D1044: bisect: add --abort flag

durin42 (Augie Fackler) phabricator at mercurial-scm.org
Mon Oct 16 21:00:30 EDT 2017


durin42 added a comment.


  In https://phab.mercurial-scm.org/D1044#18780, @quark wrote:
  
  > I think with `--command`, an option to restore to the original commit is a useful feature in `bisect` itself. If we cannot reach an agreement before freeze, could we push the first 3 commits (https://phab.mercurial-scm.org/D947+https://phab.mercurial-scm.org/D948+https://phab.mercurial-scm.org/D949)? They are safe and solve a real issue when log template is changed.
  
  
  I'm confused what you think the state of this is: the proximate goal of this series (as I understood it!) was to get run-tests able to restore to an original revision after doing its bisection, a goal which I wholeheartedly endorse.
  
  While I also agree that some built-in-to-bisect mechanism for "take be back where I was before bisection began" probably makes sense, I disagree that it should be named --abort. I've not looked at any of the earlier diffs in the series because @ryanmce requested changes, so much like a commented V1 on the list, I wasn't going to spend time on them until you sent a V2 (aka updated the diff, or otherwise cleared the "changes requested" state.) Does that make sense? I'm happy to land 947-949 assuming they're ready to go, but it looks like at least 948 needs some minor indentation cleanup?

REPOSITORY
  rHG Mercurial

REVISION DETAIL
  https://phab.mercurial-scm.org/D1044

To: quark, #hg-reviewers, ryanmce
Cc: kulshrax, durin42, ryanmce, dlax, mercurial-devel


More information about the Mercurial-devel mailing list