Hello!
We have been working on a Java API for reading Wikipedia XML dumps for sometime and it's now reasonably functional. Check out:
http://code.google.com/p/wikixmlj/
*Features:*
- Easy access to important elements of a Wikipedia page - Also provides interfaces for Wiki text parsing. - Memory efficient - SAX interface for parsing - Lazy loading of files for DOM - Callback support with DOM - Directly operate on compressed wikipedia dumps (gzip/bzip2/native xml supported)
Best, Delip