1 package Text::Tradition::Collation;
4 use Encode qw( decode_utf8 );
8 use IPC::Run qw( run binary );
10 use Text::Tradition::Collation::Data;
11 use Text::Tradition::Collation::Reading;
12 use Text::Tradition::Collation::RelationshipStore;
13 use Text::Tradition::Error;
14 use XML::Easy::Syntax qw( $xml10_namestartchar_rx $xml10_namechar_rx );
16 use XML::LibXML::XPathContext;
20 isa => 'Text::Tradition::Collation::Data',
58 isa => 'Text::Tradition',
59 writer => '_set_tradition',
65 Text::Tradition::Collation - a software model for a text collation
70 my $t = Text::Tradition->new(
71 'name' => 'this is a text',
73 'file' => '/path/to/tei_parallel_seg_file.xml' );
75 my $c = $t->collation;
76 my @readings = $c->readings;
77 my @paths = $c->paths;
78 my @relationships = $c->relationships;
80 my $svg_variant_graph = $t->collation->as_svg();
84 Text::Tradition is a library for representation and analysis of collated
85 texts, particularly medieval ones. The Collation is the central feature of
86 a Tradition, where the text, its sequence of readings, and its relationships
87 between readings are actually kept.
93 The constructor. Takes a hash or hashref of the following arguments:
97 =item * tradition - The Text::Tradition object to which the collation
100 =item * linear - Whether the collation should be linear; that is, whether
101 transposed readings should be treated as two linked readings rather than one,
102 and therefore whether the collation graph is acyclic. Defaults to true.
104 =item * baselabel - The default label for the path taken by a base text
105 (if any). Defaults to 'base text'.
107 =item * wit_list_separator - The string to join a list of witnesses for
108 purposes of making labels in display graphs. Defaults to ', '.
110 =item * ac_label - The extra label to tack onto a witness sigil when
111 representing another layer of path for the given witness - that is, when
112 a text has more than one possible reading due to scribal corrections or
113 the like. Defaults to ' (a.c.)'.
115 =item * wordsep - The string used to separate words in the original text.
126 =head2 wit_list_separator
134 Simple accessors for collation attributes.
138 The meta-reading at the start of every witness path.
142 The meta-reading at the end of every witness path.
146 Returns all Reading objects in the graph.
148 =head2 reading( $id )
150 Returns the Reading object corresponding to the given ID.
152 =head2 add_reading( $reading_args )
154 Adds a new reading object to the collation.
155 See L<Text::Tradition::Collation::Reading> for the available arguments.
157 =head2 del_reading( $object_or_id )
159 Removes the given reading from the collation, implicitly removing its
160 paths and relationships.
162 =head2 has_reading( $id )
164 Predicate to see whether a given reading ID is in the graph.
166 =head2 reading_witnesses( $object_or_id )
168 Returns a list of sigils whose witnesses contain the reading.
172 Returns all reading paths within the document - that is, all edges in the
173 collation graph. Each path is an arrayref of [ $source, $target ] reading IDs.
175 =head2 add_path( $source, $target, $sigil )
177 Links the given readings in the collation in sequence, under the given witness
178 sigil. The readings may be specified by object or ID.
180 =head2 del_path( $source, $target, $sigil )
182 Links the given readings in the collation in sequence, under the given witness
183 sigil. The readings may be specified by object or ID.
185 =head2 has_path( $source, $target );
187 Returns true if the two readings are linked in sequence in any witness.
188 The readings may be specified by object or ID.
192 Returns all Relationship objects in the collation.
194 =head2 add_relationship( $reading, $other_reading, $options )
196 Adds a new relationship of the type given in $options between the two readings,
197 which may be specified by object or ID. Returns a value of ( $status, @vectors)
198 where $status is true on success, and @vectors is a list of relationship edges
199 that were ultimately added.
200 See L<Text::Tradition::Collation::Relationship> for the available options.
205 my ( $class, @args ) = @_;
206 my %args = @args == 1 ? %{ $args[0] } : @args;
207 # TODO determine these from the Moose::Meta object
208 my @delegate_attrs = qw(sequence relations readings wit_list_separator baselabel
209 linear wordsep start end cached_table _graphcalc_done);
211 for my $attr (@delegate_attrs) {
212 $data_args{$attr} = delete $args{$attr} if exists $args{$attr};
214 $args{_data} = Text::Tradition::Collation::Data->new(%data_args);
220 $self->_set_relations( Text::Tradition::Collation::RelationshipStore->new( 'collation' => $self ) );
221 $self->_set_start( $self->add_reading(
222 { 'collation' => $self, 'is_start' => 1, 'init' => 1 } ) );
223 $self->_set_end( $self->add_reading(
224 { 'collation' => $self, 'is_end' => 1, 'init' => 1 } ) );
227 sub register_relationship_type {
229 my %args = @_ == 1 ? %{$_[0]} : @_;
230 if( $self->relations->has_type( $args{name} ) ) {
231 throw( 'Relationship type ' . $args{name} . ' already registered' );
233 $self->relations->add_type( %args );
236 sub get_relationship_type {
237 my( $self, $name ) = @_;
238 return $self->relations->has_type( $name )
239 ? $self->relations->type( $name ) : undef;
242 ### Reading construct/destruct functions
245 my( $self, $reading ) = @_;
246 unless( ref( $reading ) eq 'Text::Tradition::Collation::Reading' ) {
247 my %args = %$reading;
248 if( $args{'init'} ) {
249 # If we are initializing an empty collation, don't assume that we
250 # have set a tradition.
251 delete $args{'init'};
252 } elsif( $self->tradition->can('language') && $self->tradition->has_language
253 && !exists $args{'language'} ) {
254 $args{'language'} = $self->tradition->language;
256 $reading = Text::Tradition::Collation::Reading->new(
257 'collation' => $self,
260 # First check to see if a reading with this ID exists.
261 if( $self->reading( $reading->id ) ) {
262 throw( "Collation already has a reading with id " . $reading->id );
264 $self->_graphcalc_done(0);
265 $self->_add_reading( $reading->id => $reading );
266 # Once the reading has been added, put it in both graphs.
267 $self->sequence->add_vertex( $reading->id );
268 $self->relations->add_reading( $reading->id );
272 around del_reading => sub {
277 if( ref( $arg ) eq 'Text::Tradition::Collation::Reading' ) {
280 # Remove the reading from the graphs.
281 $self->_graphcalc_done(0);
282 $self->_clear_cache; # Explicitly clear caches to GC the reading
283 $self->sequence->delete_vertex( $arg );
284 $self->relations->delete_reading( $arg );
287 $self->$orig( $arg );
290 =head2 merge_readings( $main, $second, $concatenate, $with_str )
292 Merges the $second reading into the $main one. If $concatenate is true, then
293 the merged node will carry the text of both readings, concatenated with either
294 $with_str (if specified) or a sensible default (the empty string if the
295 appropriate 'join_*' flag is set on either reading, or else $self->wordsep.)
297 The first two arguments may be either readings or reading IDs.
304 my $cxfile = 't/data/Collatex-16.xml';
305 my $t = Text::Tradition->new(
307 'input' => 'CollateX',
310 my $c = $t->collation;
312 my $rno = scalar $c->readings;
313 # Split n21 ('unto') for testing purposes
314 my $new_r = $c->add_reading( { 'id' => 'n21p0', 'text' => 'un', 'join_next' => 1 } );
315 my $old_r = $c->reading( 'n21' );
316 $old_r->alter_text( 'to' );
317 $c->del_path( 'n20', 'n21', 'A' );
318 $c->add_path( 'n20', 'n21p0', 'A' );
319 $c->add_path( 'n21p0', 'n21', 'A' );
320 $c->add_relationship( 'n21', 'n22', { type => 'collated', scope => 'local' } );
322 ok( $c->reading( 'n21p0' ), "New reading exists" );
323 is( scalar $c->readings, $rno, "Reading add offset by flatten_ranks" );
325 # Combine n3 and n4 ( with his )
326 $c->merge_readings( 'n3', 'n4', 1 );
327 ok( !$c->reading('n4'), "Reading n4 is gone" );
328 is( $c->reading('n3')->text, 'with his', "Reading n3 has both words" );
330 # Collapse n9 and n10 ( rood / root )
331 $c->merge_readings( 'n9', 'n10' );
332 ok( !$c->reading('n10'), "Reading n10 is gone" );
333 is( $c->reading('n9')->text, 'rood', "Reading n9 has an unchanged word" );
335 # Try to combine n21 and n21p0. This should break.
336 my $remaining = $c->reading('n21');
337 $remaining ||= $c->reading('n22'); # one of these should still exist
339 $c->merge_readings( 'n21p0', $remaining, 1 );
340 ok( 0, "Bad reading merge changed the graph" );
341 } catch( Text::Tradition::Error $e ) {
342 like( $e->message, qr/neither concatenated nor collated/, "Expected exception from bad concatenation" );
344 ok( 0, "Unexpected error on bad reading merge: $@" );
348 $c->calculate_ranks();
349 ok( 1, "Graph is still evidently whole" );
350 } catch( Text::Tradition::Error $e ) {
351 ok( 0, "Caught a rank exception: " . $e->message );
362 my( $kept_obj, $del_obj, $combine, $combine_char ) = $self->_objectify_args( @_ );
363 my $mergemeta = $kept_obj->is_meta;
364 throw( "Cannot merge meta and non-meta reading" )
365 unless ( $mergemeta && $del_obj->is_meta )
366 || ( !$mergemeta && !$del_obj->is_meta );
368 throw( "Cannot merge with start or end node" )
369 if( $kept_obj eq $self->start || $kept_obj eq $self->end
370 || $del_obj eq $self->start || $del_obj eq $self->end );
371 throw( "Cannot combine text of meta readings" ) if $combine;
373 # We can only merge readings in a linear graph if:
374 # - they are contiguous with only one edge between them, OR
375 # - they are at equivalent ranks in the graph.
