1 package Regexp::Wildcards;
7 use Text::Balanced qw/extract_bracketed/;
11 Regexp::Wildcards - Converts wildcard expressions to Perl regular expressions.
19 use vars qw/$VERSION/;
26 use Regexp::Wildcards;
28 my $rw = Regexp::Wildcards->new(type => 'unix');
31 $re = $rw->convert('a{b?,c}*'); # Do it Unix shell style.
32 $re = $rw->convert('a?,b*', 'win32'); # Do it Windows shell style.
33 $re = $rw->convert('*{x,y}?', 'jokers'); # Process the jokers and escape the rest.
34 $re = $rw->convert('%a_c%', 'sql'); # Turn SQL wildcards into regexps.
36 $rw = Regexp::Wildcards->new(
37 do => [ qw/jokers brackets/ ], # Do jokers and brackets.
38 capture => [ qw/any greedy/ ], # Capture *'s greedily.
41 $rw->do(add => 'groups'); # Don't escape groups.
42 $rw->capture(rem => [ qw/greedy/ ]); # Actually we want non-greedy matches.
43 $re = $rw->convert('*a{,(b)?}?c*'); # '(.*?)a(?:|(b).).c(.*?)'
44 $rw->capture(); # No more captures.
48 In many situations, users may want to specify patterns to match but don't need the full power of regexps. Wildcards make one of those sets of simplified rules. This module converts wildcard expressions to Perl regular expressions, so that you can use them for matching.
50 It handles the C<*> and C<?> jokers, as well as Unix bracketed alternatives C<{,}>, but also C<%> and C<_> SQL wildcards. If required, it can also keep original C<(...)> groups or C<^> and C<$> anchors. Backspace (C<\>) is used as an escape character.
52 Typesets that mimic the behaviour of Windows and Unix shells are also provided.
59 croak 'First argument isn\'t a valid ' . __PACKAGE__ . ' object'
60 unless ref $_[0] and $_[0]->isa(__PACKAGE__);
64 jokers => [ qw/jokers/ ],
66 commas => [ qw/commas/ ],
67 brackets => [ qw/brackets/ ],
68 unix => [ qw/jokers brackets/ ],
69 win32 => [ qw/jokers commas/ ],
71 $types{$_} = $types{win32} for qw/dos os2 MSWin32 cygwin/;
72 $types{$_} = $types{unix} for qw/linux
74 aix irix hpux dgux dynixptx
76 svr4 solaris sunos dec_osf
77 sco_sv unicos unicosmk/;
89 single => sub { $_[1] ? '(.)' : '.' },
90 any => sub { $_[1] ? ($_[0]->{greedy} ? '(.*)'
93 brackets => sub { $_[1] ? '(' : '(?:'; },
102 $old = { } unless defined $old;
105 $c = { set => $_[0] };
107 croak 'Arguments must be passed as an unique scalar or as key => value pairs';
110 $c = { map { (exists $args{$_}) ? ($_ => $args{$_}) : () } qw/set add rem/ };
112 for (qw/set add rem/) {
114 next unless defined $v;
116 '' => sub { +{ ($_[0] => 1) x (exists $valid->{$_[0]}) } },
117 'ARRAY' => sub { +{ map { ($_ => 1) x (exists $valid->{$_}) } @{$_[0]} } },
118 'HASH' => sub { +{ map { ($_ => $_[0]->{$_}) x (exists $valid->{$_}) }
121 croak 'Wrong option set' unless $cb;
122 $c->{$_} = $cb->($v);
124 my $config = (exists $c->{set}) ? $c->{set} : $old;
125 $config->{$_} = $c->{add}->{$_} for grep $c->{add}->{$_},
126 keys %{$c->{add} || {}};
127 delete $config->{$_} for grep $c->{rem}->{$_}, keys %{$c->{rem} || {}};
134 $config->{do} = $self->_validate(\%escapes, $self->{do}, @_);
135 $config->{escape} = '';
136 $config->{escape} .= $escapes{$_} for keys %{$config->{do}};
137 $config->{escape} = quotemeta $config->{escape};
144 my $config = $self->_do(@_);
145 $self->{$_} = $config->{$_} for keys %$config;
