On Thu, 27 Nov 2003, Christopher Lenz <[EMAIL PROTECTED]> wrote:
> With automatic typo correction ('x' -> 'n') in the introspection
> code ;-)
Cool, eh?
> Stefan Bodewig wrote:
>
>> via f.getFragment()
>
> which will return a org.w3c.dom.DocumentFragment object,
exactly.
> Any reason why the introspection doesn't look for that interface
> directly, instead of XMLFragment?
XMLFragment doesn't implement DocumentFragment, it just builds one.
To make your suggestion work, it would be add(DocumentFragment)
instead and the XMLFragment class would have to be typedeffed in
addition to implementing DocumentFragment.
> So the namespace URI and the local name are currently actually
> passed to the DynamicConfigurator in the form
> namespace-uri:local-name?
Actually, I thought so from reading the code, but given that the tests
still pass when I add some namespace information to it, I may be wrong
...
> I wasn't aware of that. Wouldn't it make more sense to add a
> DynamicConfiguratorNS interface, possibly extending
> DynamicConfigurator, that would pass all the required information
> explicitly?
Probably yes.
Stefan
---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
