Anthony Williams has posted an excellent summary of the two major language design questions going into next month’s ISO C++ meeting in Batavia, IL, USA.
As we wind down C++0x, we are still working on an ever-decreasing set of open issues. Unsurprisingly, they’re in the newest features, as we bake them and see how they integrate with the rest of the standard.
One thought on “C++0x Current Hot Issues”
I’m still confused about a certain aspect of the override specifiers discussion. Why is there base_check? It seems that writing base_check makes sense for any derived class. Requiring programmers to write base_check explicitly bears the risk that people will never adopt the practice of using the override specifiers, because it requires actively activating the mechanism.
I am aware that activating the base check by default would break code. But this can be avoided by continuing to allow implicit hiding/override. Compilers will be able to emit a warning that an override specifier should be used in such cases.
As I see it, making override specifiers optional (but reminding the programmer via a warning) is the best way of maintaining backward compatibility, while still making the programmers use that new feature. No need for the base_check clutter, right?
Comments are closed.