version-release for different branches
On Wed, Sep 12, 2012 at 10:51:32AM +0200, Stanislav Ochotnicky wrote:
> > FWIW, I beg to differ. There is no problem at all if you skip over
> > release number 3 in some branch, but having a specfile changelog that
> > lies about what happened in that branch is a bad idea. It'll lead to
> > confusion later.
> If the commit was anything else than raising release tag and adding a
> changelog entry I'd completely agree. As originally stated, I am aware
> of opinions like yours and even if I don't agree...it's up to maintainer
> to decide in the end (unless one of us brings the issue in front of
> FPC/FESCO :-) ).
+1. In this case, the changelog entry simply serves to explain what happened
to the missing number -- there was a build with no changes made. If it
described changes, that'd be a different issue. (It happens that that
particular build didn't really happen on this branch, but from the POV of
what a rpm changelog is useful for, that's irrelevant.
Matthew Miller <firstname.lastname@example.org>
Senior Systems Architect -- SEAS Computing
Harvard School of Engineering & Applied Sciences
packaging mailing list