376 if( $self->linear ) {
377 my @delpred = $del_obj->predecessors;
378 my @keptsuc = $kept_obj->successors;
379 unless ( @delpred == 1 && $delpred[0] eq $kept_obj
380 && @keptsuc == 1 && $keptsuc[0] eq $del_obj ) {
381 my( $is_ok, $msg ) = $self->relations->relationship_valid(
382 $kept_obj, $del_obj, 'collated' );
384 throw( "Readings $kept_obj and $del_obj can be neither concatenated nor collated" );
389 # We only need the IDs for adding paths to the graph, not the reading
390 # objects themselves.
391 my $kept = $kept_obj->id;
392 my $deleted = $del_obj->id;
393 $self->_graphcalc_done(0);
395 # The kept reading should inherit the paths and the relationships
396 # of the deleted reading.
397 foreach my $path ( $self->sequence->edges_at( $deleted ) ) {
398 my @vector = ( $kept );
399 push( @vector, $path->[1] ) if $path->[0] eq $deleted;
400 unshift( @vector, $path->[0] ) if $path->[1] eq $deleted;
401 next if $vector[0] eq $vector[1]; # Don't add a self loop
402 my %wits = %{$self->sequence->get_edge_attributes( @$path )};
403 $self->sequence->add_edge( @vector );
404 my $fwits = $self->sequence->get_edge_attributes( @vector );
405 @wits{keys %$fwits} = values %$fwits;
406 $self->sequence->set_edge_attributes( @vector, \%wits );
408 $self->relations->merge_readings( $kept, $deleted, $combine );
410 # Do the deletion deed.
412 # Combine the text of the readings
413 my $joinstr = $combine_char;
414 unless( defined $joinstr ) {
415 $joinstr = '' if $kept_obj->join_next || $del_obj->join_prior;
416 $joinstr = $self->wordsep unless defined $joinstr;
418 $kept_obj->_combine( $del_obj, $joinstr );
420 $self->del_reading( $deleted );
423 =head2 merge_related( @relationship_types )
425 Merge all readings linked with the relationship types given. If any of the selected type(s) is not a colocation, the graph will no longer be linear. The majority/plurality reading in each case will be the one kept.
427 WARNING: This operation cannot be undone.
439 $t = Text::Tradition->new( 'input' => 'Self', 'file' => 't/data/legendfrag.xml' );
440 } [qr/Cannot set relationship on a meta reading/],
441 "Got expected relationship drop warning on parse";
443 my $c = $t->collation;
446 map { $rdg_ids{$_} = 1 } $c->readings;
447 $c->merge_related( 'orthographic' );
448 is( scalar( $c->readings ), keys( %rdg_ids ) - 8,
449 "Successfully collapsed orthographic variation" );
450 map { $rdg_ids{$_} = undef } qw/ r13.3 r11.4 r8.5 r8.2 r7.7 r7.5 r7.4 r7.1 /;
451 foreach my $rid ( keys %rdg_ids ) {
452 my $exp = $rdg_ids{$rid};
453 is( !$c->reading( $rid ), !$exp, "Reading $rid correctly " .
454 ( $exp ? "retained" : "removed" ) );
456 ok( $c->linear, "Graph is still linear" );
458 $c->calculate_ranks; # This should succeed
459 ok( 1, "Can still calculate ranks on the new graph" );
461 ok( 0, "Rank calculation on merged graph failed: $@" );
464 # Now add some transpositions
465 $c->add_relationship( 'r8.4', 'r10.4', { type => 'transposition' } );
466 $c->merge_related( 'transposition' );
467 is( scalar( $c->readings ), keys( %rdg_ids ) - 9,
468 "Transposed relationship is merged away" );
469 ok( !$c->reading('r8.4'), "Correct transposed reading removed" );
470 ok( !$c->linear, "Graph is no longer linear" );
472 $c->calculate_ranks; # This should fail
473 ok( 0, "Rank calculation happened on nonlinear graph?!" );
474 } catch ( Text::Tradition::Error $e ) {
475 is( $e->message, 'Cannot calculate ranks on a non-linear graph',
476 "Rank calculation on merged graph threw an error" );
483 # TODO: there should be a way to display merged without affecting the underlying data!
488 map { $reltypehash{$_} = 1 } @_;
490 # Set up the filter for finding related readings
492 exists $reltypehash{$_[0]->type};
495 # Go through all readings looking for related ones
496 foreach my $r ( $self->readings ) {
497 next unless $self->reading( "$r" ); # might have been deleted meanwhile
498 my @related = $self->related_readings( $r, $filter );
500 push( @related, $r );
502 scalar $b->witnesses <=> scalar $a->witnesses
504 my $keep = shift @related;
505 foreach my $delr ( @related ) {
507 unless( $self->get_relationship( $keep, $delr )->colocated );
508 $self->merge_readings( $keep, $delr );
514 =head2 compress_readings
516 Where possible in the graph, compresses plain sequences of readings into a
517 single reading. The sequences must consist of readings with no
518 relationships to other readings, with only a single witness path between
519 them and no other witness paths from either that would skip the other. The
520 readings must also not be marked as nonsense or bad grammar.
522 WARNING: This operation cannot be undone.
526 sub compress_readings {
528 # Anywhere in the graph that there is a reading that joins only to a single
529 # successor, and neither of these have any relationships, just join the two
531 foreach my $rdg ( sort { $a->rank <=> $b->rank } $self->readings ) {
532 # Now look for readings that can be joined to their successors.
533 next unless $rdg->is_combinable;
535 while( $self->sequence->successors( $rdg ) == 1 ) {
536 my( $next ) = $self->reading( $self->sequence->successors( $rdg ) );
537 throw( "Infinite loop" ) if $seen{$next->id};
538 $seen{$next->id} = 1;
539 last if $self->sequence->predecessors( $next ) > 1;
540 last unless $next->is_combinable;
541 say "Joining readings $rdg and $next";
542 $self->merge_readings( $rdg, $next, 1 );
545 # Make sure we haven't screwed anything up
546 foreach my $wit ( $self->tradition->witnesses ) {
547 my $pathtext = $self->path_text( $wit->sigil );
548 my $origtext = join( ' ', @{$wit->text} );
549 throw( "Text differs for witness " . $wit->sigil )
550 unless $pathtext eq $origtext;
551 if( $wit->is_layered ) {
552 $pathtext = $self->path_text( $wit->sigil.$self->ac_label );
553 $origtext = join( ' ', @{$wit->layertext} );
554 throw( "Ante-corr text differs for witness " . $wit->sigil )
555 unless $pathtext eq $origtext;
559 $self->relations->rebuild_equivalence();
560 $self->calculate_ranks();
563 # Helper function for manipulating the graph.
564 sub _stringify_args {
565 my( $self, $first, $second, @args ) = @_;
567 if ref( $first ) eq 'Text::Tradition::Collation::Reading';
568 $second = $second->id
569 if ref( $second ) eq 'Text::Tradition::Collation::Reading';
570 return( $first, $second, @args );
573 # Helper function for manipulating the graph.
574 sub _objectify_args {
575 my( $self, $first, $second, $arg ) = @_;
576 $first = $self->reading( $first )
577 unless ref( $first ) eq 'Text::Tradition::Collation::Reading';
578 $second = $self->reading( $second )
579 unless ref( $second ) eq 'Text::Tradition::Collation::Reading';
580 return( $first, $second, $arg );
583 =head2 duplicate_reading( $reading, @witlist )
585 Split the given reading into two, so that the new reading is in the path for
586 the witnesses given in @witlist. If the result is that certain non-colocated
587 relationships (e.g. transpositions) are no longer valid, these will be removed.
588 Returns the newly-created reading.
592 use Test::More::UTF8;
596 my $st = Text::Tradition->new( 'input' => 'Self', 'file' => 't/data/collatecorr.xml' );
597 is( ref( $st ), 'Text::Tradition', "Got a tradition from test file" );
598 ok( $st->has_witness('Ba96'), "Tradition has the affected witness" );
600 my $sc = $st->collation;
602 ok( $sc->reading('n131'), "Tradition has the affected reading" );
603 is( scalar( $sc->readings ), $numr, "There are $numr readings in the graph" );
604 is( $sc->end->rank, 14, "There are fourteen ranks in the graph" );
606 # Detach the erroneously collated reading
607 my( $newr, @del_rdgs ) = $sc->duplicate_reading( 'n131', 'Ba96' );
608 ok( $newr, "New reading was created" );
609 ok( $sc->reading('n131_0'), "Detached the bad collation with a new reading" );
610 is( scalar( $sc->readings ), $numr + 1, "A reading was added to the graph" );
611 is( $sc->end->rank, 10, "There are now only ten ranks in the graph" );
612 my $csucc = $sc->common_successor( 'n131', 'n131_0' );
613 is( $csucc->id, 'n136', "Found correct common successor to duped reading" );
615 # Check that the bad transposition is gone
616 is( scalar @del_rdgs, 1, "Deleted reading was returned by API call" );
617 is( $sc->get_relationship( 'n130', 'n135' ), undef, "Bad transposition relationship is gone" );
619 # The collation should not be fixed
620 my @pairs = $sc->identical_readings();
621 is( scalar @pairs, 0, "Not re-collated yet" );
623 ok( $sc->merge_readings( 'n124', 'n131_0' ), "Collated the readings correctly" );
624 @pairs = $sc->identical_readings( start => 'n124', end => $csucc->id );
625 is( scalar @pairs, 3, "Found three more identical readings" );
626 is( $sc->end->rank, 11, "The ranks shifted appropriately" );
627 $sc->flatten_ranks();
628 is( scalar( $sc->readings ), $numr - 3, "Now we are collated correctly" );
630 # Check that we can't "duplicate" a reading with no wits or with all wits
632 my( $badr, @del_rdgs ) = $sc->duplicate_reading( 'n124' );
633 ok( 0, "Reading duplication without witnesses throws an error" );
634 } catch( Text::Tradition::Error $e ) {
635 like( $e->message, qr/Must specify one or more witnesses/,
636 "Reading duplication without witnesses throws the expected error" );
638 ok( 0, "Reading duplication without witnesses threw the wrong error" );
642 my( $badr, @del_rdgs ) = $sc->duplicate_reading( 'n124', 'Ba96', 'Mü11475' );
643 ok( 0, "Reading duplication with all witnesses throws an error" );
644 } catch( Text::Tradition::Error $e ) {
645 like( $e->message, qr/Cannot join all witnesses/,
646 "Reading duplication with all witnesses throws the expected error" );
648 ok( 0, "Reading duplication with all witnesses threw the wrong error" );
652 $sc->calculate_ranks();
653 ok( 1, "Graph is still evidently whole" );
654 } catch( Text::Tradition::Error $e ) {
655 ok( 0, "Caught a rank exception: " . $e->message );
662 sub duplicate_reading {
663 my( $self, $r, @wits ) = @_;
664 # Check that we are not doing anything unwise.
