RFC: Calligra press kit

[Jaroslaw Staniek]

On 7 December 2011 23:53, Markus Slopianka <markus.s at kdemail.net> wrote:
> On Mittwoch 07 Dezember 2011 22:54:11 C. Boemann wrote:
>> I think it looks generally very nice. The part about version 3.0 is
>> incorrect however. We plan for at least 2.5 and 2.6 before 3.0 neither of
>> which will include support for new qt or kde. Our release schedule is
>> every 4 months from now on
>
> Well, until Jaroslaw changed it very recently,
> http://community.kde.org/index.php?title=Calligra&action=historysubmit&diff=16611&oldid=16017
> said nothing about further 2.x releases. ;-)

I did that exactly to have a placeholder for proposed schedules and
especially release plans.

> Anyhow, Calligra (and KOffice before) isn't really the most successful project to keep
> roadmaps and schedules. I don't believe in a 4 months cycle before I see it. ;-)

If we do not split feature plans to more frequent releases, we'll not
move much further...

> That said with 2.4 pushed to January, maybe February, even when keeping the release cycle,
> 2.6 would be released in September/October which is already months after the KF5 release
> in July/August.
>
> The recent thread started by Boudewijn about Qt3Support removal sounded a lot like Qt5
> support is coming earlier rather than later with only Kexi being in question. (I suggested
> porting the Qt3Support parts required by Kexi to Qt5 but didn't get a reply.)

I think you already get the clue - this is planned but not top priority.
If it works, do not touch.

I recently found out that things are even worse. QtScript, as I know
it, slowly disappears in Qt 5 because of the port to Google's V8
backend, the API going to be a bit different. Such things interfere with
planning in projects like Calligra. So I discovered that I have to
delay all the scripting tasks (not just for Kexi forms/reports but for
generic ODF stuff too) to minimize losing of the resources.
No wonder then, that the transition may take more time.

One can shorten the time by contributing more of course :)

> Frankly: No way I believe you telling me about a Qt4-based Calligra 3.0 release in
> January/February *2013*, roughly a year after Qt4 was EOLed.
> Maybe at some point that was agreed on that schedule but didn't know about the Qt5 roadmap
> then.

Boemann said about no use of Qt 5 in 2.5 and 2.6. Nothing definitive
is declared for 3.0.
This is normal for projects where workforce per app and lib is hard to
estimate, right?
Sometimes relatively small fix needs calendar weeks to materialize...

> Boudewijn's thread (with replies also by Cyrille) is definitely the more recent
> information. And as Cyrille wrote there: When depending on Qt4, you have to do the
> maintenance work for it, too. Somehow I don't think anybody is interested in that...

All depends what is more costly, port to Qt 5 or keep the Qt 4 branch
alive. This can differ from case to case.
If you know the industry, Qt 4 will be on the market quite many years from now.

> Well, my document doesn't need to be finished now. If a clear roadmap for 2012 hasn't been
> decided until a few days before the 2.4 release, the outlook part can also be removed.
> It's nice to have but non-vital.

Yes, just let the developers stabilize the apps and libs; transition
can take less time given the Qt 4->5 change is not big from their
perspective and they (as in Kexi tables) find resources to find
temporary solutions to make unported code work with Qt 5 without file
format/feature degradation. There are certain priorities - 'new' is
sometimes enemy of 'stable'.

-- 
regards / pozdrawiam, Jaroslaw Staniek
 http://www.linkedin.com/in/jstaniek
 Kexi & Calligra (kexi-project.org, identi.ca/kexi, calligra-suite.org)
 KDE Software Development Platform on MS Windows (windows.kde.org)