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 $rw->capture(); # No more captures.
48 In many situations, users may want to specify patterns to match but don't need the full power of regexps.
49 Wildcards make one of those sets of simplified rules.
50 This module converts wildcard expressions to Perl regular expressions, so that you can use them for matching.
52 It handles the C<*> and C<?> jokers, as well as Unix bracketed alternatives C<{,}>, but also C<%> and C<_> SQL wildcards.
53 If required, it can also keep original C<(...)> groups or C<^> and C<$> anchors.
54 Backspace (C<\>) is used as an escape character.
56 Typesets that mimic the behaviour of Windows and Unix shells are also provided.
63 croak 'First argument isn\'t a valid ' . __PACKAGE__ . ' object'
64 unless ref $_[0] and $_[0]->isa(__PACKAGE__);
68 jokers => [ qw/jokers/ ],
70 commas => [ qw/commas/ ],
71 brackets => [ qw/brackets/ ],
72 unix => [ qw/jokers brackets/ ],
73 win32 => [ qw/jokers commas/ ],
75 $types{$_} = $types{win32} for qw/dos os2 MSWin32 cygwin/;
76 $types{$_} = $types{unix} for qw/linux
78 aix irix hpux dgux dynixptx
80 svr4 solaris sunos dec_osf
81 sco_sv unicos unicosmk/;
93 single => sub { $_[1] ? '(.)' : '.' },
94 any => sub { $_[1] ? ($_[0]->{greedy} ? '(.*)'
97 brackets => sub { $_[1] ? '(' : '(?:'; },
106 $old = { } unless defined $old;
109 $c = { set => $_[0] };
111 croak 'Arguments must be passed as an unique scalar or as key => value pairs';
114 $c = { map { (exists $args{$_}) ? ($_ => $args{$_}) : () } qw/set add rem/ };
116 for (qw/set add rem/) {
118 next unless defined $v;
120 '' => sub { +{ ($_[0] => 1) x (exists $valid->{$_[0]}) } },
121 'ARRAY' => sub { +{ map { ($_ => 1) x (exists $valid->{$_}) } @{$_[0]} } },
122 'HASH' => sub { +{ map { ($_ => $_[0]->{$_}) x (exists $valid->{$_}) }
125 croak 'Wrong option set' unless $cb;
126 $c->{$_} = $cb->($v);
128 my $config = (exists $c->{set}) ? $c->{set} : $old;
129 $config->{$_} = $c->{add}->{$_} for grep $c->{add}->{$_},
130 keys %{$c->{add} || {}};
131 delete $config->{$_} for grep $c->{rem}->{$_}, keys %{$c->{rem} || {}};
138 $config->{do} = $self->_validate(\%escapes, $self->{do}, @_);
139 $config->{escape} = '';
140 $config->{escape} .= $escapes{$_} for keys %{$config->{do}};
141 $config->{escape} = quotemeta $config->{escape};
148 my $config = $self->_do(@_);
149 $self->{$_} = $config->{$_} for keys %$config;
156 $config->{capture} = $self->_validate(\%captures, $self->{capture}, @_);
157 $config->{greedy} = delete $config->{capture}->{greedy};
158 for (keys %captures) {
159 $config->{'c_' . $_} = $captures{$_}->($config, $config->{capture}->{$_})
160 if $captures{$_}; # Skip 'greedy'
168 my $config = $self->_capture(@_);
169 $self->{$_} = $config->{$_} for keys %$config;
174 my ($self, $type) = @_;
175 $type = 'unix' unless defined $type;
176 croak 'Wrong type' unless exists $types{$type};
177 my $config = $self->_do($types{$type});
178 $config->{type} = $type;
185 my $config = $self->_type(@_);
186 $self->{$_} = $config->{$_} for keys %$config;
192 $class = ref($class) || $class || __PACKAGE__;
193 croak 'Optional arguments must be passed as key => value pairs' if @_ % 2;
197 if (defined $args{do}) {
198 $self->do($args{do});
200 $self->type($args{type});
202 $self->capture($args{capture});
205 =head2 C<< new [ do => $what E<verbar> type => $type ], capture => $captures >>
