[Aptitude-devel] Removing GTK+, Qt from the tree [was Re: Bug#537858: aptitude: comments on i18n]

[Axel Beckert]

Hi,

sorry for my silence the past few days. I caught myself a cough (maybe
at FOSDEM, maybe at work) and sleep a lot currently.

Manuel A. Fernandez Montecelo wrote:
> 2014-02-09 Daniel Hartwig <mandyke at gmail.com>:
> > On 8 February 2014 23:29, Manuel A. Fernandez Montecelo
> > <manuel.montezelo at gmail.com> wrote:
> >> - I marked as not translatable the ones in GTK GUI and fixed the one
> >> which was not using ngettext, but GTK is disabled since long ago and
> >> never was in good shape.  Same for Qt GUI.
> >>
> >> Is there a common/easy way to disable source code directories for
> >> gettext when generating the .pot files, so at least new translators
> >> don't waste time to translate things that will possibly not be enabled
> >> again?

Good question.

> > It is clear by now no one is going to update these interfaces.

Yeah, nobody even argued about aptitude-gtk's removal from the package
builds.

> > I think the time has come to finally remove them from the tree.

Indeed.

> Now that they've been several years inactive, I would prefer to wait
> until after the next freeze or so.

Since they're not built anyway, where's the relation to the freeze? Or
do you fear that the removal will be bumpy and cause side effects?
Can't imagine any besides FTBFS which would be obvious and should be
quickly fixed.

I rather think that the 0.7 bump would be a good place. (Of course it
could come after the freeze, too.) But I don't care much actually. I'd
be even ok if we leave it in their. But I do see a that it might ease
development.

I just would tag the commit of the removal accordingly (e.g. as
"aptitude-gui-removal") so that it can be easily found in case someone
really tries to get it working again later. Reverting that commit
would then be a start for such a project.

> I think that this would be the last release with Qt4 and GTK2, at that
> point the early prototypes would require first adapting to the new
> major versions of frameworks before continuing, which I think that
> it's not a small undertaking.

Indeed.

> LXDE project is moving from GTK2 to Qt instead to GTK3 because since
> they have to make huge changes for GTK2->3 anyway, they prefer to move
> to Qt.

Yeah, AFAIK the same counts for the evolution from Maemo to Meego, Mer
and SailfishOS.

> So if nobody comes around in the next few months, it will be even more
> clear that continuing with the early prototypes might not even have
> any advantage from starting afresh,

Good point.

> and the code will be present in repositories to take as an example
> anyway.

Yeah, they're not lost after the removal.

		Regards, Axel
-- 
 ,''`.  |  Axel Beckert <abe at debian.org>, http://people.debian.org/~abe/
: :' :  |  Debian Developer, ftp.ch.debian.org Admin
`. `'   |  1024D: F067 EA27 26B9 C3FC 1486  202E C09E 1D89 9593 0EDE
  `-    |  4096R: 2517 B724 C5F6 CA99 5329  6E61 2FF9 CD59 6126 16B5