> Thinking further about reasons it didn't improved for you - It may > happen that InputStream.read + DocumentBuilder.parse for huge response > takes so much of the whole process , so other parts just do not > important and any difference is just a measurement error. Most changes > were done in order to rather improve continuous operations running many > times.
I agree with that, too. I've been meaning to (but haven't had the time) to try to characterize the time spent in the client receive code path. It would be interesting to know, for various response payloads, what amounts or percentages of time were spent... - in the reading steps before the DOM parser - in the DOM parser itself - in the deserialization after the DOM parser Knowing this stuff would give us a good hint whether it would make any great difference do implement a SAX-based deserialization model, for example. [[As a side note, it seems to me that doing a SAX deserialization model is no more difficult than the DOM-based model we have now. I completely understand why nobody has stepped up to that as an *additional* effort for Apache SOAP, which is more or less in pure maintenance mode. It's a bigger mystery to me why the Apache Axis folks didn't do SAX parsing from the start and not bother with DOM.]] -- [EMAIL PROTECTED] (WJCarpenter) PGP 0x91865119 38 95 1B 69 C9 C6 3D 25 73 46 32 04 69 D6 ED F3 -- To unsubscribe, e-mail: <mailto:[EMAIL PROTECTED]> For additional commands, e-mail: <mailto:[EMAIL PROTECTED]>
