Hi, I currently have a wrapper written around xerces and xalan. However, it ends up converting xalan nodes to xerces nodes. In some cases, I return NodeRefSet - but when I step through the nodeset, I again revert to the bride and return the xerces node. I guess this has a performance penalty.
Since, I have a wrapper anyway, is it not better to just use xalan all the way (from loading the document to creating new nodes, querying etc.). One hack I keep having to do is - If I start a document from scratch and later try to query it (xpath), it always fails, until I save and reload that document. Can this step be avoided if I use xalan exclusively? Thanks for any info/pointers. ===== __________________________________________________ Do You Yahoo!? Yahoo! Health - Feel better, live better http://health.yahoo.com
