[Bug java/46983] New: Diagnostic about change in meaning of name in class misses some cases

["jason at gcc dot gnu.org" <gcc-bugzilla@gcc.gnu.org>]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=46983

           Summary: Diagnostic about change in meaning of name in class
                    misses some cases
           Product: gcc
           Version: 4.6.0
            Status: UNCONFIRMED
          Keywords: diagnostic
          Severity: normal
          Priority: P3
         Component: java
        AssignedTo: unassigned@gcc.gnu.org
        ReportedBy: jason@gcc.gnu.org
                CC: jason@gcc.gnu.org, rth@gcc.gnu.org


In C++, it is ill-formed, no diagnostic required, for a lookup to have
different meanings at the point of lookup and in the context of the complete
class.  G++ gives diagnostics about some instances of this, but misses this
one:

template <class T>
struct A
{
  typedef typename T::Y Z;
};

struct B
{
  typedef int Y;
};

struct C: B
{
  struct N: A<C> { };
  typedef char Y;
};

int ar[sizeof(C::N::Z) == sizeof(int) ? 1 : -1];

Here when we instantiate A<C>, C is not yet complete, so when we lookup C::Y we
find B::Y.  If we were to do the lookup in the full class, it would find C::Y
instead.