ASCII, but
97 # a weird way to spell cat
98 Regexes are treated mostly as double-quoted strings, so variable
100 substitution works:
101 $foo = 'house';
103 'cathouse' =~ /cat$foo/; # matches
104 'housecat' =~ /${foo}cat/; # matches
105 With all of the regexes above, if the regex matched anywhere in the
107 string, it was considered a match. To specify where it should match,
108 we would use the anchor metacharacters "^" and "$". The anchor "^"
109 means match at the beginning of the string and the anchor "$" means
110 match at the end of the string, or before a newline at the end of the
111 string. Some examples:
112 "housekeeper" =~ /keeper/; # matches
114 "housekeeper" =~ /^keeper/; # doesn't match
115 "housekeeper" =~ /keeper$/; # matches
116 "housekeeper\n" =~ /keeper$/; # matches
117 "housekeeper" =~ /^housekeeper$/; # matches
118 Using character classes
120 A character class allows a set of possible characters, rather than just
121 a single character, to match at a particular point in a regex. There
122 are a number of different types of character classes, but usually when
123 people use this term, they are referring to the type described in this
124 section, which are technically called "Bracketed character classes",
125 because they are denoted by brackets "[...]", with the set of
126 characters to be possibly matched inside. But we'll drop the
127 "bracketed" below to correspond with common usage. Here are some
128 examples of (bracketed) character classes:
129 /cat/; # matches 'cat'
131 /[bcr]at/; # matches 'bat', 'cat', or 'rat'
132 "abc" =~ /[cab]/; # matches 'a'
133 In the last statement, even though 'c' is the first character in the
135 class, the earliest point at which the regex can match is 'a'.
136 /[yY][eE][sS]/; # match 'yes' in a case-insensitive way
138 # 'yes', 'Yes', 'YES', etc.
139 /yes/i; # also match 'yes' in a case-insensitive way
140 The last example shows a match with an 'i' modifier, which makes the
142 match case-insensitive.
143 Character classes also have ordinary and special characters, but the
145 sets of ordinary and special characters inside a character class are
146 different than those outside a character class. The special characters
147 for a character class are "-]\^$" and are matched using an escape:
148 /[\]c]def/; # matches ']def' or 'cdef'
150 $x = 'bcr';
151 /[$x]at/; # matches 'bat, 'cat', or 'rat'
152 /[\$x]at/; # matches '$at' or 'xat'
153 /[\\$x]at/; # matches '\at', 'bat, 'cat', or 'rat'
154 The special character '-' acts as a range operator within character
156 classes, so that the unwieldy "[0123456789]" and "[abc...xyz]" become
157 the svelte "[0-9]" and "[a-z]":
158 /item[0-9]/; # matches 'item0' or ... or 'item9'
160 /[0-9a-fA-F]/; # matches a hexadecimal digit
161 If '-' is the first or last character in a character class, it is
163 treated as an ordinary character.
164 The special character "^" in the first position of a character class
166 denotes a negated character class, which matches any character but
167 those in the brackets. Both "[...]" and "[^...]" must match a
168 character, or the match fails. Then
169 /[^a]at/; # doesn't match 'aat' or 'at', but matches
171 # all other 'bat', 'cat, '0at', '%at', etc.
172 /[^0-9]/; # matches a non-numeric character
173 /[a^]at/; # matches 'aat' or '^at'; here '^' is ordinary
174 Perl has several abbreviations for common character classes. (These
176 definitions are those that Perl uses in ASCII-safe mode with the "/a"
177 modifier. Otherwise they could match many more non-ASCII Unicode
178 characters as well. See "Backslash sequences" in perlrecharclass for
179 details.)
