X-Git-Url: http://git.shadowcat.co.uk/gitweb/gitweb.cgi?a=blobdiff_plain;f=lib%2FText%2FTradition%2FCollation%2FReading.pm;h=29a8aa17891441840d94a407a58ce617246f376a;hb=c3e04fb5fc2dfb0a8451c2a12538440548ccedbd;hp=fa9ce8b49d3fa598337ada8fb40360664940a978;hpb=1d3104950074a3d7470f01ef0ec8e9046d95b124;p=scpubgit%2Fstemmatology.git diff --git a/lib/Text/Tradition/Collation/Reading.pm b/lib/Text/Tradition/Collation/Reading.pm index fa9ce8b..29a8aa1 100644 --- a/lib/Text/Tradition/Collation/Reading.pm +++ b/lib/Text/Tradition/Collation/Reading.pm @@ -1,13 +1,15 @@ package Text::Tradition::Collation::Reading; use Moose; -use overload '""' => \&_stringify; -use Text::Tradition::Collation; +use Module::Load; +use YAML::XS; +use overload '""' => \&_stringify, 'fallback' => 1; =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 @@ -20,12 +22,13 @@ a particular sequence =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. @@ -37,7 +40,11 @@ Options include: =item is_lacuna - The 'reading' represents a known gap in the text. -=item rank - The sequence number of the reading. This should probably not be set manually. +=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. =back @@ -78,8 +85,15 @@ has 'text' => ( is => 'ro', isa => 'Str', required => 1, + writer => 'alter_text', ); - + +has 'language' => ( + is => 'ro', + isa => 'Str', + predicate => 'has_language', + ); + has 'is_start' => ( is => 'ro', isa => 'Bool', @@ -97,12 +111,60 @@ has 'is_lacuna' => ( isa => 'Bool', default => undef, ); + +has 'is_ph' => ( + is => 'ro', + isa => 'Bool', + default => undef, + ); + +has 'is_common' => ( + is => 'rw', + isa => 'Bool', + default => undef, + ); has 'rank' => ( 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 lexemes for the reading. +has 'reading_lexemes' => ( + traits => ['Array'], + isa => 'ArrayRef[Text::Tradition::Collation::Reading::Lexeme]', + handles => { + lexemes => 'elements', + has_lexemes => 'count', + clear_lexemes => 'clear', + add_lexeme => 'push', + }, + default => sub { [] }, + ); + +## 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 { @@ -114,12 +176,11 @@ around BUILDARGS => sub { } else { $args = { @_ }; } - + # If one of our special booleans is set, we change the text and the # ID to match. - if( exists $args->{'is_lacuna'} && !exists $args->{'text'} ) { - $args->{'text'} = sprintf( "#LACUNA_%s#", $args->{'id'} ); + $args->{'text'} = '#LACUNA#'; } elsif( exists $args->{'is_start'} ) { $args->{'id'} = '#START#'; # Change the ID to ensure we have only one $args->{'text'} = '#START#'; @@ -127,6 +188,8 @@ around BUILDARGS => sub { } elsif( exists $args->{'is_end'} ) { $args->{'id'} = '#END#'; # Change the ID to ensure we have only one $args->{'text'} = '#END#'; + } elsif( exists $args->{'is_ph'} ) { + $args->{'text'} = $args->{'id'}; } $class->$orig( $args ); @@ -142,15 +205,64 @@ of text found in a witness. sub is_meta { my $self = shift; - return $self->is_start || $self->is_end || $self->is_lacuna; + return $self->is_start || $self->is_end || $self->is_lacuna || $self->is_ph; } -# Some syntactic sugar +=head1 Convenience methods + +=head2 related_readings + +Calls Collation's related_readings with $self as the first argument. + +=cut + sub related_readings { my $self = shift; 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. + +=cut + +sub predecessors { + my $self = shift; + my @pred = $self->collation->sequence->predecessors( $self->id ); + return map { $self->collation->reading( $_ ) } @pred; +} + +=head2 successors + +Returns a list of Reading objects that immediately follow $self in the collation. + +=cut + +sub successors { + my $self = shift; + my @succ = $self->collation->sequence->successors( $self->id ); + return map { $self->collation->reading( $_ ) } @succ; +} + +=head2 set_identical( $other_reading) + +Backwards compatibility method, to add a transposition relationship +between $self and $other_reading. Don't use this. + +=cut + sub set_identical { my( $self, $other ) = @_; return $self->collation->add_relationship( $self, $other, @@ -162,8 +274,79 @@ sub _stringify { return $self->id; } +=head1 MORPHOLOGY + +Methods for the morphological information (if any) attached to readings. +A reading may be made up of multiple lexemes; the concatenated lexeme +strings ought to match the reading's normalized form. + +See L for more information +on Lexeme objects and their attributes. + +=head2 has_lexemes + +Returns a true value if the reading has any attached lexemes. + +=head2 lexemes + +Returns the Lexeme objects (if any) attached to the reading. + +=head2 clear_lexemes + +Wipes any associated Lexeme objects out of the reading. + +=head2 add_lexeme( $lexobj ) + +Adds the Lexeme in $lexobj to the list of lexemes. + +=head2 lemmatize + +If the language of the reading is set, this method will use the appropriate +Language model to determine the lexemes that belong to this reading. See +L if you wish to lemmatize an entire tradition. + +=cut + +sub lemmatize { + my $self = shift; + unless( $self->has_language ) { + warn "Please set a language to lemmatize a tradition"; + return; + } + my $mod = "Text::Tradition::Language::" . $self->language; + load( $mod ); + $mod->can( 'reading_lookup' )->( $self ); + +} + +# For graph serialization. Return a string representation of the associated +# reading lexemes. +sub _serialize_lexemes { + my $self = shift; + my @lexstrs; + foreach my $l ( $self->lexemes ) { + my @mf; + foreach my $wf ( $l->matching_forms ) { + push( @mf, $wf->to_string ); + } + my $form = $l->form ? $l->form->to_string : ''; + push( @lexstrs, join( '|L|', $l->language, $l->string, $form, + join( '|M|', @mf ) ) ); + } + return join( '|R|', @lexstrs ); +} + + +## Utility methods + +sub TO_JSON { + my $self = shift; + return $self->text; +} + +## TODO will need a throw() here + no Moose; __PACKAGE__->meta->make_immutable; 1; -