Vivek Goyal wrote:
> What is fast forwarding commits from f21 to f20. I guess you are saying
> there are bunch of commits in master branch and you want to now apply
> those commits to f20 branch too?
"Fast-forwarding" is git jargon for doing a merge that simply accepts ALL
commits from master into f20, under the precondition that no non-master
commits have been made to f20. So it means you build the exact same thing
for both master and f20, and you never commit anything to f20 that isn't
merged from master (or the other way round, but you need to make sure (by
merging, if necessary) that the branches are in sync before committing to a
branch, or fast-forwarding won't work anymore, so it's easier if you always
do the merges in the same direction, which in Fedora is normally
master/Rawhide first).
For some packages, this workflow makes a lot of sense. For others, not so
much. It depends on whether you need to make changes to one branch that you
DON'T want in the other (and that it doesn't make sense to use "%if
0%{?fedora} > n" conditionals on). But note that there are only few cases
where you want to do release-branch-only changes, normally you always want
to do the changes either only in master (which implies maintaining the
branches separately) or in both.
If you really need to maintain the branches separately, then yes, your
versioning scheme makes a lot of sense and is in fact already recommended by
our guidelines!
https://fedoraproject.org/wiki/Packaging:NamingGuidelines#Minor_release_bumps_for_old_branches
(Yes, Michael Schwendt has already posted that link.)
If you really need to make more changes in the release than in master (which
usually should not happen, but sometimes there's a legitimate reason for
that), this is even REQUIRED, because you MUST NOT push n+1.fc20 when
Rawhide has n.fc21, as that would break the upgrade path.
Kevin Kofler
--
devel mailing list
devel@xxxxxxxxxxxxxxxxxxxxxxx
https://admin.fedoraproject.org/mailman/listinfo/devel
Fedora Code of Conduct: http://fedoraproject.org/code-of-conduct