180 • \d is a digit and represents
182 [0-9]
184 • \s is a whitespace character and represents
186 [\ \t\r\n\f]
188 • \w is a word character (alphanumeric or _) and represents
190 [0-9a-zA-Z_]
192 • \D is a negated \d; it represents any character but a digit
194 [^0-9]
196 • \S is a negated \s; it represents any non-whitespace character
198 [^\s]
200 • \W is a negated \w; it represents any non-word character
202 [^\w]
204 • The period '.' matches any character but "\n"
206 The "\d\s\w\D\S\W" abbreviations can be used both inside and outside of
208 character classes. Here are some in use:
209 /\d\d:\d\d:\d\d/; # matches a hh:mm:ss time format
211 /[\d\s]/; # matches any digit or whitespace character
212 /\w\W\w/; # matches a word char, followed by a
213 # non-word char, followed by a word char
214 /..rt/; # matches any two chars, followed by 'rt'
215 /end\./; # matches 'end.'
216 /end[.]/; # same thing, matches 'end.'
217 The word anchor "\b" matches a boundary between a word character and a
219 non-word character "\w\W" or "\W\w":
220 $x = "Housecat catenates house and cat";
222 $x =~ /\bcat/; # matches cat in 'catenates'
223 $x =~ /cat\b/; # matches cat in 'housecat'
224 $x =~ /\bcat\b/; # matches 'cat' at end of string
225 In the last example, the end of the string is considered a word
227 boundary.
228 For natural language processing (so that, for example, apostrophes are
230 included in words), use instead "\b{wb}"
231 "don't" =~ / .+? \b{wb} /x; # matches the whole string
233 Matching this or that
235 We can match different character strings with the alternation
236 metacharacter '|'. To match "dog" or "cat", we form the regex
237 "dog|cat". As before, Perl will try to match the regex at the earliest
238 possible point in the string. At each character position, Perl will
239 first try to match the first alternative, "dog". If "dog" doesn't
240 match, Perl will then try the next alternative, "cat". If "cat"
241 doesn't match either, then the match fails and Perl moves to the next
242 position in the string. Some examples:
243 "cats and dogs" =~ /cat|dog|bird/; # matches "cat"
245 "cats and dogs" =~ /dog|cat|bird/; # matches "cat"
246 Even though "dog" is the first alternative in the second regex, "cat"
248 is able to match earlier in the string.
249 "cats" =~ /c|ca|cat|cats/; # matches "c"
251 "cats" =~ /cats|cat|ca|c/; # matches "cats"
252 At a given character position, the first alternative that allows the
254 regex match to succeed will be the one that matches. Here, all the
255 alternatives match at the first string position, so the first matches.
256 Grouping things and hierarchical matching
258 The grouping metacharacters "()" allow a part of a regex to be treated
259 as a single unit. Parts of a regex are grouped by enclosing them in
260 parentheses. The regex "house(cat|keeper)" means match "house"
261 followed by either "cat" or "keeper". Some more examples are
262 /(a|b)b/; # matches 'ab' or 'bb'
264 /(^a|b)c/; # matches 'ac' at start of string or 'bc' anywhere
265 /house(cat|)/; # matches either 'housecat' or 'house'
267 /house(cat(s|)|)/; # matches either 'housecats' or 'housecat' or
268 # 'house'. Note groups can be nested.
269 "20" =~ /(19|20|)\d\d/; # matches the null alternative '()\d\d',
271 # because '20\d\d' can't match
272 Extracting matches
274 The grouping metacharacters "()" also allow the extraction of the parts
275 of a string that matched. For each grouping, the part that matched
276 inside goes into the special variables $1, $2, etc. They can be used
277 just as ordinary variables:
278 # extract hours, minutes, seconds
280 $time =~ /(\d\d):(\d\d):(\d\d)/; # match hh:mm:ss format
281 $hours = $1;
282 $minutes = $2;
283 $seconds = $3;
284 In list context, a match "/regex/" with groupings will return the list
286 of matched values "($1,$2,...)". So we could rewrite it as
287 ($hours, $minutes, $second) = ($time =~ /(\d\d):(\d\d):(\d\d)/);
289 If the groupings in a regex are nested, $1 gets the group with the
291 leftmost opening parenthesis, $2 the next opening parenthesis, etc.
