lib/MOBY/MobyXMLRetriever.pl

   1 #!/usr/bin/perl -w
   2 use XML::LibXML;
   3 my $parser = XML::LibXML->new;
   4
   5 ## Parsing large CDATA sections is very slow, so replace each
   6 ## CDATA section with a placeholder tag before parsing the document.
   7 my $xml = '';
   8
   9 my @cdata;
  10 $xml =~ s{<!\[CDATA\[((?>[^\]]+))\]\]>}
  11 {
  12   push @cdata, $1;
  13   '<cdata/>';
  14 }eg;
  15
  16 my $doc = eval { $parser->parse_string($xml) };
  17 if ($@)
  18 {
  19   $@ =~ s/ at \S+ line \d+$//;
  20   die $@;
  21 }
  22
  23 ## <document>
  24 my $root_node = $doc->getDocumentElement;
  25
  26 ## Replace the cdata placeholder tag with a CDATASection node.
  27 for my $cdata_node ($root_node->getElementsByTagName('cdata'))
  28 {
  29   $cdata_node->replaceNode(XML::LibXML::CDATASection->new(shift @cdata));
  30 }