665 throw( "Must specify one or more witnesses for the duplicated reading" )
667 unless( ref( $r ) eq 'Text::Tradition::Collation::Reading' ) {
668 $r = $self->reading( $r );
670 throw( "Cannot duplicate a meta-reading" )
672 throw( "Cannot join all witnesses to the new reading" )
673 if scalar( @wits ) == scalar( $r->witnesses );
675 # Get all the reading attributes and duplicate them.
676 my $rmeta = Text::Tradition::Collation::Reading->meta;
678 foreach my $attr( $rmeta->get_all_attributes ) {
679 next if $attr->name =~ /^_/;
680 my $acc = $attr->get_read_method;
681 if( !$acc && $attr->has_applied_traits ) {
682 my $tr = $attr->applied_traits;
683 if( $tr->[0] =~ /::(Array|Hash)$/ ) {
685 my %methods = reverse %{$attr->handles};
686 $acc = $methods{elements};
687 $args{$attr->name} = $which eq 'Array'
688 ? [ $r->$acc ] : { $r->$acc };
691 $args{$attr->name} = $r->$acc if $acc;
694 # By definition the new reading will no longer be common.
695 $args{is_common} = 0;
696 # The new reading also needs its own ID.
697 $args{id} = $self->_generate_dup_id( $r->id );
699 # Try to make the new reading.
700 my $newr = $self->add_reading( \%args );
701 # The old reading is also no longer common.
704 # For each of the witnesses, dissociate from the old reading and
705 # associate with the new.
706 foreach my $wit ( @wits ) {
707 my $prior = $self->prior_reading( $r, $wit );
708 my $next = $self->next_reading( $r, $wit );
709 $self->del_path( $prior, $r, $wit );
710 $self->add_path( $prior, $newr, $wit );
711 $self->del_path( $r, $next, $wit );
712 $self->add_path( $newr, $next, $wit );
715 # If the graph is ranked, we need to look for relationships that are now
716 # invalid (i.e. 'non-colocation' types that might now be colocated) and
717 # remove them. If not, we can skip it.
720 my @deleted_relations;
721 if( $self->end->has_rank ) {
722 # Find the point where we can stop checking
723 $succ = $self->common_successor( $r, $newr );
725 # Hash the existing ranks
726 foreach my $rdg ( $self->readings ) {
727 $rrk{$rdg->id} = $rdg->rank;
729 # Calculate the new ranks
730 $self->calculate_ranks();
732 # Check for invalid non-colocated relationships among changed-rank readings
733 # from where the ranks start changing up to $succ
734 my $lastrank = $succ->rank;
735 foreach my $rdg ( $self->readings ) {
736 next if $rdg->rank > $lastrank;
737 next if $rdg->rank == $rrk{$rdg->id};
738 my @noncolo = $rdg->related_readings( sub { !$_[0]->colocated } );
739 next unless @noncolo;
740 foreach my $nc ( @noncolo ) {
741 unless( $self->relations->verify_or_delete( $rdg, $nc ) ) {
742 push( @deleted_relations, [ $rdg->id, $nc->id ] );
747 return ( $newr, @deleted_relations );
750 sub _generate_dup_id {
751 my( $self, $rid ) = @_;
756 if( $self->has_reading( $newid ) ) {
769 # We only need the IDs for adding paths to the graph, not the reading
770 # objects themselves.
771 my( $source, $target, $wit ) = $self->_stringify_args( @_ );
773 $self->_graphcalc_done(0);
774 # Connect the readings
775 unless( $self->sequence->has_edge( $source, $target ) ) {
776 $self->sequence->add_edge( $source, $target );
777 $self->relations->add_equivalence_edge( $source, $target );
779 # Note the witness in question
780 $self->sequence->set_edge_attribute( $source, $target, $wit, 1 );
786 if( ref( $_[0] ) eq 'ARRAY' ) {
793 # We only need the IDs for removing paths from the graph, not the reading
794 # objects themselves.
795 my( $source, $target, $wit ) = $self->_stringify_args( @args );
797 $self->_graphcalc_done(0);
798 if( $self->sequence->has_edge_attribute( $source, $target, $wit ) ) {
799 $self->sequence->delete_edge_attribute( $source, $target, $wit );
801 unless( $self->sequence->has_edge_attributes( $source, $target ) ) {
802 $self->sequence->delete_edge( $source, $target );
803 $self->relations->delete_equivalence_edge( $source, $target );
808 # Extra graph-alike utility
811 my( $source, $target, $wit ) = $self->_stringify_args( @_ );
812 return undef unless $self->sequence->has_edge( $source, $target );
813 return $self->sequence->has_edge_attribute( $source, $target, $wit );
816 =head2 clear_witness( @sigil_list )
818 Clear the given witnesses out of the collation entirely, removing references
819 to them in paths, and removing readings that belong only to them. Should only
820 be called via $tradition->del_witness.
825 my( $self, @sigils ) = @_;
827 $self->_graphcalc_done(0);
828 # Clear the witness(es) out of the paths
829 foreach my $e ( $self->paths ) {
830 foreach my $sig ( @sigils ) {
831 $self->del_path( $e, $sig );
835 # Clear out the newly unused readings
836 foreach my $r ( $self->readings ) {
837 unless( $self->reading_witnesses( $r ) ) {
838 $self->del_reading( $r );
843 sub add_relationship {
845 my( $source, $target, $opts ) = $self->_stringify_args( @_ );
846 my( @vectors ) = $self->relations->add_relationship( $source, $target, $opts );
847 foreach my $v ( @vectors ) {
848 next unless $self->get_relationship( $v )->colocated;
849 if( $self->reading( $v->[0] )->has_rank && $self->reading( $v->[1] )->has_rank
850 && $self->reading( $v->[0] )->rank ne $self->reading( $v->[1] )->rank ) {
851 $self->_graphcalc_done(0);
859 around qw/ get_relationship del_relationship / => sub {
863 if( @args == 1 && ref( $args[0] ) eq 'ARRAY' ) {
866 my @stringargs = $self->_stringify_args( @args );
867 $self->$orig( @stringargs );
870 =head2 reading_witnesses( $reading )
872 Return a list of sigils corresponding to the witnesses in which the reading appears.
876 sub reading_witnesses {
877 my( $self, $reading ) = @_;
878 # We need only check either the incoming or the outgoing edges; I have
879 # arbitrarily chosen "incoming". Thus, special-case the start node.
880 if( $reading eq $self->start ) {
881 return map { $_->sigil } grep { $_->is_collated } $self->tradition->witnesses;
884 foreach my $e ( $self->sequence->edges_to( $reading ) ) {
885 my $wits = $self->sequence->get_edge_attributes( @$e );
886 @all_witnesses{ keys %$wits } = 1;
888 my $acstr = $self->ac_label;
889 foreach my $acwit ( grep { $_ =~ s/^(.*)\Q$acstr\E$/$1/ } keys %all_witnesses ) {
890 delete $all_witnesses{$acwit.$acstr} if exists $all_witnesses{$acwit};
892 return keys %all_witnesses;
895 =head1 OUTPUT METHODS
897 =head2 as_svg( \%options )
899 Returns an SVG string that represents the graph, via as_dot and graphviz.
900 See as_dot for a list of options. Must have GraphViz (dot) installed to run.
905 my( $self, $opts ) = @_;
906 throw( "Need GraphViz installed to output SVG" )
907 unless File::Which::which( 'dot' );
908 my $want_subgraph = exists $opts->{'from'} || exists $opts->{'to'};
909 $self->calculate_ranks()
910 unless( $self->_graphcalc_done || $opts->{'nocalc'} || !$self->linear );
911 my @cmd = qw/dot -Tsvg/;
913 my $dotfile = File::Temp->new();
915 # $dotfile->unlink_on_destroy(0);
916 binmode $dotfile, ':utf8';
917 print $dotfile $self->as_dot( $opts );
918 push( @cmd, $dotfile->filename );
919 run( \@cmd, ">", binary(), \$svg );
920 $svg = decode_utf8( $svg );
925 =head2 as_dot( \%options )
927 Returns a string that is the collation graph expressed in dot
928 (i.e. GraphViz) format. Options include:
943 my( $self, $opts ) = @_;
944 my $startrank = $opts->{'from'} if $opts;
945 my $endrank = $opts->{'to'} if $opts;
946 my $color_common = $opts->{'color_common'} if $opts;
947 my $STRAIGHTENHACK = !$startrank && !$endrank && $self->end->rank
948 && $self->end->rank > 100;
949 $STRAIGHTENHACK = 1 if $opts->{'straight'}; # even for subgraphs or small graphs
951 # Check the arguments
953 return if $endrank && $startrank > $endrank;
954 return if $startrank > $self->end->rank;
956 if( defined $endrank ) {
957 return if $endrank < 0;
958 $endrank = undef if $endrank == $self->end->rank;
961 my $graph_name = $self->tradition->name;
962 $graph_name =~ s/[^\w\s]//g;
963 $graph_name = join( '_', split( /\s+/, $graph_name ) );
971 'fillcolor' => 'white',
976 'arrowhead' => 'open',
977 'color' => '#000000',
978 'fontcolor' => '#000000',
981 my $dot = sprintf( "digraph %s {\n", $graph_name );
982 $dot .= "\tgraph " . _dot_attr_string( \%graph_attrs ) . ";\n";
983 $dot .= "\tnode " . _dot_attr_string( \%node_attrs ) . ";\n";
985 # Output substitute start/end readings if necessary
987 $dot .= "\t\"__SUBSTART__\" [ label=\"...\",id=\"__START__\" ];\n";
990 $dot .= "\t\"__SUBEND__\" [ label=\"...\",id=\"__END__\" ];\n";
992 if( $STRAIGHTENHACK ) {
994 my $startlabel = $startrank ? '__SUBSTART__' : '__START__';
995 $dot .= "\tsubgraph { rank=same \"$startlabel\" \"#SILENT#\" }\n";
996 $dot .= "\t\"#SILENT#\" [ shape=diamond,color=white,penwidth=0,label=\"\" ];"
998 my %used; # Keep track of the readings that actually appear in the graph
999 # Sort the readings by rank if we have ranks; this speeds layout.
