X-Git-Url: http://git.shadowcat.co.uk/gitweb/gitweb.cgi?a=blobdiff_plain;f=lib%2FText%2FTradition%2FCollation.pm;h=77a367efd6bab9f37378704c12d170046b142773;hb=861c3e272c65c7553ad7c03cca51cbdd561f126c;hp=3cc85b8ecc5e71dcc1edeca1e0bd0965cce936ac;hpb=c84275ff42c4d3e6f7fbc13140101975c990101a;p=scpubgit%2Fstemmatology.git diff --git a/lib/Text/Tradition/Collation.pm b/lib/Text/Tradition/Collation.pm index 3cc85b8..77a367e 100644 --- a/lib/Text/Tradition/Collation.pm +++ b/lib/Text/Tradition/Collation.pm @@ -66,12 +66,6 @@ has 'linear' => ( default => 1, ); -has 'collapse_punctuation' => ( - is => 'rw', - isa => 'Bool', - default => 1, - ); - has 'ac_label' => ( is => 'rw', isa => 'Str', @@ -92,20 +86,146 @@ has 'end' => ( weak_ref => 1, ); -# The collation can be created two ways: -# 1. Collate a set of witnesses (with CollateX I guess) and process -# the results as in 2. -# 2. Read a pre-prepared collation in one of a variety of formats, -# and make the graph from that. - -# The graph itself will (for now) be immutable, and the positions -# within the graph will also be immutable. We need to calculate those -# positions upon graph construction. The equivalences between graph -# nodes will be mutable, entirely determined by the user (or possibly -# by some semantic pre-processing provided by the user.) So the -# constructor should just make an empty equivalences object. The -# constructor will also need to make the witness objects, if we didn't -# come through option 1. +=head1 NAME + +Text::Tradition::Collation - a software model for a text collation + +=head1 SYNOPSIS + + use Text::Tradition; + my $t = Text::Tradition->new( + 'name' => 'this is a text', + 'input' => 'TEI', + 'file' => '/path/to/tei_parallel_seg_file.xml' ); + + my $c = $t->collation; + my @readings = $c->readings; + my @paths = $c->paths; + my @relationships = $c->relationships; + + my $svg_variant_graph = $t->collation->as_svg(); + +=head1 DESCRIPTION + +Text::Tradition is a library for representation and analysis of collated +texts, particularly medieval ones. The Collation is the central feature of +a Tradition, where the text, its sequence of readings, and its relationships +between readings are actually kept. + +=head1 CONSTRUCTOR + +=head2 new + +The constructor. Takes a hash or hashref of the following arguments: + +=over + +=item * tradition - The Text::Tradition object to which the collation +belongs. Required. + +=item * linear - Whether the collation should be linear; that is, whether +transposed readings should be treated as two linked readings rather than one, +and therefore whether the collation graph is acyclic. Defaults to true. + +=item * baselabel - The default label for the path taken by a base text +(if any). Defaults to 'base text'. + +=item * wit_list_separator - The string to join a list of witnesses for +purposes of making labels in display graphs. Defaults to ', '. + +=item * ac_label - The extra label to tack onto a witness sigil when +representing another layer of path for the given witness - that is, when +a text has more than one possible reading due to scribal corrections or +the like. Defaults to ' (a.c.)'. + +=back + +=head1 ACCESSORS + +=head2 tradition + +=head2 linear + +=head2 wit_list_separator + +=head2 baselabel + +=head2 ac_label + +Simple accessors for collation attributes. + +=head2 start + +The meta-reading at the start of every witness path. + +=head2 end + +The meta-reading at the end of every witness path. + +=head2 readings + +Returns all Reading objects in the graph. + +=head2 reading( $id ) + +Returns the Reading object corresponding to the given ID. + +=head2 add_reading( $reading_args ) + +Adds a new reading object to the collation. +See L for the available arguments. + +=head2 del_reading( $object_or_id ) + +Removes