Stripping illegal characters out of an XML document

Hi there. I'm working with RDF/XML that is strict on what characters are allowed within the elements and attributes. I was wondering if anyone had a script that processed a string and replaced all illegal-characters with their HTML code, for example "&" is converted to & and " to ". It should also work for characters like "é". It would be possible to process the strings before they are inserted into the XML document - if that is easier. If there isn't a script that'll do this I'll have to write my own - which could be quite time consuming ;( Thanks Dan