Bio-Phylo

 view release on metacpan or  search on metacpan

lib/Bio/Phylo/Parsers/Cdao.pm  view on Meta::CPAN

    my ( $self, $predicate ) = @_;
    # attempt to split URI in namespace and term
    my ( $ns, $term );
    
    # this is for cases where the term is referenced as somewhere inside
    # an ontology using an anchor '#', e.g. in CDAO
    if ( $predicate =~ m/^(.+#)(.+?)$/ ) {
        ( $ns, $term ) = ( $1, $2 );
    }
    
    # this is for cases where the term is a path fragment inside a namespace,
    # i.e. preceded by a '/', as in dublin core
    elsif ( $predicate =~ m/^(.+\/)([^\/]+?)$/ ) {
        ( $ns, $term ) = ( $1, $2 );
    }
    
    # this is for cases where the term is relative to a urn:, i.e. preceded
    # by a ':', as in the uBio predicates
    elsif ( $predicate =~ m/^(.+:)([^:]+?)$/ ) {
        ( $ns, $term ) = ( $1, $2 );
    }

lib/Bio/Phylo/Parsers/Nexus.pm  view on Meta::CPAN

=begin comment

 Type    : Method
 Title   : _tokenize()
 Usage   : $nexus->_tokenize($lines);
 Function: Tokenizes lines in $lines array ref
 Returns : Two dimensional ARRAY
 Args    : An array ref of lines (e.g. read from an input file);
 Comments: This method accepts an array ref holding lines that may contain
           single quotes, double quotes or square brackets. Line breaks and
           spaces inside these quoted/bracketed fragments are ignored, otherwise
           it is split, e.g.:

           [
               [ '#NEXUS' ],
               [ 'BEGIN TAXA; [taxablock comment]' ],
               [ 'DIMENSIONS NTAX=3;' ],
               [ 'TAXLABELS "Taxon \' A" \'Taxon B\' TAXON[comment]C' ],
               ...etc...
           ]

lib/Bio/Phylo/Parsers/Nexus.pm  view on Meta::CPAN

        ']' => '[',
        ')' => '(',
    );

    # tokenize
  LINE: for my $line ( @{$lines} ) {
        $LineCount++;
      TOKEN: while ( $line =~ /\S/ ) {

            # line in file has no quoting/bracketing characters, and
            # is no extension of a quoted/bracketed fragment starting
            # on a previous line
            if ( $line !~ $QUOTES_OR_BRACKETS && !$INSIDE_QUOTE ) {
                if ($continue) {
                    push @{ $tokens[-1] }, $line;
                    $continue = 0;
                }
                else {
                    push @tokens, [$line];
                }
                my $logline = join( ' ', @{ $tokens[-1] } );
                chomp($logline);
                $self->_logger->debug("Tokenized line $LineCount: $logline");
                next LINE;
            }

            # line in file has opening quoting/bracketing characters, and
            # is no extension of a quoted/bracketed fragment starting
            # on a previous line
            elsif ( $line =~ $OPENING_QUOTE_OR_BRACKET && !$INSIDE_QUOTE ) {
                my ( $start, $quoted ) = ( $1, $2 );
                push @tokens, [$start];
                $line    = $quoted;
                $extract = $quoted;
                $INSIDE_QUOTE++;
                $continue = 1;
                $QuoteContext = substr( $quoted, 0, 1 );
                $self->_logger->debug("Line $LineCount contains $QuoteContext");
                $QuoteStartLine      = $LineCount;
                $CONTEXT_QB_AT_START = qr/^(\Q$QuoteContext\E)(.*)$/;
                my $context_closer = $CLOSE_CHAR{$QuoteContext};
                $CONTEXT_CLOSER = qr/^(.*?)(\Q$context_closer\E)(.*)$/;
                next TOKEN;
            }

            # line in file has no quoting/bracketing characters, and
            # is an extension of a quoted/bracketed fragment starting
            # on a previous line
            elsif ( $line !~ $CONTEXT_CLOSER && $INSIDE_QUOTE ) {
                $self->_logger->debug(
                    "Line $LineCount extends quote or comment");
                $extract .= $line;
                next LINE;
            }
            elsif ( $line =~ $CONTEXT_QB_AT_START && $INSIDE_QUOTE ) {
                my ( $q, $remainder ) = ( $1, $1 . $2 );
                if ( $q eq '"' || $q eq "'" ) {

lib/Bio/Phylo/Parsers/Nexus.pm  view on Meta::CPAN

    # an exception here means that an opening quote symbol " ' [
    # ($QuoteContext) was encountered at input file/string line $QuoteStartLine.
    # This can happen if any of these symbols is used in an illegal
    # way, e.g. by using double quotes as gap symbols in matrices.
    if ($INSIDE_QUOTE) {
        throw 'BadArgs' =>
          "Unbalanced $QuoteContext starting at line $QuoteStartLine";
    }

    # final split: non-quoted/bracketed fragments are split on whitespace,
    # others are preserved verbatim
    $self->_logger->info(
        "going to split non-quoted/commented fragments on whitespace");
    foreach my $line (@tokens) {
        my @line;
        foreach my $word (@$line) {
            if ( $word !~ $QUOTES_OR_BRACKETS ) {
                $word =~ s/(=|;|,)/ $1 /g;
                push @line, grep { /\S/ } split /\s+/, $word;
            }
            else {
                push @line, $word;
            }

lib/Bio/Phylo/Util/Logger.pm  view on Meta::CPAN

 Type    : Constructor
 Title   : new
 Usage   : my $logger = Bio::Phylo::Util::Logger->new;
 Function: Instantiates a logger
 Returns : a Bio::Phylo::Util::Logger object
 Args    : -level  => Bio::Phylo::Util::Logger::INFO (DEBUG/INFO/WARN/ERROR/FATAL)
           -class  => a package (or array ref) for which to set verbosity (optional)
           -method => a sub name (or array ref) for which to set verbosity (optional)
           -file   => a file to which to append logging messages
           -listeners => array ref of subs that handle logging messages
           -prefix    => a path fragment to strip from the paths in logging messages
           

=back

=head2 VERBOSITY LEVELS

=over

=item FATAL