>>>Hey George,
>>>
>>>I was off fighting other fires (I know you can't
>>>relate to that :-) In any event, it was really
>>>more of a general question based on a recent
>>>problem that I'm having and the seeming C5
>>>epidemic out there.
>>>
>>>Having read a number of things at some of the
>>>sites mentioned in my previous post, it seems
>>>that there are still quite a few things which
>>>cause it (most of which seem to be related to
>>>OO or COM features).
>>>
>>>From your responses I gather that you're either
>>>not having any such problems, or you are still
>>>pretty confident about deploying even with the
>>>potential of encountering them. Is that accurate?
>>
>>Yes, I'm confident. I've found 6.0 to be more stable than any of its predecessors. I've run into the C...5 problem too, but in each case I was always able to trace it back to the programmer (me) not VFP. Were there issues in this regard? Yes, but SP3 seems to have address the vast majority.
>>
>>One of the things that I would urge you to do is post the code. That way one of two things will happen. If it's something that you're doing, then perhaps the folks here can help. If it is a real issue then Microsoft knows and can address it. The fact that Robert Green replied should be an indicator as to how serious they feel about it.
>>
>>Regards,
>
>The actualy statement that it blows up on is a
>simple IF VAL() = 0. The code is a little
>involved to post. Besides the fact that the
>error is intermittent. It is for a datascrubbing
>application where we're using DataTech's
>(formerly Group 1's) AccuMail product. I wrote
>a wrapper class that does oall of the address
>scrubbing including a call to the AccuMail API.
>I can run the program once and get the error or
>run it 10 times in succession and not get it.
>
>I'm reluctant to post anything because whatever
>I post would either be too much or not enough.
>
>I do indeed understand the significane of Bob
>taking the time out to respond and it is my
>plan of attack to beat tech-support up Monday.
>I have no idea of how they can approach this one
>but I'll keep y'all posted.
>
>Thanks for your help George.
>
> ...kt
Keith,

About the only thing that I can think of here is to look at what the VAL() function is trying to resolve. For example, if it's an object reference (especially from a server), then that might be the root cause. If so, I'd try simply to assign the property to a variable and check from there. Of course, this is an obvious SWAG on my part, but it would seem to be the most likely. I've never seen the VAL() function casuing a problem like this.