Hi Tim, Thanks for replying here as well. For the sake of discoverability, this was discussed in a bit more detail on the bnd issue tracker
https://github.com/bndtools/bnd/issues/2112 Robert On Wed, 2017-08-16 at 15:24 +0000, Timothy Ward wrote: > I agree that there is no behavioural change, which is why the > baseline reports a micro change, rather than a minor or major > change. > > The important thing is that the class file has actually changed, it > used to be annotated with one thing, now it is annotated with > another. According to semantic versioning rules this change must be > reflected by a change in the version number, therefore the baseline > detection is doing the right thing in this case. > > Regards, > > Tim > > > > On 2 Aug 2017, at 16:20, Robert Munteanu <romb...@apache.org> > > wrote: > > > > Hi, > > > > I am trying to migrate a project from the bnd versioning annotation > > to > > the OSGi ones. I am stuck with the maven-bundle-plugin complaining > > when > > baselining runs since the annotations have changed: > > > > [INFO] --- maven-bundle-plugin:3.3.0:baseline (baseline) @ oak-api > > --- > > [INFO] Baseline Report - Generated by Apache Felix Maven Bundle > > Plugin > > on 2017-08-02T18:13Z based on Bnd - see http://www.aqute.biz/Bnd/Bn > > d > > [INFO] Comparing bundle oak-api version 1.8-SNAPSHOT to version > > 1.6.2 > > [INFO] > > [INFO] PACKAGE_NAME DELTA > > > > CUR_VER BASE_VER REC_VER WARNINGS > > [INFO] = ================================================== > > ========== > > ========== ========== ========== ========== > > [INFO] * > > org.apache.jackrabbit.oak.api changed 3.1.0 > > > > 3.1.0 3.1.1 Version increase required > > [INFO] ~ interface org.apache.jackrabbit.oak.api.QueryEngine > > [INFO] - annotated aQute.bnd.annotation.ProviderType > > [INFO] + annotated > > org.osgi.annotation.versioning.ProviderType > > > > > > I don't see why this is an issue since both annotations have CLASS > > retention policy so they would not cause any behaviour changes at > > runtime. > > > > Has anyone already run into this? Alternatively, are there any > > workarounds to convince the baselining check that this is OK? > > > > Thanks, > > > > Robert > > > > ----------------------------------------------------------------- > > ---- > > To unsubscribe, e-mail: users-unsubscr...@felix.apache.org > > For additional commands, e-mail: users-h...@felix.apache.org > > > > > --------------------------------------------------------------------- > To unsubscribe, e-mail: users-unsubscr...@felix.apache.org > For additional commands, e-mail: users-h...@felix.apache.org > --------------------------------------------------------------------- To unsubscribe, e-mail: users-unsubscr...@felix.apache.org For additional commands, e-mail: users-h...@felix.apache.org