152 $config->{capture} = $self->_validate(\%captures, $self->{capture}, @_);
153 $config->{greedy} = delete $config->{capture}->{greedy};
154 for (keys %captures) {
155 $config->{'c_' . $_} = $captures{$_}->($config, $config->{capture}->{$_})
156 if $captures{$_}; # Skip 'greedy'
164 my $config = $self->_capture(@_);
165 $self->{$_} = $config->{$_} for keys %$config;
170 my ($self, $type) = @_;
171 $type = 'unix' unless defined $type;
172 croak 'Wrong type' unless exists $types{$type};
173 my $config = $self->_do($types{$type});
174 $config->{type} = $type;
181 my $config = $self->_type(@_);
182 $self->{$_} = $config->{$_} for keys %$config;
188 $class = ref($class) || $class || __PACKAGE__;
189 croak 'Optional arguments must be passed as key => value pairs' if @_ % 2;
193 if (defined $args{do}) {
194 $self->do($args{do});
196 $self->type($args{type});
198 $self->capture($args{capture});
201 =head2 C<< new [ do => $what E<verbar> type => $type ], capture => $captures >>
203 Constructs a new L<Regexp::Wildcard> object.
205 C<do> lists all features that should be enabled when converting wildcards to regexps. Refer to L</do> for details on what can be passed in C<$what>.
207 The C<type> specifies a predefined set of C<do> features to use. See L</type> for details on which types are valid. The C<do> option overrides C<type>.
209 C<capture> lists which atoms should be capturing. Refer to L</capture> for more details.
211 =head2 C<< do [ $what E<verbar> set => $c1, add => $c2, rem => $c3 ] >>
213 Specifies the list of metacharacters to convert or to prevent for escaping.
214 They fit into six classes :
220 C<'jokers'> converts C<?> to C<.> and C<*> to C<.*> ;
222 'a**\\*b??\\?c' ==> 'a.*\\*b..\\?c'
226 C<'sql'> converts C<_> to C<.> and C<%> to C<.*> ;
228 'a%%\\%b__\\_c' ==> 'a.*\\%b..\\_c'
232 C<'commas'> converts all C<,> to C<|> and puts the complete resulting regular expression inside C<(?: ... )> ;
234 'a,b{c,d},e' ==> '(?:a|b\\{c|d\\}|e)'
238 C<'brackets'> converts all matching C<{ ... , ... }> brackets to C<(?: ... | ... )> alternations. If some brackets are unbalanced, it tries to substitute as many of them as possible, and then escape the remaining unmatched C<{> and C<}>. Commas outside of any bracket-delimited block are also escaped ;
240 'a,b{c,d},e' ==> 'a\\,b(?:c|d)\\,e'
241 '{a\\{b,c}d,e}' ==> '(?:a\\{b|c)d\\,e\\}'
242 '{a{b,c\\}d,e}' ==> '\\{a\\{b\\,c\\}d\\,e\\}'
246 C<'groups'> keeps the parenthesis C<( ... )> of the original string without escaping them. Currently, no check is done to ensure that the parenthesis are matching.
248 'a(b(c))d\\(\\)' ==> (no change)
252 C<'anchors'> prevents the I<beginning-of-line> C<^> and I<end-of-line> C<$> anchors to be escaped. Since C<[...]> character class are currently escaped, a C<^> will always be interpreted as I<beginning-of-line>.
254 'a^b$c' ==> (no change)
258 Each C<$c> can be any of :
264 A hash reference, with wanted metacharacter group names (described above) as keys and booleans as values ;
268 An array reference containing the list of wanted metacharacter classes ;
272 A plain scalar, when only one group is required.
276 When C<set> is present, the classes given as its value replace the current object options. Then the C<add> classes are added, and the C<rem> classes removed.
