[scpubgit/stemmatology.git] / lib / Text / Tradition / Language / English.pm

package Text::Tradition::Language::English;

use strict;
use warnings;
use Text::Tradition::Language::Base qw/ lemmatize_treetagger reading_lookup_treetagger /;
use TryCatch;

=head1 NAME

Text::Tradition::Language::English - language-specific module for English

=head1 DESCRIPTION

Implements morphology lookup for English words in context.  This module
depends on the TreeTagger software
(L<http://www.ims.uni-stuttgart.de/projekte/corplex/TreeTagger/>), which is
(for now) expected to be installed in $MORPHDIR/TreeTagger.

=head1 SUBROUTINES

=head2 lemmatize( $text )

Evaluates the string using the TreeTagger, and returns the results.

=begin testing

binmode STDOUT, ':utf8';
use Text::Tradition;
use_ok( 'Text::Tradition::Language::English' );

=end testing

=cut

sub lemmatize {
	my $tradition = shift;
	my %opts = ( 
		'language' => 'French', 
		'callback' => sub { _parse_wordform( @_ ) } 
		);
	return lemmatize_treetagger( $tradition, %opts );
}

=head2 reading_lookup( $rdg[, $rdg, ...] )

Looks up one or more readings using the Flemm package, and returns the
possible results.  This uses the same logic as L<lemmatize> above for the
entire tradition, but can also be used to (re-)analyze individual readings.

=cut

sub reading_lookup {
	my( @path ) = @_;
	my %opts = ( 
		'language' => 'French',
		'callback' => sub { _parse_wordform( @_ ) },
		'path' => \@path,
		);
	return reading_lookup_treetagger( %opts );
}

# Utility function to turn a TreeTagger result into a WordForm
sub _parse_wordform {
	my $tagresult = shift;
	my( $orig, $tag, $lemma ) = split( /\t/, $tagresult );
	my $morphobj = Lingua::TagSet::TreeTagger->tag2structure( $tag );
	if( $morphobj ) {
		return Text::Tradition::Collation::Reading::WordForm->new(
			'language' => 'English',
			'lemma' => $lemma,
			'morphology' => $morphobj,
			);
	} else {
		warn "No morphology found for word: $_";
	}
}

1;

=head2 TODO

=over

=item * Tests!

=back

=head1 LICENSE

This package is free software and is provided "as is" without express
or implied warranty.  You can redistribute it and/or modify it under
the same terms as Perl itself.

=head1 AUTHOR

Tara L Andrews E<lt>aurum@cpan.orgE<gt>
Commit	Line	Data
f4b6b4d0	1	package Text::Tradition::Language::English;
	2
	3	use strict;
	4	use warnings;
e0f6836a	5	use Text::Tradition::Language::Base qw/ lemmatize_treetagger reading_lookup_treetagger /;
f4b6b4d0	6	use TryCatch;
f4b6b4d0	7
f4b6b4d0	8	=head1 NAME
	9
	10	Text::Tradition::Language::English - language-specific module for English
	11
	12	=head1 DESCRIPTION
	13
	14	Implements morphology lookup for English words in context. This module
	15	depends on the TreeTagger software
	16	(L<http://www.ims.uni-stuttgart.de/projekte/corplex/TreeTagger/>), which is
	17	(for now) expected to be installed in $MORPHDIR/TreeTagger.
	18
	19	=head1 SUBROUTINES
	20
	21	=head2 lemmatize( $text )
	22
	23	Evaluates the string using the TreeTagger, and returns the results.
	24
	25	=begin testing
	26
	27	binmode STDOUT, ':utf8';
	28	use Text::Tradition;
	29	use_ok( 'Text::Tradition::Language::English' );
	30
	31	=end testing
	32
	33	=cut
	34
	35	sub lemmatize {
	36	my $tradition = shift;
e0f6836a	37	my %opts = (
	38	'language' => 'French',
	39	'callback' => sub { _parse_wordform( @_ ) }
	40	);
	41	return lemmatize_treetagger( $tradition, %opts );
f4b6b4d0	42	}
	43
	44	=head2 reading_lookup( $rdg[, $rdg, ...] )
	45
	46	Looks up one or more readings using the Flemm package, and returns the
	47	possible results. This uses the same logic as L<lemmatize> above for the
	48	entire tradition, but can also be used to (re-)analyze individual readings.
	49
	50	=cut
	51
	52	sub reading_lookup {
e0f6836a	53	my( @path ) = @_;
	54	my %opts = (
	55	'language' => 'French',
	56	'callback' => sub { _parse_wordform( @_ ) },
	57	'path' => \@path,
	58	);
	59	return reading_lookup_treetagger( %opts );
f4b6b4d0	60	}
	61
	62	# Utility function to turn a TreeTagger result into a WordForm
	63	sub _parse_wordform {
	64	my $tagresult = shift;
	65	my( $orig, $tag, $lemma ) = split( /\t/, $tagresult );
	66	my $morphobj = Lingua::TagSet::TreeTagger->tag2structure( $tag );
	67	if( $morphobj ) {
	68	return Text::Tradition::Collation::Reading::WordForm->new(
	69	'language' => 'English',
	70	'lemma' => $lemma,
	71	'morphology' => $morphobj,
	72	);
	73	} else {
	74	warn "No morphology found for word: $_";
	75	}
	76	}
	77
	78	1;
	79
	80	=head2 TODO
	81
	82	=over
	83
e0f6836a	84	=item * Tests!
f4b6b4d0	85
	86	=back
	87
	88	=head1 LICENSE
	89
	90	This package is free software and is provided "as is" without express
	91	or implied warranty. You can redistribute it and/or modify it under
	92	the same terms as Perl itself.
	93
	94	=head1 AUTHOR
	95
	96	Tara L Andrews E<lt>aurum@cpan.orgE<gt>