1000 my @all_readings = $self->end->has_rank
1001 ? sort { $a->rank <=> $b->rank } $self->readings
1003 # TODO Refrain from outputting lacuna nodes - just grey out the edges.
1004 foreach my $reading ( @all_readings ) {
1005 # Only output readings within our rank range.
1006 next if $startrank && $reading->rank < $startrank;
1007 next if $endrank && $reading->rank > $endrank;
1008 $used{$reading->id} = 1;
1009 # Need not output nodes without separate labels
1010 next if $reading->id eq $reading->text;
1012 my $label = $reading->text;
1013 $label .= '-' if $reading->join_next;
1014 $label = "-$label" if $reading->join_prior;
1015 $label =~ s/\"/\\\"/g;
1016 $rattrs->{'label'} = $label;
1017 $rattrs->{'id'} = $reading->id;
1018 $rattrs->{'fillcolor'} = '#b3f36d' if $reading->is_common && $color_common;
1019 $dot .= sprintf( "\t\"%s\" %s;\n", $reading->id, _dot_attr_string( $rattrs ) );
1022 # Add the real edges. Need to weight one edge per rank jump, in a
1024 # my $weighted = $self->_add_edge_weights;
1025 my @edges = $self->paths;
1026 my( %substart, %subend );
1027 foreach my $edge ( @edges ) {
1028 # Do we need to output this edge?
1029 if( $used{$edge->[0]} && $used{$edge->[1]} ) {
1030 my $label = $self->_path_display_label( $opts,
1031 $self->path_witnesses( $edge ) );
1032 my $variables = { %edge_attrs, 'label' => $label };
1034 # Account for the rank gap if necessary
1035 my $rank0 = $self->reading( $edge->[0] )->rank
1036 if $self->reading( $edge->[0] )->has_rank;
1037 my $rank1 = $self->reading( $edge->[1] )->rank
1038 if $self->reading( $edge->[1] )->has_rank;
1039 if( defined $rank0 && defined $rank1 && $rank1 - $rank0 > 1 ) {
1040 $variables->{'minlen'} = $rank1 - $rank0;
1043 # Add the calculated edge weights
1044 # if( exists $weighted->{$edge->[0]}
1045 # && $weighted->{$edge->[0]} eq $edge->[1] ) {
1046 # # $variables->{'color'} = 'red';
1047 # $variables->{'weight'} = 3.0;
1050 # EXPERIMENTAL: make edge width reflect no. of witnesses
1051 my $extrawidth = scalar( $self->path_witnesses( $edge ) ) * 0.2;
1052 $variables->{'penwidth'} = $extrawidth + 0.8; # gives 1 for a single wit
1054 my $varopts = _dot_attr_string( $variables );
1055 $dot .= sprintf( "\t\"%s\" -> \"%s\" %s;\n",
1056 $edge->[0], $edge->[1], $varopts );
1057 } elsif( $used{$edge->[0]} ) {
1058 $subend{$edge->[0]} = $edge->[1];
1059 } elsif( $used{$edge->[1]} ) {
1060 $substart{$edge->[1]} = $edge->[0];
1064 # If we are asked to, add relationship links
1065 if( exists $opts->{show_relations} ) {
1066 my $filter = $opts->{show_relations}; # can be 'transposition' or 'all'
1067 if( $filter eq 'transposition' ) {
1068 $filter =~ qr/^transposition$/;
1071 my @types = sort( map { $_->name } $self->relations->types );
1072 if( exists $opts->{graphcolors} ) {
1073 foreach my $tdx ( 0 .. $#types ) {
1074 $typecolors{$types[$tdx]} = $opts->{graphcolors}->[$tdx];
1077 map { $typecolors{$_} = '#FFA14F' } @types;
1079 foreach my $redge ( $self->relationships ) {
1080 if( $used{$redge->[0]} && $used{$redge->[1]} ) {
1081 my $rel = $self->get_relationship( $redge );
1082 next unless $filter eq 'all' || $rel->type =~ /$filter/;
1084 arrowhead => 'none',
1085 color => $typecolors{$rel->type},
1086 constraint => 'false',
1089 unless( exists $opts->{graphcolors} ) {
1090 $variables->{label} = uc( substr( $rel->type, 0, 4 ) ),
1092 $dot .= sprintf( "\t\"%s\" -> \"%s\" %s;\n",
1093 $redge->[0], $redge->[1], _dot_attr_string( $variables ) );
1098 # Add substitute start and end edges if necessary
1099 foreach my $node ( keys %substart ) {
1100 my $witstr = $self->_path_display_label( $opts,
1101 $self->path_witnesses( $substart{$node}, $node ) );
1102 my $variables = { %edge_attrs, 'label' => $witstr };
1103 my $nrdg = $self->reading( $node );
1104 if( $nrdg->has_rank && $nrdg->rank > $startrank ) {
1105 # Substart is actually one lower than $startrank
1106 $variables->{'minlen'} = $nrdg->rank - ( $startrank - 1 );
1108 my $varopts = _dot_attr_string( $variables );
1109 $dot .= "\t\"__SUBSTART__\" -> \"$node\" $varopts;\n";
1111 foreach my $node ( keys %subend ) {
1112 my $witstr = $self->_path_display_label( $opts,
1113 $self->path_witnesses( $node, $subend{$node} ) );
1114 my $variables = { %edge_attrs, 'label' => $witstr };
1115 my $varopts = _dot_attr_string( $variables );
1116 $dot .= "\t\"$node\" -> \"__SUBEND__\" $varopts;\n";
1119 if( $STRAIGHTENHACK ) {
1120 my $endlabel = $endrank ? '__SUBEND__' : '__END__';
1121 $dot .= "\t\"$endlabel\" -> \"#SILENT#\" [ color=white,penwidth=0 ];\n";
1128 sub _dot_attr_string {
1131 foreach my $k ( sort keys %$hash ) {
1132 my $v = $hash->{$k};
1133 push( @attrs, $k.'="'.$v.'"' );
1135 return( '[ ' . join( ', ', @attrs ) . ' ]' );
1138 sub _add_edge_weights {
1140 # Walk the graph from START to END, choosing the successor node with
1141 # the largest number of witness paths each time.
1143 my $curr = $self->start->id;
1144 my $ranked = $self->end->has_rank;
1145 while( $curr ne $self->end->id ) {
1146 my $rank = $ranked ? $self->reading( $curr )->rank : 0;
1147 my @succ = sort { $self->path_witnesses( $curr, $a )
1148 <=> $self->path_witnesses( $curr, $b ) }
1149 $self->sequence->successors( $curr );
1150 my $next = pop @succ;
1151 my $nextrank = $ranked ? $self->reading( $next )->rank : 0;
1152 # Try to avoid lacunae in the weighted path.
1154 ( $self->reading( $next )->is_lacuna ||
1155 $nextrank - $rank > 1 ) ){
1158 $weighted->{$curr} = $next;
1164 =head2 path_witnesses( $edge )
1166 Returns the list of sigils whose witnesses are associated with the given edge.
1167 The edge can be passed as either an array or an arrayref of ( $source, $target ).
1171 sub path_witnesses {
1172 my( $self, @edge ) = @_;
1173 # If edge is an arrayref, cope.
1174 if( @edge == 1 && ref( $edge[0] ) eq 'ARRAY' ) {
1175 my $e = shift @edge;
1178 my @wits = keys %{$self->sequence->get_edge_attributes( @edge )};
1182 # Helper function. Make a display label for the given witnesses, showing a.c.
1183 # witnesses only where the main witness is not also in the list.
1184 sub _path_display_label {
1188 map { $wits{$_} = 1 } @_;
1190 # If an a.c. wit is listed, remove it if the main wit is also listed.
1191 # Otherwise keep it for explicit listing.
1192 my $aclabel = $self->ac_label;
1194 foreach my $w ( sort keys %wits ) {
1195 if( $w =~ /^(.*)\Q$aclabel\E$/ ) {
1196 if( exists $wits{$1} ) {
1199 push( @disp_ac, $w );
1204 if( $opts->{'explicit_wits'} ) {
1205 return join( ', ', sort keys %wits );
1207 # See if we are in a majority situation.
1208 my $maj = scalar( $self->tradition->witnesses ) * 0.6;
1209 $maj = $maj > 5 ? $maj : 5;
1210 if( scalar keys %wits > $maj ) {
1211 unshift( @disp_ac, 'majority' );
1212 return join( ', ', @disp_ac );
1214 return join( ', ', sort keys %wits );
1219 =head2 readings_at_rank( $rank )
1221 Returns a list of readings at a given rank, taken from the alignment table.
1225 sub readings_at_rank {
1226 my( $self, $rank ) = @_;
1227 my $table = $self->alignment_table;
1228 # Table rank is real rank - 1.
