The ultimate quest of this module is to produce from non-XML text
text, that will will most probably pass throught any XML parser one
could find.

Basic cleaning is just XML tag matching (for every opening tag there
will be closing tag as well, and they will form a tree structure).

When you add some extra parameters, you will receive complete XML
text, including XML head and root element (if none were defined in
text, then some will be added).

WWW: http://search.cpan.org/~pkubanek/XML-Clean/