view release on metacpan or  search on metacpan

lib/Lingua/IdSplitter.pm  view on Meta::CPAN

    foreach (@{$self->{hard}}) {
      $str .= "Technique: $_->{tech}\n";
      $str .= "Terms: ".join(',',@{$_->{terms}});
      $str .= "\n";
    }
  }

  if ( $self->{explain_rank}) {
    $str .= "\n## soft split rank(s):\n";
    $str .= join("\n", @{$self->{explain_rank}});
  }

  return $str;
}

sub _explain_rank {
  my ($self) = @_;

  my $r;
  foreach (@{$self->{rank}}) {
    my @parts;
    foreach (@{$_->{terms}}) {
      if ($_->{t} eq $_->{s}) {
        push @parts, $_->{t};
      }
      else {
        push @parts, "$_->{t}(<-$_->{s})";
      }
    }
    $r .= join(',',@parts) . " ---> $_->{expr} = $_->{score}\n";
  }

  return $r;
}

1;

__END__

=pod

=encoding UTF-8

=head1 NAME

Lingua::IdSplitter - split identifiers into words

=head1 VERSION

version 0.03

=head1 SYNOPSIS

    use Lingua::IdSplitter;

    my $splitter = Lingua::IdSplitter->new;
    $splitter->split($identifier);

=head1 DESCRIPTION

This module implements and algorithm to identify and split multi-word
identifier in their individual words. For example, "UserFind" in "user"
and "find", or "timesort" in "time" and "sort".

For more details on the algorithm check the following
L<article|http://www.sciencedirect.com/science/article/pii/S0164121214002179>
(also available L<here|http://hdl.handle.net/10198/11577>).

=head1 FUNCTIONS

=head2 new

Create a new splitter object. A list of specific dictionaries is optional,
check the C<bin/id-splitter> command for an example on how to use more
dictionaries.

=head2 soft_split

Perform a soft split of the identifier, ie split words without using
explicit markers (eg, the underscore character, or CamelCase notation).

=head2 hard_split

Perform a hard split of the identifier, ie split words using
explicit markers (eg, the underscore character, or CamelCase notation).

=head2 split

Perform a split applying first a hard split, and the applying a soft split
to the resulting set of the first split.

=head2 explain

Show the computed ranked (including scores) for a split.

=head1 AUTHOR

Nuno Carvalho <smash@cpan.org>

=head1 COPYRIGHT AND LICENSE

This software is copyright (c) 2014-2015 by Project Natura <natura@natura.di.uminho.pt>.

This is free software; you can redistribute it and/or modify it under
the same terms as the Perl 5 programming language system itself.

=cut