You could also look at HTML::TokeParser on CPAN. It is a great little module for parsing HTML elements. I Personally like to aviod trying parse something like HTML there are way too many gotchas for what i know of Regexs