[Qgis-user] Unstable 1.7.3 - one specific point is WMS

[Werner Macho]

Hi!

On 12/26/2011 04:36 PM, Jürgen E. Fischer wrote:
> Hi Werner,
> The bug itself wasn't even related to WMS.  The raster layer properties dialog
> just tried to access a radio button for a property that WMS layers don't have.
> Introduced by a fix related to the GDAL provider IIRC.   That was quickly
> backported before the breakage of master was even noticed.
>
> I think the wrong scale stuff in the last "stable" release, was also introduced
> by a quick backport of a commit that broke master.
>
Well ok - so it seems to be some kind of communication problem .. For
now we take the things marked as "bugfix" in master and try to backport
them to the "stable" release .. (assuming a bugfix fixes a bug).
I think it's kind of naturally that bugfixes are introducing new bugs ..
> So apparently we are backporting stuff too quick without proper review and
> testing.
>
correct .. as (it seems) we are not so many people to test ..
> That said, all the talk about quality assurance stuff is nice, but I (still)
> don't think that we have enough resources to properly maintain a stable branch.
>
> And if that's the case trying to might even waste some of it and do more harm
> than good to the project.
I don't see it that way .. I see even the efforts of trying to establish
a more or less stable branch with more releases very appreciated out
there.. There is still a lot of room for improvement and we are only in
the very beginnings of that QA way ...
> I also have the impression - I might be mistaken, that the people actually
> working on master, including me, therefore didn't participate much in the QA
> discussion.
yeah - that's a sentence I fully support ... We need you .. We need you
to tell us how you like backports be done - we need you to tell us which
commit actually can be backported - we need you to give us some kind of
workflow for the redmine - issue thing ..
It's us that have to support you as developer to make your life as easy
as possible .. but therefore we need info and some advices - otherwise
we try it our own way - but at least (IMHO) doing something is better
than doing nothing .. In that way there is always movement (which does
not necessarily mean improvement) and I am pretty sure we'll find a way
to improve that things..

There is an ongoing discussion (big parts also on IRC) about howto
correctly report a bug and help the developers quickly finding the most
annoying bugs (sure - for every person an other bug is more important)..
but as I said already - there should be a developer involved to work out
such a workflow and make it easy for a user to report a bug "correct" ..

Probably we can make a IRC meeting in the next few days about that topic?

Thanks for your mail - I hope we will solve that issues soon (shall I
file a bugreport on this? ;) )

Merry Christmas and relaxing free days with all best wishes for a happy
new year
kind regards
Werner