3 perlrebackslash - Perl Regular Expression Backslash Sequences and Escapes
7 The top level documentation about Perl regular expressions
10 This document describes all backslash and escape sequences. After
11 explaining the role of the backslash, it lists all the sequences that have
12 a special meaning in Perl regular expressions (in alphabetical order),
13 then describes each of them.
15 Most sequences are described in detail in different documents; the primary
16 purpose of this document is to have a quick reference guide describing all
17 backslash and escape sequences.
22 In a regular expression, the backslash can perform one of two tasks:
23 it either takes away the special meaning of the character following it
24 (for instance, C<\|> matches a vertical bar, it's not an alternation),
25 or it is the start of a backslash or escape sequence.
27 The rules determining what it is are quite simple: if the character
28 following the backslash is an ASCII punctuation (non-word) character (that is,
29 anything that is not a letter, digit or underscore), then the backslash just
30 takes away the special meaning (if any) of the character following it.
32 If the character following the backslash is an ASCII letter or an ASCII digit,
33 then the sequence may be special; if so, it's listed below. A few letters have
34 not been used yet, so escaping them with a backslash doesn't change them to be
35 special. A future version of Perl may assign a special meaning to them, so if
36 you have warnings turned on, Perl will issue a warning if you use such a
39 It is however guaranteed that backslash or escape sequences never have a
40 punctuation character following the backslash, not now, and not in a future
41 version of Perl 5. So it is safe to put a backslash in front of a non-word
44 Note that the backslash itself is special; if you want to match a backslash,
45 you have to escape the backslash with a backslash: C</\\/> matches a single
52 There is one exception. If you use an alphanumerical character as the
53 delimiter of your pattern (which you probably shouldn't do for readability
54 reasons), you will have to escape the delimiter if you want to match
55 it. Perl won't warn then. See also L<perlop/Gory details of parsing
61 =head2 All the sequences and escapes
63 Those not usable within a bracketed character class (like C<[\da-z]>) are marked
66 \000 Octal escape sequence.
67 \1 Absolute backreference. Not in [].
69 \A Beginning of string. Not in [].
70 \b Word/non-word boundary. (Backspace in []).
71 \B Not a word/non-word boundary. Not in [].
72 \cX Control-X (X can be any ASCII character).
73 \C Single octet, even under UTF-8. Not in [].
74 \d Character class for digits.
75 \D Character class for non-digits.
77 \E Turn off \Q, \L and \U processing. Not in [].
79 \g{}, \g1 Named, absolute or relative backreference. Not in [].
80 \G Pos assertion. Not in [].
81 \h Character class for horizontal whitespace.
82 \H Character class for non horizontal whitespace.
83 \k{}, \k<>, \k'' Named backreference. Not in [].
84 \K Keep the stuff left of \K. Not in [].
85 \l Lowercase next character. Not in [].
86 \L Lowercase till \E. Not in [].
87 \n (Logical) newline character.
88 \N Any character but newline. Experimental. Not in [].
89 \N{} Named or numbered (Unicode) character.
90 \p{}, \pP Character with the given Unicode property.
91 \P{}, \PP Character without the given Unicode property.
92 \Q Quotemeta till \E. Not in [].
94 \R Generic new line. Not in [].
95 \s Character class for whitespace.
96 \S Character class for non whitespace.
98 \u Titlecase next character. Not in [].
99 \U Uppercase till \E. Not in [].
100 \v Character class for vertical whitespace.
101 \V Character class for non vertical whitespace.
102 \w Character class for word characters.
103 \W Character class for non-word characters.
104 \x{}, \x00 Hexadecimal escape sequence.
105 \X Unicode "extended grapheme cluster". Not in [].
106 \z End of string. Not in [].
107 \Z End of string. Not in [].
109 =head2 Character Escapes
111 =head3 Fixed characters
113 A handful of characters have a dedicated I<character escape>. The following
114 table shows them, along with their ASCII code points (in decimal and hex),
115 their ASCII name, the control escape (see below) and a short description.
