Community
Participate
Working Groups
For both, the 4.5.2 and 4.6 stream the baseline must be 4.5.1.
I think I am always confused about this, due to some early experience, where I was told the baseline always stays as the "main" release. I can't find it now, in a bug, so maybe I misunderstood at the time. So to be explicit, once a release comes out, it becomes the new baseline for any new work being done. Right? Is that true for performance work, as well as API and Deprecation work? That procedure makes sense to me ... just have a vague memory I had been told differently in the past. Thanks.
(In reply to David Williams from comment #1) > I think I am always confused about this, due to some early experience, where > I was told the baseline always stays as the "main" release. I can't find it > now, in a bug, so maybe I misunderstood at the time. So to be explicit, once > a release comes out, it becomes the new baseline for any new work being > done. Right? Yes, see https://wiki.eclipse.org/Version_Numbering#API_Baseline_in_API_Tools > Is that true for performance work, as well as API and Deprecation work? I'd say so, though at the end of the main release we would also want to compare to the previous main release. And actually that is also true for the other two reports since one is also interested in that delta.
I've made the change in the 4.6.0 (Neon) stream: http://git.eclipse.org/c/platform/eclipse.platform.releng.aggregator.git/commit/?id=3a5a20791df5f8f9ae8f161acaebc965b10c24f9 This should cover the API reports, performance tests, and p2 tests that use "previous version". I'll see how this does in a Nightly (or, I-build?) before applying to M-build stream ... in case I missed something or changed something too much.
It seemed to me the reports and labels were correct, so I have also put the same change into R4_5_maintenance.
