atom:content @type="html" contents

Hi,

I want to process the atom:content element using xslt 2, however the
atom spec says that when @type="html", the content shall contain escaped
markup, if any.  In fact the document I'm dealing with actually has the
whole content wrapped in a CDATA section.

So is there a way to up-convert that content to xml such that I can
access it with xpath?

I'm considering using xpath2 string manipulation functions to achieve
what I need.  Please advise if this is the best way to go.

Thanks!

Peter Rushforth