Updating our tools

[Michael Pyne]

On Sun, August 9, 2015 09:58:26 Allen Winter wrote:
> On Sunday, August 09, 2015 09:35:06 PM Ben Cooksley wrote:
> > On Sun, Aug 9, 2015 at 3:15 AM, Allen Winter <winter at kde.org> wrote:
> > > On Saturday, August 08, 2015 04:59:49 PM Elvis Angelaccio wrote:
> > >> Sorry to bump this old thread, but it looks like Krazy still complains
> > >> about kdelibs4 errors even if an application is now KF5 based.
> > >> For instance consider again Kanagram:
> > >> http://ebn.kde.org/krazy/reports/kde-4.x/kdeedu/kanagram/index.html
> > >> 
> > >> Am I missing something? Is there another page showing KF5-related
> > >> issues
> > >> for KF5-ready apps?
> > > 
> > > If Krazy is running in the kde-4.x component, then it does use the KDE4
> > > checkers.
> > > 
> > > For now, Krazy only knows its looking at KF5 code if it's running in the
> > > frameworks 5 component in
> > > http://ebn.kde.org/krazy/index.php?component=frameworks&module=framewor
> > > ks5
> > > 
> > > I don't see kanagram listed in the frameworks5 component.
> > > 
> > > The KDE projects also lists kanagram in kde-4.x only, as far as I can
> > > tell.
> > > So first step is to get kanagram listed as a frameworks project.
> > 
> > To my knowledge there isn't anything on projects.kde.org which states
> > a project is kde-4.x only. Where is this information coming from?
> 
> the projects db
> 
> I see that kanagram master is kf5 based but is not part of frameworks.
> therefore I will need to invent a way to detect that a project is kf5 based.
> do we have an easy way to detect kde4 vs. kf5 projects?

The only way I'm aware of is to use the kde-build-metadata/logical-module-
structure metadata (this is used by kdesrc-build and the CI). This is much 
more annoying than a simple lookup but there are at least a couple of tools in 
kde-build-metadata that can be used to help.

I believe i18n tracks the 'unstable' and 'stable' branches for each module and 
'unstable' might mean "KF5" by now. This information *is* available with the 
project data directly. But maybe Albert or Luigi could better explain how the 
i18n infra is setup and what we can conclude from these variables.

Regards,
 - Michael Pyne