1229 my @elements = map { $_->{'tokens'}->[$rank-1] } @{$table->{'alignment'}};
1231 foreach my $e ( @elements ) {
1232 next unless ref( $e ) eq 'HASH';
1233 next unless exists $e->{'t'};
1234 $readings{$e->{'t'}->id} = $e->{'t'};
1236 return values %readings;
1241 Returns a GraphML representation of the collation. The GraphML will contain
1242 two graphs. The first expresses the attributes of the readings and the witness
1243 paths that link them; the second expresses the relationships that link the
1244 readings. This is the native transfer format for a tradition.
1248 use Text::Tradition;
1254 my $datafile = 't/data/florilegium_tei_ps.xml';
1255 my $tradition = Text::Tradition->new( 'input' => 'TEI',
1257 'file' => $datafile,
1260 ok( $tradition, "Got a tradition object" );
1261 is( scalar $tradition->witnesses, 13, "Found all witnesses" );
1262 ok( $tradition->collation, "Tradition has a collation" );
1264 my $c = $tradition->collation;
1265 is( scalar $c->readings, $READINGS, "Collation has all readings" );
1266 is( scalar $c->paths, $PATHS, "Collation has all paths" );
1267 is( scalar $c->relationships, 0, "Collation has all relationships" );
1269 # Add a few relationships
1270 $c->add_relationship( 'w123', 'w125', { 'type' => 'collated' } );
1271 $c->add_relationship( 'w193', 'w196', { 'type' => 'collated' } );
1272 $c->add_relationship( 'w257', 'w262', { 'type' => 'transposition' } );
1274 # Now write it to GraphML and parse it again.
1276 my $graphml = $c->as_graphml;
1277 my $st = Text::Tradition->new( 'input' => 'Self', 'string' => $graphml );
1278 is( scalar $st->collation->readings, $READINGS, "Reparsed collation has all readings" );
1279 is( scalar $st->collation->paths, $PATHS, "Reparsed collation has all paths" );
1280 is( scalar $st->collation->relationships, 3, "Reparsed collation has new relationships" );
1282 # Now add a stemma, write to GraphML, and look at the output.
1284 skip "Analysis module not present", 3 unless $tradition->can( 'add_stemma' );
1285 my $stemma = $tradition->add_stemma( 'dotfile' => 't/data/florilegium.dot' );
1286 is( ref( $stemma ), 'Text::Tradition::Stemma', "Parsed dotfile into stemma" );
1287 is( $tradition->stemmata, 1, "Tradition now has the stemma" );
1288 $graphml = $c->as_graphml;
1289 like( $graphml, qr/digraph/, "Digraph declaration exists in GraphML" );
1296 ## TODO MOVE this to Tradition.pm and modularize it better
1298 my( $self, $options ) = @_;
1299 $self->calculate_ranks unless $self->_graphcalc_done;
1301 my $start = $options->{'from'}
1302 ? $self->reading( $options->{'from'} ) : $self->start;
1303 my $end = $options->{'to'}
1304 ? $self->reading( $options->{'to'} ) : $self->end;
1305 if( $start->has_rank && $end->has_rank && $end->rank < $start->rank ) {
1306 throw( 'Start node must be before end node' );
1308 # The readings need to be ranked for this to work.
1309 $start = $self->start unless $start->has_rank;
1310 $end = $self->end unless $end->has_rank;
1312 unless( $start eq $self->start ) {
1313 $rankoffset = $start->rank - 1;
1318 my $graphml_ns = 'http://graphml.graphdrawing.org/xmlns';
1319 my $xsi_ns = 'http://www.w3.org/2001/XMLSchema-instance';
1320 my $graphml_schema = 'http://graphml.graphdrawing.org/xmlns ' .
1321 'http://graphml.graphdrawing.org/xmlns/1.0/graphml.xsd';
1323 # Create the document and root node
1324 require XML::LibXML;
1325 my $graphml = XML::LibXML->createDocument( "1.0", "UTF-8" );
1326 my $root = $graphml->createElementNS( $graphml_ns, 'graphml' );
1327 $graphml->setDocumentElement( $root );
1328 $root->setNamespace( $xsi_ns, 'xsi', 0 );
1329 $root->setAttributeNS( $xsi_ns, 'schemaLocation', $graphml_schema );
1331 # List of attribute types to save on our objects and their corresponding
1336 'Bool' => 'boolean',
1337 'ReadingID' => 'string',
1338 'RelationshipType' => 'string',
1339 'RelationshipScope' => 'string',
1342 # Add the data keys for the graph. Include an extra key 'version' for the
1343 # GraphML output version.
1344 my %graph_data_keys;
1346 my %graph_attributes = ( 'version' => 'string' );
1347 # Graph attributes include those of Tradition and those of Collation.
1349 # TODO Use meta introspection method from duplicate_reading to do this
1350 # instead of naming custom keys.
1351 my $tmeta = $self->tradition->meta;
1352 my $cmeta = $self->meta;
1353 map { $gattr_from{$_->name} = 'Tradition' } $tmeta->get_all_attributes;
1354 map { $gattr_from{$_->name} = 'Collation' } $cmeta->get_all_attributes;
1355 foreach my $attr ( ( $tmeta->get_all_attributes, $cmeta->get_all_attributes ) ) {
1356 next if $attr->name =~ /^_/;
1357 next unless $save_types{$attr->type_constraint->name};
1358 $graph_attributes{$attr->name} = $save_types{$attr->type_constraint->name};
1360 # Extra custom keys for complex objects that should be saved in some form.
1361 # The subroutine should return a string, or undef/empty.
1362 if( $tmeta->has_method('stemmata') ) {
1363 $graph_attributes{'stemmata'} = sub {
1365 map { push( @stemstrs, $_->editable( {linesep => ''} ) ) }
1366 $self->tradition->stemmata;
1367 join( "\n", @stemstrs );
1371 if( $tmeta->has_method('user') ) {
1372 $graph_attributes{'user'} = sub {
1373 $self->tradition->user ? $self->tradition->user->id : undef
1377 foreach my $datum ( sort keys %graph_attributes ) {
1378 $graph_data_keys{$datum} = 'dg'.$gdi++;
1379 my $key = $root->addNewChild( $graphml_ns, 'key' );
1380 my $dtype = ref( $graph_attributes{$datum} ) ? 'string'
1381 : $graph_attributes{$datum};
1382 $key->setAttribute( 'attr.name', $datum );
1383 $key->setAttribute( 'attr.type', $dtype );
1384 $key->setAttribute( 'for', 'graph' );
1385 $key->setAttribute( 'id', $graph_data_keys{$datum} );
1388 # Add the data keys for reading nodes
1389 my %reading_attributes;
1390 my $rmeta = Text::Tradition::Collation::Reading->meta;
1391 foreach my $attr( $rmeta->get_all_attributes ) {
1392 next if $attr->name =~ /^_/;
1393 next unless $save_types{$attr->type_constraint->name};
1394 $reading_attributes{$attr->name} = $save_types{$attr->type_constraint->name};
1396 if( $self->start->does('Text::Tradition::Morphology' ) ) {
1397 # Extra custom key for the reading morphology
1398 $reading_attributes{'lexemes'} = 'string';
1403 foreach my $datum ( sort keys %reading_attributes ) {
1404 $node_data_keys{$datum} = 'dn'.$ndi++;
1405 my $key = $root->addNewChild( $graphml_ns, 'key' );
1406 $key->setAttribute( 'attr.name', $datum );
1407 $key->setAttribute( 'attr.type', $reading_attributes{$datum} );
1408 $key->setAttribute( 'for', 'node' );
1409 $key->setAttribute( 'id', $node_data_keys{$datum} );
1412 # Add the data keys for edges, that is, paths and relationships. Path
1413 # data does not come from a Moose class so is here manually.
1416 my %edge_attributes = (
1417 witness => 'string', # ID/label for a path
1418 extra => 'boolean', # Path key
1420 my @path_attributes = keys %edge_attributes; # track our manual additions
1421 my $pmeta = Text::Tradition::Collation::Relationship->meta;
1422 foreach my $attr( $pmeta->get_all_attributes ) {
1423 next if $attr->name =~ /^_/;
1424 next unless $save_types{$attr->type_constraint->name};
1425 $edge_attributes{$attr->name} = $save_types{$attr->type_constraint->name};
1427 foreach my $datum ( sort keys %edge_attributes ) {
1428 $edge_data_keys{$datum} = 'de'.$edi++;
1429 my $key = $root->addNewChild( $graphml_ns, 'key' );
1430 $key->setAttribute( 'attr.name', $datum );
1431 $key->setAttribute( 'attr.type', $edge_attributes{$datum} );
1432 $key->setAttribute( 'for', 'edge' );
1433 $key->setAttribute( 'id', $edge_data_keys{$datum} );
1436 # Add the collation graph itself. First, sanitize the name to a valid XML ID.
1437 my $xmlidname = $self->tradition->name;
1438 $xmlidname =~ s/(?!$xml10_namechar_rx)./_/g;
1439 if( $xmlidname !~ /^$xml10_namestartchar_rx/ ) {
1440 $xmlidname = '_'.$xmlidname;
1442 my $sgraph = $root->addNewChild( $graphml_ns, 'graph' );
1443 $sgraph->setAttribute( 'edgedefault', 'directed' );
1444 $sgraph->setAttribute( 'id', $xmlidname );
1445 $sgraph->setAttribute( 'parse.edgeids', 'canonical' );
1446 $sgraph->setAttribute( 'parse.edges', 0 ); # fill in later
1447 $sgraph->setAttribute( 'parse.nodeids', 'canonical' );
1448 $sgraph->setAttribute( 'parse.nodes', 0 ); # fill in later
1449 $sgraph->setAttribute( 'parse.order', 'nodesfirst' );
1451 # Tradition/collation attribute data
1452 foreach my $datum ( keys %graph_attributes ) {
1454 if( $datum eq 'version' ) {
1456 } elsif( ref( $graph_attributes{$datum} ) ) {
1457 my $sub = $graph_attributes{$datum};
1459 } elsif( $gattr_from{$datum} eq 'Tradition' ) {
1460 $value = $self->tradition->$datum;
1462 $value = $self->$datum;
1464 _add_graphml_data( $sgraph, $graph_data_keys{$datum}, $value );
1469 # Add our readings to the graph
1470 foreach my $n ( sort { $a->id cmp $b->id } $self->readings ) {
1471 next if $n->has_rank && $n ne $self->start && $n ne $self->end &&
1472 ( $n->rank < $start->rank || $n->rank > $end->rank );
1473 $use_readings{$n->id} = 1;
1474 # Add to the main graph
1475 my $node_el = $sgraph->addNewChild( $graphml_ns, 'node' );
1476 my $node_xmlid = 'n' . $node_ctr++;
1477 $node_hash{ $n->id } = $node_xmlid;
1478 $node_el->setAttribute( 'id', $node_xmlid );
1479 foreach my $d ( keys %reading_attributes ) {
1481 # Custom serialization
1482 if( $d eq 'lexemes' ) {
1483 # If nval is a true value, we have lexemes so we need to
1484 # serialize them. Otherwise set nval to undef so that the
1485 # key is excluded from this reading.