207 Constructs a new L<Regexp::Wildcard> object.
209 C<do> lists all features that should be enabled when converting wildcards to regexps.
210 Refer to L</do> for details on what can be passed in C<$what>.
212 The C<type> specifies a predefined set of C<do> features to use.
213 See L</type> for details on which types are valid.
214 The C<do> option overrides C<type>.
216 C<capture> lists which atoms should be capturing.
217 Refer to L</capture> for more details.
219 =head2 C<< do [ $what E<verbar> set => $c1, add => $c2, rem => $c3 ] >>
221 Specifies the list of metacharacters to convert or to prevent for escaping.
222 They fit into six classes :
228 C<'jokers'> converts C<?> to C<.> and C<*> to C<.*> ;
230 'a**\\*b??\\?c' ==> 'a.*\\*b..\\?c'
234 C<'sql'> converts C<_> to C<.> and C<%> to C<.*> ;
236 'a%%\\%b__\\_c' ==> 'a.*\\%b..\\_c'
240 C<'commas'> converts all C<,> to C<|> and puts the complete resulting regular expression inside C<(?: ... )> ;
242 'a,b{c,d},e' ==> '(?:a|b\\{c|d\\}|e)'
246 C<'brackets'> converts all matching C<{ ... , ... }> brackets to C<(?: ... | ... )> alternations.
247 If some brackets are unbalanced, it tries to substitute as many of them as possible, and then escape the remaining unmatched C<{> and C<}>.
248 Commas outside of any bracket-delimited block are also escaped ;
250 'a,b{c,d},e' ==> 'a\\,b(?:c|d)\\,e'
251 '{a\\{b,c}d,e}' ==> '(?:a\\{b|c)d\\,e\\}'
252 '{a{b,c\\}d,e}' ==> '\\{a\\{b\\,c\\}d\\,e\\}'
256 C<'groups'> keeps the parenthesis C<( ... )> of the original string without escaping them.
257 Currently, no check is done to ensure that the parenthesis are matching.
259 'a(b(c))d\\(\\)' ==> (no change)
263 C<'anchors'> prevents the I<beginning-of-line> C<^> and I<end-of-line> C<$> anchors to be escaped.
264 Since C<[...]> character class are currently escaped, a C<^> will always be interpreted as I<beginning-of-line>.
266 'a^b$c' ==> (no change)
270 Each C<$c> can be any of :
276 A hash reference, with wanted metacharacter group names (described above) as keys and booleans as values ;
280 An array reference containing the list of wanted metacharacter classes ;
284 A plain scalar, when only one group is required.
288 When C<set> is present, the classes given as its value replace the current object options.
289 Then the C<add> classes are added, and the C<rem> classes removed.
291 Passing a sole scalar C<$what> is equivalent as passing C<< set => $what >>.
292 No argument means C<< set => [ ] >>.
294 $rw->do(set => 'jokers'); # Only translate jokers.
295 $rw->do('jokers'); # Same.
296 $rw->do(add => [ qw/sql commas/ ]); # Translate also SQL and commas.
297 $rw->do(rem => 'jokers'); # Specifying both 'sql' and 'jokers' is useless.
298 $rw->do(); # Translate nothing.
302 Notifies to convert the metacharacters that corresponds to the predefined type C<$type>.
303 C<$type> can be any of :
309 C<'jokers'>, C<'sql'>, C<'commas'>, C<'brackets'>
311 Singleton types that enable the corresponding C<do> classes.
317 Covers typical Unix shell globbing features (effectively C<'jokers'> and C<'brackets'>).
321 C<$^O> values for common Unix systems
323 Wrap to C<'unix'> (see L<perlport> for the list).
329 Defaults to C<'unix'>.
335 Covers typical Windows shell globbing features (effectively C<'jokers'> and C<'commas'>).
339 C<'dos'>, C<'os2'>, C<'MSWin32'>, C<'cygwin'>
345 In particular, you can usually pass C<$^O> as the C<$type> and get the corresponding shell behaviour.
347 $rw->type('win32'); # Set type to win32.
348 $rw->type($^O); # Set type to unix on Unices and win32 on Windows
349 $rw->type(); # Set type to unix.
351 The C<type> method returns the L<Regexp::Wildcards> object.
353 =head2 C<< capture [ $captures E<verbar> set => $c1, add => $c2, rem => $c3 ] >>
355 Specifies the list of atoms to capture.
356 This method works like L</do>, except that the classes are different :
362 C<'single'> will capture all unescaped I<"exactly one"> metacharacters, i.e. C<?> for wildcards or C<_> for SQL ;
364 'a???b\\??' ==> 'a(.)(.)(.)b\\?(.)'
365 'a___b\\__' ==> 'a(.)(.)(.)b\\_(.)'
369 C<'any'> will capture all unescaped I<"any"> metacharacters, i.e. C<*> for wildcards or C<%> for SQL ;
371 'a***b\\**' ==> 'a(.*)b\\*(.*)'
372 'a%%%b\\%%' ==> 'a(.*)b\\%(.*)'
376 C<'greedy'>, when used in conjunction with C<'any'>, will make the C<'any'> captures greedy (by default they are not) ;
378 'a***b\\**' ==> 'a(.*?)b\\*(.*?)'
