This is a live mirror of the Perl 5 development currently hosted at https://github.com/perl/perl5
\N is no longer experimental
[perl5.git] / pod / perlrecharclass.pod
CommitLineData
8a118206 1=head1 NAME
ea449505 2X<character class>
8a118206
RGS
3
4perlrecharclass - Perl Regular Expression Character Classes
5
6=head1 DESCRIPTION
7
8The top level documentation about Perl regular expressions
9is found in L<perlre>.
10
11This manual page discusses the syntax and use of character
6b83a163 12classes in Perl regular expressions.
8a118206 13
6b83a163 14A character class is a way of denoting a set of characters
8a118206 15in such a way that one character of the set is matched.
6b83a163 16It's important to remember that: matching a character class
8a118206
RGS
17consumes exactly one character in the source string. (The source
18string is the string the regular expression is matched against.)
19
20There are three types of character classes in Perl regular
6b83a163 21expressions: the dot, backslash sequences, and the form enclosed in square
ea449505 22brackets. Keep in mind, though, that often the term "character class" is used
6b83a163 23to mean just the bracketed form. Certainly, most Perl documentation does that.
8a118206
RGS
24
25=head2 The dot
26
27The dot (or period), C<.> is probably the most used, and certainly
28the most well-known character class. By default, a dot matches any
5db9882c 29character, except for the newline. That default can be changed to
6b83a163
KW
30add matching the newline by using the I<single line> modifier: either
31for the entire regular expression with the C</s> modifier, or
2171640d 32locally with C<(?s)>. (The C<\N> backslash sequence, described
6b83a163
KW
33below, matches any character except newline without regard to the
34I<single line> modifier.)
8a118206
RGS
35
36Here are some examples:
37
38 "a" =~ /./ # Match
39 "." =~ /./ # Match
40 "" =~ /./ # No match (dot has to match a character)
41 "\n" =~ /./ # No match (dot does not match a newline)
42 "\n" =~ /./s # Match (global 'single line' modifier)
43 "\n" =~ /(?s:.)/ # Match (local 'single line' modifier)
44 "ab" =~ /^.$/ # No match (dot matches one character)
45
6b83a163 46=head2 Backslash sequences
82206b5e 47X<\w> X<\W> X<\s> X<\S> X<\d> X<\D> X<\p> X<\P>
ea449505
KW
48X<\N> X<\v> X<\V> X<\h> X<\H>
49X<word> X<whitespace>
8a118206 50
6b83a163
KW
51A backslash sequence is a sequence of characters, the first one of which is a
52backslash. Perl ascribes special meaning to many such sequences, and some of
53these are character classes. That is, they match a single character each,
54provided that the character belongs to the specific set of characters defined
55by the sequence.
8a118206 56
6b83a163
KW
57Here's a list of the backslash sequences that are character classes. They
58are discussed in more detail below. (For the backslash sequences that aren't
59character classes, see L<perlrebackslash>.)
8a118206 60
6b83a163
KW
61 \d Match a decimal digit character.
62 \D Match a non-decimal-digit character.
8a118206
RGS
63 \w Match a "word" character.
64 \W Match a non-"word" character.
ea449505
KW
65 \s Match a whitespace character.
66 \S Match a non-whitespace character.
67 \h Match a horizontal whitespace character.
68 \H Match a character that isn't horizontal whitespace.
ea449505
KW
69 \v Match a vertical whitespace character.
70 \V Match a character that isn't vertical whitespace.
6b83a163
KW
71 \N Match a character that isn't a newline. Experimental.
72 \pP, \p{Prop} Match a character that has the given Unicode property.
6c5a041f 73 \PP, \P{Prop} Match a character that doesn't have the Unicode property
8a118206 74
1433f837
KW
75=head3 \N
76
2171640d 77C<\N>, available starting in v5.12, like the dot, matches any
1433f837
KW
78character that is not a newline. The difference is that C<\N> is not influenced
79by the I<single line> regular expression modifier (see L</The dot> above). Note
80that the form C<\N{...}> may mean something completely different. When the
81C<{...}> is a L<quantifier|perlre/Quantifiers>, it means to match a non-newline
82character that many times. For example, C<\N{3}> means to match 3
83non-newlines; C<\N{5,}> means to match 5 or more non-newlines. But if C<{...}>
84is not a legal quantifier, it is presumed to be a named character. See
85L<charnames> for those. For example, none of C<\N{COLON}>, C<\N{4F}>, and
86C<\N{F4}> contain legal quantifiers, so Perl will try to find characters whose
87names are respectively C<COLON>, C<4F>, and C<F4>.
88
8a118206
RGS
89=head3 Digits
90
b6538e4f 91C<\d> matches a single character considered to be a decimal I<digit>.
5db9882c 92If the C</a> regular expression modifier is in effect, it matches [0-9].
582da942 93Otherwise, it
82206b5e
KW
94matches anything that is matched by C<\p{Digit}>, which includes [0-9].
95(An unlikely possible exception is that under locale matching rules, the
96current locale might not have [0-9] matched by C<\d>, and/or might match
97other characters whose code point is less than 256. Such a locale
98definition would be in violation of the C language standard, but Perl
99doesn't currently assume anything in regard to this.)
100
101What this means is that unless the C</a> modifier is in effect C<\d> not
102only matches the digits '0' - '9', but also Arabic, Devanagari, and
103digits from other languages. This may cause some confusion, and some
104security issues.
105
106Some digits that C<\d> matches look like some of the [0-9] ones, but
107have different values. For example, BENGALI DIGIT FOUR (U+09EA) looks
108very much like an ASCII DIGIT EIGHT (U+0038). An application that
109is expecting only the ASCII digits might be misled, or if the match is
110C<\d+>, the matched string might contain a mixture of digits from
111different writing systems that look like they signify a number different
67592e11 112than they actually do. L<Unicode::UCD/num()> can
e397bccf 113be used to safely
82206b5e
KW
114calculate the value, returning C<undef> if the input string contains
115such a mixture.
116
117What C<\p{Digit}> means (and hence C<\d> except under the C</a>
118modifier) is C<\p{General_Category=Decimal_Number}>, or synonymously,
119C<\p{General_Category=Digit}>. Starting with Unicode version 4.1, this
120is the same set of characters matched by C<\p{Numeric_Type=Decimal}>.
