Re: C++ PATCH for c++/44909 (confusion with circular lazy declaration)

[IainS <developer at sandoe-acoustics dot co dot uk>]

On 15 Jul 2010, at 21:45, Jason Merrill wrote:

> On 07/15/2010 04:25 PM, Jason Merrill wrote:
> > I'm about to revert that patch.
>
> As promised, here's a completely different approach: rather than  
> rely on the incompleteness of the parameter type, we just disqualify  
> any constructor taking a different type.  This isn't quite right,  
> but it avoids a regression, and a proper fix will need more  
> discussion.

thanks, as you say that squashes the ObjC++ regression.

I'm trying to review what's being done in ObjC (on the basis that some  
of it might no longer be correct/optimal).

Iain.