[Commented On] D9256: backout: don't assume that tip as what we just committed

[danchr (Dan Villiom Podlaski Christiansen)]

danchr added a comment.


  In D9256#139031 <https://phab.mercurial-scm.org/D9256#139031>, @durin42 wrote:
  
  > Does this fix a defect? The test makes it seem like it does?
  
  Strictly speaking, yes it does, but like the other changeset, the nature of that bug is rather esoteric, and I discovered it by reading the code: If you back out a commit already backed out, and set the time (and so on) such that the resulting commit already existed, and is not tip, the output will be wrong. With an unchanging date, you cannot assume that the result of commit is always tip.

REPOSITORY
  rHG Mercurial

CHANGES SINCE LAST ACTION
  https://phab.mercurial-scm.org/D9256/new/

REVISION DETAIL
  https://phab.mercurial-scm.org/D9256

To: danchr, #hg-reviewers
Cc: durin42, mercurial-patches
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.mercurial-scm.org/pipermail/mercurial-patches/attachments/20201029/9e88898c/attachment-0002.html>