1PCRE2COMPAT(3) Library Functions Manual PCRE2COMPAT(3)
2
6 PCRE2 - Perl-compatible regular expressions (revised API)
7
9 This document describes some of the differences in the ways that PCRE2
11 and Perl handle regular expressions. The differences described here are
12 with respect to Perl version 5.34.0, but as both Perl and PCRE2 are
13 continually changing, the information may at times be out of date.
14 1. When PCRE2_DOTALL (equivalent to Perl's /s qualifier) is not set,
16 the behaviour of the '.' metacharacter differs from Perl. In PCRE2, '.'
17 matches the next character unless it is the start of a newline se‐
18 quence. This means that, if the newline setting is CR, CRLF, or NUL,
19 '.' will match the code point LF (0x0A) in ASCII/Unicode environments,
20 and NL (either 0x15 or 0x25) when using EBCDIC. In Perl, '.' appears
21 never to match LF, even when 0x0A is not a newline indicator.
22 2. PCRE2 has only a subset of Perl's Unicode support. Details of what
24 it does have are given in the pcre2unicode page.
25 3. Like Perl, PCRE2 allows repeat quantifiers on parenthesized asser‐
27 tions, but they do not mean what you might think. For example, (?!a){3}
28 does not assert that the next three characters are not "a". It just as‐
29 serts that the next character is not "a" three times (in principle;
30 PCRE2 optimizes this to run the assertion just once). Perl allows some
31 repeat quantifiers on other assertions, for example, \b* , but these do
32 not seem to have any use. PCRE2 does not allow any kind of quantifier
33 on non-lookaround assertions.
34 4. Capture groups that occur inside negative lookaround assertions are
36 counted, but their entries in the offsets vector are set only when a
37 negative assertion is a condition that has a matching branch (that is,
38 the condition is false). Perl may set such capture groups in other
39 circumstances.
40 5. The following Perl escape sequences are not supported: \F, \l, \L,
42 \u, \U, and \N when followed by a character name. \N on its own, match‐
43 ing a non-newline character, and \N{U+dd..}, matching a Unicode code
44 point, are supported. The escapes that modify the case of following
45 letters are implemented by Perl's general string-handling and are not
46 part of its pattern matching engine. If any of these are encountered by
47 PCRE2, an error is generated by default. However, if either of the
48 PCRE2_ALT_BSUX or PCRE2_EXTRA_ALT_BSUX options is set, \U and \u are
49 interpreted as ECMAScript interprets them.
50 6. The Perl escape sequences \p, \P, and \X are supported only if PCRE2
52 is built with Unicode support (the default). The properties that can be
53 tested with \p and \P are limited to the general category properties
54 such as Lu and Nd, script names such as Greek or Han, Bidi_Class,
55 Bidi_Control, and the derived properties Any and LC (synonym L&). Both
56 PCRE2 and Perl support the Cs (surrogate) property, but in PCRE2 its
57 use is limited. See the pcre2pattern documentation for details. The
58 long synonyms for property names that Perl supports (such as \p{Let‐
59 ter}) are not supported by PCRE2, nor is it permitted to prefix any of
60 these properties with "Is".
61 7. PCRE2 supports the \Q...\E escape for quoting substrings. Characters
63 in between are treated as literals. However, this is slightly different
64 from Perl in that $ and @ are also handled as literals inside the
65 quotes. In Perl, they cause variable interpolation (PCRE2 does not have
66 variables). Also, Perl does "double-quotish backslash interpolation" on
67 any backslashes between \Q and \E which, its documentation says, "may
68 lead to confusing results". PCRE2 treats a backslash between \Q and \E
69 just like any other character. Note the following examples:
70 Pattern PCRE2 matches Perl matches
72 \Qabc$xyz\E abc$xyz abc followed by the
74 contents of $xyz
75 \Qabc\$xyz\E abc\$xyz abc\$xyz
76 \Qabc\E\$\Qxyz\E abc$xyz abc$xyz
77 \QA\B\E A\B A\B
78 \Q\\E \ \\E
79 The \Q...\E sequence is recognized both inside and outside character
81 classes by both PCRE2 and Perl.
82 8. Fairly obviously, PCRE2 does not support the (?{code}) and
84 (??{code}) constructions. However, PCRE2 does have a "callout" feature,
85 which allows an external function to be called during pattern matching.
86 See the pcre2callout documentation for details.
87 9. Subroutine calls (whether recursive or not) were treated as atomic
89 groups up to PCRE2 release 10.23, but from release 10.30 this changed,
90 and backtracking into subroutine calls is now supported, as in Perl.
91 10. In PCRE2, if any of the backtracking control verbs are used in a
93 group that is called as a subroutine (whether or not recursively),
94 their effect is confined to that group; it does not extend to the sur‐
95 rounding pattern. This is not always the case in Perl. In particular,
96 if (*THEN) is present in a group that is called as a subroutine, its
97 action is limited to that group, even if the group does not contain any
98 | characters. Note that such groups are processed as anchored at the
99 point where they are tested.
100 11. If a pattern contains more than one backtracking control verb, the
102 first one that is backtracked onto acts. For example, in the pattern
103 A(*COMMIT)B(*PRUNE)C a failure in B triggers (*COMMIT), but a failure
104 in C triggers (*PRUNE). Perl's behaviour is more complex; in many cases
105 it is the same as PCRE2, but there are cases where it differs.
