>>Sorry, but if it doesn't fire with VFP's error routine it should not fire with mine. That makes SET SAFETY OFF inconsistent. I can't agree it's a NABAF (Not A Bug, A Feature!)
>>
>>Just my two cents...
>>
>Kevin,
>
>You can disagree all you want, but Microsoft has said that this behavior is by design. You don't have to like it, but if it works the way it is designed to work then it is not a bug. lord know VFP has enough of those already. Myself, I am very happy it works as it does because it lets me set safety off so a bunch on dialgos don't impose themselves on my interface and I can still trap for missing indexes. I can't imagine when I would want a table that has a structural index to be openned without it.
>
>If I ever wanted to do that then I just trap 1707 in my error handler and do nothing about it.

Jim,

I agree that you want it to trigger an error when your error hanlder is active. It seems, though, that it should also trigger an error when you don't have an error handler (like almost all VFP errors do). It may be by design, but I think this is an example of inconsistent design.