292 For example, here is a complex regex and the matching variables
293 indicated below it:
294 /(ab(cd|ef)((gi)|j))/;
296 1 2 34
297 Associated with the matching variables $1, $2, ... are the
299 backreferences "\g1", "\g2", ... Backreferences are matching variables
300 that can be used inside a regex:
301 /(\w\w\w)\s\g1/; # find sequences like 'the the' in string
303 $1, $2, ... should only be used outside of a regex, and "\g1", "\g2",
305 ... only inside a regex.
306 Matching repetitions
308 The quantifier metacharacters "?", "*", "+", and "{}" allow us to
309 determine the number of repeats of a portion of a regex we consider to
310 be a match. Quantifiers are put immediately after the character,
311 character class, or grouping that we want to specify. They have the
312 following meanings:
313 • "a?" = match 'a' 1 or 0 times
315 • "a*" = match 'a' 0 or more times, i.e., any number of times
317 • "a+" = match 'a' 1 or more times, i.e., at least once
319 • "a{n,m}" = match at least "n" times, but not more than "m" times.
321 • "a{n,}" = match at least "n" or more times
323 • "a{n}" = match exactly "n" times
325 Here are some examples:
327 /[a-z]+\s+\d*/; # match a lowercase word, at least some space, and
329 # any number of digits
330 /(\w+)\s+\g1/; # match doubled words of arbitrary length
331 $year =~ /^\d{2,4}$/; # make sure year is at least 2 but not more
332 # than 4 digits
333 $year =~ /^\d{4}$|^\d{2}$/; # better match; throw out 3 digit dates
334 These quantifiers will try to match as much of the string as possible,
336 while still allowing the regex to match. So we have
337 $x = 'the cat in the hat';
339 $x =~ /^(.*)(at)(.*)$/; # matches,
340 # $1 = 'the cat in the h'
341 # $2 = 'at'
342 # $3 = '' (0 matches)
343 The first quantifier ".*" grabs as much of the string as possible while
345 still having the regex match. The second quantifier ".*" has no string
346 left to it, so it matches 0 times.
347 More matching
349 There are a few more things you might want to know about matching
350 operators. The global modifier "/g" allows the matching operator to
351 match within a string as many times as possible. In scalar context,
352 successive matches against a string will have "/g" jump from match to
353 match, keeping track of position in the string as it goes along. You
354 can get or set the position with the "pos()" function. For example,
355 $x = "cat dog house"; # 3 words
357 while ($x =~ /(\w+)/g) {
358 print "Word is $1, ends at position ", pos $x, "\n";
359 }
360 prints
362 Word is cat, ends at position 3
364 Word is dog, ends at position 7
365 Word is house, ends at position 13
366 A failed match or changing the target string resets the position. If
368 you don't want the position reset after failure to match, add the "/c",
369 as in "/regex/gc".
370 In list context, "/g" returns a list of matched groupings, or if there
372 are no groupings, a list of matches to the whole regex. So
373 @words = ($x =~ /(\w+)/g); # matches,
375 # $word[0] = 'cat'
376 # $word[1] = 'dog'
377 # $word[2] = 'house'
378 Search and replace
380 Search and replace is performed using "s/regex/replacement/modifiers".
381 The "replacement" is a Perl double-quoted string that replaces in the
382 string whatever is matched with the "regex". The operator "=~" is also
383 used here to associate a string with "s///". If matching against $_,
384 the "$_ =~" can be dropped. If there is a match, "s///" returns the
385 number of substitutions made; otherwise it returns false. Here are a
386 few examples:
387 $x = "Time to feed the cat!";
389 $x =~ s/cat/hacker/; # $x contains "Time to feed the hacker!"