117 Seq. Code Point ASCII Cntr Description.
119 \a 7 07 BEL \cG alarm or bell
120 \b 8 08 BS \cH backspace [1]
121 \e 27 1B ESC \c[ escape character
122 \f 12 0C FF \cL form feed
123 \n 10 0A LF \cJ line feed [2]
124 \r 13 0D CR \cM carriage return
131 C<\b> is only the backspace character inside a character class. Outside a
132 character class, C<\b> is a word/non-word boundary.
136 C<\n> matches a logical newline. Perl will convert between C<\n> and your
137 OSses native newline character when reading from or writing to text files.
143 $str =~ /\t/; # Matches if $str contains a (horizontal) tab.
145 =head3 Control characters
147 C<\c> is used to denote a control character; the character following C<\c>
148 is the name of the control character. For instance, C</\cM/> matches the
149 character I<control-M> (a carriage return, code point 13). The case of the
150 character following C<\c> doesn't matter: C<\cM> and C<\cm> match the same
153 Mnemonic: I<c>ontrol character.
157 $str =~ /\cK/; # Matches if $str contains a vertical tab (control-K).
159 =head3 Named or numbered characters
161 All Unicode characters have a Unicode name and numeric ordinal value. Use the
162 C<\N{}> construct to specify a character by either of these values.
164 To specify by name, the name of the character goes between the curly braces.
165 In this case, you have to C<use charnames> to load the Unicode names of the
166 characters, otherwise Perl will complain.
168 To specify by Unicode ordinal number, use the form
169 C<\N{U+I<wide hex character>}>, where I<wide hex character> is a number in
170 hexadecimal that gives the ordinal number that Unicode has assigned to the
171 desired character. It is customary (but not required) to use leading zeros to
172 pad the number to 4 digits. Thus C<\N{U+0041}> means
173 C<Latin Capital Letter A>, and you will rarely see it written without the two
174 leading zeros. C<\N{U+0041}> means C<A> even on EBCDIC machines (where the
175 ordinal value of C<A> is not 0x41).
177 It is even possible to give your own names to characters, and even to short
178 sequences of characters. For details, see L<charnames>.
180 (There is an expanded internal form that you may see in debug output:
181 C<\N{U+I<wide hex character>.I<wide hex character>...}>.
182 The C<...> means any number of these I<wide hex character>s separated by dots.
183 This represents the sequence formed by the characters. This is an internal
184 form only, subject to change, and you should not try to use it yourself.)
186 Mnemonic: I<N>amed character.
188 Note that a character that is expressed as a named or numbered character is
189 considered as a character without special meaning by the regex engine, and will
194 use charnames ':full'; # Loads the Unicode names.
195 $str =~ /\N{THAI CHARACTER SO SO}/; # Matches the Thai SO SO character
197 use charnames 'Cyrillic'; # Loads Cyrillic names.
198 $str =~ /\N{ZHE}\N{KA}/; # Match "ZHE" followed by "KA".
202 Octal escapes consist of a backslash followed by two or three octal digits
203 matching the code point of the character you want to use. This allows for
204 512 characters (C<\00> up to C<\777>) that can be expressed this way (but
205 anything above C<\377> is deprecated).
206 Enough in pre-Unicode days, but most Unicode characters cannot be escaped
209 Note that a character that is expressed as an octal escape is considered
210 as a character without special meaning by the regex engine, and will match
213 =head4 Examples (assuming an ASCII platform)
216 $str =~ /\120/; # Match, "\120" is "P".
217 $str =~ /\120+/; # Match, "\120" is "P", it is repeated at least once.
218 $str =~ /P\053/; # No match, "\053" is "+" and taken literally.
222 Octal escapes potentially clash with backreferences. They both consist
223 of a backslash followed by numbers. So Perl has to use heuristics to
224 determine whether it is a backreference or an octal escape. Perl uses
231 If the backslash is followed by a single digit, it's a backreference.
235 If the first digit following the backslash is a 0, it's an octal escape.
239 If the number following the backslash is N (decimal), and Perl already has
240 seen N capture groups, Perl will consider this to be a backreference.
241 Otherwise, it will consider it to be an octal escape. Note that if N > 999,
242 Perl only takes the first three digits for the octal escape; the rest is
248 /^($pat)\1000$/; # Matches 'aa'; there are 1000 capture groups.
