GNOME Bugzilla – Bug 793416
Moving to GitLab
Last modified: 2021-06-14 15:18:34 UTC
Since GNOME is moving to GitLab I plan to move this module next week, so we can work on it and review in a better way. I'll work on a meson port soon, so would be good to be already on GitLab when that happens. Thoughts?
It is OK to move gnome-autoar to GitLab although it means I have to familiarize myself with GitLab this week. I still don't know what is the proper way to leave a reference to the merge request in commit messages when fast-forward merge is used because commits are created before the merge request. I assume that it is possible for me to always rewrite commit messages before pushing, so it doesn't seem to be a big issue. Meson port is nice, but I am going to keep autotools build longer than other modules because of the library path handling problem in meson. Meson assumes -Wl,-rpath can override LD_LIBRARY_PATH, but it is unreliable because linkers on some operating systems automatically convert -Wl,-rpath to RUNPATH. I will port the autotools build to AX_COMPILER_FLAGS in order to drop the dependency on gnome-common.
Sounds good, I'll migrate it to GitLab in two weeks. Regarding Meson, I didn't see this error, is it reported somewhere? But yeah I guess we can live with both systems for some time.
I just remembered only maintainers can set settings in the GitLab repo (although I'm admin of GitLab I prefer to use admin privileges as less as possible), did we agree I could be co-maintainer of autoar? If so, I'll update the doap file before moving to GitLab since we extract the data from there.
(In reply to Carlos Soriano from comment #2) > Regarding Meson, I didn't see this error, is it reported somewhere? But yeah > I guess we can live with both systems for some time. I discovered the problem when building colord on FreeBSD: https://github.com/hughsie/colord/pull/68#commits-pushed-0388f9b I reported the bug to meson after making a workaround patch for colord: https://github.com/mesonbuild/meson/issues/2881 There are also other similar bugs reported by other people: https://github.com/mesonbuild/meson/issues/1383 https://github.com/mesonbuild/meson/issues/1635
(In reply to Carlos Soriano from comment #3) > I just remembered only maintainers can set settings in the GitLab repo > (although I'm admin of GitLab I prefer to use admin privileges as less as > possible), did we agree I could be co-maintainer of autoar? There was no agreement but I think, yes, you can add yourself as a co-maintainer of gnome-autoar. > If so, I'll update the doap file before moving to GitLab since we extract > the data from there. Will the data extraction from doap file only be done at this time during the move, or it will be done every time the doap file is updated?
> I discovered the problem when building colord on FreeBSD: > https://github.com/hughsie/colord/pull/68#commits-pushed-0388f9b > Oh I see, FreeBSD, that might explain why I didn't hear about it. FWIW this year Meson might be doing GSoC under GNOME umbrella, so these things that are important for us at GNOME might be interesting. > There was no agreement but I think, yes, you can add yourself as a co-maintainer of gnome-autoar. Sounds good, thanks > Will the data extraction from doap file only be done at this time during the move, or it will be done every time the doap file is updated? Actually, it's done every hour.
Bug 771632 and bug 778804 are still open tickets here instead of in GitLab. Ting-Wei Lan, Carlos Soriano: Do you plan to copy these tickets over?
I think that this can be closed as gnome-autoar has been migrated to GitLab years ago...