the given reading from the collation, implicitly removing its +paths and relationships. + +=head2 merge_readings( $main, $second ) + +Merges the $second reading into the $main one. +The arguments may be either readings or reading IDs. + +=head2 has_reading( $id ) + +Predicate to see whether a given reading ID is in the graph. + +=head2 reading_witnesses( $object_or_id ) + +Returns a list of sigils whose witnesses contain the reading. + +=head2 paths + +Returns all reading paths within the document - that is, all edges in the +collation graph. Each path is an arrayref of [ $source, $target ] reading IDs. + +=head2 add_path( $source, $target, $sigil ) + +Links the given readings in the collation in sequence, under the given witness +sigil. The readings may be specified by object or ID. + +=head2 del_path( $source, $target, $sigil ) + +Links the given readings in the collation in sequence, under the given witness +sigil. The readings may be specified by object or ID. + +=head2 has_path( $source, $target ); + +Returns true if the two readings are linked in sequence in any witness. +The readings may be specified by object or ID. + +=head2 relationships + +Returns all Relationship objects in the collation. + +=head2 add_relationship( $reading, $other_reading, $options ) + +Adds a new relationship of the type given in $options between the two readings, +which may be specified by object or ID. Returns a value of ( $status, @vectors) +where $status is true on success, and @vectors is a list of relationship edges +that were ultimately added. +See L for the available options. + +=cut sub BUILD { my $self = shift; @@ -197,6 +317,15 @@ sub _stringify_args { return( $first, $second, $arg ); } +# Helper function for manipulating the graph. +sub _objectify_args { + my( $self, $first, $second, $arg ) = @_; + $first = $self->reading( $first ) + unless ref( $first ) eq 'Text::Tradition::Collation::Reading'; + $second = $self->reading( $second ) + unless ref( $second ) eq 'Text::Tradition::Collation::Reading'; + return( $first, $second, $arg ); +} ### Path logic sub add_path { @@ -243,16 +372,31 @@ sub has_path { return $self->sequence->has_edge_attribute( $source, $target, $wit ); } -=head2 add_relationship( $reading1, $reading2, $definition ) +=head2 clear_witness( @sigil_list ) -Adds the specified relationship between the two readings. A relationship -is transitive (i.e. undirected); the options for its definition may be found -in Text::Tradition::Collation::Relationship. +Clear the given witnesses out of the collation entirely, removing references +to them in paths, and removing readings that belong only to them. Should only +be called via $tradition->del_witness. =cut -# Wouldn't it be lovely if edges could be objects, and all this type checking -# and attribute management could be done via Moose? +sub clear_witness { + my( $self, @sigils ) = @_; + + # Clear the witness(es) out of the paths + foreach my $e ( $self->paths ) { + foreach my $sig ( @sigils ) { + $self->del_path( $e, $sig ); + } + } + + # Clear out the newly unused readings + foreach my $r ( $self->readings ) { + unless( $self->reading_witnesses( $r ) ) { + $self->del_reading( $r ); + } + } +} sub add_relationship { my $self = shift; @@ -285,13 +429,9 @@ sub reading_witnesses { return keys %all_witnesses; } -=head2 Output method(s) +=head1 OUTPUT METHODS -=over - -=item B - -print $collation->as_svg(); +=head2 as_svg Returns an SVG string that represents the graph, via as_dot and graphviz. @@ -313,9 +453,7 @@ sub as_svg { return $svg; } -=item B - -print $collation->svg_subgraph( $from, $to ) +=head2 svg_subgraph( $from, $to ) Returns an SVG string that represents the portion of the graph given by the specified range. The $from and $to variables refer to ranks within the