6b83a163
KW
121But Unicode also has a different property with a similar name,
122C<\p{Numeric_Type=Digit}>, which matches a completely different set of
82206b5e
KW
123characters. These characters are things such as C<CIRCLED DIGIT ONE>
124or subscripts, or are from writing systems that lack all ten digits.
6b83a163 125
82206b5e
KW
126The design intent is for C<\d> to exactly match the set of characters
127that can safely be used with "normal" big-endian positional decimal
128syntax, where, for example 123 means one 'hundred', plus two 'tens',
129plus three 'ones'. This positional notation does not necessarily apply
130to characters that match the other type of "digit",
131C<\p{Numeric_Type=Digit}>, and so C<\d> doesn't match them.
6b83a163 132
e2cfb18c 133The Tamil digits (U+0BE6 - U+0BEF) can also legally be
82206b5e
KW
134used in old-style Tamil numbers in which they would appear no more than
135one in a row, separated by characters that mean "times 10", "times 100",
136etc. (See L<http://www.unicode.org/notes/tn21>.)
8a118206 137
b6538e4f 138Any character not matched by C<\d> is matched by C<\D>.
8a118206
RGS
139
140=head3 Word characters
141
ea449505 142A C<\w> matches a single alphanumeric character (an alphabetic character, or a
41805eb9
KW
143decimal digit); or a connecting punctuation character, such as an
144underscore ("_"); or a "mark" character (like some sort of accent) that
145attaches to one of those. It does not match a whole word. To match a
146whole word, use C<\w+>. This isn't the same thing as matching an
147English word, but in the ASCII range it is the same as a string of
148Perl-identifier characters.
82206b5e
KW
149
150=over
151
152=item If the C</a> modifier is in effect ...
153
154C<\w> matches the 63 characters [a-zA-Z0-9_].
155
156=item otherwise ...
157
158=over
159
160=item For code points above 255 ...
161
162C<\w> matches the same as C<\p{Word}> matches in this range. That is,
163it matches Thai letters, Greek letters, etc. This includes connector
d35dd6c6 164punctuation (like the underscore) which connect two words together, or
b6538e4f 165diacritics, such as a C<COMBINING TILDE> and the modifier letters, which
82206b5e
KW
166are generally used to add auxiliary markings to letters.
167
168=item For code points below 256 ...
169
170=over
171
172=item if locale rules are in effect ...
173
174C<\w> matches the platform's native underscore character plus whatever
175the locale considers to be alphanumeric.
176
177=item if Unicode rules are in effect or if on an EBCDIC platform ...
178
179C<\w> matches exactly what C<\p{Word}> matches.
180
181=item otherwise ...
182
183C<\w> matches [a-zA-Z0-9_].
184
185=back
186
187=back
188
189=back
190
191Which rules apply are determined as described in L<perlre/Which character set modifier is in effect?>.
8a118206 192
6b83a163
KW
193There are a number of security issues with the full Unicode list of word
194characters. See L<http://unicode.org/reports/tr36>.
195
196Also, for a somewhat finer-grained set of characters that are in programming
197language identifiers beyond the ASCII range, you may wish to instead use the
e2cfb18c
KW
198more customized L</Unicode Properties>, C<\p{ID_Start}>,
199C<\p{ID_Continue}>, C<\p{XID_Start}>, and C<\p{XID_Continue}>. See
200L<http://unicode.org/reports/tr31>.
6b83a163 201
b6538e4f 202Any character not matched by C<\w> is matched by C<\W>.
8a118206 203
ea449505
KW
204=head3 Whitespace
205
82206b5e
KW
206C<\s> matches any single character considered whitespace.
207
208=over
209
210=item If the C</a> modifier is in effect ...
211
d28d8023
KW
212In all Perl versions, C<\s> matches the 5 characters [\t\n\f\r ]; that
213is, the horizontal tab,
214the newline, the form feed, the carriage return, and the space.
215Starting in Perl v5.18, experimentally, it also matches the vertical tab, C<\cK>.
216See note C<[1]> below for a discussion of this.
82206b5e
KW
217
218=item otherwise ...
219
220=over
221
222=item For code points above 255 ...
223
224C<\s> matches exactly the code points above 255 shown with an "s" column
225in the table below.
226
227=item For code points below 256 ...
228
229=over
230
231=item if locale rules are in effect ...
232
d28d8023 233C<\s> matches whatever the locale considers to be whitespace.
82206b5e
KW
234
235=item if Unicode rules are in effect or if on an EBCDIC platform ...
236
237C<\s> matches exactly the characters shown with an "s" column in the
238table below.
239
240=item otherwise ...
241
d28d8023
KW
242C<\s> matches [\t\n\f\r\cK ] and, starting, experimentally in Perl
243v5.18, the vertical tab, C<\cK>.
244(See note C<[1]> below for a discussion of this.)
82206b5e
KW
245Note that this list doesn't include the non-breaking space.
246
247=back
248
249=back
250
251=back
252
253Which rules apply are determined as described in L<perlre/Which character set modifier is in effect?>.
8a118206 254
b6538e4f 255Any character not matched by C<\s> is matched by C<\S>.
8a118206 256
b6538e4f 257C<\h> matches any character considered horizontal whitespace;
8129baca 258this includes the platform's space and tab characters and several others
b6538e4f 259listed in the table below. C<\H> matches any character
8129baca
KW
260not considered horizontal whitespace. They use the platform's native
261character set, and do not consider any locale that may otherwise be in
262use.
ea449505 263
b6538e4f 264C<\v> matches any character considered vertical whitespace;
8129baca 265this includes the platform's carriage return and line feed characters (newline)
b6538e4f
TC
266plus several other characters, all listed in the table below.
267C<\V> matches any character not considered vertical whitespace.
8129baca
KW
268They use the platform's native character set, and do not consider any
269locale that may otherwise be in use.
8a118206
RGS
270
271C<\R> matches anything that can be considered a newline under Unicode
272rules. It's not a character class, as it can match a multi-character
273sequence. Therefore, it cannot be used inside a bracketed character
8129baca
KW
274class; use C<\v> instead (vertical whitespace). It uses the platform's
275native character set, and does not consider any locale that may
276otherwise be in use.
ea449505 277Details are discussed in L<perlrebackslash>.
