> Or are you saying that you don't see a
> need for grammar caching? If it's the latter, then I find
> this comment extremely short-sighted.
No, I agree that the grammar caching is necessary. I have no doubt to it
at all. I just meant that I don't think the parser has to provide an
explicit pool, where grammars can be stored and accessed through some
kind of keys. I think caching can be done without using a pool.
> Going from the assumption that grammar caching is needed,
> how else would you locate a grammar that is defined with
> a target namespace?
I think locating a whole grammar is enough. In this way, grammar caching
is necessary but namespace-level caching is unnecessary.
regards,
--
Kohsuke KAWAGUCHI +1 650 786 0721
Sun Microsystems [EMAIL PROTECTED]
---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