graph. @@ -345,16 +483,11 @@ sub svg_subgraph { } -=item B - -print $collation->as_dot(); +=head2 as_dot( $from, $to ) Returns a string that is the collation graph expressed in dot -(i.e. GraphViz) format. The 'view' argument determines what kind of -graph is produced. - * 'path': a graph of witness paths through the collation (DEFAULT) - * 'relationship': a graph of how collation readings relate to - each other +(i.e. GraphViz) format. If $from or $to is passed, as_dot creates +a subgraph rather than the entire graph. =cut @@ -375,11 +508,26 @@ sub as_dot { my $graph_name = $self->tradition->name; $graph_name =~ s/[^\w\s]//g; $graph_name = join( '_', split( /\s+/, $graph_name ) ); + + my %graph_attrs = ( + 'rankdir' => 'LR', + 'bgcolor' => 'none', + ); + my %node_attrs = ( + 'fontsize' => 11, + 'fillcolor' => 'white', + 'style' => 'filled', + 'shape' => 'ellipse' + ); + my %edge_attrs = ( + 'arrowhead' => 'open', + 'color' => '#000000', + 'fontcolor' => '#000000', + ); + my $dot = sprintf( "digraph %s {\n", $graph_name ); - $dot .= "\tedge [ arrowhead=open ];\n"; - $dot .= "\tgraph [ rankdir=LR,bgcolor=none ];\n"; - $dot .= sprintf( "\tnode [ fontsize=%d, fillcolor=%s, style=%s, shape=%s ];\n", - 11, "white", "filled", "ellipse" ); + $dot .= "\tgraph " . _dot_attr_string( \%graph_attrs ) . ";\n"; + $dot .= "\tnode " . _dot_attr_string( \%node_attrs ) . ";\n"; # Output substitute start/end readings if necessary if( $startrank ) { @@ -389,60 +537,71 @@ sub as_dot { $dot .= "\t\"#SUBEND#\" [ label=\"...\" ];\n"; } my %used; # Keep track of the readings that actually appear in the graph - my %subedges; - my %subend; foreach my $reading ( $self->readings ) { # Only output readings within our rank range. next if $startrank && $reading->rank < $startrank; next if $endrank && $reading->rank > $endrank; $used{$reading->id} = 1; - $subedges{$reading->id} = '#SUBSTART#' - if $startrank && $startrank == $reading->rank; - $subedges{$reading->id} = '#SUBEND#' - if $endrank && $endrank == $reading->rank; # Need not output nodes without separate labels next if $reading->id eq $reading->text; - my $label = $reading->punctuated_form; + my $label = $reading->text; $label =~ s/\"/\\\"/g; $dot .= sprintf( "\t\"%s\" [ label=\"%s\" ];\n", $reading->id, $label ); } - # Add substitute start and end edges if necessary - foreach my $node ( keys %subedges ) { - my @vector = ( $subedges{$node}, $node ); - @vector = reverse( @vector ) if $vector[0] =~ /END/; - my $witstr = join( ', ', sort $self->reading_witnesses( $self->reading( $node ) ) ); - my %variables = ( 'color' => '#000000', - 'fontcolor' => '#000000', - 'label' => $witstr, - ); - my $varopts = join( ', ', map { $_.'="'.$variables{$_}.'"' } sort keys %variables ); - $dot .= sprintf( "\t\"%s\" -> \"%s\" [ %s ];\n", @vector, $varopts ); - } - # Add the real edges my @edges = $self->paths; + my( %substart, %subend ); foreach my $edge ( @edges ) { # Do we need to output this edge? if( $used{$edge->[0]} && $used{$edge->[1]} ) {; - my %variables = ( 'color' => '#000000', - 'fontcolor' => '#000000', - 'label' => join( ', ', $self->path_display_label( $edge ) ), - ); - my $varopts = join( ', ', map { $_.'="'.$variables{$_}.'"' } sort keys %variables ); + my $label = $self->path_display_label( $self->path_witnesses( $edge ) ); + my $variables = { %edge_attrs, 'label' => $label }; # Account for the rank gap if necessary - my $rankgap = $self->reading( $edge->[1] )->rank + if( $self->reading( $edge->[1] )->has_rank + && $self->reading( $edge->[0] )->has_rank + && $self->reading( $edge->[1] )->rank + - $self->reading( $edge->[0] )->rank > 1 ) { + $variables->{'minlen'} = $self->reading( $edge->[1] )->rank - $self->reading( $edge->[0] )->rank; - $varopts .= ", minlen=$rankgap" if $rankgap > 1; - $dot .