8a118206 278
82206b5e 279Note that unlike C<\s> (and C<\d> and C<\w>), C<\h> and C<\v> always match
8129baca
KW
280the same characters, without regard to other factors, such as the active
281locale or whether the source string is in UTF-8 format.
8a118206 282
d28d8023
KW
283One might think that C<\s> is equivalent to C<[\h\v]>. This is indeed true
284starting in Perl v5.18, but prior to that, the sole difference was that the
285vertical tab (C<"\cK">) was not matched by C<\s>.
8a118206
RGS
286
287The following table is a complete listing of characters matched by
82206b5e 288C<\s>, C<\h> and C<\v> as of Unicode 6.0.
8a118206 289
582da942 290The first column gives the Unicode code point of the character (in hex format),
8a118206 291the second column gives the (Unicode) name. The third column indicates
ea449505
KW
292by which class(es) the character is matched (assuming no locale or EBCDIC code
293page is in effect that changes the C<\s> matching).
8a118206 294
fc28d2a3
KW
295 0x0009 CHARACTER TABULATION h s
296 0x000a LINE FEED (LF) vs
d28d8023 297 0x000b LINE TABULATION vs [1]
fc28d2a3
KW
298 0x000c FORM FEED (FF) vs
299 0x000d CARRIAGE RETURN (CR) vs
300 0x0020 SPACE h s
d28d8023
KW
301 0x0085 NEXT LINE (NEL) vs [2]
302 0x00a0 NO-BREAK SPACE h s [2]
fc28d2a3
KW
303 0x1680 OGHAM SPACE MARK h s
304 0x180e MONGOLIAN VOWEL SEPARATOR h s
305 0x2000 EN QUAD h s
306 0x2001 EM QUAD h s
307 0x2002 EN SPACE h s
308 0x2003 EM SPACE h s
309 0x2004 THREE-PER-EM SPACE h s
310 0x2005 FOUR-PER-EM SPACE h s
311 0x2006 SIX-PER-EM SPACE h s
312 0x2007 FIGURE SPACE h s
313 0x2008 PUNCTUATION SPACE h s
314 0x2009 THIN SPACE h s
315 0x200a HAIR SPACE h s
316 0x2028 LINE SEPARATOR vs
317 0x2029 PARAGRAPH SEPARATOR vs
318 0x202f NARROW NO-BREAK SPACE h s
319 0x205f MEDIUM MATHEMATICAL SPACE h s
320 0x3000 IDEOGRAPHIC SPACE h s
8a118206
RGS
321
322=over 4
323
324=item [1]
325
d28d8023
KW
326Prior to Perl v5.18, C<\s> did not match the vertical tab. The change
327in v5.18 is considered an experiment, which means it could be backed out
328in v5.20 or v5.22 if experience indicates that it breaks too much
329existing code. If this change adversely affects you, send email to
330C<perlbug@perl.org>; if it affects you positively, email
331C<perlthanks@perl.org>. In the meantime, C<[^\S\cK]> (obscurely)
332matches what C<\s> traditionally did.
333
334=item [2]
335
82206b5e
KW
336NEXT LINE and NO-BREAK SPACE may or may not match C<\s> depending
337on the rules in effect. See
338L<the beginning of this section|/Whitespace>.
8a118206
RGS
339
340=back
341
8a118206
RGS
342=head3 Unicode Properties
343
c1c4ae3a
KW
344C<\pP> and C<\p{Prop}> are character classes to match characters that fit given
345Unicode properties. One letter property names can be used in the C<\pP> form,
346with the property name following the C<\p>, otherwise, braces are required.
347When using braces, there is a single form, which is just the property name
348enclosed in the braces, and a compound form which looks like C<\p{name=value}>,
b6538e4f 349which means to match if the property "name" for the character has that particular
c1c4ae3a 350"value".
e1b711da
KW
351For instance, a match for a number can be written as C</\pN/> or as
352C</\p{Number}/>, or as C</\p{Number=True}/>.
353Lowercase letters are matched by the property I<Lowercase_Letter> which
e2cfb18c 354has the short form I<Ll>. They need the braces, so are written as C</\p{Ll}/> or
e1b711da
KW
355C</\p{Lowercase_Letter}/>, or C</\p{General_Category=Lowercase_Letter}/>
356(the underscores are optional).
357C</\pLl/> is valid, but means something different.
8a118206
RGS
358It matches a two character string: a letter (Unicode property C<\pL>),
359followed by a lowercase C<l>.
360
bc943be5 361If locale rules are not in effect, the use of
82206b5e 362a Unicode property will force the regular expression into using Unicode
bc943be5 363rules, if it isn't already.
82206b5e 364
56ca34ca
KW
365Note that almost all properties are immune to case-insensitive matching.
366That is, adding a C</i> regular expression modifier does not change what
82206b5e 367they match. There are two sets that are affected. The first set is
56ca34ca
KW
368C<Uppercase_Letter>,
369C<Lowercase_Letter>,
370and C<Titlecase_Letter>,
371all of which match C<Cased_Letter> under C</i> matching.
b6538e4f 372The second set is
56ca34ca
KW
373C<Uppercase>,
374C<Lowercase>,
375and C<Titlecase>,
376all of which match C<Cased> under C</i> matching.
377(The difference between these sets is that some things, such as Roman
e2cfb18c 378numerals, come in both upper and lower case, so they are C<Cased>, but
b6538e4f 379aren't considered to be letters, so they aren't C<Cased_Letter>s. They're
82206b5e
KW
380actually C<Letter_Number>s.)
381This set also includes its subsets C<PosixUpper> and C<PosixLower>, both
e2cfb18c 382of which under C</i> match C<PosixAlpha>.
56ca34ca
KW
383
384For more details on Unicode properties, see L<perlunicode/Unicode
385Character Properties>; for a
e1b711da 386complete list of possible properties, see
56ca34ca
KW
387L<perluniprops/Properties accessible through \p{} and \P{}>,
388which notes all forms that have C</i> differences.
e1b711da 389It is also possible to define your own properties. This is discussed in
8a118206
RGS
390L<perlunicode/User-Defined Character Properties>.
391
94b42e47
KW
392Unicode properties are defined (surprise!) only on Unicode code points.
