On Thu, Apr 06, 2006 at 01:43:30AM +0200, Rhialto wrote:
> Normally not perhaps, but if it disambiguates an ambiguous function
> call, it may help.

Ok. I missesd the overloading. So if you make sure that the casted
variant calls into the right/intended variant, the change is OK.
Otherwise I would suggest to explicitly pass a temporary object.

>  But since, on amd64 NULL is __null I do get the error, and
> since on i386 it is 0, people there don't get the error.

I don't think this explains it. In g++, NULL should always be __null.
If not, that would be a bug in some header file. (Try assigning NULL to
a function pointer for one case where 0 vs. __null makes a difference
even if sizeof(int) == sizeof(void*))

Martin