On Thu, 01 Oct 2009 11:25:03 -0500, Andrei Alexandrescu <[email protected]> wrote:
>I am becoming increasingly aware that we need to provide some means to >define certain members (data and functions) for each class as if they >were pasted there. > I am glad you have finally bumped into the problem. It has been waiting for a solution for quite some time. For example, the upcoming reflection mechanism in QtD (which uses its own metaobjects instead of the limited classinfo) requires manual insertion of a mixin into derived classes unless the mixin is inserted implicitly when a signal/property/etc is declared. A couple of votes for the feature.
