David Williams (djw@smooth1.demon.co.uk) wrote:
>Date: Sat, 26 Apr 1997 03:15:00 +0100
>X-Informix-List-Id: <news.37162>
>
>Jonathan Leffler (johnl@informix.com) writes:
>>Paul Mosser (pmosser@primenet.com) asks:
>>>Date: 25 Apr 1997 07:49:02 -0700
>>>X-Informix-List-Id: <news.37130>
>>>
>>>[...3 levels of embedded message removed...]
>>>
>>>I have heard that the "notfound" syntax (as in 'if status=notfound...') is
>>>no longer valid in 6.x. Can anyone confirm / refute that???
>>
>>Consider it refuted!
>>[...refutation removed...]
>
> Using 4GL 6.00 on an SCO 3.2.4cv2 box status=notfound always returns
> false even when no rows are found!!! 6.03 fixes this bug....
If this was a real bug (and I have no reason to dispute it, though I find
it difficult to believe that it escaped QA testing), it is different from
the question which I was answering. The question is about the *syntax*;
the syntax is still valid, even though the resulting behaviour is wrong.
And assuming that the bug really happened on the SCO platform quoted, it
would have been a major problem indeed. I'd also be very surprised indeed
if it was not fixed considerably prior to the 6.03 release -- I would have
expected a 6.00.UF1 or thereabouts maintenance release to fix the 6.00.UE1
problem. (If the 6.00 version was earlier than 6.00.UE1, then it was a
pre-release that should not have been in use one the 6.00.UE1 release was
made.)
I'm probably nit-picking unnecessarily.
Yours pedantically,
Jonathan Leffler (johnl@informix.com) #include <disclaimer.h>