393A warning is raised and all matches fail on non-Unicode code points
394(those above the legal Unicode maximum of 0x10FFFF). This can be
395somewhat surprising,
396
397 chr(0x110000) =~ \p{ASCII_Hex_Digit=True} # Fails.
398 chr(0x110000) =~ \p{ASCII_Hex_Digit=False} # Also fails!
399
400Even though these two matches might be thought of as complements, they
401are so only on Unicode code points.
402
8a118206
RGS
403=head4 Examples
404
405 "a" =~ /\w/ # Match, "a" is a 'word' character.
406 "7" =~ /\w/ # Match, "7" is a 'word' character as well.
407 "a" =~ /\d/ # No match, "a" isn't a digit.
408 "7" =~ /\d/ # Match, "7" is a digit.
ea449505 409 " " =~ /\s/ # Match, a space is whitespace.
8a118206
RGS
410 "a" =~ /\D/ # Match, "a" is a non-digit.
411 "7" =~ /\D/ # No match, "7" is not a non-digit.
ea449505 412 " " =~ /\S/ # No match, a space is not non-whitespace.
8a118206 413
ea449505
KW
414 " " =~ /\h/ # Match, space is horizontal whitespace.
415 " " =~ /\v/ # No match, space is not vertical whitespace.
416 "\r" =~ /\v/ # Match, a return is vertical whitespace.
8a118206
RGS
417
418 "a" =~ /\pL/ # Match, "a" is a letter.
419 "a" =~ /\p{Lu}/ # No match, /\p{Lu}/ matches upper case letters.
420
421 "\x{0e0b}" =~ /\p{Thai}/ # Match, \x{0e0b} is the character
422 # 'THAI CHARACTER SO SO', and that's in
423 # Thai Unicode class.
ea449505 424 "a" =~ /\P{Lao}/ # Match, as "a" is not a Laotian character.
8a118206 425
82206b5e
KW
426It is worth emphasizing that C<\d>, C<\w>, etc, match single characters, not
427complete numbers or words. To match a number (that consists of digits),
428use C<\d+>; to match a word, use C<\w+>. But be aware of the security
429considerations in doing so, as mentioned above.
8a118206
RGS
430
431=head2 Bracketed Character Classes
432
433The third form of character class you can use in Perl regular expressions
6b83a163 434is the bracketed character class. In its simplest form, it lists the characters
c1c4ae3a 435that may be matched, surrounded by square brackets, like this: C<[aeiou]>.
ea449505 436This matches one of C<a>, C<e>, C<i>, C<o> or C<u>. Like the other
1f59b283 437character classes, exactly one character is matched.* To match
ea449505 438a longer string consisting of characters mentioned in the character
6b83a163 439class, follow the character class with a L<quantifier|perlre/Quantifiers>. For
b6538e4f 440instance, C<[aeiou]+> matches one or more lowercase English vowels.
8a118206
RGS
441
442Repeating a character in a character class has no
443effect; it's considered to be in the set only once.
444
445Examples:
446
447 "e" =~ /[aeiou]/ # Match, as "e" is listed in the class.
448 "p" =~ /[aeiou]/ # No match, "p" is not listed in the class.
449 "ae" =~ /^[aeiou]$/ # No match, a character class only matches
450 # a single character.
451 "ae" =~ /^[aeiou]+$/ # Match, due to the quantifier.
452
1f59b283
KW
453 -------
454
df0e3973 455* There is an exception to a bracketed character class matching a
1cecf2c0 456single character only. When the class is to match caselessly under C</i>
9d53c457
KW
457matching rules, and a character that is explicitly mentioned inside the
458class matches a
1f59b283
KW
459multiple-character sequence caselessly under Unicode rules, the class
460(when not L<inverted|/Negation>) will also match that sequence. For
461example, Unicode says that the letter C<LATIN SMALL LETTER SHARP S>
462should match the sequence C<ss> under C</i> rules. Thus,
463
464 'ss' =~ /\A\N{LATIN SMALL LETTER SHARP S}\z/i # Matches
465 'ss' =~ /\A[aeioust\N{LATIN SMALL LETTER SHARP S}]\z/i # Matches
466
9d53c457
KW
467For this to happen, the character must be explicitly specified, and not
468be part of a multi-character range (not even as one of its endpoints).
469(L</Character Ranges> will be explained shortly.) Therefore,
470
471 'ss' =~ /\A[\0-\x{ff}]\z/i # Doesn't match
472 'ss' =~ /\A[\0-\N{LATIN SMALL LETTER SHARP S}]\z/i # No match
473 'ss' =~ /\A[\xDF-\xDF]\z/i # Matches on ASCII platforms, since \XDF
474 # is LATIN SMALL LETTER SHARP S, and the
475 # range is just a single element
476
477Note that it isn't a good idea to specify these types of ranges anyway.
478
8a118206
RGS
479=head3 Special Characters Inside a Bracketed Character Class
480
481Most characters that are meta characters in regular expressions (that
df225385 482is, characters that carry a special meaning like C<.>, C<*>, or C<(>) lose
8a118206
RGS
483their special meaning and can be used inside a character class without
484the need to escape them. For instance, C<[()]> matches either an opening
485parenthesis, or a closing parenthesis, and the parens inside the character
486class don't group or capture.
487
488Characters that may carry a special meaning inside a character class are:
489C<\>, C<^>, C<->, C<[> and C<]>, and are discussed below. They can be
490escaped with a backslash, although this is sometimes not needed, in which
491case the backslash may be omitted.
492
493The sequence C<\b> is special inside a bracketed character class. While
6b83a163 494outside the character class, C<\b> is an assertion indicating a point
8a118206
RGS
495that does not have either two word characters or two non-word characters
496on either side, inside a bracketed character class, C<\b> matches a
497backspace character.
498
df225385
KW
499The sequences
500C<\a>,
501C<\c>,
502C<\e>,
503C<\f>,
504C<\n>,
e526e8bb 505C<\N{I<NAME>}>,
765fa144 506C<\N{U+I<hex char>}>,
df225385
KW
507C<\r>,
508C<\t>,
509and
510C<\x>
06ee63cd
KW
511are also special and have the same meanings as they do outside a
512bracketed character class. (However, inside a bracketed character
513class, if C<\N{I<NAME>}> expands to a sequence of characters, only the first
514one in the sequence is used, with a warning.)
df225385 515
ea449505
KW
516Also, a backslash followed by two or three octal digits is considered an octal
517number.
df225385 518
6b83a163
KW
519A C<[> is not special inside a character class, unless it's the start of a
520POSIX character class (see L</POSIX Character Classes> below). It normally does
521not need escaping.
