pcplantdb

[Richard Morris <webmaster@pfaf.org>]

John Schinnerer wrote:

> Aloha,
>
> > Not quite sure what you mean by "regression in next build".
> >
> > This seems to indicate that we need a list of "things to check"
> > with each build. Is this the general idea? Keep the bug in system so 
> > we can review it later?
>
>
> This is how the basic dev/test cycle works:
>
> Developer(s) release a build.
> Tester(s) test the build.
> Bugs are found and entered against that specific build.
> Developer(s) try to find and fix open bugs - in development code, *not* 
> in any release code!
> When a developer has made a fix, they set the bug status to 'resolved' 
> and assign it back to the tester who entered it.
> When the next build is released, the tester goes through all bugs marked 
> 'resolved' and assigned to them and then "regresses" them (tries to 
> reproduce them again) in the new build.
> If fixed, tester closes bug.
> If not fixed, tester sets status back to open and assigns back to 
> developer to try again in next build.

Now I understand!

> I am gathering that I need to hustle to the extent possible on getting 
> testing How-To info linked on the testing page on the wiki...

Yep, that would be great. A page on issue tracker usage, explaining all 
the different catagories could be good as well.

Rich