re: 4GL problem

Path: emory!wupost!cs.utexas.edu!tamsun!lstar31!rjsparks From: rjsparks@lstar31.tamu.edu (Robert Sparks) Newsgroups: comp.databases.informix Message-ID: <7508@tamsun.tamu.edu> Date: 10 Jan 92 22:56:28 GMT References: <1676AAC10.M22367@mwvm.mitre.org> Sender: usenet@tamsun.tamu.edu Organization: LoanSTAR In article <1676AAC10.M22367@mwvm.mitre.org>, M22367@mwvm.mitre.org writes: |> Is your C function returning correct values to the 4GL? Double check |> the resulting calculations to be sure. |> |> I'm not familiar with the "ret..." function (heck, I'm only passing familiar |> with C) but (excuse my niavete) are you popping as much as you're pushing? |> If not, the program stack is definitely getting hosed - which would explain the |> sudden and unexplained termination. It might also explain why you only see |> it after a large number of iterations, ya know? |> |> Push 3, pop 2, push 3, pop 2, push 3, pop 2... |> The calculations would still chugg along, but the results would not be what |> you want and you'd be leaving junk on the stack after each call. |> |> Is this anywhere in the ballpark? Thanks to everyone who responded by email and by newspost. Yeah, I've been playing long enough to look for botched stacks. I ran simplified versions of the program. I did the normal debugging routine. Now to save other people from burning time on my problem, its time for me to hang my head in shame. The problem was, in fact, in the C function. Its fixed now and everything is fine. You think at my age, I'd quit leaving myself open to divide-by-zero eh? Boom! rjsparks@loanstar.tamu.edu