Hi Sandy,
> I didn't make myself clear in my earlier message. I agree with your
> proposed init4List and init4Union. The "BIG CHANGE" means we can only have
>
> init4List(XSSimpleTypeDecl itemDecl);
> init4Union(XSSimpleTypeDecl [] memberDecls)
>
> instead of
>
> init4List(String itemUri, int itemIdx)
> init4Union(String [] memberUri, int [] memberIdx)
>
> because we only use object reference instead of index.
Great, we both thought in different directions and came to the same conclusion.
I wanted getting the proper grammar from the resolver and getting proper
declaration using index from that grammar to be done by traverser. But
we dont have index now and its good if we are left only with that option, helps
us to remove dependency of Grammar from XSSimpleTypeDecl, as its very clean.
> I'm not sure about this one yet. I think we should stick to the current
> implementation, and postpone the decision after we switch to the new
> design. It's a minor implementation issue, and we can always change it
> after the fact.
OK, i agree.
> Sorry, I forgot that there is already that method on TypeValidator. But we
> still need one on XSSimpleTypeDecl, which will be called from
> SchemaValidator.
right now i am not able to think in ur terms, as validate() of
XSSimpleTypeDecl is calling isEqual() of different DV's classes.
and SchemaValidator will make a call to validate().
I would look into it tomorrow.
regards
Neeraj Bajaj
---------------------
Sun Microsystems, inc.
Ph.91-80-2298989 x87425.
---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
