[cfe-commits] [PATCH] Improved handling of member pointers

[Douglas Gregor]

On Nov 1, 2009, at 1:11 AM, Peter Collingbourne wrote:

> Hello,
>
> This patch modifies Sema to store the unresolved class type
> in MemberPointerType's Class field, and fixes a bug where an
> assertion would fail if the last element of a member pointer type's
> nested-name-specifier is an identifier.

Very nice, thanks! I've committed this here:

	http://lists.cs.uiuc.edu/pipermail/cfe-commits/Week-of-Mon-20091102/023095.html

The only change I made was to replace the cascading if's (to decompose  
the NestedNameSpecifier) with a switch statement, so that any future  
extensions to NestedNameSpecifier will trigger a warning here.

	- Doug