1PCRE2COMPAT(3) Library Functions Manual PCRE2COMPAT(3)
2
6 PCRE2 - Perl-compatible regular expressions (revised API)
7
9 This document describes the differences in the ways that PCRE2 and Perl
11 handle regular expressions. The differences described here are with
12 respect to Perl versions 5.26, but as both Perl and PCRE2 are continu‐
13 ally changing, the information may sometimes be out of date.
14 1. PCRE2 has only a subset of Perl's Unicode support. Details of what
16 it does have are given in the pcre2unicode page.
17 2. Like Perl, PCRE2 allows repeat quantifiers on parenthesized asser‐
19 tions, but they do not mean what you might think. For example, (?!a){3}
20 does not assert that the next three characters are not "a". It just
21 asserts that the next character is not "a" three times (in principle;
22 PCRE2 optimizes this to run the assertion just once). Perl allows some
23 repeat quantifiers on other assertions, for example, \b* (but not
24 \b{3}), but these do not seem to have any use.
25 3. Capture groups that occur inside negative lookaround assertions are
27 counted, but their entries in the offsets vector are set only when a
28 negative assertion is a condition that has a matching branch (that is,
29 the condition is false).
30 4. The following Perl escape sequences are not supported: \F, \l, \L,
32 \u, \U, and \N when followed by a character name. \N on its own, match‐
33 ing a non-newline character, and \N{U+dd..}, matching a Unicode code
34 point, are supported. The escapes that modify the case of following
35 letters are implemented by Perl's general string-handling and are not
36 part of its pattern matching engine. If any of these are encountered by
37 PCRE2, an error is generated by default. However, if either of the
38 PCRE2_ALT_BSUX or PCRE2_EXTRA_ALT_BSUX options is set, \U and \u are
39 interpreted as ECMAScript interprets them.
40 5. The Perl escape sequences \p, \P, and \X are supported only if PCRE2
42 is built with Unicode support (the default). The properties that can be
43 tested with \p and \P are limited to the general category properties
44 such as Lu and Nd, script names such as Greek or Han, and the derived
45 properties Any and L&. PCRE2 does support the Cs (surrogate) property,
46 which Perl does not; the Perl documentation says "Because Perl hides
47 the need for the user to understand the internal representation of Uni‐
48 code characters, there is no need to implement the somewhat messy con‐
49 cept of surrogates."
50 6. PCRE2 supports the \Q...\E escape for quoting substrings. Characters
52 in between are treated as literals. However, this is slightly different
53 from Perl in that $ and @ are also handled as literals inside the
54 quotes. In Perl, they cause variable interpolation (but of course PCRE2
55 does not have variables). Also, Perl does "double-quotish backslash
56 interpolation" on any backslashes between \Q and \E which, its documen‐
57 tation says, "may lead to confusing results". PCRE2 treats a backslash
58 between \Q and \E just like any other character. Note the following
59 examples:
60 Pattern PCRE2 matches Perl matches
62 \Qabc$xyz\E abc$xyz abc followed by the
64 contents of $xyz
65 \Qabc\$xyz\E abc\$xyz abc\$xyz
66 \Qabc\E\$\Qxyz\E abc$xyz abc$xyz
67 \QA\B\E A\B A\B
68 \Q\\E \ \\E
69 The \Q...\E sequence is recognized both inside and outside character
71 classes.
72 7. Fairly obviously, PCRE2 does not support the (?{code}) and
74 (??{code}) constructions. However, PCRE2 does have a "callout" feature,
75 which allows an external function to be called during pattern matching.
76 See the pcre2callout documentation for details.
77 8. Subroutine calls (whether recursive or not) were treated as atomic
79 groups up to PCRE2 release 10.23, but from release 10.30 this changed,
80 and backtracking into subroutine calls is now supported, as in Perl.
81 9. If any of the backtracking control verbs are used in a group that is
83 called as a subroutine (whether or not recursively), their effect is
84 confined to that group; it does not extend to the surrounding pattern.
85 This is not always the case in Perl. In particular, if (*THEN) is
86 present in a group that is called as a subroutine, its action is lim‐
87 ited to that group, even if the group does not contain any | charac‐
88 ters. Note that such groups are processed as anchored at the point
89 where they are tested.
90 10. If a pattern contains more than one backtracking control verb, the
92 first one that is backtracked onto acts. For example, in the pattern
93 A(*COMMIT)B(*PRUNE)C a failure in B triggers (*COMMIT), but a failure
94 in C triggers (*PRUNE). Perl's behaviour is more complex; in many cases
95 it is the same as PCRE2, but there are cases where it differs.
96 11. Most backtracking verbs in assertions have their normal actions.