249 /^$pat\1000$/; # Matches 'a@0'; there are 999 capture groups
250 # and \1000 is seen as \100 (a '@') and a '0'.
254 =head3 Hexadecimal escapes
256 Hexadecimal escapes start with C<\x> and are then either followed by a
257 two digit hexadecimal number, or a hexadecimal number of arbitrary length
258 surrounded by curly braces. The hexadecimal number is the code point of
259 the character you want to express.
261 Note that a character that is expressed as a hexadecimal escape is considered
262 as a character without special meaning by the regex engine, and will match
265 Mnemonic: heI<x>adecimal.
267 =head4 Examples (assuming an ASCII platform)
270 $str =~ /\x50/; # Match, "\x50" is "P".
271 $str =~ /\x50+/; # Match, "\x50" is "P", it is repeated at least once.
272 $str =~ /P\x2B/; # No match, "\x2B" is "+" and taken literally.
274 /\x{2603}\x{2602}/ # Snowman with an umbrella.
275 # The Unicode character 2603 is a snowman,
276 # the Unicode character 2602 is an umbrella.
277 /\x{263B}/ # Black smiling face.
278 /\x{263b}/ # Same, the hex digits A - F are case insensitive.
282 A number of backslash sequences have to do with changing the character,
283 or characters following them. C<\l> will lowercase the character following
284 it, while C<\u> will uppercase (or, more accurately, titlecase) the
285 character following it. (They perform similar functionality as the
286 functions C<lcfirst> and C<ucfirst>).
288 To uppercase or lowercase several characters, one might want to use
289 C<\L> or C<\U>, which will lowercase/uppercase all characters following
290 them, until either the end of the pattern, or the next occurrence of
291 C<\E>, whatever comes first. They perform similar functionality as the
292 functions C<lc> and C<uc> do.
294 C<\Q> is used to escape all characters following, up to the next C<\E>
295 or the end of the pattern. C<\Q> adds a backslash to any character that
296 isn't a letter, digit or underscore. This will ensure that any character
297 between C<\Q> and C<\E> is matched literally, and will not be interpreted
298 by the regexp engine.
300 Mnemonic: I<L>owercase, I<U>ppercase, I<Q>uotemeta, I<E>nd.
306 $miranda = "(Miranda)";
307 $str =~ /\u$sid/; # Matches 'Sid'
308 $str =~ /\L$greg/; # Matches 'greg'
309 $str =~ /\Q$miranda\E/; # Matches '(Miranda)', as if the pattern
310 # had been written as /\(Miranda\)/
312 =head2 Character classes
314 Perl regular expressions have a large range of character classes. Some of
315 the character classes are written as a backslash sequence. We will briefly
316 discuss those here; full details of character classes can be found in
319 C<\w> is a character class that matches any single I<word> character (letters,
320 digits, underscore). C<\d> is a character class that matches any decimal digit,
321 while the character class C<\s> matches any whitespace character.
322 New in perl 5.10.0 are the classes C<\h> and C<\v> which match horizontal
323 and vertical whitespace characters.
325 The uppercase variants (C<\W>, C<\D>, C<\S>, C<\H>, and C<\V>) are
326 character classes that match any character that isn't a word character,
327 digit, whitespace, horizontal whitespace nor vertical whitespace.
329 Mnemonics: I<w>ord, I<d>igit, I<s>pace, I<h>orizontal, I<v>ertical.
331 =head3 Unicode classes
333 C<\pP> (where C<P> is a single letter) and C<\p{Property}> are used to
334 match a character that matches the given Unicode property; properties
335 include things like "letter", or "thai character". Capitalizing the
336 sequence to C<\PP> and C<\P{Property}> make the sequence match a character
337 that doesn't match the given Unicode property. For more details, see
338 L<perlrecharclass/Backslashed sequences> and
339 L<perlunicode/Unicode Character Properties>.
341 Mnemonic: I<p>roperty.
346 If capturing parenthesis are used in a regular expression, we can refer
347 to the part of the source string that was matched, and match exactly the
348 same thing. There are three ways of referring to such I<backreference>:
349 absolutely, relatively, and by name.
