MyBB Community Forums

Full Version: [Feedback] More rigorous PostgreSQL testing
You're currently viewing a stripped down version of our content. View the full version with proper formatting.
Hey everyone,

It seems to me that ever since PgSQL has been supported in MyBB it has been pretty much neglected. Release after release I find that many parts of the ACP still throw SQL errors on PgSQL systems. I seriously think there needs to be in-depth PgSQL testing in MyBB BEFORE "stable" releases are made because clearly these releases are not stable while running anything other than MySQL.

I will say that many of the issues which I have run into are related to primary keys and sequences. As usual I will continue submitting/following bug reports and their respective code revisions to make sure things are getting fixed (which they are). However, I find it very troubling when I can't simply take a stable release of MyBB and run it without many problems. Just a thought.

In addition, I am more than happy to contribute to the project in whatever way possible.

Alec
One of the main gripes I have with PostgreSQL is that it is such an unintuitive SQL language and it's complex and confusing to do something simple on MySQL sometimes.

That being said, yes, more testing needs to be done with PostgreSQL. What environment setup should I ask SQA to setup to test on for PostgreSQL testing?
Well I am running both 8.3.x and 8.4.x series servers and they both have basically the same problems. I am finding though that since I installed a fresh copy of 1.6, its been basically unusable so I'm not sure quite what to do at this point. All the errors I see are related to duplicate primary key issues and sequences as I mentioned earlier.
Have you reported these bugs to the development tracker? http://dev.mybb.com/
Yeah, I'm working on it.
(2010-09-27, 08:09 PM)maddog39 Wrote: [ -> ]All the errors I see are related to duplicate primary key issues and sequences as I mentioned earlier.

It's a problem I've noticed too, but put it down to my setup (which is pretty wierd). It's very strange that the sequences are set too low.