278 Passing a sole scalar C<$what> is equivalent as passing C<< set => $what >>.
279 No argument means C<< set => [ ] >>.
281 $rw->do(set => 'jokers'); # Only translate jokers.
282 $rw->do('jokers'); # Same.
283 $rw->do(add => [ qw/sql commas/ ]); # Translate also SQL and commas.
284 $rw->do(rem => 'jokers'); # Specifying both 'sql' and 'jokers' is useless.
285 $rw->do(); # Translate nothing.
289 Notifies to convert the metacharacters that corresponds to the predefined type C<$type>.
290 C<$type> can be any of :
296 C<'jokers'>, C<'sql'>, C<'commas'>, C<'brackets'>
298 Singleton types that enable the corresponding C<do> classes.
304 Covers typical Unix shell globbing features (effectively C<'jokers'> and C<'brackets'>).
308 C<$^O> values for common Unix systems
310 Wrap to C<'unix'> (see L<perlport> for the list).
316 Defaults to C<'unix'>.
322 Covers typical Windows shell globbing features (effectively C<'jokers'> and C<'commas'>).
326 C<'dos'>, C<'os2'>, C<'MSWin32'>, C<'cygwin'>
332 In particular, you can usually pass C<$^O> as the C<$type> and get the corresponding shell behaviour.
334 $rw->type('win32'); # Set type to win32.
335 $rw->type($^O); # Set type to unix on Unices and win32 on Windows
336 $rw->type(); # Set type to unix.
338 The C<type> method returns the L<Regexp::Wildcards> object.
340 =head2 C<< capture [ $captures E<verbar> set => $c1, add => $c2, rem => $c3 ] >>
342 Specifies the list of atoms to capture.
343 This method works like L</do>, except that the classes are different :
349 C<'single'> will capture all unescaped I<"exactly one"> metacharacters, i.e. C<?> for wildcards or C<_> for SQL ;
351 'a???b\\??' ==> 'a(.)(.)(.)b\\?(.)'
352 'a___b\\__' ==> 'a(.)(.)(.)b\\_(.)'
356 C<'any'> will capture all unescaped I<"any"> metacharacters, i.e. C<*> for wildcards or C<%> for SQL ;
358 'a***b\\**' ==> 'a(.*)b\\*(.*)'
359 'a%%%b\\%%' ==> 'a(.*)b\\%(.*)'
363 C<'greedy'>, when used in conjunction with C<'any'>, will make the C<'any'> captures greedy (by default they are not) ;
365 'a***b\\**' ==> 'a(.*?)b\\*(.*?)'
366 'a%%%b\\%%' ==> 'a(.*?)b\\%(.*?)'
370 C<'brackets'> will capture matching C<{ ... , ... }> alternations.
372 'a{b\\},\\{c}' ==> 'a(b\\}|\\{c)'
376 $rw->capture(set => 'single'); # Only capture "exactly one" metacharacters.
377 $rw->capture('single'); # Same.
378 $rw->capture(add => [ qw/any greedy/ ]); # Also greedily capture "any" metacharacters.
379 $rw->capture(rem => 'greedy'); # No more greed please.
380 $rw->capture(); # Capture nothing.
382 =head2 C<convert $wc [ , $type ]>
384 Converts the wildcard expression C<$wc> into a regular expression according to the options stored into the L<Regexp::Wildcards> object, or to C<$type> if it's supplied. It successively escapes all unprotected regexp special characters that doesn't hold any meaning for wildcards, then replace C<'jokers'> or C<'sql'> and C<'commas'> or C<'brackets'> (depending on the L</do> or L</type> options), all of this by applying the C<'capture'> rules specified in the constructor or by L</capture>.
