C++ PATCH: PR c++/7228

[David Edelsohn]

>>>>> Mike Stump writes:

Mike> I'd be interested in know if we can just check to see if the field 
Mike> exists first before grabbing it.  From a random 2 second search:

Mike> (I just worry about chewing up more memory, when maybe we don't have 
Mike> to.)

	I agree about not wasting memory, but there are lots of test of
IS_AGGR_TYPE.  I do not want this failure to arise again in other
settings.  I thought about LANG_TYPE_CLASS_CHECK testing that the pointer
is non-null, but I do not see any other checks that do that.  It seems
that LANG_TYPE_CLASS_CHECK is asserting that the lang_type structure
exists, which is why I proposed a patch adding it.  Assuming the check is
correct, it seems that the lack of the structure is a fundamental failure
exposed by the check.  However, this is where I do not know enough about
typename and its implementation in G++ to know if that structure really
should be present for all IS_AGGR_TYPE objects.

David