1486 $nval = $nval ? $n->_serialize_lexemes : undef;
1487 } elsif( $d eq 'normal_form' && $n->normal_form eq $n->text ) {
1490 if( $rankoffset && $d eq 'rank' && $n ne $self->start ) {
1491 # Adjust the ranks within the subgraph.
1492 $nval = $n eq $self->end ? $end->rank - $rankoffset + 1
1493 : $nval - $rankoffset;
1495 _add_graphml_data( $node_el, $node_data_keys{$d}, $nval )
1500 # Add the path edges to the sequence graph
1502 foreach my $e ( sort { $a->[0] cmp $b->[0] } $self->sequence->edges() ) {
1503 # We add an edge in the graphml for every witness in $e.
1504 next unless( $use_readings{$e->[0]} || $use_readings{$e->[1]} );
1505 my @edge_wits = sort $self->path_witnesses( $e );
1506 $e->[0] = $self->start->id unless $use_readings{$e->[0]};
1507 $e->[1] = $self->end->id unless $use_readings{$e->[1]};
1508 # Skip any path from start to end; that witness is not in the subgraph.
1509 next if ( $e->[0] eq $self->start->id && $e->[1] eq $self->end->id );
1510 foreach my $wit ( @edge_wits ) {
1511 my( $id, $from, $to ) = ( 'e'.$edge_ctr++,
1512 $node_hash{ $e->[0] },
1513 $node_hash{ $e->[1] } );
1514 my $edge_el = $sgraph->addNewChild( $graphml_ns, 'edge' );
1515 $edge_el->setAttribute( 'source', $from );
1516 $edge_el->setAttribute( 'target', $to );
1517 $edge_el->setAttribute( 'id', $id );
1519 # It's a witness path, so add the witness
1521 my $key = $edge_data_keys{'witness'};
1522 # Is this an ante-corr witness?
1523 my $aclabel = $self->ac_label;
1524 if( $wit =~ /^(.*)\Q$aclabel\E$/ ) {
1525 # Keep the base witness
1527 # ...and record that this is an 'extra' reading path
1528 _add_graphml_data( $edge_el, $edge_data_keys{'extra'}, $aclabel );
1530 _add_graphml_data( $edge_el, $edge_data_keys{'witness'}, $base );
1534 # Report the actual number of nodes and edges that went in
1535 $sgraph->setAttribute( 'parse.edges', $edge_ctr );
1536 $sgraph->setAttribute( 'parse.nodes', $node_ctr );
1538 # Add the relationship graph to the XML
1539 map { delete $edge_data_keys{$_} } @path_attributes;
1540 $self->relations->_as_graphml( $graphml_ns, $root, \%node_hash,
1541 $node_data_keys{'id'}, \%edge_data_keys );
1543 # Save and return the thing
1544 my $result = decode_utf8( $graphml->toString(1) );
1548 sub _add_graphml_data {
1549 my( $el, $key, $value ) = @_;
1550 return unless defined $value;
1551 my $data_el = $el->addNewChild( $el->namespaceURI, 'data' );
1552 $data_el->setAttribute( 'key', $key );
1553 $data_el->appendText( $value );
1558 Returns a CSV alignment table representation of the collation graph, one
1559 row per witness (or witness uncorrected.)
1563 Returns a tab-separated alignment table representation of the collation graph,
1564 one row per witness (or witness uncorrected.)
1568 use Text::Tradition;
1576 my $datafile = 't/data/florilegium_tei_ps.xml';
1577 my $tradition = Text::Tradition->new( 'input' => 'TEI',
1579 'file' => $datafile,
1582 my $c = $tradition->collation;
1583 # Export the thing to CSV
1584 my $csvstr = $c->as_csv();
1586 my $csv = Text::CSV->new({ sep_char => ',', binary => 1 });
1587 my @lines = split(/\n/, $csvstr );
1588 ok( $csv->parse( $lines[0] ), "Successfully parsed first line of CSV" );
1589 is( scalar( $csv->fields ), $WITS + $WITAC, "CSV has correct number of witness columns" );
1590 my @q_ac = grep { $_ eq 'Q'.$c->ac_label } $csv->fields;
1591 ok( @q_ac, "Found a layered witness" );
1593 my $t2 = Text::Tradition->new( input => 'Tabular',
1597 is( scalar $t2->collation->readings, $READINGS, "Reparsed CSV collation has all readings" );
1598 is( scalar $t2->collation->paths, $PATHS, "Reparsed CSV collation has all paths" );
1600 # Now do it with TSV
1601 my $tsvstr = $c->as_tsv();
1602 my $t3 = Text::Tradition->new( input => 'Tabular',
1606 is( scalar $t3->collation->readings, $READINGS, "Reparsed TSV collation has all readings" );
1607 is( scalar $t3->collation->paths, $PATHS, "Reparsed TSV collation has all paths" );
1609 my $table = $c->alignment_table;
1610 my $noaccsv = $c->as_csv({ noac => 1 });
1611 my @noaclines = split(/\n/, $noaccsv );
1612 ok( $csv->parse( $noaclines[0] ), "Successfully parsed first line of no-ac CSV" );
1613 is( scalar( $csv->fields ), $WITS, "CSV has correct number of witness columns" );
1614 is( $c->alignment_table, $table, "Request for CSV did not alter the alignment table" );
1616 my $safecsv = $c->as_csv({ safe_ac => 1});
1617 my @safelines = split(/\n/, $safecsv );
1618 ok( $csv->parse( $safelines[0] ), "Successfully parsed first line of safe CSV" );
1619 is( scalar( $csv->fields ), $WITS + $WITAC, "CSV has correct number of witness columns" );
1620 @q_ac = grep { $_ eq 'Q__L' } $csv->fields;
1621 ok( @q_ac, "Found a sanitized layered witness" );
1622 is( $c->alignment_table, $table, "Request for CSV did not alter the alignment table" );
1629 my( $self, $opts ) = @_;
1630 my $table = $self->alignment_table( $opts );
1631 my $csv_options = { binary => 1, quote_null => 0 };
1632 $csv_options->{'sep_char'} = $opts->{fieldsep};
1633 if( $opts->{fieldsep} eq "\t" ) {
1634 # If it is really tab separated, nothing is an escape char.
1635 $csv_options->{'quote_char'} = undef;
1636 $csv_options->{'escape_char'} = '';
1638 my $csv = Text::CSV->new( $csv_options );
1640 # Make the header row
1641 $csv->combine( map { $_->{'witness'} } @{$table->{'alignment'}} );
1642 push( @result, $csv->string );
1643 # Make the rest of the rows
1644 foreach my $idx ( 0 .. $table->{'length'} - 1 ) {
1645 my @rowobjs = map { $_->{'tokens'}->[$idx] } @{$table->{'alignment'}};
1646 my @row = map { $_ ? $_->{'t'}->text : $_ } @rowobjs;
1647 $csv->combine( @row );
1648 push( @result, $csv->string );
1650 return join( "\n", @result );
1655 my $opts = shift || {};
1656 $opts->{fieldsep} = ',';
1657 return $self->_tabular( $opts );
1662 my $opts = shift || {};
1663 $opts->{fieldsep} = "\t";
1664 return $self->_tabular( $opts );
1667 =head2 alignment_table
1669 Return a reference to an alignment table, in a slightly enhanced CollateX
1670 format which looks like this:
1672 $table = { alignment => [ { witness => "SIGIL",
1673 tokens => [ { t => "TEXT" }, ... ] },
1674 { witness => "SIG2",
1675 tokens => [ { t => "TEXT" }, ... ] },
1677 length => TEXTLEN };
1681 sub alignment_table {
1682 my( $self, $opts ) = @_;
1683 if( $self->has_cached_table ) {
1684 return $self->cached_table
1685 unless $opts->{noac} || $opts->{safe_ac};
1688 # Make sure we can do this
1689 throw( "Need a linear graph in order to make an alignment table" )
1690 unless $self->linear;
1691 $self->calculate_ranks()
1692 unless $self->_graphcalc_done && $self->end->has_rank;
1694 my $table = { 'alignment' => [], 'length' => $self->end->rank - 1 };
1695 my @all_pos = ( 1 .. $self->end->rank - 1 );
1696 foreach my $wit ( sort { $a->sigil cmp $b->sigil } $self->tradition->witnesses ) {
1697 # say STDERR "Making witness row(s) for " . $wit->sigil;
1698 my @wit_path = $self->reading_sequence( $self->start, $self->end, $wit->sigil );
1699 my @row = _make_witness_row( \@wit_path, \@all_pos );
1700 my $witobj = { 'witness' => $wit->sigil, 'tokens' => \@row };
1701 $witobj->{'identifier'} = $wit->identifier if $wit->identifier;
1702 push( @{$table->{'alignment'}}, $witobj );
1703 if( $wit->is_layered && !$opts->{noac} ) {
1704 my @wit_ac_path = $self->reading_sequence( $self->start, $self->end,
1705 $wit->sigil.$self->ac_label );
1706 my @ac_row = _make_witness_row( \@wit_ac_path, \@all_pos );
1707 my $witlabel = $opts->{safe_ac}
1708 ? $wit->sigil . '__L' : $wit->sigil.$self->ac_label;
1709 my $witacobj = { 'witness' => $witlabel,
1710 'tokens' => \@ac_row };
1711 $witacobj->{'identifier'} = $wit->identifier if $wit->identifier;
1712 push( @{$table->{'alignment'}}, $witacobj );
1715 unless( $opts->{noac} || $opts->{safe_ac} ) {
1716 $self->cached_table( $table );
1721 sub _make_witness_row {
1722 my( $path, $positions ) = @_;
1724 map { $char_hash{$_} = undef } @$positions;
1726 foreach my $rdg ( @$path ) {
1727 say STDERR "rank " . $rdg->rank if $debug;
1728 # say STDERR "No rank for " . $rdg->id unless defined $rdg->rank;
1729 $char_hash{$rdg->rank} = { 't' => $rdg };
1731 my @row = map { $char_hash{$_} } @$positions;
1732 # Fill in lacuna markers for undef spots in the row
1733 my $last_el = shift @row;
1734 my @filled_row = ( $last_el );
1735 foreach my $el ( @row ) {
1736 # If we are using node reference, make the lacuna node appear many times
1737 # in the table. If not, use the lacuna tag.
