On Dec 28, 2009, at 1:51 AM, Chandler Carruth wrote:
> Attached is a patch that seems to fix PR5751 by looking through using
> decls when resolving overloaded candidate functions. Does this look
> correct?

We don't care about the declaring context when collecting associated 
namespaces;  that's a miswording in the standard.  I've made a comment in the 
bug to that effect.

> Should this perhaps be factored into UnresolvedLookupExpr?

No;  clients should be forced to handle using decls.  If the typesystem let me 
assert this, I would do it.

> In ResolveAddressOfOverloadedFunction, I'm applying the logic even for
> member functions, is that correct?

Yes, I think so.

Looks good overall.

John.
_______________________________________________
cfe-commits mailing list
[email protected]
http://lists.cs.uiuc.edu/mailman/listinfo/cfe-commits

Reply via email to