Hi David, I took a look at Javalution, but I'm a little bit skeptical. First, they completely changed the StAX API in an incompatabile way which was unnecessary. We can develop backward compatabile extensions to enable efficient access of ints/longs/etc. Not that there is that much performance value in that (at least from my limited tests). However, that limits our ability to use it within CXF - we could develop something to enable it, but that would be a lot of work and....
Second, I don't see any verification that their parser is actually faster than Woodstox on their website. Tatu, the woodstox author, did some benchmarking though. Woodstox is actually 20-25% faster than the Javalution parser. :-) Cheers, - Dan On 2/20/07, David Greco <[EMAIL PROTECTED]> wrote:
Hi all, looking at the Javolution initiative www.javolution.org, I was really amazed by the performances of their pull parser, so I wonder if it's possible to use the Javolution parser, I think it could give CXF an excellent burst in terms of performance. Regards, David
-- Dan Diephouse Envoi Solutions http://envoisolutions.com | http://netzooid.com/blog
