Hi all, I'm not sure if my question should be addressed to the Tomcat or dom4j list, so this is going to both.
I'm running Tomcat for my company intranet and most of my Java applications return their results as XML so that I can style the data with XSLT. I am using dom4j to dynamically build up the DOM (dom4j Document) and I pass that to a Transformer in a servlet. Since I want to slice'n'dice these results with XSLT, I am keeping hold of the Document on the server using an HttpSession entry so it's sitting there as I apply the different stylesheets (specified as a servlet parameter). I've read that dom4j in-memory representations are 8-to-10 times the size of the XML text - so I am very aware that my solution is not scalable (my current project results are about 1MB on disc). Is there a "best practice" for this sort of thing? Should I be serializing the Document to a DB or something and just keeping a handle to that in the HttpSessions? Should I be using SAX in some clever way? I'm sure it's a common task to handle medium-sized results from Tomcat. Any input would be gratefully received. Best regards Chris -- Chris Ward, Horizon Asset Limited <mailto:[EMAIL PROTECTED]> Tel +44 (20) 7367 7028, Fax 7367 7029 --------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
