Am Freitag 22 September 2006 16:55 schrieb Elie Morisse:
A new version of my refcounting patch, lighter and honoring Stefan Dösinger proposal.
Is there any reason why this patch wasn't applied? It looks ok to me, and it complies with the wording of the msdn(complex surfaces can only be destroyed by destroying the root).
It might be good to have a test case, but how can we test if something has been destroyed?