8a118206 522
6b83a163
KW
523A C<]> is normally either the end of a POSIX character class (see
524L</POSIX Character Classes> below), or it signals the end of the bracketed
525character class. If you want to include a C<]> in the set of characters, you
526must generally escape it.
b6538e4f 527
8a118206
RGS
528However, if the C<]> is the I<first> (or the second if the first
529character is a caret) character of a bracketed character class, it
530does not denote the end of the class (as you cannot have an empty class)
531and is considered part of the set of characters that can be matched without
532escaping.
533
534Examples:
535
536 "+" =~ /[+?*]/ # Match, "+" in a character class is not special.
537 "\cH" =~ /[\b]/ # Match, \b inside in a character class
c1c4ae3a 538 # is equivalent to a backspace.
8a118206
RGS
539 "]" =~ /[][]/ # Match, as the character class contains.
540 # both [ and ].
541 "[]" =~ /[[]]/ # Match, the pattern contains a character class
542 # containing just ], and the character class is
543 # followed by a ].
544
545=head3 Character Ranges
546
547It is not uncommon to want to match a range of characters. Luckily, instead
b6538e4f 548of listing all characters in the range, one may use the hyphen (C<->).
8a118206 549If inside a bracketed character class you have two characters separated
b6538e4f 550by a hyphen, it's treated as if all characters between the two were in
8a118206 551the class. For instance, C<[0-9]> matches any ASCII digit, and C<[a-m]>
e2cfb18c 552matches any lowercase letter from the first half of the ASCII alphabet.
8a118206
RGS
553
554Note that the two characters on either side of the hyphen are not
765fa144 555necessarily both letters or both digits. Any character is possible,
8a118206 556although not advisable. C<['-?]> contains a range of characters, but
b6538e4f 557most people will not know which characters that means. Furthermore,
8a118206
RGS
558such ranges may lead to portability problems if the code has to run on
559a platform that uses a different character set, such as EBCDIC.
560
ea449505
KW
561If a hyphen in a character class cannot syntactically be part of a range, for
562instance because it is the first or the last character of the character class,
b6538e4f
TC
563or if it immediately follows a range, the hyphen isn't special, and so is
564considered a character to be matched literally. If you want a hyphen in
565your set of characters to be matched and its position in the class is such
566that it could be considered part of a range, you must escape that hyphen
567with a backslash.
8a118206
RGS
568
569Examples:
570
571 [a-z] # Matches a character that is a lower case ASCII letter.
c1c4ae3a
KW
572 [a-fz] # Matches any letter between 'a' and 'f' (inclusive) or
573 # the letter 'z'.
8a118206
RGS
574 [-z] # Matches either a hyphen ('-') or the letter 'z'.
575 [a-f-m] # Matches any letter between 'a' and 'f' (inclusive), the
576 # hyphen ('-'), or the letter 'm'.
577 ['-?] # Matches any of the characters '()*+,-./0123456789:;<=>?
578 # (But not on an EBCDIC platform).
579
580
581=head3 Negation
582
583It is also possible to instead list the characters you do not want to
584match. You can do so by using a caret (C<^>) as the first character in the
b6538e4f 585character class. For instance, C<[^a-z]> matches any character that is not a
e2cfb18c
KW
586lowercase ASCII letter, which therefore includes more than a million
587Unicode code points. The class is said to be "negated" or "inverted".
8a118206
RGS
588
589This syntax make the caret a special character inside a bracketed character
590class, but only if it is the first character of the class. So if you want
82206b5e 591the caret as one of the characters to match, either escape the caret or
e2cfb18c 592else don't list it first.
8a118206 593
1f59b283 594In inverted bracketed character classes, Perl ignores the Unicode rules
56e1c5aa
KW
595that normally say that certain characters should match a sequence of
596multiple characters under caseless C</i> matching. Following those
597rules could lead to highly confusing situations:
1f59b283 598
582da942 599 "ss" =~ /^[^\xDF]+$/ui; # Matches!
1f59b283
KW
600
601This should match any sequences of characters that aren't C<\xDF> nor
602what C<\xDF> matches under C</i>. C<"s"> isn't C<\xDF>, but Unicode
603says that C<"ss"> is what C<\xDF> matches under C</i>. So which one
604"wins"? Do you fail the match because the string has C<ss> or accept it
582da942
KW
605because it has an C<s> followed by another C<s>? Perl has chosen the
606latter.
1f59b283 607
8a118206
RGS
608Examples:
609
610 "e" =~ /[^aeiou]/ # No match, the 'e' is listed.
611 "x" =~ /[^aeiou]/ # Match, as 'x' isn't a lowercase vowel.
612 "^" =~ /[^^]/ # No match, matches anything that isn't a caret.
613 "^" =~ /[x^]/ # Match, caret is not special here.
614
615=head3 Backslash Sequences
616
ea449505 617You can put any backslash sequence character class (with the exception of
765fa144 618C<\N> and C<\R>) inside a bracketed character class, and it will act just
b6538e4f
TC
619as if you had put all characters matched by the backslash sequence inside the
620character class. For instance, C<[a-f\d]> matches any decimal digit, or any
6b83a163
KW
621of the lowercase letters between 'a' and 'f' inclusive.
622
623C<\N> within a bracketed character class must be of the forms C<\N{I<name>}>
765fa144 624or C<\N{U+I<hex char>}>, and NOT be the form that matches non-newlines,
6b83a163
KW
625for the same reason that a dot C<.> inside a bracketed character class loses
626its special meaning: it matches nearly anything, which generally isn't what you
627want to happen.
df225385 628
8a118206
RGS
629
630Examples:
631
632 /[\p{Thai}\d]/ # Matches a character that is either a Thai
633 # character, or a digit.
634 /[^\p{Arabic}()]/ # Matches a character that is neither an Arabic
635 # character, nor a parenthesis.
636
637Backslash sequence character classes cannot form one of the endpoints
6b83a163
KW
638of a range. Thus, you can't say:
639
640 /[\p{Thai}-\d]/ # Wrong!
