Parsing very large XML documents (and a bit more) in java

Stax is the right way. I would recommend looking at Woodstox

Leave a Comment