> On 04 Jul 2016, at 14:34, Oliver Lietz <[email protected]> wrote: > > On Monday 04 July 2016 14:17:13 Konrad Windszus wrote: >> On 04 Jul 2016, at 14:03, Oliver Lietz <[email protected]> wrote: >>> On Monday 04 July 2016 10:04:10 Carsten Ziegeler wrote: >>>> Konrad Windszus wrote >>>> >>>>> It was changed from 2.1.1 to 2.2.0 in r1714786 (for SLING-5301), but >>>>> there >>>>> was a release in between. Therefore the change from 2.2.0 to 2.3.0 for >>>>> SLING-5665 seems right to me. >>>> >>>> Yes, sorry for the noise - somehow FishEye displayed the change as >>>> directly from 2.1.1 to 2.3.0 - I verified with svn that your change is >>>> totally correct >>> >>> Thanks. What about org.apache.sling.api.resource and changes in >>> SLING-5757? >>> Version was 2.9.0 before and is now 2.9.2 but should be 2.10.0, right? >> >> According to semantic versioning: >> >> A difference in the micro part does not signal any backward compatibility >> issues. The micro number is used to fix bugs that do not affect either >> consumers or providers of the API. >> >> In this change I don't see any backwards compatibility issue (although >> NonExistingResource.getParent() behaves differently with the fix). But code >> being able to deal with null in the past can almost certainly also deal >> with a non-null return value. > > Broken client code could stuck in a loop with this change when expecting to > get a null for parent sooner or later (and not handling non-existing > properly). When you reached the root, you will still get null.
> > So version should be 2.9.1 but not 2.9.2 (previous is 2.9.0). You are right, I fixed it with r1751265. > > O. > >> Any other opinions? >> Konrad >> >>> O. >>> >>>> Carsten > >
