=head1 NAME
-Text::Tradition::Collation::Reading - represents a reading (usually a word) in a collation.
-
+Text::Tradition::Collation::Reading - represents a reading (usually a word)
+in a collation.
+
=head1 DESCRIPTION
Text::Tradition is a library for representation and analysis of collated
=head2 new
-Creates a new reading in the given collation with the given attributes.
+Creates a new reading in the given collation with the given attributes.
Options include:
=over 4
-=item collation - The Text::Tradition::Collation object to which this reading belongs. Required.
+=item collation - The Text::Tradition::Collation object to which this
+reading belongs. Required.
=item id - A unique identifier for this reading. Required.
=item is_lacuna - The 'reading' represents a known gap in the text.
-=item is_ph - A temporary placeholder for apparatus parsing purposes. Do not use unless you know what you are doing.
+=item is_ph - A temporary placeholder for apparatus parsing purposes. Do
+not use unless you know what you are doing.
-=item rank - The sequence number of the reading. This should probably not be set manually.
+=item rank - The sequence number of the reading. This should probably not
+be set manually.
=back
writer => 'alter_text',
);
+has 'language' => (
+ is => 'ro',
+ isa => 'Str',
+ default => 'Default',
+ );
+
has 'is_start' => (
is => 'ro',
isa => 'Bool',
is => 'rw',
isa => 'Int',
predicate => 'has_rank',
+ clearer => 'clear_rank',
);
+
+## For morphological analysis
+
+has 'normal_form' => (
+ is => 'rw',
+ isa => 'Str',
+ predicate => 'has_normal_form',
+ );
+
+# Holds the word form. If is_disambiguated is true, the form at index zero
+# is the correct one.
+has 'lexemes' => (
+ traits => ['Array'],
+ isa => 'ArrayRef[Text::Tradition::Collation::Lexeme]',
+ handles => {
+ lexemes => 'elements',
+ has_lexemes => 'count',
+ _clear_lexemes => 'clear',
+ _add_lexeme => 'push',
+ },
+ );
+
+## For prefix/suffix readings
+
+has 'join_prior' => (
+ is => 'ro',
+ isa => 'Bool',
+ default => undef,
+ );
+
+has 'join_next' => (
+ is => 'ro',
+ isa => 'Bool',
+ default => undef,
+ );
around BUILDARGS => sub {
return $self->collation->related_readings( $self, @_ );
}
+=head2 witnesses
+
+Calls Collation's reading_witnesses with $self as the first argument.
+
+=cut
+
+sub witnesses {
+ my $self = shift;
+ return $self->collation->reading_witnesses( $self, @_ );
+}
+
=head2 predecessors
Returns a list of Reading objects that immediately precede $self in the collation.
return $self->id;
}
+=head1 MORPHOLOGY
+
+A few methods to try to tack on morphological information.
+
+=head2 use_lexemes
+
+TBD
+
+=cut
+
+# sub use_lexemes {
+# my( $self, @lexemes ) = @_;
+# # The lexemes need to be the same as $self->text.
+# my $cmpstr = $self->has_normal_form ? lc( $self->normal_form ) : lc( $self->text );
+# $cmpstr =~ s/[\s-]+//g;
+# my $lexstr = lc( join( '', @lexemes ) );
+# $lexstr =~ s/[\s-]+//g;
+# unless( $lexstr eq $cmpstr ) {
+# warn "Cannot split " . $self->text . " into " . join( '.', @lexemes );
+# return;
+# }
+# $self->_clear_morph;
+# map { $self->_add_morph( { $_ => [] } ) } @lexemes;
+# }
+#
+# sub add_morphological_tag {
+# my( $self, $lexeme, $opts ) = @_;
+# my $struct;
+# unless( $opts ) {
+# # No lexeme was passed; use reading text.
+# $opts = $lexeme;
+# $lexeme = $self->text;
+# $self->use_lexemes( $lexeme );
+# }
+# # Get the correct container
+# ( $struct ) = grep { exists $_->{$lexeme} } $self->lexemes;
+# unless( $struct ) {
+# warn "No lexeme $lexeme exists in this reading";
+# return;
+# }
+# # Now make the morph object and add it to this lexeme.
+# my $morph_obj = Text::Tradition::Collation::Reading::Morphology->new( $opts );
+# # TODO Check for existence
+# push( @{$struct->{$lexeme}}, $morph_obj );
+# }
+
+## Utility methods
+
+sub TO_JSON {
+ my $self = shift;
+ return $self->text;
+}
+
+## TODO will need a throw() here
+
+no Moose;
+__PACKAGE__->meta->make_immutable;
+
+###################################################
+### Morphology objects, to be attached to readings
+###################################################
+
+package Text::Tradition::Collation::Reading::Morphology;
+
+use Moose;
+
+has 'lemma' => (
+ is => 'ro',
+ isa => 'Str',
+ required => 1,
+ );
+
+has 'code' => (
+ is => 'ro',
+ isa => 'Str',
+ required => 1,
+ );
+
+has 'language' => (
+ is => 'ro',
+ isa => 'Str',
+ required => 1,
+ );
+
+## Transmute codes into comparison arrays for our various languages.
+
+around BUILDARGS => sub {
+ my $orig = shift;
+ my $class = shift;
+ my $args;
+ if( @_ == 1 && ref( $_[0] ) ) {
+ $args = shift;
+ } else {
+ $args = { @_ };
+ }
+ if( exists( $args->{'serial'} ) ) {
+ my( $lemma, $code ) = split( /!!/, delete $args->{'serial'} );
+ $args->{'lemma'} = $lemma;
+ $args->{'code'} = $code;
+ }
+ $class->$orig( $args );
+};
+
+sub serialization {
+ my $self = shift;
+ return join( '!!', $self->lemma, $self->code );
+};
+
+sub comparison_array {
+ my $self = shift;
+ if( $self->language eq 'French' ) {
+ my @array;
+ my @bits = split( /\+/, $self->code );
+ # First push the non k/v parts.
+ while( @bits && $bits[0] !~ /=/ ) {
+ push( @array, shift @bits );
+ }
+ while( @array < 2 ) {
+ push( @array, undef );
+ }
+ # Now push the k/v parts in a known order.
+ my @fields = qw/ Pers Nb Temps Genre Spec Fonc /;
+ my %props;
+ map { my( $k, $v ) = split( /=/, $_ ); $props{$k} = $v; } @bits;
+ foreach my $k ( @fields ) {
+ push( @array, $props{$k} );
+ }
+ # Give the answer.
+ return @array;
+ } elsif( $self->language eq 'English' ) {
+ # Do something as yet undetermined
+ } else {
+ # Latin or Greek or Armenian, just split the chars
+ return split( '', $self->code );
+ }
+};
+
no Moose;
__PACKAGE__->meta->make_immutable;
projects / scpubgit/stemmatology.git / blobdiff