= sprintf( "\t\"%s\" -> \"%s\" [ %s ];\n", - $edge->[0], $edge->[1], $varopts ); + } + my $varopts = _dot_attr_string( $variables ); + $dot .= sprintf( "\t\"%s\" -> \"%s\" %s;\n", + $edge->[0], $edge->[1], $varopts ); + } elsif( $used{$edge->[0]} ) { + $subend{$edge->[0]} = 1; + } elsif( $used{$edge->[1]} ) { + $substart{$edge->[1]} = 1; } } - + # Add substitute start and end edges if necessary + foreach my $node ( keys %substart ) { + my $witstr = $self->path_display_label ( $self->reading_witnesses( $self->reading( $node ) ) ); + my $variables = { %edge_attrs, 'label' => $witstr }; + my $varopts = _dot_attr_string( $variables ); + $dot .= "\t\"#SUBSTART#\" -> \"$node\" $varopts;"; + } + foreach my $node ( keys %subend ) { + my $witstr = $self->path_display_label ( $self->reading_witnesses( $self->reading( $node ) ) ); + my $variables = { %edge_attrs, 'label' => $witstr }; + my $varopts = _dot_attr_string( $variables ); + $dot .= "\t\"$node\" -> \"#SUBEND#\" $varopts;"; + } + $dot .= "}\n"; return $dot; } +sub _dot_attr_string { + my( $hash ) = @_; + my @attrs; + foreach my $k ( sort keys %$hash ) { + my $v = $hash->{$k}; + push( @attrs, $k.'="'.$v.'"' ); + } + return( '[ ' . join( ', ', @attrs ) . ' ]' ); +} + sub path_witnesses { my( $self, @edge ) = @_; # If edge is an arrayref, cope. @@ -455,10 +614,10 @@ sub path_witnesses { } sub path_display_label { - my( $self, $edge ) = @_; - my @wits = $self->path_witnesses( $edge ); + my( $self, @wits ) = @_; my $maj = scalar( $self->tradition->witnesses ) * 0.6; if( scalar @wits > $maj ) { + # TODO break out a.c. wits return 'majority'; } else { return join( ', ', @wits ); @@ -466,14 +625,49 @@ sub path_display_label { } -=item B +=head2 as_graphml + +Returns a GraphML representation of the collation. The GraphML will contain +two graphs. The first expresses the attributes of the readings and the witness +paths that link them; the second expresses the relationships that link the +readings. This is the native transfer format for a tradition. + +=begin testing + +use Text::Tradition; + +my $READINGS = 311; +my $PATHS = 361; + +my $datafile = 't/data/florilegium_tei_ps.xml'; +my $tradition = Text::Tradition->new( 'input' => 'TEI', + 'name' => 'test0', + 'file' => $datafile, + 'linear' => 1 ); + +ok( $tradition, "Got a tradition object" ); +is( scalar $tradition->witnesses, 13, "Found all witnesses" ); +ok( $tradition->collation, "Tradition has a collation" ); -print $collation->as_graphml( $recalculate ) +my $c = $tradition->collation; +is( scalar $c->readings, $READINGS, "Collation has all readings" ); +is( scalar $c->paths, $PATHS, "Collation has all paths" ); +is( scalar $c->relationships, 0, "Collation has all relationships" ); -Returns a GraphML representation of the collation graph, with -transposition information and position information. Unless -$recalculate is passed (and is a true value), the method will return a -cached copy of the SVG after the first call to the method. +# Add a few relationships +$c->add_relationship( 'w123', 'w125', { 'type' => 'collated' } ); +$c->add_relationship( 'w193', 'w196', { 'type' => 'collated' } ); +$c->add_relationship( 'w257', 'w262', { 'type' => 'transposition' } ); + +# Now write it to GraphML and parse it again. + +my $graphml = $c->as_graphml; +my $st = Text::Tradition->new( 'input' => 'Self', 'string' => $graphml ); +is( scalar $st->collation->readings, $READINGS, "Reparsed collation has all readings" ); +is( scalar $st->collation->paths, $PATHS, "Reparsed collation has all paths" ); +is( scalar $st->collation->relationships, 3, "Reparsed collation has new relationships" ); + +=end testing =cut @@ -574,7 +768,6 @@ sub as_graphml { $node_el->setAttribute( 'id', $node_xmlid ); foreach my $d ( keys %node_data ) { my $nval = $n->$d; - $nval = $n->punctuated_form if $d eq 'text'; _add_graphml_data( $node_el, $node_data_keys{$d}, $nval ) if defined $nval; } @@ -610,7 +803,8 @@ sub as_graphml { } # Add the relationship graph to the XML - $self->relations->as_graphml( $root, $graphml_ns, \%node_hash, \%edge_data_keys ); + $self->relations->as_graphml( $graphml_ns, $root, \%node_hash, + $node_data_keys{'id'}, \%edge_data_keys ); # Save and return the thing my $result = decode_utf8( $graphml->toString(1) ); @@ -625,9 +819,7 @@ sub _add_graphml_data { $data_el->appendText( $value ); } -=item B - -print $collation->as_csv( $recalculate ) +=head2 as_csv Returns a CSV alignment table representation of the collation graph, one row per witness (or witness uncorrected.) @@ -652,23 +844,22 @@ sub as_csv { return join( "\n", @result ); } -=item B - -my $table = $collation->make_alignment_table( $use_refs, \@wits_to_include ) +=head2 make_alignment_table( $use_refs, $include_witnesses ) Return a reference to an alignment table, in a slightly enhanced CollateX format which looks like this: $table = { alignment => [ { witness => "SIGIL", - tokens => [ { t => "READINGTEXT" }, ... ] }, + tokens => [ { t => "TEXT" }, ... ] }, { witness => "SIG2", - tokens => [ { t => "READINGTEXT" }, ... ] }, + tokens => [ { t => "TEXT" }, ... ] }, ... ], length => TEXTLEN }; If $use_refs is set to 1, the reading object is returned in the table instead of READINGTEXT; if not, the text of the reading is returned. -If $wits_to_include is set to a hashref, only the witnesses whose sigil + +If $include_witnesses is set to a hashref, only the witnesses whose sigil keys have a true hash value will be included. =cut @@ -692,7 +883,7 @@ sub make_alignment_table { { 'witness' => $wit->sigil, 'tokens' => \@row } ); if( $wit->is_layered ) { my @wit_ac_path = $self->reading_sequence( $self->start, $self->end, - $wit->sigil.$self->ac_label, $wit->sigil ); + $wit->sigil.$self->ac_label ); my @ac_row = _make_witness_row( \@wit_ac_path, \@all_pos, $noderefs ); push( @{$table->{'alignment'}}, { 'witness' => $wit->sigil.$self->ac_label, 'tokens' => \@ac_row } ); @@ -754,39 +945,23 @@ sub _turn_table { return $result; } -=back +=head1 NAVIGATION METHODS -=head2 Navigation methods - -=over - -=item B - -my $beginning = $collation->start(); - -Returns the beginning of the collation, a meta-reading with label '#START#'. - -=item B - -my $end = $collation->end(); - -Returns the end of the collation, a meta-reading with label '#END#'. - - -=item B - -my @readings = $collation->reading_sequence( $first, $last, $path[, $alt_path] ); +=head2 reading_sequence( $first, $last, $sigil, $backup ) Returns the ordered list of readings, starting with $first and ending -with $last, along the given witness path. If no path is specified, -assume that the path is that of the base text (if any.) +with $last, for the witness given in $sigil. If a $backup sigil is +specified (e.g. when walking a layered witness), it will be used wherever +no $sigil path exists. If there is a base text reading, that will be +used wherever no path exists for $sigil or $backup. =cut # TODO Think about returning some lazy-eval iterator. +# TODO Get rid of backup; we should know from what witness is whether we need it. sub reading_sequence { - my( $self, $start, $end, $witness, $backup ) = @_; + my( $self, $start, $end, $witness ) = @_; $witness = $self->baselabel unless $witness; my @readings = ( $start ); @@ -799,7 +974,7 @@ sub reading_sequence { } $seen{$n->id} = 1; - my $next = $self->next_reading( $n, $witness, $backup ); + my $next = $self->next_reading( $n, $witness ); unless( $next ) { warn "Did not find any path for $witness from reading " . $n->id; last; @@ -816,9 +991,7 @@ sub reading_sequence { return @readings; } -=item B - -my $next_reading = $collation->next_reading( $reading, $witpath ); +=head2 next_reading( $reading, $sigil ); Returns the reading that follows the given reading along the given witness path. @@ -833,9 +1006,7 @@ sub next_reading { return $self->reading( $answer ); } -=item B - -my $prior_reading = $collation->prior_reading( $reading, $witpath ); +=head2 prior_reading( $reading, $sigil ) Returns the reading that precedes the given reading along the given witness path. @@ -850,7 +1021,15 @@ sub prior_reading { } sub _find_linked_reading { - my( $self, $direction, $node, $path, $alt_path ) = @_; + my( $self, $direction, $node, $path ) = @_; + + # Get a backup if we are dealing with a layered witness + my $alt_path; + my $aclabel = $self->ac_label; + if( $path && $path =~ /^(.*)\Q$aclabel\E$/ ) { + $alt_path = $1; + } + my @linked_paths = $direction eq 'next' ? $self->sequence->edges_from( $node ) : $self->sequence->edges_to( $node ); @@ -859,8 +1038,8 @@ sub _find_linked_reading { # We have to find the linked path that contains all of the # witnesses supplied in $path. my( @path_wits, @alt_path_wits ); - @path_wits = sort( $self->witnesses_of_label( $path ) ) if $path; - @alt_path_wits = sort( $self->witnesses_of_label( $alt_path ) ) if $alt_path; + @path_wits = sort( $self->_witnesses_of_label( $path ) ) if $path; + @alt_path_wits = sort( $self->_witnesses_of_label( $alt_path ) ) if $alt_path; my $base_le; my $alt_le; foreach my $le ( @linked_paths ) { @@ -899,8 +1078,48 @@ sub _is_within { return $ret; } +# Return the string that joins together a list of witnesses for +# display on a single path. +sub _witnesses_of_label { + my( $self, $label ) = @_; + my $regex = $self->wit_list_separator; + my @answer = split( /\Q$regex\E/, $label ); + return @answer; +} + +=head2 path_text( $sigil, $mainsigil [, $start, $end ] ) + +Returns the text of a witness (plus its backup, if we are using a layer) +as stored in the collation. The text is returned as a string, where the +individual readings are joined with spaces and the meta-readings (e.g. +lacunae) are omitted. Optional specification of $start and $end allows +the generation of a subset of the witness text. + +=cut + +sub path_text { + my( $self, $wit, $start, $end ) = @_; + $start = $self->start unless $start; + $end = $self->end unless $end; + my @path = grep { !$_->is_meta } $self->reading_sequence( $start, $end, $wit ); + return join( ' ', map { $_->text } @path ); +} + +=head1 INITIALIZATION METHODS + +These are mostly for use by parsers. + +=head2 make_witness_path( $witness ) -## INITIALIZATION METHODS - for use by parsers +Link the array of readings contained in $witness->path (and in +$witness->uncorrected_path if it exists) into collation paths. +Clear out the arrays when finished. + +=head2 make_witness_paths + +Call make_witness_path for all witnesses in the tradition. + +=cut # For use when a collation is constructed from a base text and an apparatus. # We have the sequences of readings and just need to add path edges. @@ -936,6 +1155,13 @@ sub make_witness_path { $wit->clear_uncorrected_path; } +=head2 calculate_ranks + +Calculate the reading ranks (that is, their aligned positions relative +to each other) for the graph. This can only be called on linear collations. + +=cut + sub calculate_ranks { my $self = shift; # Walk a version of the graph where every node linked by a relationship @@ -968,7 +1194,7 @@ sub calculate_ranks { foreach my $n ( $self->sequence->successors( $r->id ) ) { my( $tfrom, $tto ) = ( $rel_containers{$r->id}, $rel_containers{$n} ); - $DB::single = 1 unless $tfrom && $tto; + # $DB::single = 1 unless $tfrom && $tto; $topo_graph->add_edge( $tfrom, $tto ); } } @@ -986,7 +1212,6 @@ sub calculate_ranks { if( defined $node_ranks->{$rel_containers{$r->id}} ) { $r->rank( $node_ranks->{$rel_containers{$r->id}} ); } else { - $DB::single = 1; die "No rank calculated for node " . $r->id . " - do you have a cycle in the graph?"; } @@ -1028,8 +1253,13 @@ sub _assign_rank { return @next_nodes; } -# Another method to make up for rough collation methods. If the same reading -# appears multiple times at the same rank, collapse the nodes. +=head2 flatten_ranks + +A convenience method for parsing collation data. Searches the graph for readings +with the same text at the same rank, and merges any that are found. + +=cut + sub flatten_ranks { my $self = shift; my %unique_rank_rdg; @@ -1038,7 +1268,7 @@ sub flatten_ranks { my $key = $rdg->rank . "||" . $rdg->text; if( exists $unique_rank_rdg{$key} ) { # Combine! - # print STDERR "Combining readings at same rank: $key\n"; + # print STDERR "Combining readings at same rank: $key\n"; $self->merge_readings( $unique_rank_rdg{$key}, $rdg ); } else { $unique_rank_rdg{$key} = $rdg; @@ -1046,18 +1276,39 @@ sub flatten_ranks { } } +=head2 text_from_paths -## Utility functions - -# Return the string that joins together a list of witnesses for -# display on a single path. -sub witnesses_of_label { - my( $self, $label ) = @_; - my $regex = $self->wit_list_separator; - my @answer = split( /\Q$regex\E/, $label ); - return @answer; -} +Calculate the text array for all witnesses from the path, for later consistency +checking. Only to be used if there is no non-graph-based way to know the +original texts. + +=cut +sub text_from_paths { + my $self = shift; + foreach my $wit ( $self->tradition->witnesses ) { + my @text = split( /\s+/, + $self->reading_sequence( $self->start, $self->end, $wit->sigil ) ); + $wit->text( \@text ); + if( $wit->is_layered ) { + my @uctext = split( /\s+/, + $self->reading_sequence( $self->start, $self->end, + $wit->sigil.$self->ac_label ) ); + $wit->text( \@uctext ); + } + } +} + +=head1 UTILITY FUNCTIONS + +=head2 common_predecessor( $reading_a, $reading_b ) + +Find the last reading that occurs in sequence before both the given readings. + +=head2 common_successor( $reading_a, $reading_b ) + +Find the first reading that occurs in sequence after both the given readings. + =begin testing use Text::Tradition; @@ -1070,14 +1321,14 @@ my $t = Text::Tradition->new( ); my $c = $t->collation; -is( $c->common_predecessor( $c->reading('n9'), $c->reading('n23') )->id, +is( $c->common_predecessor( 'n9', 'n23' )->id, 'n20', "Found correct common predecessor" ); -is( $c->common_successor( $c->reading('n9'), $c->reading('n23') )->id, +is( $c->common_successor( 'n9', 'n23' )->id, '#END#', "Found correct common successor" ); -is( $c->common_predecessor( $c->reading('n19'), $c->reading('n17') )->id, +is( $c->common_predecessor( 'n19', 'n17' )->id, 'n16', "Found correct common predecessor for readings on same path" ); -is( $c->common_successor( $c->reading('n21'), $c->reading('n26') )->id, +is( $c->common_successor( 'n21', 'n26' )->id, '#END#', "Found correct common successor for readings on same path" ); =end testing @@ -1087,12 +1338,14 @@ is( $c->common_successor( $c->reading('n21'), $c->reading('n26') )->id, ## Return the closest reading that is a predecessor of both the given readings. sub common_predecessor { my $self = shift; - return $self->common_in_path( @_, 'predecessors' ); + my( $r1, $r2 ) = $self->_objectify_args( @_ ); + return $self->common_in_path( $r1, $r2, 'predecessors' ); } sub common_successor { my $self = shift; - return $self->common_in_path( @_, 'successors' ); + my( $r1, $r2 ) = $self->_objectify_args( @_ ); + return $self->common_in_path( $r1, $r2, 'successors' ); } sub common_in_path { @@ -1127,6 +1380,6 @@ __PACKAGE__->meta->make_immutable; =over -=item * Think about making Relationship objects again +=item * Get rid of $backup in reading_sequence =back