351 =for later add link to perlrecapture
353 =head3 Absolute referencing
355 A backslash sequence that starts with a backslash and is followed by a
356 number is an absolute reference (but be aware of the caveat mentioned above).
357 If the number is I<N>, it refers to the Nth set of parentheses - whatever
358 has been matched by that set of parenthesis has to be matched by the C<\N>
363 /(\w+) \1/; # Finds a duplicated word, (e.g. "cat cat").
364 /(.)(.)\2\1/; # Match a four letter palindrome (e.g. "ABBA").
367 =head3 Relative referencing
369 New in perl 5.10.0 is a different way of referring to capture buffers: C<\g>.
370 C<\g> takes a number as argument, with the number in curly braces (the
371 braces are optional). If the number (N) does not have a sign, it's a reference
372 to the Nth capture group (so C<\g{2}> is equivalent to C<\2> - except that
373 C<\g> always refers to a capture group and will never be seen as an octal
374 escape). If the number is negative, the reference is relative, referring to
375 the Nth group before the C<\g{-N}>.
377 The big advantage of C<\g{-N}> is that it makes it much easier to write
378 patterns with references that can be interpolated in larger patterns,
379 even if the larger pattern also contains capture groups.
388 \g{-1} # Refers to buffer 3 (B)
389 \g{-3} # Refers to buffer 1 (A)
391 /x; # Matches "ABBA".
393 my $qr = qr /(.)(.)\g{-2}\g{-1}/; # Matches 'abab', 'cdcd', etc.
394 /$qr$qr/ # Matches 'ababcdcd'.
396 =head3 Named referencing
398 Also new in perl 5.10.0 is the use of named capture buffers, which can be
399 referred to by name. This is done with C<\g{name}>, which is a
400 backreference to the capture buffer with the name I<name>.
402 To be compatible with .Net regular expressions, C<\g{name}> may also be
403 written as C<\k{name}>, C<< \k<name> >> or C<\k'name'>.
405 Note that C<\g{}> has the potential to be ambiguous, as it could be a named
406 reference, or an absolute or relative reference (if its argument is numeric).
407 However, names are not allowed to start with digits, nor are they allowed to
408 contain a hyphen, so there is no ambiguity.
412 /(?<word>\w+) \g{word}/ # Finds duplicated word, (e.g. "cat cat")
413 /(?<word>\w+) \k{word}/ # Same.
414 /(?<word>\w+) \k<word>/ # Same.
415 /(?<letter1>.)(?<letter2>.)\g{letter2}\g{letter1}/
416 # Match a four letter palindrome (e.g. "ABBA")
420 Assertions are conditions that have to be true; they don't actually
421 match parts of the substring. There are six assertions that are written as
428 C<\A> only matches at the beginning of the string. If the C</m> modifier
429 isn't used, then C</\A/> is equivalent with C</^/>. However, if the C</m>
430 modifier is used, then C</^/> matches internal newlines, but the meaning
431 of C</\A/> isn't changed by the C</m> modifier. C<\A> matches at the beginning
432 of the string regardless whether the C</m> modifier is used.
436 C<\z> and C<\Z> match at the end of the string. If the C</m> modifier isn't
437 used, then C</\Z/> is equivalent with C</$/>, that is, it matches at the
438 end of the string, or before the newline at the end of the string. If the
439 C</m> modifier is used, then C</$/> matches at internal newlines, but the
440 meaning of C</\Z/> isn't changed by the C</m> modifier. C<\Z> matches at
441 the end of the string (or just before a trailing newline) regardless whether
442 the C</m> modifier is used.
444 C<\z> is just like C<\Z>, except that it will not match before a trailing
445 newline. C<\z> will only match at the end of the string - regardless of the
446 modifiers used, and not before a newline.
450 C<\G> is usually only used in combination with the C</g> modifier. If the
451 C</g> modifier is used (and the match is done in scalar context), Perl will
452 remember where in the source string the last match ended, and the next time,
453 it will start the match from where it ended the previous time.
455 C<\G> matches the point where the previous match ended, or the beginning
456 of the string if there was no previous match.
