Subject | Re: [IBO] TIBOQuery & kbmMemTable: access violation |
---|---|
Author | Lester Caine |
Post date | 2003-01-04T17:49:19Z |
>>And IBO is not fast enough?My point is that I started using kbmMemTab for much the same
>
> little by little I will have to tell you all the specs of my current
> work :)
>
> I am asking the way to go from NY to Washington.
>
> I need no advice to go to LA, instead :)
reason, but now I use IBO4 directly and don't have any 'time
problems'.
>>Would not help youI'm not using the kbmMemTab directly from TIBOQuery, I just
>
> instead, this is exactly what I need.
>
> If you really used kbmMemTable and TIBOQuery in IBO3 and Delphi3 and
> want to be helpful, please show me the few lines you used.
>
> If you have no D3 and IBO3 at hand, please post the lines as you
> remember them and I will test them.
copy the data from the original TIBQuery into a ready built
kbmMemTab table. Only 10-100 records. Using Builder5 and
IBO3 or 4, but those that I have switched to IBO4 have lost
the kbmMemTab stuff altogether - did not need the
complexity, and the improved buffering in IBO4 was faster
than IBO3.
I would suspect that the problem with going direct will be
one that I seem to remember seeing with IBO3 some time back.
You should be able to trap the error by single stepping into
the souce code of kbmMemTab and IBO3 - there were a few
points where an automatic operation was being run twice
because of the 'bug fixes' to get TQuery working at all, but
which TIBOQuery did not want to see. I had to dump years of
work done trying to make BDE work stabily at all when I
finally commited to IBO. Wish I had switched a LOT earlier <g>.
--
Lester Caine
-----------------------------
L.S.Caine Electronic Services