Like in this HTML snippet:
<div class="c1"><span class="c2"><b class="c3"/></span></div>
The expected hierarchy structure of b
element should be: div.c1 span.c2 b.c3
There is the parentNode
method in XML::LibXML::Node, which simply returns the parent node. So you can locate your node of interest (b
)
and then 'drill' upwards to the top of the tree, collecting suitable information about nodes. For the desired element.class format:
use warnings;
use strict;
use feature 'say';
use XML::LibXML;
my $xml = q(<div class="c1"><span class="c2"><b class="c3"/></span></div>);
my $doc = XML::LibXML->load_xml(string => $xml);
my @hier;
my ($node) = $doc->findnodes('//b'); # only first such node assigned
unshift @hier, join '.', $node->nodeName, $node->getAttribute('class');
while (my $parent = $node->parentNode) {
last if $parent->nodeType == XML_DOCUMENT_NODE; # top, <?xml ...
unshift @hier, join '.', $parent->nodeName, $parent->getAttribute('class');
$node = $parent;
}
say for @hier;
The getAttribute
method is in the XML::LibXML::Element class.