106 12. There are some differences that are concerned with the settings of
108 captured strings when part of a pattern is repeated. For example,
109 matching "aba" against the pattern /^(a(b)?)+$/ in Perl leaves $2 un‐
110 set, but in PCRE2 it is set to "b".
111 13. PCRE2's handling of duplicate capture group numbers and names is
113 not as general as Perl's. This is a consequence of the fact the PCRE2
114 works internally just with numbers, using an external table to trans‐
115 late between numbers and names. In particular, a pattern such as
116 (?|(?<a>A)|(?<b>B)), where the two capture groups have the same number
117 but different names, is not supported, and causes an error at compile
118 time. If it were allowed, it would not be possible to distinguish which
119 group matched, because both names map to capture group number 1. To
120 avoid this confusing situation, an error is given at compile time.
121 14. Perl used to recognize comments in some places that PCRE2 does not,
123 for example, between the ( and ? at the start of a group. If the /x
124 modifier is set, Perl allowed white space between ( and ? though the
125 latest Perls give an error (for a while it was just deprecated). There
126 may still be some cases where Perl behaves differently.
127 15. Perl, when in warning mode, gives warnings for character classes
129 such as [A-\d] or [a-[:digit:]]. It then treats the hyphens as liter‐
130 als. PCRE2 has no warning features, so it gives an error in these cases
131 because they are almost certainly user mistakes.
132 16. In PCRE2, the upper/lower case character properties Lu and Ll are
134 not affected when case-independent matching is specified. For example,
135 \p{Lu} always matches an upper case letter. I think Perl has changed in
136 this respect; in the release at the time of writing (5.34), \p{Lu} and
137 \p{Ll} match all letters, regardless of case, when case independence is
138 specified.
139 17. From release 5.32.0, Perl locks out the use of \K in lookaround as‐
141 sertions. From release 10.38 PCRE2 does the same by default. However,
142 there is an option for re-enabling the previous behaviour. When this
143 option is set, \K is acted on when it occurs in positive assertions,
144 but is ignored in negative assertions.
145 18. PCRE2 provides some extensions to the Perl regular expression fa‐
147 cilities. Perl 5.10 included new features that were not in earlier
148 versions of Perl, some of which (such as named parentheses) were in
149 PCRE2 for some time before. This list is with respect to Perl 5.34:
150 (a) Although lookbehind assertions in PCRE2 must match fixed length
152 strings, each alternative toplevel branch of a lookbehind assertion can
153 match a different length of string. Perl used to require them all to
154 have the same length, but the latest version has some variable length
155 support.
156 (b) From PCRE2 10.23, backreferences to groups of fixed length are sup‐
158 ported in lookbehinds, provided that there is no possibility of refer‐
159 encing a non-unique number or name. Perl does not support backrefer‐
160 ences in lookbehinds.
161 (c) If PCRE2_DOLLAR_ENDONLY is set and PCRE2_MULTILINE is not set, the
163 $ meta-character matches only at the very end of the string.
164 (d) A backslash followed by a letter with no special meaning is
166 faulted. (Perl can be made to issue a warning.)
167 (e) If PCRE2_UNGREEDY is set, the greediness of the repetition quanti‐
169 fiers is inverted, that is, by default they are not greedy, but if fol‐
170 lowed by a question mark they are.
171 (f) PCRE2_ANCHORED can be used at matching time to force a pattern to
173 be tried only at the first matching position in the subject string.
174 (g) The PCRE2_NOTBOL, PCRE2_NOTEOL, PCRE2_NOTEMPTY and
176 PCRE2_NOTEMPTY_ATSTART options have no Perl equivalents.
177 (h) The \R escape sequence can be restricted to match only CR, LF, or
179 CRLF by the PCRE2_BSR_ANYCRLF option.
180 (i) The callout facility is PCRE2-specific. Perl supports codeblocks
182 and variable interpolation, but not general hooks on every match.
183 (j) The partial matching facility is PCRE2-specific.
185 (k) The alternative matching function (pcre2_dfa_match() matches in a
187 different way and is not Perl-compatible.
188 (l) PCRE2 recognizes some special sequences such as (*CR) or (*NO_JIT)
190 at the start of a pattern. These set overall options that cannot be
191 changed within the pattern.
192 (m) PCRE2 supports non-atomic positive lookaround assertions. This is
194 an extension to the lookaround facilities. The default, Perl-compatible
195 lookarounds are atomic.
196 19. The Perl /a modifier restricts /d numbers to pure ascii, and the
198 /aa modifier restricts /i case-insensitive matching to pure ascii, ig‐
199 noring Unicode rules. This separation cannot be represented with
200 PCRE2_UCP.
201 20. Perl has different limits than PCRE2. See the pcre2limit documenta‐
203 tion for details. Perl went with 5.10 from recursion to iteration keep‐
204 ing the intermediate matches on the heap, which is ~10% slower but does
205 not fall into any stack-overflow limit. PCRE2 made a similar change at
206 release 10.30, and also has many build-time and run-time customizable
207 limits.
208
210 Philip Hazel
212 Retired from University Computing Service
213 Cambridge, England.
214
216 Last updated: 08 December 2021
218 Copyright (c) 1997-2021 University of Cambridge.
219PCRE2 10.40 08 December 2021 PCRE2COMPAT(3)