8a118206 641
6b83a163 642=head3 POSIX Character Classes
ea449505 643X<character class> X<\p> X<\p{}>
ea449505
KW
644X<alpha> X<alnum> X<ascii> X<blank> X<cntrl> X<digit> X<graph>
645X<lower> X<print> X<punct> X<space> X<upper> X<word> X<xdigit>
8a118206 646
6b83a163
KW
647POSIX character classes have the form C<[:class:]>, where I<class> is
648name, and the C<[:> and C<:]> delimiters. POSIX character classes only appear
8a118206 649I<inside> bracketed character classes, and are a convenient and descriptive
82206b5e 650way of listing a group of characters.
6b83a163
KW
651
652Be careful about the syntax,
8a118206
RGS
653
654 # Correct:
655 $string =~ /[[:alpha:]]/
656
657 # Incorrect (will warn):
658 $string =~ /[:alpha:]/
659
660The latter pattern would be a character class consisting of a colon,
661and the letters C<a>, C<l>, C<p> and C<h>.
82206b5e 662POSIX character classes can be part of a larger bracketed character class.
b6538e4f 663For example,
ea449505
KW
664
665 [01[:alpha:]%]
666
667is valid and matches '0', '1', any alphabetic character, and the percent sign.
8a118206
RGS
668
669Perl recognizes the following POSIX character classes:
670
ea449505 671 alpha Any alphabetical character ("[A-Za-z]").
b6538e4f 672 alnum Any alphanumeric character. ("[A-Za-z0-9]")
ea449505 673 ascii Any character in the ASCII character set.
ea8b8ad2 674 blank A GNU extension, equal to a space or a horizontal tab ("\t").
ea449505
KW
675 cntrl Any control character. See Note [2] below.
676 digit Any decimal digit ("[0-9]"), equivalent to "\d".
677 graph Any printable character, excluding a space. See Note [3] below.
678 lower Any lowercase character ("[a-z]").
679 print Any printable character, including a space. See Note [4] below.
c1c4ae3a 680 punct Any graphical character excluding "word" characters. Note [5].
d28d8023
KW
681 space Any whitespace character. "\s" including the vertical tab
682 ("\cK").
ea449505
KW
683 upper Any uppercase character ("[A-Z]").
684 word A Perl extension ("[A-Za-z0-9_]"), equivalent to "\w".
685 xdigit Any hexadecimal digit ("[0-9a-fA-F]").
686
687Most POSIX character classes have two Unicode-style C<\p> property
688counterparts. (They are not official Unicode properties, but Perl extensions
689derived from official Unicode properties.) The table below shows the relation
690between POSIX character classes and these counterparts.
691
692One counterpart, in the column labelled "ASCII-range Unicode" in
b6538e4f 693the table, matches only characters in the ASCII character set.
ea449505
KW
694
695The other counterpart, in the column labelled "Full-range Unicode", matches any
696appropriate characters in the full Unicode character set. For example,
b6538e4f 697C<\p{Alpha}> matches not just the ASCII alphabetic characters, but any
82206b5e 698character in the entire Unicode character set considered alphabetic.
582da942 699An entry in the column labelled "backslash sequence" is a (short)
5db9882c 700equivalent.
ea449505 701
cbc24f92
KW
702 [[:...:]] ASCII-range Full-range backslash Note
703 Unicode Unicode sequence
ea449505 704 -----------------------------------------------------
cbc24f92
KW
705 alpha \p{PosixAlpha} \p{XPosixAlpha}
706 alnum \p{PosixAlnum} \p{XPosixAlnum}
82206b5e 707 ascii \p{ASCII}
cbc24f92
KW
708 blank \p{PosixBlank} \p{XPosixBlank} \h [1]
709 or \p{HorizSpace} [1]
710 cntrl \p{PosixCntrl} \p{XPosixCntrl} [2]
711 digit \p{PosixDigit} \p{XPosixDigit} \d
712 graph \p{PosixGraph} \p{XPosixGraph} [3]
713 lower \p{PosixLower} \p{XPosixLower}
714 print \p{PosixPrint} \p{XPosixPrint} [4]
715 punct \p{PosixPunct} \p{XPosixPunct} [5]
716 \p{PerlSpace} \p{XPerlSpace} \s [6]
717 space \p{PosixSpace} \p{XPosixSpace} [6]
718 upper \p{PosixUpper} \p{XPosixUpper}
719 word \p{PosixWord} \p{XPosixWord} \w
82206b5e 720 xdigit \p{PosixXDigit} \p{XPosixXDigit}
8a118206
RGS
721
722=over 4
723
ea449505
KW
724=item [1]
725
726C<\p{Blank}> and C<\p{HorizSpace}> are synonyms.
727
728=item [2]
8a118206 729
ea449505 730Control characters don't produce output as such, but instead usually control
b6538e4f 731the terminal somehow: for example, newline and backspace are control characters.
82206b5e 732In the ASCII range, characters whose code points are between 0 and 31 inclusive,
ea449505 733plus 127 (C<DEL>) are control characters.
8a118206 734
c1c4ae3a
KW
735On EBCDIC platforms, it is likely that the code page will define C<[[:cntrl:]]>
736to be the EBCDIC equivalents of the ASCII controls, plus the controls
bc943be5 737that in Unicode have code points from 128 through 159.
ea449505
KW
738
739=item [3]
8a118206
RGS
740
741Any character that is I<graphical>, that is, visible. This class consists
b6538e4f 742of all alphanumeric characters and all punctuation characters.
8a118206 743
ea449505 744=item [4]
8a118206 745
b6538e4f
TC
746All printable characters, which is the set of all graphical characters
747plus those whitespace characters which are not also controls.
ea449505 748
b6dac59a 749=item [5]
ea449505 750
b6538e4f 751C<\p{PosixPunct}> and C<[[:punct:]]> in the ASCII range match all
ea449505
KW
752non-controls, non-alphanumeric, non-space characters:
753C<[-!"#$%&'()*+,./:;<=E<gt>?@[\\\]^_`{|}~]> (although if a locale is in effect,
754it could alter the behavior of C<[[:punct:]]>).
