view release on metacpan or  search on metacpan

lib/Algorithm/MLCS.pm  view on Meta::CPAN

use vars qw/ $VERSION @ISA @EXPORT /;

require Exporter;

@ISA     = qw/ Exporter /;
@EXPORT  = qw/ lcs /;
$VERSION = '1.02';

# Gets arrayref of sequences (arrayrefs) and return LCS array in list context
# or length of LCS in scalar context
sub lcs {
    my ( @seq, @lcs ) = map { _build_seq($_) } _get_dict( $_[0] );

    while ( @seq && !( grep { !@$_ } @seq ) ) {
        my %dict = ( %{ $seq[0][0] } );

        for my $s ( @seq[ 1 .. $#seq ] ) {
            %dict = map {
                      $_ => $dict{$_} > $s->[0]{$_}
                    ? $s->[0]{$_} : $dict{$_}

lib/Algorithm/MLCS.pm  view on Meta::CPAN

    use Algorithm::MLCS;

    my @seqs = (
        [ qw/a b c d f g h j q z/ ],
        [ qw/a b c d f g h j q z/ ],
        [ qw/a b c x f h j q z/   ],
        [ qw/a b c f g j q z/     ],
    );

    my @lcs = lcs( \@seqs );
    my $lcs_length = lcs( \@seqs );
    print Dumper( \@lcs );

=head1 ABSTRACT

Finding the longest common subsequence (LCS) for the general case of an arbitrary
number of input sequences is an NP-hard problem. Algorithm::MLCS implements a fast
heuristic algorithm that addresses the general case of multiple sequences.
It is able to extract common subsequence that is close to the optimal ones.

=head1 METHODS

=head2 lcs ( \@seqs )

Finds a Longest Common Subsequence of multiple sequences given by @seqs arrayref.
Each element of @seqs is arrayref that represents the one of multiple sequences
(e.g. [ ['a', 'b', 'c'], ['a', 'c', 'd', 'e'], ... ]). In list context it returns
LCS array, in scalar - the length of LCS.

=head1 SEE ALSO

Algorithm::LCS

=head1 AUTHOR

Slava Moiseev, C<< <slava.moiseev at yahoo.com> >>

=head1 LICENSE AND COPYRIGHT