>David,
>
>It is a difference in VFP, but it's not really that big a deal. The only places where this is an issue is for the event methods where the VFP baseclasses actually do something (like KeyPress, MouseDown). The VFP designers gave us this shortcut for this default behavior to simply occur at the end of all of our code. NODEFAULT allows us to prevent this from happening.
>
>There'd probably be near as many problem posts from people complaining why the keys they type are not showing up in their textboxes anymore. I did C++ before VFP and I at first preferred the explicitness of calling the ParentClass code using ::, but now I'm really happy with the way this detail is just "handled" by VFP.

---

As I said, I understand the advantages of this design decision. Fact is, having the base default behavior occur is probably desirable 99.99% of the time, so this divergence from the pure class inheritance protocol could simply be modeled as inherited from some hidden virtual base class from which all VFP base classes descend. (....ahhh! my consistency hobgoblin feels better now!)

However, for this kind of deviation from classic inheritance protocol, I believe that the VFP documentation should be very clear and specific about what's going on (which it is not). You really have to have HackFox in hand (see 'A Gala Event') to get the full explanation of what's going on in event behavior inheritance vis-a-vis events inherited from a base class.