390 $y = "'quoted words'";
391 $y =~ s/^'(.*)'$/$1/; # strip single quotes,
392 # $y contains "quoted words"
393 With the "s///" operator, the matched variables $1, $2, etc. are
395 immediately available for use in the replacement expression. With the
396 global modifier, "s///g" will search and replace all occurrences of the
397 regex in the string:
398 $x = "I batted 4 for 4";
400 $x =~ s/4/four/; # $x contains "I batted four for 4"
401 $x = "I batted 4 for 4";
402 $x =~ s/4/four/g; # $x contains "I batted four for four"
403 The non-destructive modifier "s///r" causes the result of the
405 substitution to be returned instead of modifying $_ (or whatever
406 variable the substitute was bound to with "=~"):
407 $x = "I like dogs.";
409 $y = $x =~ s/dogs/cats/r;
410 print "$x $y\n"; # prints "I like dogs. I like cats."
411 $x = "Cats are great.";
413 print $x =~ s/Cats/Dogs/r =~ s/Dogs/Frogs/r =~
414 s/Frogs/Hedgehogs/r, "\n";
415 # prints "Hedgehogs are great."
416 @foo = map { s/[a-z]/X/r } qw(a b c 1 2 3);
418 # @foo is now qw(X X X 1 2 3)
419 The evaluation modifier "s///e" wraps an "eval{...}" around the
421 replacement string and the evaluated result is substituted for the
422 matched substring. Some examples:
423 # reverse all the words in a string
425 $x = "the cat in the hat";
426 $x =~ s/(\w+)/reverse $1/ge; # $x contains "eht tac ni eht tah"
427 # convert percentage to decimal
429 $x = "A 39% hit rate";
430 $x =~ s!(\d+)%!$1/100!e; # $x contains "A 0.39 hit rate"
431 The last example shows that "s///" can use other delimiters, such as
433 "s!!!" and "s{}{}", and even "s{}//". If single quotes are used
434 "s'''", then the regex and replacement are treated as single-quoted
435 strings.
436 The split operator
438 "split /regex/, string" splits "string" into a list of substrings and
439 returns that list. The regex determines the character sequence that
440 "string" is split with respect to. For example, to split a string into
441 words, use
442 $x = "Calvin and Hobbes";
444 @word = split /\s+/, $x; # $word[0] = 'Calvin'
445 # $word[1] = 'and'
446 # $word[2] = 'Hobbes'
447 To extract a comma-delimited list of numbers, use
449 $x = "1.618,2.718, 3.142";
451 @const = split /,\s*/, $x; # $const[0] = '1.618'
452 # $const[1] = '2.718'
453 # $const[2] = '3.142'
454 If the empty regex "//" is used, the string is split into individual
456 characters. If the regex has groupings, then the list produced
457 contains the matched substrings from the groupings as well:
458 $x = "/usr/bin";
460 @parts = split m!(/)!, $x; # $parts[0] = ''
461 # $parts[1] = '/'
462 # $parts[2] = 'usr'
463 # $parts[3] = '/'
464 # $parts[4] = 'bin'
465 Since the first character of $x matched the regex, "split" prepended an
467 empty initial element to the list.
468 "use re 'strict'"
470 New in v5.22, this applies stricter rules than otherwise when compiling
471 regular expression patterns. It can find things that, while legal, may
472 not be what you intended.
473 See 'strict' in re.
475
477 None.
478
480 This is just a quick start guide. For a more in-depth tutorial on
481 regexes, see perlretut and for the reference page, see perlre.
482
484 Copyright (c) 2000 Mark Kvale All rights reserved.
485 This document may be distributed under the same terms as Perl itself.
487 Acknowledgments
489 The author would like to thank Mark-Jason Dominus, Tom Christiansen,
490 Ilya Zakharevich, Brad Hughes, and Mike Giroux for all their helpful
491 comments.
492perl v5.32.1 2021-05-31 PERLREQUICK(1)