I traced this back to revision 1822 (64df422), and there's no explanation why we
would prefer to error out instead of waiting for the locks.
Details
Details
- Reviewers
marmoute - Group Reviewers
hg-reviewers - Commits
- rHGa9ff0742c8ea: locking: wait for locks in `hg cp` and `hg mv`
Diff Detail
Diff Detail
- Repository
- rHG Mercurial
- Lint
Automatic diff as part of commit; lint not applicable. - Unit
Automatic diff as part of commit; unit tests not applicable.
Event Timeline
Comment Actions
Maybe because you can't be sure the situation will be the same as the one before you ran the command ? However, normal race condition could still happens so that would not seems like a strong enough reason ? With the current code coverning the full command execution, I feel like this change is fine.