Neil Graham wrote:
modify XMLLocator. In fact, I'll admit to a bit of curiosity as to why
XMLLocator is an interface at all; seems like it's sufficiently simple and
generic that it might have been better had it been made a class (or
abstract class). Oh well; water under the bridge.
I think it needs to be an interface. It's just a lot easier for
a parser implementation to manage it's own data and then expose
that through an interface than require that the implementation
manage the data through the interface. Does that make sense?
But, even knowing I'll have to update all of my parser impls, I'm
much happier with making this change to the existing XMLLocator
interface.
Also, I wonder if it's time to follow the convention we seem to have
adopted in the manifest file of speaking of XNI as having versions in our
documentation. The manifest file says XNI is at version 1.1, and after all
these changes I guess it'll go to 1.2; but I don't believe our
documentation makes this clear anywhere.
I agree. We should have done a better job of documenting XNI
changes through time. Oh well... We can get away with it, for
the most part, because XNI is an internal interface. But we
should try to be better moving forward.
--
Andy Clark * [EMAIL PROTECTED]
---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
