martinvonz added a comment.
In https://phab.mercurial-scm.org/D6255#92356, @mharbison72 wrote: > This might break `--pure` without `--local` in the annotate tests. No idea if that's a valid combination, but the buildbots (mostly) use that. In fairness, it seems that this combination had an error where `_filecommit()` was given too many arguments in the direct ancestors, so maybe the real breakage occurred in there. But there seems to be extra output here. > > https://buildbot.mercurial-scm.org/builders/macOS%2010.12%20hg%20tests/builds/828/steps/pure/logs/stdio > > It seems similar to `--pure` without `--local` failing in test-repo-compengines.t recently. > > https://www.mercurial-scm.org/pipermail/mercurial-devel/2019-April/130762.html It's apparently the pure file merge code (simplemerge.py, I think) that works differently. The common ancestor's content is: a a a The local side is: a z a The other side is: a a a b4 c b6 The pure version thinks they conflict and gives result: a z a <<<<<<< working copy: b80e3e32f75a - test: c ||||||| base a ======= a b4 c b5 >>>>>>> merge rev: 64afcdf8e29e - test: mergeb I'll see if I can work around it by changing the contents of the file a bit. REPOSITORY rHG Mercurial REVISION DETAIL https://phab.mercurial-scm.org/D6255 To: martinvonz, #hg-reviewers Cc: mharbison72, marmoute, mjpieters, mercurial-devel _______________________________________________ Mercurial-devel mailing list Mercurial-devel@mercurial-scm.org https://www.mercurial-scm.org/mailman/listinfo/mercurial-devel