1738 if( $last_el && $last_el->{'t'}->is_lacuna && !defined $el ) {
1741 push( @filled_row, $el );
1748 =head1 NAVIGATION METHODS
1750 =head2 reading_sequence( $first, $last, $sigil, $backup )
1752 Returns the ordered list of readings, starting with $first and ending
1753 with $last, for the witness given in $sigil. If a $backup sigil is
1754 specified (e.g. when walking a layered witness), it will be used wherever
1755 no $sigil path exists. If there is a base text reading, that will be
1756 used wherever no path exists for $sigil or $backup.
1760 # TODO Think about returning some lazy-eval iterator.
1761 # TODO Get rid of backup; we should know from what witness is whether we need it.
1763 sub reading_sequence {
1764 my( $self, $start, $end, $witness ) = @_;
1766 $witness = $self->baselabel unless $witness;
1767 my @readings = ( $start );
1770 while( $n && $n->id ne $end->id ) {
1771 if( exists( $seen{$n->id} ) ) {
1772 throw( "Detected loop for $witness at " . $n->id );
1776 my $next = $self->next_reading( $n, $witness );
1778 throw( "Did not find any path for $witness from reading " . $n->id );
1780 push( @readings, $next );
1783 # Check that the last reading is our end reading.
1784 my $last = $readings[$#readings];
1785 throw( "Last reading found from " . $start->text .
1786 " for witness $witness is not the end!" ) # TODO do we get this far?
1787 unless $last->id eq $end->id;
1792 =head2 next_reading( $reading, $sigil );
1794 Returns the reading that follows the given reading along the given witness
1800 # Return the successor via the corresponding path.
1802 my $answer = $self->_find_linked_reading( 'next', @_ );
1803 return undef unless $answer;
1804 return $self->reading( $answer );
1807 =head2 prior_reading( $reading, $sigil )
1809 Returns the reading that precedes the given reading along the given witness
1815 # Return the predecessor via the corresponding path.
1817 my $answer = $self->_find_linked_reading( 'prior', @_ );
1818 return $self->reading( $answer );
1821 sub _find_linked_reading {
1822 my( $self, $direction, $node, $path ) = @_;
1824 # Get a backup if we are dealing with a layered witness
1826 my $aclabel = $self->ac_label;
1827 if( $path && $path =~ /^(.*)\Q$aclabel\E$/ ) {
1831 my @linked_paths = $direction eq 'next'
1832 ? $self->sequence->edges_from( $node )
1833 : $self->sequence->edges_to( $node );
1834 return undef unless scalar( @linked_paths );
1836 # We have to find the linked path that contains all of the
1837 # witnesses supplied in $path.
1838 my( @path_wits, @alt_path_wits );
1839 @path_wits = sort( $self->_witnesses_of_label( $path ) ) if $path;
1840 @alt_path_wits = sort( $self->_witnesses_of_label( $alt_path ) ) if $alt_path;
1843 foreach my $le ( @linked_paths ) {
1844 if( $self->sequence->has_edge_attribute( @$le, $self->baselabel ) ) {
1847 my @le_wits = sort $self->path_witnesses( $le );
1848 if( _is_within( \@path_wits, \@le_wits ) ) {
1849 # This is the right path.
1850 return $direction eq 'next' ? $le->[1] : $le->[0];
1851 } elsif( _is_within( \@alt_path_wits, \@le_wits ) ) {
1855 # Got this far? Return the alternate path if it exists.
1856 return $direction eq 'next' ? $alt_le->[1] : $alt_le->[0]
1859 # Got this far? Return the base path if it exists.
1860 return $direction eq 'next' ? $base_le->[1] : $base_le->[0]
1863 # Got this far? We have no appropriate path.
1864 warn "Could not find $direction node from " . $node->id
1865 . " along path $path";
1871 my( $set1, $set2 ) = @_;
1872 my $ret = @$set1; # will be 0, i.e. false, if set1 is empty
1873 foreach my $el ( @$set1 ) {
1874 $ret = 0 unless grep { /^\Q$el\E$/ } @$set2;
1879 # Return the string that joins together a list of witnesses for
1880 # display on a single path.
1881 sub _witnesses_of_label {
1882 my( $self, $label ) = @_;
1883 my $regex = $self->wit_list_separator;
1884 my @answer = split( /\Q$regex\E/, $label );
1888 =head2 common_readings
1890 Returns the list of common readings in the graph (i.e. those readings that are
1891 shared by all non-lacunose witnesses.)
1895 sub common_readings {
1897 my @common = grep { $_->is_common } $self->readings;
1901 =head2 path_text( $sigil, [, $start, $end ] )
1903 Returns the text of a witness (plus its backup, if we are using a layer)
1904 as stored in the collation. The text is returned as a string, where the
1905 individual readings are joined with spaces and the meta-readings (e.g.
1906 lacunae) are omitted. Optional specification of $start and $end allows
1907 the generation of a subset of the witness text.
1912 my( $self, $wit, $start, $end ) = @_;
1913 $start = $self->start unless $start;
1914 $end = $self->end unless $end;
1915 my @path = grep { !$_->is_meta } $self->reading_sequence( $start, $end, $wit );
1918 foreach my $r ( @path ) {
1919 unless ( $r->join_prior || !$last || $last->join_next ) {
1922 $pathtext .= $r->text;
1928 =head1 INITIALIZATION METHODS
1930 These are mostly for use by parsers.
1932 =head2 make_witness_path( $witness )
1934 Link the array of readings contained in $witness->path (and in
1935 $witness->uncorrected_path if it exists) into collation paths.
1936 Clear out the arrays when finished.
1938 =head2 make_witness_paths
1940 Call make_witness_path for all witnesses in the tradition.
1944 # For use when a collation is constructed from a base text and an apparatus.
1945 # We have the sequences of readings and just need to add path edges.
1946 # When we are done, clear out the witness path attributes, as they are no
1948 # TODO Find a way to replace the witness path attributes with encapsulated functions?
1950 sub make_witness_paths {
1952 foreach my $wit ( $self->tradition->witnesses ) {
1953 # say STDERR "Making path for " . $wit->sigil;
1954 $self->make_witness_path( $wit );
1958 sub make_witness_path {
1959 my( $self, $wit ) = @_;
1960 my @chain = @{$wit->path};
1961 my $sig = $wit->sigil;
1962 # Add start and end if necessary
1963 unshift( @chain, $self->start ) unless $chain[0] eq $self->start;
1964 push( @chain, $self->end ) unless $chain[-1] eq $self->end;
1965 foreach my $idx ( 0 .. $#chain-1 ) {
1966 $self->add_path( $chain[$idx], $chain[$idx+1], $sig );
1968 if( $wit->is_layered ) {
1969 @chain = @{$wit->uncorrected_path};
1970 unshift( @chain, $self->start ) unless $chain[0] eq $self->start;
1971 push( @chain, $self->end ) unless $chain[-1] eq $self->end;
1972 foreach my $idx( 0 .. $#chain-1 ) {
1973 my $source = $chain[$idx];
1974 my $target = $chain[$idx+1];
1975 $self->add_path( $source, $target, $sig.$self->ac_label )
1976 unless $self->has_path( $source, $target, $sig );
1980 $wit->clear_uncorrected_path;
1983 =head2 calculate_ranks
1985 Calculate the reading ranks (that is, their aligned positions relative
1986 to each other) for the graph. This can only be called on linear collations.
1990 use Text::Tradition;
1992 my $cxfile = 't/data/Collatex-16.xml';
1993 my $t = Text::Tradition->new(
1995 'input' => 'CollateX',
1998 my $c = $t->collation;
2001 my $table = $c->alignment_table;
2002 ok( $c->has_cached_table, "Alignment table was cached" );
2003 is( $c->alignment_table, $table, "Cached table returned upon second call" );
2004 $c->calculate_ranks;
2005 is( $c->alignment_table, $table, "Cached table retained with no rank change" );
2006 $c->add_relationship( 'n13', 'n23', { type => 'repetition' } );
2007 is( $c->alignment_table, $table, "Alignment table unchanged after non-colo relationship add" );
2008 $c->add_relationship( 'n24', 'n23', { type => 'spelling' } );
2009 isnt( $c->alignment_table, $table, "Alignment table changed after colo relationship add" );
2015 sub calculate_ranks {
2017 # Save the existing ranks, in case we need to invalidate the cached SVG.
2018 throw( "Cannot calculate ranks on a non-linear graph" )
2019 unless $self->linear;
2021 map { $existing_ranks{$_} = $_->rank } $self->readings;
2023 # Do the rankings based on the relationship equivalence graph, starting
2024 # with the start node.
