8 use Text::Tradition::Stemma;
10 binmode STDERR, ":utf8";
11 binmode STDOUT, ":utf8";
12 eval { no warnings; binmode $DB::OUT, ":utf8"; };
14 my( $informat, $inbase, $outformat, $help, $linear, $HACK )
15 = ( '', '', '', '', 1, 0 );
17 GetOptions( 'i|in=s' => \$informat,
18 'b|base=s' => \$inbase,
19 'o|out=s' => \$outformat,
20 'l|linear!' => \$linear,
29 unless( $informat =~ /^(CSV|CTE|KUL|Self|TEI|CollateX|tab(ular)?)$/i ) {
30 help( "Input format must be one of CollateX, CSV, CTE, Self, TEI" );
32 $informat = 'CollateX' if $informat =~ /^c(ollate)?x$/i;
33 $informat = 'KUL' if $informat =~ /^kul$/i;
34 $informat = 'CTE' if $informat =~ /^cte$/i;
35 $informat = 'Self' if $informat =~ /^self$/i;
36 $informat = 'TEI' if $informat =~ /^tei$/i;
37 $informat = 'Tabular' if $informat =~ /^tab$/i;
39 unless( $outformat =~ /^(graphml|svg|dot|stemma|csv)$/ ) {
40 help( "Output format must be one of graphml, svg, csv, stemma, or dot" );
43 # Do we have a base if we need it?
44 if( $informat eq 'KUL' && !$inbase ) {
45 help( "$informat input needs a base text" );
48 # CSV parsing requires a filename; XML parsing requires a string.
50 unless( $informat eq 'KUL' || $informat eq 'CSV' ) {
52 open( INFILE, "$input" ) or die "Could not read $input";
53 binmode INFILE, ':utf8';
56 $input = join( '', @lines );
59 # First: read the base. Make a graph, but also note which
60 # nodes represent line beginnings.
61 my %args = ( $informat => $input,
62 'linear' => $linear );
63 $args{'base'} = $inbase if $inbase;
64 my $tradition = Text::Tradition->new( %args );
67 # Remove witnesses C, E, G in the Matthew text
69 foreach( $tradition->collation->paths() ) {
70 $tradition->collation->del_path( $_ ) if $_->label =~ /^[ceg]$/i;
72 foreach( $tradition->collation->readings() ) {
73 if( !$_->outgoing() && !$_->incoming() ) {
74 print STDERR "Deleting reading " . $_->label . "\n";
75 $tradition->collation->del_reading( $_ );
80 # Now output what we have been asked to.
81 if( $outformat eq 'stemma' ) {
82 my $stemma = Text::Tradition::Stemma->new(
83 'collation' => $tradition->collation );
84 my( $result, $tree ) = $stemma->run_phylip_pars();
88 print STDERR "Bad result: $tree";
91 my $output = "as_$outformat";
92 print $tradition->collation->$output();
98 Usage: $0 -i [format] -o [format] (--base [filename]) (--(no)linear) [inputfile]
99 i, input: Format of the input file. Must be one of CollateX, CSV, CTE, Self, TEI.
100 o, output: Format of the output. Must be one of svg, dot, graphml, csv, stemma.
101 b, base: Filename that contains a base text. Needed for CSV input.
102 l, linear: Treat transposed readings separately, producing a linear graph.
103 If nolinear, treat transposed readings as the same node.
104 h, help: Print this message.
108 print STDERR "$msg\n";
110 exit ($msg ? 1 : 0 );