98 They are not confined to the assertion.
99 12. There are some differences that are concerned with the settings of
101 captured strings when part of a pattern is repeated. For example,
102 matching "aba" against the pattern /^(a(b)?)+$/ in Perl leaves $2
103 unset, but in PCRE2 it is set to "b".
104 13. PCRE2's handling of duplicate capture group numbers and names is
106 not as general as Perl's. This is a consequence of the fact the PCRE2
107 works internally just with numbers, using an external table to trans‐
108 late between numbers and names. In particular, a pattern such as
109 (?|(?<a>A)|(?<b>B), where the two capture groups have the same number
110 but different names, is not supported, and causes an error at compile
111 time. If it were allowed, it would not be possible to distinguish which
112 group matched, because both names map to capture group number 1. To
113 avoid this confusing situation, an error is given at compile time.
114 14. Perl used to recognize comments in some places that PCRE2 does not,
116 for example, between the ( and ? at the start of a group. If the /x
117 modifier is set, Perl allowed white space between ( and ? though the
118 latest Perls give an error (for a while it was just deprecated). There
119 may still be some cases where Perl behaves differently.
120 15. Perl, when in warning mode, gives warnings for character classes
122 such as [A-\d] or [a-[:digit:]]. It then treats the hyphens as liter‐
123 als. PCRE2 has no warning features, so it gives an error in these cases
124 because they are almost certainly user mistakes.
125 16. In PCRE2, the upper/lower case character properties Lu and Ll are
127 not affected when case-independent matching is specified. For example,
128 \p{Lu} always matches an upper case letter. I think Perl has changed in
129 this respect; in the release at the time of writing (5.24), \p{Lu} and
130 \p{Ll} match all letters, regardless of case, when case independence is
131 specified.
132 17. PCRE2 provides some extensions to the Perl regular expression
134 facilities. Perl 5.10 includes new features that are not in earlier
135 versions of Perl, some of which (such as named parentheses) were in
136 PCRE2 for some time before. This list is with respect to Perl 5.26:
137 (a) Although lookbehind assertions in PCRE2 must match fixed length
139 strings, each alternative branch of a lookbehind assertion can match a
140 different length of string. Perl requires them all to have the same
141 length.
142 (b) From PCRE2 10.23, backreferences to groups of fixed length are sup‐
144 ported in lookbehinds, provided that there is no possibility of refer‐
145 encing a non-unique number or name. Perl does not support backrefer‐
146 ences in lookbehinds.
147 (c) If PCRE2_DOLLAR_ENDONLY is set and PCRE2_MULTILINE is not set, the
149 $ meta-character matches only at the very end of the string.
150 (d) A backslash followed by a letter with no special meaning is
152 faulted. (Perl can be made to issue a warning.)
153 (e) If PCRE2_UNGREEDY is set, the greediness of the repetition quanti‐
155 fiers is inverted, that is, by default they are not greedy, but if fol‐
156 lowed by a question mark they are.
157 (f) PCRE2_ANCHORED can be used at matching time to force a pattern to
159 be tried only at the first matching position in the subject string.
160 (g) The PCRE2_NOTBOL, PCRE2_NOTEOL, PCRE2_NOTEMPTY and
162 PCRE2_NOTEMPTY_ATSTART options have no Perl equivalents.
163 (h) The \R escape sequence can be restricted to match only CR, LF, or
165 CRLF by the PCRE2_BSR_ANYCRLF option.
166 (i) The callout facility is PCRE2-specific. Perl supports codeblocks
168 and variable interpolation, but not general hooks on every match.
169 (j) The partial matching facility is PCRE2-specific.
171 (k) The alternative matching function (pcre2_dfa_match() matches in a
173 different way and is not Perl-compatible.
174 (l) PCRE2 recognizes some special sequences such as (*CR) or (*NO_JIT)
176 at the start of a pattern that set overall options that cannot be
177 changed within the pattern.
178 18. The Perl /a modifier restricts /d numbers to pure ascii, and the
180 /aa modifier restricts /i case-insensitive matching to pure ascii,
181 ignoring Unicode rules. This separation cannot be represented with
182 PCRE2_UCP.
183 19. Perl has different limits than PCRE2. See the pcre2limit documenta‐
185 tion for details. Perl went with 5.10 from recursion to iteration keep‐
186 ing the intermediate matches on the heap, which is ~10% slower but does
187 not fall into any stack-overflow limit. PCRE2 made a similar change at
188 release 10.30, and also has many build-time and run-time customizable
189 limits.
190
192 Philip Hazel
194 University Computing Service
195 Cambridge, England.
196
198 Last updated: 12 February 2019
200 Copyright (c) 1997-2019 University of Cambridge.
201PCRE2 10.33 12 February 2019 PCRE2COMPAT(3)