quick hack to make CTE-derived reading IDs valid XML names
[scpubgit/stemmatology.git] / lib / Text / Tradition / Collation / Reading / WordForm.pm
CommitLineData
cca4f996 1package Text::Tradition::Collation::Reading::WordForm;
2
70745e70 3use Lingua::Features::Structure;
7604424b 4use JSON ();
5use Moose;
cca4f996 6
7=head1 NAME
8
9Text::Tradition::Collation::Reading::WordForm - represents a
10language/lemma/morphology triplet that can be associated with a Reading.
11
12=head1 DESCRIPTION
13
14Text::Tradition is a library for representation and analysis of collated
15texts, particularly medieval ones. A word form is used for the analysis of
16Reading objects; it consists of a lemma, a language, and a code to
17represent its part of speech. In general the word forms for a particular
18language should be read from / written to some morphological database.
19
20=head1 METHODS
21
22=head2 new
23
24Creates a new word form from the passed options.
25
26=head2 language
27
28Returns the language to which this word form belongs.
29
30=head2 lemma
31
32Returns the lemma for the word form.
33
34=head2 morphology
35
36Returns an array representing this word's morphology. The contents of the
37array depend on the language being used.
38
39=cut
40
41has 'language' => (
42 is => 'ro',
43 isa => 'Str',
44 required => 1,
45 );
46
47# TODO do we need this?
48has 'form' => (
49 is => 'ro',
50 isa => 'Str',
51 # required => 1,
52 );
53
54has 'lemma' => (
55 is => 'ro',
56 isa => 'Str',
57 required => 1,
58 );
59
60has 'morphology' => (
61 is => 'ro',
6ad2ce78 62 isa => 'Lingua::Features::Structure',
cca4f996 63 required => 1,
64 );
65
70745e70 66around BUILDARGS => sub {
67 my $orig = shift;
68 my $class = shift;
69 my $args = @_ == 1 ? $_[0] : { @_ };
7604424b 70 if( exists $args->{'JSON'} ) {
71 my $data = $args->{'JSON'};
72 my $morph = Lingua::Features::Structure->from_string( $data->{'morphology'} );
73 $data->{'morphology'} = $morph;
74 $args = $data;
70745e70 75 }
76 $class->$orig( $args );
77};
78
6ad2ce78 79=head2 to_string
80
81Returns a string combination of language/lemma/morphology that can be used
82in equivalence testing.
83
84=cut
85
86sub to_string {
cca4f996 87 my $self = shift;
7604424b 88 return JSON->new->convert_blessed(1)->encode( $self );
89}
90
91sub TO_JSON {
92 my $self = shift;
93 return {
94 'language' => $self->language,
95 'lemma' => $self->lemma,
96 'morphology' => $self->morphology->to_string
97 };
cca4f996 98}
6ad2ce78 99
cca4f996 100no Moose;
101__PACKAGE__->meta->make_immutable;
102
1031;
104
105=head1 LICENSE
106
107This package is free software and is provided "as is" without express
108or implied warranty. You can redistribute it and/or modify it under
109the same terms as Perl itself.
110
111=head1 AUTHOR
112
113Tara L Andrews E<lt>aurum@cpan.orgE<gt>