On the caching thing, I'm not sure if the TraxTransformer caches the Template objects from Xalan, but that would be the way to go for speeding up that part of things. We used that technique on another project (pre-cocoon)
David
Neil wrote:
I work for a company that has a web service. I do development for that web service.
We have currently run into a stumbling block. We are using cocoon on our web server and when a user requests a page it is dynamically generated by our backend server as xml. We want the web server to use a different transform depending on the value of a single tag in the xml or a directive in the xml. We also want these transforms to be separate so that if there is an error in one the rest are still functional, aka not to compile them into one big transform. We also want cocoon to keep as many of these transforms precompiled and in memory as demand seems fit.
Does anyone have any ideas?
The pipeline seems to only allow for either always using the same transform or selecting one based on a request parameter but we need it to select based on a response value.
We are using a custom generator to send a request to the backend server and receive a response to pass down the pipeline.
I would appreciate anyone's help.
--------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
--------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