458 =for later add link to perlremodifiers
464 C<\b> matches at any place between a word and a non-word character; C<\B>
465 matches at any place between characters where C<\b> doesn't match. C<\b>
466 and C<\B> assume there's a non-word character before the beginning and after
467 the end of the source string; so C<\b> will match at the beginning (or end)
468 of the source string if the source string begins (or ends) with a word
469 character. Otherwise, C<\B> will match.
471 Mnemonic: I<b>oundary.
477 "cat" =~ /\Acat/; # Match.
478 "cat" =~ /cat\Z/; # Match.
479 "cat\n" =~ /cat\Z/; # Match.
480 "cat\n" =~ /cat\z/; # No match.
482 "cat" =~ /\bcat\b/; # Matches.
483 "cats" =~ /\bcat\b/; # No match.
484 "cat" =~ /\bcat\B/; # No match.
485 "cats" =~ /\bcat\B/; # Match.
487 while ("cat dog" =~ /(\w+)/g) {
488 print $1; # Prints 'catdog'
490 while ("cat dog" =~ /\G(\w+)/g) {
491 print $1; # Prints 'cat'
496 Here we document the backslash sequences that don't fall in one of the
497 categories above. They are:
503 C<\C> always matches a single octet, even if the source string is encoded
504 in UTF-8 format, and the character to be matched is a multi-octet character.
505 C<\C> was introduced in perl 5.6.
511 This is new in perl 5.10.0. Anything that is matched left of C<\K> is
512 not included in C<$&> - and will not be replaced if the pattern is
513 used in a substitution. This will allow you to write C<s/PAT1 \K PAT2/REPL/x>
514 instead of C<s/(PAT1) PAT2/${1}REPL/x> or C<s/(?<=PAT1) PAT2/REPL/x>.
520 This is a new experimental feature in perl 5.12.0. It matches any character
521 that is not a newline. It is a short-hand for writing C<[^\n]>, and is
522 identical to the C<.> metasymbol, except under the C</s> flag, which changes
523 the meaning of C<.>, but not C<\N>.
525 Note that C<\N{...}> can mean a
526 L<named or numbered character|/Named or numbered characters>.
528 Mnemonic: Complement of I<\n>.
533 C<\R> matches a I<generic newline>, that is, anything that is considered
534 a newline by Unicode. This includes all characters matched by C<\v>
535 (vertical whitespace), and the multi character sequence C<"\x0D\x0A">
536 (carriage return followed by a line feed, aka the network newline, or
537 the newline used in Windows text files). C<\R> is equivalent to
538 C<< (?>\x0D\x0A)|\v) >>. Since C<\R> can match a sequence of more than one
539 character, it cannot be put inside a bracketed character class; C</[\R]/> is an
540 error; use C<\v> instead. C<\R> was introduced in perl 5.10.0.
542 Mnemonic: none really. C<\R> was picked because PCRE already uses C<\R>,
543 and more importantly because Unicode recommends such a regular expression
544 metacharacter, and suggests C<\R> as the notation.
549 This matches a Unicode I<extended grapheme cluster>.
551 C<\X> matches quite well what normal (non-Unicode-programmer) usage
552 would consider a single character. As an example, consider a G with some sort
553 of diacritic mark, such as an arrow. There is no such single character in
554 Unicode, but one can be composed by using a G followed by a Unicode "COMBINING
555 UPWARDS ARROW BELOW", and would be displayed by Unicode-aware software as if it
556 were a single character.
558 Mnemonic: eI<X>tended Unicode character.
564 "\x{256}" =~ /^\C\C$/; # Match as chr (256) takes 2 octets in UTF-8.
566 $str =~ s/foo\Kbar/baz/g; # Change any 'bar' following a 'foo' to 'baz'.
567 $str =~ s/(.)\K\1//g; # Delete duplicated characters.
569 "\n" =~ /^\R$/; # Match, \n is a generic newline.
570 "\r" =~ /^\R$/; # Match, \r is a generic newline.
571 "\r\n" =~ /^\R$/; # Match, \r\n is a generic newline.
573 "P\x{0307}" =~ /^\X$/ # \X matches a P with a dot above.