[Qgis-developer] Failing tests consider blockers

[Radim Blazek]

On Wed, Feb 19, 2014 at 10:53 AM, Jürgen E. <jef at norbit.de> wrote:
> Hi Radim,
>
> On Wed, 19. Feb 2014 at 09:14:43 +0100, Radim Blazek wrote:
>> On Sun, Feb 16, 2014 at 11:41 PM, Jürgen E. <jef at norbit.de> wrote:
>> > What I wanted to say is that there are a bunch of tests that fail because
>> > they are not up to date or fail for other reasons that are not related to
>> > actual bugs in qgis itself. For example render tests, which fail because of
>> > fonts or other differences (eg. removed renderers).  I think that are the
>> > majority of tests that currently fail.
>
>> Platform dependent tests (fonts) don't make sense for me and should be
>> disabled until get fixed (e.g. using font set shipped with QGIS).  Renderer
>> test should be updated and only real bugs will remain and those should be
>> blockers, I think.
>
> Well, there are platform dependant bugs and therefore there also could be
> platform dependant tests.  But those would need to be clearly marked as such
> and only run on the corresponding platform.
>
> Do we need to discuss that failing tests - or any other bug - should be fixed?
> I doubt that we do.
>
> I'm just objecting to planing ahead too much

Exactly, so why to insist exactly on release day planned 3 months ago?

> - let's first fix the tests and then decide further steps.

That won't work, we must be forced to fix them.

>> > Another thing is that we don't have the concept of actual blockers anymore.
>> > We release every four months what we have at that point.  Blockers are just
>> > highest priority for bugs.
>
>> Blockers do not block stable release?! I don't think anything like that was
>> ever explicitly said nor there is a general consensus about that. I read your
>> release plan again and I only found:
>
> We just have releases.  The plan is to release what we have on release day.  Of
> course that should be in the best possible shape, but whatever we do it will
> have known und unknown bugs.

Whatever we do it will have unknown bugs but does not have to have known bugs.

I am strongly against releasing with known blockers.

Radim