Hi!

I have to parse a lot of xmls to convert them from xml to text, and I´m looking for the fastest way to do it.

I´m not sure whether to build a xslt and run it with Saxon, or parsing xmls with a Java implementation of Sax or Stax APIs and produce text output.

Please, I´d need your experience. Thanks!

Writing it in Java rather than XSLT is unlikely to give you more than a 10% improvement, and to achieve that you will have to write a lot more code and probably do some performance tuning.

There's no intrinsic difference between the performance of SAX and StAX parsers. Woodstox is probably the fastest conformant parser around, but it's probably only 5% to 10% faster than Apache Xerces on most workloads.

My recommendation would be to use Saxon with the default parser; if that's not fast enough for you, the cheapest way to improve the performance is to buy faster hardware.

__________________
Michael Kay
http://www.saxonica.com/
Author, XSLT 2.0 and XPath 2.0 Programmer\'s Reference

Thank you very much for the response accuracy, Michael.

Best regards!