389 my ($self, $wc, $type) = @_;
393 $config = $self->_type($type);
397 return unless defined $wc;
398 my $do = $config->{do};
399 my $e = $config->{escape};
400 $wc =~ s/(?<!\\)((?:\\\\)*[^\w\s\\$e])/\\$1/g;
402 $wc = $self->_jokers($wc);
403 } elsif ($do->{sql}) {
404 $wc = $self->_sql($wc);
406 if ($do->{brackets}) {
407 $wc = $self->_bracketed($wc);
408 } elsif ($do->{commas}) {
409 if ($wc =~ /(?<!\\)(?:\\\\)*,/) { # win32 allows comma-separated lists
410 $wc = $self->{'c_brackets'} . $self->_commas($wc) . ')';
418 An object module shouldn't export any function, and so does this one.
422 L<Carp> (core module since perl 5), L<Text::Balanced> (since 5.7.3).
426 This module does not implement the strange behaviours of Windows shell that result from the special handling of the three last characters (for the file extension). For example, Windows XP shell matches C<*a> like C<.*a>, C<*a?> like C<.*a.?>, C<*a??> like C<.*a.{0,2}> and so on.
434 Vincent Pit, C<< <perl at profvince.com> >>, L<http://www.profvince.com>.
436 You can contact me by mail or on C<irc.perl.org> (vincent).
440 Please report any bugs or feature requests to C<bug-regexp-wildcards at rt.cpan.org>, or through the web interface at L<http://rt.cpan.org/NoAuth/ReportBug.html?Queue=Regexp-Wildcards>. I will be notified, and then you'll automatically be notified of progress on your bug as I make changes.
444 You can find documentation for this module with the perldoc command.
446 perldoc Regexp::Wildcards
448 Tests code coverage report is available at L<http://www.profvince.com/perl/cover/Regexp-Wildcards>.
450 =head1 COPYRIGHT & LICENSE
452 Copyright 2007-2008 Vincent Pit, all rights reserved.
454 This program is free software; you can redistribute it and/or modify it
455 under the same terms as Perl itself.
459 sub _extract ($) { extract_bracketed $_[0], '{', qr/.*?(?<!\\)(?:\\\\)*(?={)/ }
464 # escape an odd number of \ that doesn't protect a regexp/wildcard special char
465 s/(?<!\\)((?:\\\\)*\\(?:[\w\s]|$))/\\$1/g;
466 # substitute ? preceded by an even number of \
467 my $s = $self->{c_single};
468 s/(?<!\\)((?:\\\\)*)\?/$1$s/g;
469 # substitute * preceded by an even number of \
471 s/(?<!\\)((?:\\\\)*)\*+/$1$s/g;
478 # escape an odd number of \ that doesn't protect a regexp/wildcard special char
479 s/(?<!\\)((?:\\\\)*\\(?:[^\W_]|\s|$))/\\$1/g;
480 # substitute _ preceded by an even number of \
481 my $s = $self->{c_single};
482 s/(?<!\\)((?:\\\\)*)_/$1$s/g;
483 # substitute % preceded by an even number of \
485 s/(?<!\\)((?:\\\\)*)%+/$1$s/g;
491 # substitute , preceded by an even number of \
492 s/(?<!\\)((?:\\\\)*),/$1|/g;
497 my ($self, $rest) = @_;
498 substr $rest, 0, 1, '';
500 my ($re, $bracket, $prefix) = ('');
501 while (do { ($bracket, $rest, $prefix) = _extract $rest; $bracket }) {
502 $re .= $self->_commas($prefix) . $self->_brackets($bracket);
504 $re .= $self->_commas($rest);
505 return $self->{c_brackets} . $re . ')';
509 my ($self, $rest) = @_;
510 my ($re, $bracket, $prefix) = ('');
511 while (do { ($bracket, $rest, $prefix) = _extract $rest; $bracket }) {
512 $re .= $prefix . $self->_brackets($bracket);
515 $re =~ s/(?<!\\)((?:\\\\)*[\{\},])/\\$1/g;
519 1; # End of Regexp::Wildcards