On Thu, 03 Aug 2023 at 07:24:17 -0400, Jeremy Bícha wrote: > glib 2.77.1 appears to be the GNOME 45 Beta release. It has > successfully migrated out of mantic-proposed in Ubuntu. I am unaware > of any remaining blockers to landing it in Testing.
GLib is usually solid enough that I'm OK with uploading betas to unstable in general (more so than GNOME Shell, for example). But, https://release.debian.org/britney/pseudo-excuses-experimental.html#glib2.0 says that if the experimental version landed in unstable now, it would be unable to migrate. Maybe some of those failures can be addressed by a new upstream of the dependent packages, and maybe some of them are flaky tests that just need enough retries and will eventually succeed - I haven't tried to analyze them yet. I think the blocker for GLib 2.77.x in unstable is looking into those. What I don't want is GLib 2.77.x sitting in unstable indefinitely, unable to reach trixie, because there are regressions but nobody is looking at them. > Would you be comfortable if we uploaded it to Unstable now? Last > cycle, the upstream maintainers were annoyed that distros didn't get > more user testing sooner. I think they were talking more about Ubuntu > and Fedora but maybe Debian can help there. I think there are two big things on the GNOME team's to-do list: * finishing GNOME 44 with the GNOME Shell 44 transition; * starting GNOME 45 beta stuff, for which GLib is the first step I don't think they necessarily have to block each other, but the release team are going to be upset if one of them gets entangled in the other and takes us a long time to sort out. The GNOME Shell transition seems to be basically ready, consisting of: * budgie-desktop * gnome-remote-desktop * gnome-shell * gnome-shell-extensions * mutter * and then when the rest is ready to migrate, kicking out any remaining packages listed in https://udd.debian.org/cgi-bin/bts-usertags.cgi?user=pkg-gnome-maintainers%40lists.alioth.debian.org&tag=gnome-shell-44 from testing Are you aware of anything else blocking those, or should I be opening a transition-tracker bug? smcv
