On Sun, Sep 27, 2026 at 01:04:33PM +0200, Fabio Valentini wrote: > On Sun, Sep 27, 2026 at 10:53 AM Zbigniew Jędrzejewski-Szmek > <[email protected]> wrote: > > > > On Sun, Sep 27, 2026 at 01:22:11AM -0700, Adam Williamson wrote: > > > On Sun, 2026-09-27 at 07:54 +0000, Zbigniew Jędrzejewski-Szmek wrote: > > > > On Sat, Sep 26, 2026 at 04:55:13PM -0000, Leigh Scott wrote: > > > > > Thanks a lot for messing up cinnamon git, I always merge from rawhide. > > > > > > > > I assume that's meant to be passive-aggressive, not sincere. > > > > > > > > The f45 branch tip was equal to rawhide and a build had been done from > > > > that commit. I needed to bump the version in f45 to do the build in > > > > f45, so I made a single commit on the f45 branch. What other approach > > > > would you have preferred? > > > > > > > > Zbyszek > > > > > > FWIW, in this situation I'd normally bump Rawhide too, both to keep the > > > branches synced and because technically the F45 package should never > > > have a higher version than the Rawhide one (though this is *less* > > > important than it used to be). > > > > The maintainer can still do this, rawhide can be fast-forwarded to > > f45. Nothing stops the maintainer from doing this if it's appropriate. > > I don't think I should be doing those extra builds as a non-maintainer. > > It's quite a bit of extra context to figure out, and at least in > > case of cinnamon, Leigh does builds every few days, so the temporary > > lag behind rawhide should be OK. > > Theoretically yes, but that requires you to notice that the f45 branch > is ahead of rawhide through no fault of your own. > > If you do the "normal" workflow of updating rawhide first and then > merging (if necessary) to other branches, as soon as you notice that > "git merge rawhide" gives you an error (or an unexpected merge > commit), it's too late, because you have already pushed a diverging > commit to rawhide. > > I agree that the commits should have been pushed to both rawhide and > f45 (and just not *built* for rawhide again) in this case to avoid > this issue.
But this adds a bogus change to rawhide, that'd actually be shown in the changelog. This is not nice. That's is why I always update all the branches first and push with --atomic. This way one gets to notice those problems your describe and fix them locally in the most appropriate way. Zbyszek -- _______________________________________________ devel mailing list -- [email protected] To unsubscribe send an email to [email protected] Fedora Code of Conduct: https://docs.fedoraproject.org/en-US/project/code-of-conduct/ List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines List Archives: https://lists.fedoraproject.org/archives/list/[email protected] Do not reply to spam, report it: https://forge.fedoraproject.org/infra/tickets/issues/new
