Hi David.

It's not a bug because when you run the query a seconf time you are not just doing the same thing as the first time becasue the first time has changed your environment. There is now a new cursor that didn't exist and you are now in a new work area. That's why it works the second time. Technically, it's probably not working the second time either, it's just that the mistake that is being made happens to produce the right answer that time.

As Sergey mentioned before, you can't safely use table references with functions in an SQL select. You said you weren't but actually you are, you are using the inherent "current workarea" table reference that many functions provide.

VFP does have some bugs, but I'm afraid this isn't one of them :-)

Ian Simcock.



>Why would it not be considered a bug in VFP if this is the case? Like I said, when I run the query a second, sequential time in code or from the command line, the results are correct?
>
>It seems to me that what we are doing is just a workaround for a VFP bug. The workaround isn't really cludgy, so I am rather OK with it, but it does bother me that it is one of those unnoticable things.
>
>David