755
cbc24f92
KW
756The similarly named property, C<\p{Punct}>, matches a somewhat different
757set in the ASCII range, namely
0be9b861
KW
758C<[-!"#%&'()*,./:;?@[\\\]_{}]>. That is, it is missing the nine
759characters C<[$+E<lt>=E<gt>^`|~]>.
6c5a041f
KW
760This is because Unicode splits what POSIX considers to be punctuation into two
761categories, Punctuation and Symbols.
762
e2cfb18c 763C<\p{XPosixPunct}> and (under Unicode rules) C<[[:punct:]]>, match what
765fa144
KW
764C<\p{PosixPunct}> matches in the ASCII range, plus what C<\p{Punct}>
765matches. This is different than strictly matching according to
766C<\p{Punct}>. Another way to say it is that
82206b5e
KW
767if Unicode rules are in effect, C<[[:punct:]]> matches all characters
768that Unicode considers punctuation, plus all ASCII-range characters that
769Unicode considers symbols.
8a118206 770
ea449505 771=item [6]
8a118206 772
d28d8023
KW
773C<\p{SpacePerl}> and C<\p{Space}> match identically starting with Perl
774v5.18. In earlier versions, these differ only in that in non-locale
775matching, C<\p{SpacePerl}> does not match the vertical tab, C<\cK>.
776Same for the two ASCII-only range forms.
8a118206
RGS
777
778=back
779
ab6199be
KW
780There are various other synonyms that can be used besides the names
781listed in the table. For example, C<\p{PosixAlpha}> can be written as
782C<\p{Alpha}>. All are listed in
783L<perluniprops/Properties accessible through \p{} and \P{}>,
784plus all characters matched by each ASCII-range property.
785
786Both the C<\p> counterparts always assume Unicode rules are in effect.
787On ASCII platforms, this means they assume that the code points from 128
788to 255 are Latin-1, and that means that using them under locale rules is
789unwise unless the locale is guaranteed to be Latin-1 or UTF-8. In contrast, the
790POSIX character classes are useful under locale rules. They are
791affected by the actual rules in effect, as follows:
792
793=over
794
795=item If the C</a> modifier, is in effect ...
796
797Each of the POSIX classes matches exactly the same as their ASCII-range
798counterparts.
799
800=item otherwise ...
801
802=over
803
804=item For code points above 255 ...
805
806The POSIX class matches the same as its Full-range counterpart.
807
808=item For code points below 256 ...
809
810=over
811
812=item if locale rules are in effect ...
813
8129baca
KW
814The POSIX class matches according to the locale, except that
815C<word> uses the platform's native underscore character, no matter what
816the locale is.
ab6199be
KW
817
818=item if Unicode rules are in effect or if on an EBCDIC platform ...
819
820The POSIX class matches the same as the Full-range counterpart.
821
822=item otherwise ...
823
824The POSIX class matches the same as the ASCII range counterpart.
825
826=back
827
828=back
829
830=back
831
832Which rules apply are determined as described in
833L<perlre/Which character set modifier is in effect?>.
834
835It is proposed to change this behavior in a future release of Perl so that
836whether or not Unicode rules are in effect would not change the
837behavior: Outside of locale or an EBCDIC code page, the POSIX classes
838would behave like their ASCII-range counterparts. If you wish to
839comment on this proposal, send email to C<perl5-porters@perl.org>.
cbc24f92 840
1f59b283 841=head4 Negation of POSIX character classes
ea449505 842X<character class, negation>
8a118206
RGS
843
844A Perl extension to the POSIX character class is the ability to
845negate it. This is done by prefixing the class name with a caret (C<^>).
846Some examples:
847
ea449505
KW
848 POSIX ASCII-range Full-range backslash
849 Unicode Unicode sequence
850 -----------------------------------------------------
cbc24f92
KW
851 [[:^digit:]] \P{PosixDigit} \P{XPosixDigit} \D
852 [[:^space:]] \P{PosixSpace} \P{XPosixSpace}
853 \P{PerlSpace} \P{XPerlSpace} \S
854 [[:^word:]] \P{PerlWord} \P{XPosixWord} \W
855
765fa144 856The backslash sequence can mean either ASCII- or Full-range Unicode,
82206b5e 857depending on various factors as described in L<perlre/Which character set modifier is in effect?>.
8a118206
RGS
858
859=head4 [= =] and [. .]
860
b6538e4f 861Perl recognizes the POSIX character classes C<[=class=]> and
82206b5e 862C<[.class.]>, but does not (yet?) support them. Any attempt to use
b6538e4f 863either construct raises an exception.
8a118206
RGS
864
865=head4 Examples
866
867 /[[:digit:]]/ # Matches a character that is a digit.
868 /[01[:lower:]]/ # Matches a character that is either a
869 # lowercase letter, or '0' or '1'.
c1c4ae3a 870 /[[:digit:][:^xdigit:]]/ # Matches a character that can be anything
bc943be5
KW
871 # except the letters 'a' to 'f' and 'A' to
872 # 'F'. This is because the main character
873 # class is composed of two POSIX character
874 # classes that are ORed together, one that
875 # matches any digit, and the other that
876 # matches anything that isn't a hex digit.
877 # The OR adds the digits, leaving only the
878 # letters 'a' to 'f' and 'A' to 'F' excluded.
572224ce
KW
879
880=head3 Extended Bracketed Character Classes
881X<character class>
882X<set operations>
883
884This is a fancy bracketed character class that can be used for more
885readable and less error-prone classes, and to perform set operations,
886such as intersection. An example is
887
888 /(?[ \p{Thai} & \p{Digit} ])/
889
890This will match all the digit characters that are in the Thai script.
891
892This is an experimental feature available starting in 5.18, and is
893subject to change as we gain field experience with it. Any attempt to
894use it will raise a warning, unless disabled via
895
896 no warnings "experimental::regex_sets";
897
898Comments on this feature are welcome; send email to
899C<perl5-porters@perl.org>.
900
901We can extend the example above:
902
903 /(?[ ( \p{Thai} + \p{Lao} ) & \p{Digit} ])/
904
905This matches digits that are in either the Thai or Laotian scripts.
906
907Notice the white space in these examples. This construct always has
908the C<E<sol>x> modifier turned on.