379 'a%%%b\\%%' ==> 'a(.*?)b\\%(.*?)'
383 C<'brackets'> will capture matching C<{ ... , ... }> alternations.
385 'a{b\\},\\{c}' ==> 'a(b\\}|\\{c)'
389 $rw->capture(set => 'single'); # Only capture "exactly one" metacharacters.
390 $rw->capture('single'); # Same.
391 $rw->capture(add => [ qw/any greedy/ ]); # Also greedily capture "any" metacharacters.
392 $rw->capture(rem => 'greedy'); # No more greed please.
393 $rw->capture(); # Capture nothing.
395 =head2 C<convert $wc [ , $type ]>
397 Converts the wildcard expression C<$wc> into a regular expression according to the options stored into the L<Regexp::Wildcards> object, or to C<$type> if it's supplied.
398 It successively escapes all unprotected regexp special characters that doesn't hold any meaning for wildcards, then replace C<'jokers'> or C<'sql'> and C<'commas'> or C<'brackets'> (depending on the L</do> or L</type> options), all of this by applying the C<'capture'> rules specified in the constructor or by L</capture>.
403 my ($self, $wc, $type) = @_;
407 $config = $self->_type($type);
411 return unless defined $wc;
412 my $do = $config->{do};
413 my $e = $config->{escape};
414 $wc =~ s/(?<!\\)((?:\\\\)*[^\w\s\\$e])/\\$1/g;
416 $wc = $self->_jokers($wc);
417 } elsif ($do->{sql}) {
418 $wc = $self->_sql($wc);
420 if ($do->{brackets}) {
421 $wc = $self->_bracketed($wc);
422 } elsif ($do->{commas}) {
423 if ($wc =~ /(?<!\\)(?:\\\\)*,/) { # win32 allows comma-separated lists
424 $wc = $self->{'c_brackets'} . $self->_commas($wc) . ')';
432 An object module shouldn't export any function, and so does this one.
436 L<Carp> (core module since perl 5), L<Text::Balanced> (since 5.7.3).
440 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).
441 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.
449 Vincent Pit, C<< <perl at profvince.com> >>, L<http://www.profvince.com>.
451 You can contact me by mail or on C<irc.perl.org> (vincent).
455 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.
459 You can find documentation for this module with the perldoc command.
461 perldoc Regexp::Wildcards
463 Tests code coverage report is available at L<http://www.profvince.com/perl/cover/Regexp-Wildcards>.
465 =head1 COPYRIGHT & LICENSE
467 Copyright 2007-2008 Vincent Pit, all rights reserved.
469 This program is free software; you can redistribute it and/or modify it
470 under the same terms as Perl itself.
474 sub _extract ($) { extract_bracketed $_[0], '{', qr/.*?(?<!\\)(?:\\\\)*(?={)/ }
479 # escape an odd number of \ that doesn't protect a regexp/wildcard special char
480 s/(?<!\\)((?:\\\\)*\\(?:[\w\s]|$))/\\$1/g;
481 # substitute ? preceded by an even number of \
482 my $s = $self->{c_single};
483 s/(?<!\\)((?:\\\\)*)\?/$1$s/g;
484 # substitute * preceded by an even number of \
486 s/(?<!\\)((?:\\\\)*)\*+/$1$s/g;
493 # escape an odd number of \ that doesn't protect a regexp/wildcard special char
494 s/(?<!\\)((?:\\\\)*\\(?:[^\W_]|\s|$))/\\$1/g;
495 # substitute _ preceded by an even number of \
496 my $s = $self->{c_single};
497 s/(?<!\\)((?:\\\\)*)_/$1$s/g;
498 # substitute % preceded by an even number of \
500 s/(?<!\\)((?:\\\\)*)%+/$1$s/g;
506 # substitute , preceded by an even number of \
507 s/(?<!\\)((?:\\\\)*),/$1|/g;
512 my ($self, $rest) = @_;
513 substr $rest, 0, 1, '';
515 my ($re, $bracket, $prefix) = ('');
516 while (do { ($bracket, $rest, $prefix) = _extract $rest; $bracket }) {
517 $re .= $self->_commas($prefix) . $self->_brackets($bracket);
519 $re .= $self->_commas($rest);
520 return $self->{c_brackets} . $re . ')';
524 my ($self, $rest) = @_;
525 my ($re, $bracket, $prefix) = ('');
526 while (do { ($bracket, $rest, $prefix) = _extract $rest; $bracket }) {
527 $re .= $prefix . $self->_brackets($bracket);
530 $re =~ s/(?<!\\)((?:\\\\)*[\{\},])/\\$1/g;
534 1; # End of Regexp::Wildcards