On Tue, Dec 14, 2010 at 4:32 AM, Pierre Schmitz <pierre@archlinux.de> wrote:

On Tue, 14 Dec 2010 09:55:22 +0100, Andrea Scarpino <andrea@archlinux.org> wrote:

...

Well, I didn't know that I can patch our scripts to enable [kde-unstable] support officially. Or better, I didn't think that I can explicitly add a [kde- unstable] case in that way because we should add a gnome-unstable and a foo- unstable case too. Or am I wrong?

I wont add any hack for a specific repo to the dbscripts. Every repos has to have the repos/$repo-$arch structure.

-- Pierre Schmitz, https://users.archlinux.de/~pierre

Andrea: If you don't want to mess up the trunk directory with the kde unstable stuff, you could at least have the files in the right places. Instead of having a kde-unstable directory in${pkgbase}, create a kde-unstable-i686 directory in the ${pkgbase}/repos/ directory where you'll do your packaging followed by manual commit and upload as I believe you already do for the kde-unstable packages. After that, copy the PKGBUILD and build files in the ${pkgbase}/repos/kde-unstable-x86_64 directory and commit them. Of course, for 'any' arch packages, just use ${pkgbase}/repos/kde-unstable-any. This way, the files will be in the standard locations were the scripts and tools expect them. So the current sourceball script will be working without any patching. And as most scripts/tools simply loops over the repos, it'll be easy to add kde-unstable support to them without adding exceptions all over the place.