Pending tasks for the 4.9 release

[David Edmundson]

On Mon, May 21, 2012 at 4:24 PM, Martin Gräßlin <mgraesslin at kde.org> wrote:
> Am 21.05.2012 15:20, schrieb David Edmundson:
>>
>> On Fri, May 18, 2012 at 10:59 AM, Anne-Marie Mahfouf
>> <annemarie.mahfouf at free.fr> wrote:
>>>
>>> On 05/18/2012 10:55 AM, Andi Fischer wrote:
>>>>
>>>>
>>>> Am 11.05.2012 20:00, schrieb Anne-Marie Mahfouf:
>>>>>
>>>>>
>>>>> List of apps to test:
>>>>>
>>>>>  * Pairs
>>>>>  * Rocs
>>>>>  * kcm which changed?
>>
>>
>> Can I highly recommend we add "KWin's Window Decoration" KCM to the
>> testing list. It's been massively rewritten,
>
> I have to disagree on the point that it is massively rewritten. In fact
> there is only a minor change.
>>
I withdraw the word "massively". I was going purely on a visual
inspection of the various KCMs between 4.8 and current master, and
this has changed the most out of all the ones I looked at.

>> and it's using some highly experimental components from oxygen-qml.
>
> There is no such thing as "oxygen-qml". It uses a scrollbar written by Nuno
> specifically for this KCM.
>>
>> I had a play last
>> night and there are a massive amount of regressions in it,
>> particularly with respect to scrolling behaviour.
>
> The issues are known to the KWin team, documented on the KWin mailinglist
> and patches are in the pipeline including dropping the scrollbar. In
> summary: nothing to test here once the patches are in which will happen
> before the beta release :-)
>
I'm not saying it will be bad by the beta release, and by the sounds
of things you have everything totally under control, which is
fantastic.

However, it's still an area that has changed, and during the
development has (as you've acknowledged) had issues.

All I'm suggesting is that in the list of parts that testers should
focus on we encourage more people to play with the decorations KCM
than would do so otherwise, it's quite a niche KCM which would
probably otherwise be completely ignored. Testing something with a
larger audience after the developer fixes it never hurts, and they'll
(hopefully) be no bugs to report so it won't use up any of your time
:). I'm just proposing adding a single line to the wiki, not make a
full checklist/test plan or anything

Regards

Dave