Do you just want to extract the links which are stored inside the permalink tags? If it is as simple as that I would just use a regex to do it -

#!/usr/bin/perl -w

while(<DATA>) {
    if (/permalink/) {
        s|<permalink>(.*)</permalink>|$1|i;
        print;
    }
}

__DATA__
<permalink> Link A </permalink>
<tempNode> value1 </tempNode>
SomeText
<permalink> Link B </permalink>
<tempNode2> value2 </tempNode2>
<permalink> Link C </permalink>
[download]

Output

 Link A
 Link B
 Link C
[download]

If you are looking to parse the XML file cleanly for other purposes then you can check out XML::DOM. I don't have XML::TokeParser on my machine to check out code but XML::DOM should be easy to use. http://search.cpan.org/~tjmather/XML-DOM-1.43/lib/XML/DOM.pm

cheers

SK