+1 makes sense to me

As far as I understand it, the odd numbered release number exist solely
because the the C++ build system can't cope with non-numerical -SNAPSHOT or
similar... so 0.21 is just a synonym for 0.22-SNAPSHOT

-- Rob


On 22 November 2012 20:59, Robbie Gemmell <[email protected]> wrote:

> Hi all,
>
> I recently made the nightly build of the Java tree that we run on Jenkins (
>
> https://builds.apache.org/view/M-R/view/Qpid/job/Qpid-Java-Artefact-Release/
> )
> start deploying our maven snapshot releases to the ASF snapshots repo (
> https://repository.apache.org/content/repositories/snapshots/). Since then
> we have created the 0.20 release branch and I put a Jenkins job (
>
> https://builds.apache.org/view/M-R/view/Qpid/job/Qpid-Java-Artefact-Release-0.20/
> )
> in place to do a nightly snapshot deployment of that.
>
> Due to the above we have now stopped deploying 0.19 snapshots and started
> deploying 0.20 and 0.21 snapshot artifacts. The result is that the 0.19
> snapshots will simply remain on the Nexus instance because we will never
> deploy 0.19 release artifacts and the automated cleanup process will simply
> leave them in place as a result. The same will happen to 0.21 snapshots
> when we make the 0.22 branch. In addition to generally cluttering thigns
> up, its also going to be annoying for anyone actually using the snapshots
> since they wont necessarily be aware of the change and could just use the
> stale output, or instead will just have to change their POMs twice (from
> 0.19 snapshot to 0.20 snapshot, then 0.20 release).
>
> As such, unless there is strong objection I plan to tweak the build system
> in the days ahead to simply avoid deploying the odd-numbered snapshots in
> future, moving it to deploy 0.22 snapshots now (then 0.24 snapshots after
> that) and cleaning the old 0.19 and 0.21 snapshots from the repo.
>
> Robbie
>

Reply via email to