[Gambas-bugtracker] Bug #1331: ODBC driver crack status 2: Result object are NULL now.

Tony Morehen tmorehen at ajm-software.com
Tue Aug 14 00:00:22 CEST 2018


Tobias, great advice.

I took a quick look at those commits and saw nothing that should give 
any problems.  I think the problems are more at the driver level.

On 2018-08-13 04:45 PM, bugtracker at gambaswiki.org wrote:
> http://gambaswiki.org/bugtracker/edit?object=BUG.1331&from=L21haW4-
>
> Comment #18 by Tobias BOEGE:
>
> Since nobody else seems to use the same database systems you do, your best bet to get this fixed
> is to put more effort into it on your side. And don't tell me that you don't have the time or
> your screen is too small or that Gambas is in cahoots with My$QL, winbuntu and the RAM waster
> industry. Remember this is a free software project and nobody owes you anything. When it comes
> down to it, you are the sole reason that something you need is not working. *You* are the only
> person using *truly* enterprise-grade databases here and you can't force others to use them too.
>
> Bisect the failure. Are you 100% sure that the three commits you mentioned above cause the
> difference in behaviour? Why those three commits? Why isn't it only one commit that introduces
> failure? It either works or doesn't work after a certain commit. That's the first thing I'd want
> to know if it affected my business. "It worked in Gambas 3.9" is also not a good enough hint for
> developers, 912 commits lie between 3.9 and 3.11.4.
>
> What you should do is fork Gambas, create a branch *off master* dedicated to the problem and
> first of all revert those three commits you suspect to be the culprit, then try to run your
> project with the modified Gambas. Be sure to check `gbx3 --version` before you run anything
> to confirm that you run the correct Gambas. Do not just check out v3.9.0 because it's already
> too far away from master. If your guess is correct, someone (not me :-)) will have an easier
> time reasoning about the problem. If your guess turns out to be incorrect, you have to use
> `git bisect` to find the first commit since 3.9 where your project breaks -- the single very
> first commit. Better yet, write a minimal project that indicates if queries work or not and
> use the power of `git bisect run` to save you time.
>
> Either that, or continue to try to find someone who wants to solve your problems for you.
> It might help if you'd at least offer to give them access to your database systems.
>
> Tobias BOEGE changed the state of the bug to: NeedsInfo.
>
>
>
> ----[ Gambas bugtracker mailing-list is hosted by https://www.hostsharing.net ]----




More information about the Bugtracker mailing list