909
910The available binary operators are:
911
912 & intersection
913 + union
914 | another name for '+', hence means union
915 - subtraction (the result matches the set consisting of those
916 code points matched by the first operand, excluding any that
917 are also matched by the second operand)
918 ^ symmetric difference (the union minus the intersection). This
919 is like an exclusive or, in that the result is the set of code
920 points that are matched by either, but not both, of the
921 operands.
922
923There is one unary operator:
924
925 ! complement
926
927All the binary operators left associate, and are of equal precedence.
928The unary operator right associates, and has higher precedence. Use
929parentheses to override the default associations. Some feedback we've
930received indicates a desire for intersection to have higher precedence
931than union. This is something that feedback from the field may cause us
932to change in future releases; you may want to parenthesize copiously to
933avoid such changes affecting your code, until this feature is no longer
934considered experimental.
935
936The main restriction is that everything is a metacharacter. Thus,
937you cannot refer to single characters by doing something like this:
938
939 /(?[ a + b ])/ # Syntax error!
940
941The easiest way to specify an individual typable character is to enclose
942it in brackets:
943
944 /(?[ [a] + [b] ])/
945
946(This is the same thing as C<[ab]>.) You could also have said the
947equivalent:
948
949 /(?[[ a b ]])/
950
951(You can, of course, specify single characters by using, C<\x{ }>,
952C<\N{ }>, etc.)
953
954This last example shows the use of this construct to specify an ordinary
955bracketed character class without additional set operations. Note the
956white space within it; C<E<sol>x> is turned on even within bracketed
957character classes, except you can't have comments inside them. Hence,
958
959 (?[ [#] ])
960
961matches the literal character "#". To specify a literal white space character,
962you can escape it with a backslash, like:
963
964 /(?[ [ a e i o u \ ] ])/
965
966This matches the English vowels plus the SPACE character.
967All the other escapes accepted by normal bracketed character classes are
968accepted here as well; but unrecognized escapes that generate warnings
969in normal classes are fatal errors here.
970
971All warnings from these class elements are fatal, as well as some
972practices that don't currently warn. For example you cannot say
973
974 /(?[ [ \xF ] ])/ # Syntax error!
975
976You have to have two hex digits after a braceless C<\x> (use a leading
977zero to make two). These restrictions are to lower the incidence of
978typos causing the class to not match what you thought it would.
979
980The final difference between regular bracketed character classes and
981these, is that it is not possible to get these to match a
982multi-character fold. Thus,
983
984 /(?[ [\xDF] ])/iu
985
986does not match the string C<ss>.
987
988You don't have to enclose POSIX class names inside double brackets,
989hence both of the following work:
990
991 /(?[ [:word:] - [:lower:] ])/
992 /(?[ [[:word:]] - [[:lower:]] ])/
993
994Any contained POSIX character classes, including things like C<\w> and C<\D>
995respect the C<E<sol>a> (and C<E<sol>aa>) modifiers.
996
997C<< (?[ ]) >> is a regex-compile-time construct. Any attempt to use
998something which isn't knowable at the time the containing regular
999expression is compiled is a fatal error. In practice, this means
1000just three limitiations:
1001
1002=over 4
1003
1004=item 1
1005
1006This construct cannot be used within the scope of
1007C<use locale> (or the C<E<sol>l> regex modifier).
1008
1009=item 2
1010
1011Any
1012L<user-defined property|perlunicode/"User-Defined Character Properties">
1013used must be already defined by the time the regular expression is
1014compiled (but note that this construct can be used instead of such
1015properties).
1016
1017=item 3
1018
1019A regular expression that otherwise would compile
1020using C<E<sol>d> rules, and which uses this construct will instead
1021use C<E<sol>u>. Thus this construct tells Perl that you don't want
1022C<E<sol>d> rules for the entire regular expression containing it.
1023
1024=back
1025
1026The C<E<sol>x> processing within this class is an extended form.
1027Besides the characters that are considered white space in normal C</x>
1028processing, there are 5 others, recommended by the Unicode standard:
1029
1030 U+0085 NEXT LINE
1031 U+200E LEFT-TO-RIGHT MARK
1032 U+200F RIGHT-TO-LEFT MARK
1033 U+2028 LINE SEPARATOR
1034 U+2029 PARAGRAPH SEPARATOR
1035
1036Note that skipping white space applies only to the interior of this
1037construct. There must not be any space between any of the characters
1038that form the initial C<(?[>. Nor may there be space between the
1039closing C<])> characters.
1040
1041Just as in all regular expressions, the pattern can can be built up by
1042including variables that are interpolated at regex compilation time.
1043Care must be taken to ensure that you are getting what you expect. For
1044example:
1045
1046 my $thai_or_lao = '\p{Thai} + \p{Lao}';
1047 ...
1048 qr/(?[ \p{Digit} & $thai_or_lao ])/;
1049
1050compiles to
1051
1052 qr/(?[ \p{Digit} & \p{Thai} + \p{Lao} ])/;
1053
1054But this does not have the effect that someone reading the code would
1055likely expect, as the intersection applies just to C<\p{Thai}>,
1056excluding the Laotian. Pitfalls like this can be avoided by
1057parenthesizing the component pieces:
1058
1059 my $thai_or_lao = '( \p{Thai} + \p{Lao} )';
1060
1061But any modifiers will still apply to all the components:
1062
1063 my $lower = '\p{Lower} + \p{Digit}';
1064 qr/(?[ \p{Greek} & $lower ])/i;
1065
1066matches upper case things. You can avoid surprises by making the
1067components into instances of this construct by compiling them:
1068
1069 my $thai_or_lao = qr/(?[ \p{Thai} + \p{Lao} ])/;
1070 my $lower = qr/(?[ \p{Lower} + \p{Digit} ])/;
1071
1072When these are embedded in another pattern, what they match does not
1073change, regardless of parenthesization or what modifiers are in effect
1074in that outer pattern.
1075
1076Due to the way that Perl parses things, your parentheses and brackets
1077may need to be balanced, even including comments. If you run into any
1078examples, please send them to C<perlbug@perl.org>, so that we can have a
1079concrete example for this man page.
1080
1081We may change it so that things that remain legal uses in normal bracketed
1082character classes might become illegal within this experimental
1083construct. One proposal, for example, is to forbid adjacent uses of the
1084same character, as in C<(?[ [aa] ])>. The motivation for such a change
1085is that this usage is likely a typo, as the second "a" adds nothing.