On 18/05/14 17:39, Urs Liska wrote: > Am 18.05.2014 17:54, schrieb James: >> On 18/05/14 15:38, Urs Liska wrote: >>> >>> As discussed previously I take it for granted that pondings don't need >>> formal review. >> >> uh .. they can still break master if you screw up your checkin (I know, >> I have done it). So yes they should have a _full_ patch test _at least_. >> >> After that then do what you will. >> >> James >> > > OK. I just did what I was told ... > > What to do now? I think that tests are done before staging is merged > into master?
No the tests are done before anyone pushes anything to anywhere (that is what trackers are for). Whether a developer wants to wait for the full review process is something they can decide for themselves (some - i.e. most - patches are beyond my knowledge so if someone says 'I'm pushing this now' then I'm not going to stop them. The point is that checking into the the tree (staging or otherwise) without a test is irresponsible. By all means you can run patchy yourself if you wish (I know David K does) but that requires a tracker and the proper git-cl / patch-new workflow. > Currently my two commits are the only ones ahead of master. > Should I reset staging to master and upload the patch for review or > should I leave it for now and do better next time? It's merged now. My desktop and probably some other machines (in Paris?) sit and quietly check and merge Staging to Master periodically. James _______________________________________________ lilypond-devel mailing list lilypond-devel@gnu.org https://lists.gnu.org/mailman/listinfo/lilypond-devel