2025 my ( $node_ranks, $rank_nodes ) = $self->relations->equivalence_ranks();
2027 # Transfer our rankings from the topological graph to the real one.
2028 foreach my $r ( $self->readings ) {
2029 if( defined $node_ranks->{$self->equivalence( $r->id )} ) {
2030 $r->rank( $node_ranks->{$self->equivalence( $r->id )} );
2032 # Die. Find the last rank we calculated.
2033 my @all_defined = sort { ( $node_ranks->{$self->equivalence( $a->id )}||-1 )
2034 <=> ( $node_ranks->{$self->equivalence( $b->id )}||-1 ) }
2036 my $last = pop @all_defined;
2037 throw( "Ranks not calculated after $last - do you have a cycle in the graph?" );
2040 # Do we need to invalidate the cached data?
2041 if( $self->has_cached_table ) {
2042 foreach my $r ( $self->readings ) {
2043 next if defined( $existing_ranks{$r} )
2044 && $existing_ranks{$r} == $r->rank;
2045 # Something has changed, so clear the cache
2046 $self->_clear_cache;
2047 # ...and recalculate the common readings.
2048 $self->calculate_common_readings();
2052 # The graph calculation information is now up to date.
2053 $self->_graphcalc_done(1);
2058 $self->wipe_table if $self->has_cached_table;
2062 =head2 flatten_ranks
2064 A convenience method for parsing collation data. Searches the graph for readings
2065 with the same text at the same rank, and merges any that are found.
2070 my ( $self, %args ) = shift;
2071 my %unique_rank_rdg;
2073 foreach my $p ( $self->identical_readings( %args ) ) {
2074 # say STDERR "Combining readings at same rank: @$p";
2076 $self->merge_readings( @$p );
2077 # TODO see if this now makes a common point.
2079 # If we merged readings, the ranks are still fine but the alignment
2080 # table is wrong. Wipe it.
2081 $self->wipe_table() if $changed;
2084 =head2 identical_readings
2085 =head2 identical_readings( start => $startnode, end => $endnode )
2086 =head2 identical_readings( startrank => $startrank, endrank => $endrank )
2088 Goes through the graph identifying all pairs of readings that appear to be
2089 identical, and therefore able to be merged into a single reading. Returns the
2090 relevant identical pairs. Can be restricted to run over only a part of the
2091 graph, specified either by node or by rank.
2095 sub identical_readings {
2096 my ( $self, %args ) = @_;
2097 # Find where we should start and end.
2098 my $startrank = $args{startrank} || 0;
2099 if( $args{start} ) {
2100 throw( "Starting reading has no rank" ) unless $self->reading( $args{start} )
2101 && $self->reading( $args{start} )->has_rank;
2102 $startrank = $self->reading( $args{start} )->rank;
2104 my $endrank = $args{endrank} || $self->end->rank;
2106 throw( "Ending reading has no rank" ) unless $self->reading( $args{end} )
2107 && $self->reading( $args{end} )->has_rank;
2108 $endrank = $self->reading( $args{end} )->rank;
2111 # Make sure the ranks are correct.
2112 unless( $self->_graphcalc_done ) {
2113 $self->calculate_ranks;
2115 # Go through the readings looking for duplicates.
2116 my %unique_rank_rdg;
2118 foreach my $rdg ( $self->readings ) {
2119 next unless $rdg->has_rank;
2120 my $rk = $rdg->rank;
2121 next if $rk > $endrank || $rk < $startrank;
2122 my $key = $rk . "||" . $rdg->text;
2123 if( exists $unique_rank_rdg{$key} ) {
2124 # Make sure they don't have different grammatical forms
2125 my $ur = $unique_rank_rdg{$key};
2126 if( $rdg->is_identical( $ur ) ) {
2127 push( @pairs, [ $ur, $rdg ] );
2130 $unique_rank_rdg{$key} = $rdg;
2138 =head2 calculate_common_readings
2140 Goes through the graph identifying the readings that appear in every witness
2141 (apart from those with lacunae at that spot.) Marks them as common and returns
2146 use Text::Tradition;
2148 my $cxfile = 't/data/Collatex-16.xml';
2149 my $t = Text::Tradition->new(
2151 'input' => 'CollateX',
2154 my $c = $t->collation;
2156 my @common = $c->calculate_common_readings();
2157 is( scalar @common, 8, "Found correct number of common readings" );
2158 my @marked = sort $c->common_readings();
2159 is( scalar @common, 8, "All common readings got marked as such" );
2160 my @expected = qw/ n1 n11 n16 n19 n20 n5 n6 n7 /;
2161 is_deeply( \@marked, \@expected, "Found correct list of common readings" );
2167 sub calculate_common_readings {
2170 map { $_->is_common( 0 ) } $self->readings;
2171 # Implicitly calls calculate_ranks
2172 my $table = $self->alignment_table;
2173 foreach my $idx ( 0 .. $table->{'length'} - 1 ) {
2174 my @row = map { $_->{'tokens'}->[$idx]
2175 ? $_->{'tokens'}->[$idx]->{'t'} : '' }
2176 @{$table->{'alignment'}};
2178 foreach my $r ( @row ) {
2180 $hash{$r->id} = $r unless $r->is_meta;
2182 $hash{'UNDEF'} = $r;
2185 if( keys %hash == 1 && !exists $hash{'UNDEF'} ) {
2186 my( $r ) = values %hash;
2188 push( @common, $r );
2194 =head2 text_from_paths
2196 Calculate the text array for all witnesses from the path, for later consistency
2197 checking. Only to be used if there is no non-graph-based way to know the
2202 sub text_from_paths {
2204 foreach my $wit ( $self->tradition->witnesses ) {
2205 my @readings = $self->reading_sequence( $self->start, $self->end, $wit->sigil );
2207 foreach my $r ( @readings ) {
2208 next if $r->is_meta;
2209 push( @text, $r->text );
2211 $wit->text( \@text );
2212 if( $wit->is_layered ) {
2213 my @ucrdgs = $self->reading_sequence( $self->start, $self->end,
2214 $wit->sigil.$self->ac_label );
2216 foreach my $r ( @ucrdgs ) {
2217 next if $r->is_meta;
2218 push( @uctext, $r->text );
2220 $wit->layertext( \@uctext );
2225 =head1 UTILITY FUNCTIONS
2227 =head2 common_predecessor( $reading_a, $reading_b )
2229 Find the last reading that occurs in sequence before both the given readings.
2230 At the very least this should be $self->start.
2232 =head2 common_successor( $reading_a, $reading_b )
2234 Find the first reading that occurs in sequence after both the given readings.
2235 At the very least this should be $self->end.
2239 use Text::Tradition;
2241 my $cxfile = 't/data/Collatex-16.xml';
2242 my $t = Text::Tradition->new(
2244 'input' => 'CollateX',
2247 my $c = $t->collation;
2249 is( $c->common_predecessor( 'n24', 'n23' )->id,
2250 'n20', "Found correct common predecessor" );
2251 is( $c->common_successor( 'n24', 'n23' )->id,
2252 '__END__', "Found correct common successor" );
2254 is( $c->common_predecessor( 'n19', 'n17' )->id,
2255 'n16', "Found correct common predecessor for readings on same path" );
2256 is( $c->common_successor( 'n21', 'n10' )->id,
2257 '__END__', "Found correct common successor for readings on same path" );
2263 ## Return the closest reading that is a predecessor of both the given readings.
2264 sub common_predecessor {
2266 my( $r1, $r2 ) = $self->_objectify_args( @_ );
2267 return $self->_common_in_path( $r1, $r2, 'predecessors' );
2270 sub common_successor {
2272 my( $r1, $r2 ) = $self->_objectify_args( @_ );
2273 return $self->_common_in_path( $r1, $r2, 'successors' );
2277 # TODO think about how to do this without ranks...
2278 sub _common_in_path {
2279 my( $self, $r1, $r2, $dir ) = @_;
2280 my $iter = $self->end->rank;
2282 my @last_r1 = ( $r1 );
2283 my @last_r2 = ( $r2 );
2284 # my %all_seen = ( $r1 => 'r1', $r2 => 'r2' );
2286 # say STDERR "Finding common $dir for $r1, $r2";
2287 while( !@candidates ) {
2288 last unless $iter--; # Avoid looping infinitely
2289 # Iterate separately down the graph from r1 and r2
2290 my( @new_lc1, @new_lc2 );
2291 foreach my $lc ( @last_r1 ) {
2292 foreach my $p ( $lc->$dir ) {
2293 if( $all_seen{$p->id} && $all_seen{$p->id} ne 'r1' ) {
2294 # say STDERR "Path candidate $p from $lc";
2295 push( @candidates, $p );
2296 } elsif( !$all_seen{$p->id} ) {
2297 $all_seen{$p->id} = 'r1';
2298 push( @new_lc1, $p );
2302 foreach my $lc ( @last_r2 ) {
2303 foreach my $p ( $lc->$dir ) {
2304 if( $all_seen{$p->id} && $all_seen{$p->id} ne 'r2' ) {
2305 # say STDERR "Path candidate $p from $lc";
2306 push( @candidates, $p );
2307 } elsif( !$all_seen{$p->id} ) {
2308 $all_seen{$p->id} = 'r2';
2309 push( @new_lc2, $p );
2313 @last_r1 = @new_lc1;
2314 @last_r2 = @new_lc2;
2316 my @answer = sort { $a->rank <=> $b->rank } @candidates;
2317 return $dir eq 'predecessors' ? pop( @answer ) : shift ( @answer );
2321 Text::Tradition::Error->throw(
2322 'ident' => 'Collation error',
2328 __PACKAGE__->meta->make_immutable;
2334 =item * Rework XML serialization in a more modular way
2340 This package is free software and is provided "as is" without express
2341 or implied warranty. You can redistribute it and/or modify it under
2342 the same terms as Perl itself.
2346 Tara L Andrews E<lt>aurum@cpan.orgE<gt>