This example shows that the main speed difference is reall Regexp::Common, which does a bit more than just match \d{1,3}\.\d{1,3}\. ..:
$ perl -MRegexp::Common=net -wle 'print $RE{net}{IPv4}'
(?:(?:25[0-5]|2[0-4][0-9]|[0-1]?[0-9]{1,2})[.](?:25[0-5]|2[0-4][0-9]|[
+0-1]?[0-9]{1,2})[.](?:25[0-5]|2[0-4][0-9]|[0-1]?[0-9]{1,2})[.](?:25[0
+-5]|2[0-4][0-9]|[0-1]?[0-9]{1,2}))
The optimzation I talked about kicks in when the string is much longer, and the literal char occurs only once or twice. Then the literal is used as an anchor, thus reducing the need for backtracking.
#!/usr/bin/perl
use strict;
use warnings;
my $line = ('a' x 500) . 'b!' . ('a' x 20);
use Benchmark qw( cmpthese );
cmpthese -3, {
literal => sub {$line =~ /a.{1,10}b!/ },
class => sub {$line =~ /a.{1,10}[b][!]/},
};
__END__
Rate class literal
class 3855/s -- -99%
literal 712766/s 18390% --
Update: added benchmark |