NAME Regexp::SAR - perl module implementing regular expression engine for handling matching events (Simple API Regexp) SYNOPSIS use Regexp::SAR; my $sar1 = new Regexp::SAR; my $matched = 0; $sar1->addRegexp('abc', sub {$matched = 1;}); $sar1->match('mm abc nn'); if ($matched) { #proc matching } ################################################# #index many regexp for single match run my @matched; my $sar2 = new Regexp::SAR; my $regexps = [ ['ab+c', 'First Match'], ['\d+', 'Second Match'], ]; my $string; foreach my $re (@$regexps) { my ($reStr, $reTitle) = @$re; $sar2->addRegexp( $reStr, sub { my ($from, $to) = @_; my $matchStr = substr($string, $from, $to - $from); push @matched, "$reTitle: $matchStr"; $sar2->continueFrom($to); } ); } $string = 'first abbbbc second 123 end'; $sar2->match(\$string); # @matched has ('First Match: abbbbc', 'Second Match: 123') ################################################# #get third match and stop my $sar3 = new Regexp::SAR; my $matchedStr3; my $matchCount = 0; my $string3 = 'aa11 bb22 cc33 dd44'; $sar3->addRegexp('\w+', sub { my ($from, $to) = @_; ++$matchCount; if ($matchCount == 3) { $matchedStr3 = substr($string3, $from, $to - $from); $sar3->stopMatch(); } else { $sar3->continueFrom($to); } }); $sar3->match($string3); # $matchCount is 3, $matchedStr3 is 'cc33' ################################################# #get match only at certain position my $sar4 = new Regexp::SAR; my $matchedStr4; my $string4 = 'aa11 bb22 cc33 dd44'; $sar4->addRegexp('\w+', sub { my ($from, $to) = @_; $matchedStr4 = substr($string4, $from, $to - $from); }); $sar4->matchAt($string4, 5); #$matchedStr4 is 'bb22' ################################################# #negative matching my $sar5 = new Regexp::SAR; $sar5->addRegexp('a\^\d+c', sub { print "Matched\n"; }); $sar5->match('axyzb'); DESCRIPTION Regexp::SAR (Simple API for Regexp) module build trie structure for many regular expressions and store match handler for each regular expression that will be called when match occurs. There is no limit for number of regular expressions. Handler called immediately on match and it get matching start and end positions in matched string. Matching can be started from any point in matching string. Match handler can decide from which point matching should continue or it can stop matching at all. METHODS new() Create new Regexp::SAR object. Every object store it's own trie structure separately. When object goes out of scope object and it's internal data structure will be cleared from memory. addRegexp Add regular expression for handling. First parameter is regular expression string. Second parameter is reference to subroutine that will be called when match on this regexp occurs. Handler subroutine get as input two integers, matching start and matching end. Matching start is position of first matching character. Matching end is position after last matching character. my $sar = new Regexp::SAR; my $string = 'a123b'; $sar->addRegexp('\d+', sub { my ($from, $to) = @_; # $from is 1 # $to is 4 $sar->stopMatch(); }); $sar->match($string); match Process matching all added regular expressions on matching string passed to match as parameter. match can accept matching string as reference to scalar, it useful when matching string is very long. matchFrom Process matching from specific position. Get two parameters: matching string and number from which start processing. match subroutine is syntactic sugar form matchFrom when second parameter is 0. matchAt Process matching from specific position and do not continue on next characters. continueFrom continueFrom subroutine called in matching handler and define from which position continue matching after it finished matching on current position. stopMatch stopMatch subroutine called in matching handler and send signal to Regexp::SAR object do not continue matching on next characters. Matching rules * Continue matching process character by character even if there was match. my $sar = new Regexp::SAR; my $string = 'a123b'; $sar->addRegexp('\d+', sub { my ($from, $to) = @_; $matchedStr = substr($string, $from, $to - $from); print "Found number is: $matchedStr\n"; }); $sar->match($string); Above code will print 3 times strings: '123', '23', '3' In case it should be matched only once use continueFrom. * Call all matching handlers that could be found from matching position. my $sar = new Regexp::SAR; $sar->addRegexp('new', sub { print "new found\n"; }); $sar->addRegexp('new york', sub { print "new york found\n"; }); $sar->match('new york'); Above code will print "new found", then print "new york found" * Call all matching handlers from different regular expressions that match same matched string. my $sar = new Regexp::SAR; $sar->addRegexp('1', sub { print "one found\n"; }); $sar->addRegexp('\d', sub { print "digit found\n"; }); $sar->match('1'); Above code will print both 'one found' and 'digit found' Character class abbreviations * '.' matches any character * '\s' matches space character (checked by internal isSPACE) * '\d' matches digit character (checked by internal isDIGIT) * '\w' matches alphanumeric character (checked by internal isALNUM) * '\a' matches alpha character (checked by internal isALPHA) * '\^' matches any character that is not followed character or class abbreviation Matching repetitions * '?' means: match 1 or 0 times * '*' means: match 0 or more times * '+' means: match 1 or more times '\' escape character For matching '\' character in matching string regular expression string should iclude it 4 times '\\\\'. my $sar = new Regexp::SAR; my $string = 'a b\c d'; $sar->addRegexp('b\\\\c', sub { print "Matched\n"; }); $sar->match($string); Unicode support Currently this module does not support unicode matching AUTHOR Pinkhas Nisanov COPYRIGHT AND LICENSE This software is copyright (c) 2012 by Pinkhas Nisanov. This is free software; you can redistribute it and/or modify it under the same terms as the Perl 5 programming language system itself. .