Parse URI from RSS feed content with Hpricot

Yesterday, I wanted to parse rlslog movie feed for a small project, and I needed some time to figure out how to extract the right tag - named content:encoded - and found a magic ruby method to extract url from a string: URI:extract.

It's far simpler than any method I've used in recent years to do the same thing over and over again (see feevy.com), but I still havn't found yet how to only fetch the CDATA from the "content:encoded" Hpricot element... if you have an idea?