Glenn Marcy wrote:
> Perhaps another useful parameter would be the name of the entity
> that you are resolving.
That's what I was referring to. For example, if your document
contained "&entity;" which is defined in the following DTD:
<!-- file:///c:/xml/grammar.dtd -->
<!ENTITY entity SYSTEM 'entity.ent'>
then the call to the entity resolver would be:
resolveEntity("entity", null, "entity.ent",
"file:///c:/xml/grammar.dtd")
I was initially thinking of adding the name parameter so that
the entity resolver could know the root element name and return
different DTDs based on the root element name. But Sandy Gao
was mentioning how he wants to enable grammar caching and then
grammar resolution (if its not in the cache). And in that case
there is entirely different information that you would need to
know. However, at this time, I don't think we can satisfy
every use-case with a general interface.
So would adding the single name parameter work for most people.
And what about Jim Gan's suggestion of adding a "type" parameter?
--
Andy Clark * IBM, TRL - Japan * [EMAIL PROTECTED]
---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
