On Wed, Jan 20, 2021 at 3:23 AM Kevin Kofler via devel <devel@xxxxxxxxxxxxxxxxxxxxxxx> wrote: > > Daniel P. Berrangé wrote: > > > On Tue, Jan 19, 2021 at 03:18:41PM +0000, Richard W.M. Jones wrote: > >> On Tue, Jan 19, 2021 at 04:12:59PM +0100, Miro Hrončok wrote: > >> > On 19. 01. 21 16:06, Richard W.M. Jones wrote: > >> > >(2) Or adding: > >> > > > >> > >%package tar-filter > >> > >Obsoletes: %{name}-tar-plugin <= %{version}-%{release} > >> > > > >> > >which says that it obsoletes but doesn't provide a replacement. Note > >> > >however I'm not sure if this will work, because they won't have > >> > >nbdkit-tar-filter installed before the upgrade (it never existed in > >> > >F32), so will the Obsoletes ever be "seen" by dnf? > >> > > >> > Yes, it will be seen and respected. > >> > > >> > Note that it in most cases, it is recommended to obsolete a specific > >> > version-release rather than %{version}-%{release}. > >> > >> To be clear, I would find the last released nbdkit-tar-plugin version > >> in F32 (https://koji.fedoraproject.org/koji/buildinfo?buildID=1605446) > >> and write this? > >> > >> %package tar-filter > >> Obsoletes: %{name}-tar-plugin = 1.20.7-1.fc32 > > > > Yes, but use <= here, rather than =, as users might be updating from an > > older install. > > > > Arguably can leave off the .fc32 part too IMHO. > > But then, <= 1.20.7-1 will do the wrong thing, which is why Miro recommends > (and I agree with him) to use < 1.20.7-2 instead. Yes. This is also in line with the Packaging Guidelines, where there's a good description of this scenario: https://docs.fedoraproject.org/en-US/packaging-guidelines/#renaming-or-replacing-existing-packages Fabio _______________________________________________ devel mailing list -- devel@xxxxxxxxxxxxxxxxxxxxxxx To unsubscribe send an email to devel-leave@xxxxxxxxxxxxxxxxxxxxxxx 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/devel@xxxxxxxxxxxxxxxxxxxxxxx