-
Suggestion
-
Resolution: Won't Fix
Recently we've noticed several old open PRs get updated without any human intervention.
This is not the best behavior given our development workflow, which is based around pull requesting from anonymous heads in a mercurial repository and does not use feature branches.
Worse, it appears that pull requests do not actually auto-update at all, instead all pending PRs get auto-updated when another pull request is merged.
For example, last night the following pull requests was merged:
https://bitbucket.org/enzo/enzo-dev/pull-request/206/adding-rotating-sphere-test-problem/activity
This caused the following pull requests to be automatically updated:
https://bitbucket.org/enzo/enzo-dev/pull-request/173/fixes-header-include-order-and-missing/activity
https://bitbucket.org/enzo/enzo-dev/pull-request/199/keep-initially-uniform-metallicity-fields/diff
https://bitbucket.org/enzo/enzo-dev/pull-request/202/adding-the-agora-isolated-disk-problem-and/diff
In all three cases this was not the desired behavior.
I don't think it's a good idea to let pull requests be updated without some sort of manual confirmation from the PR issuer. I would appreciate a way to revert to the old behavior.
- was cloned as
-
- Closed
| Form Name | |||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|
[BCLOUD-9361] Ability to disable pull requests auto-updating
Disabling the auto-updating feature of pull requests is essential to our Team's workflow. I'd like to see this issue re-opened and more seriously considered by the Bitbucket Cloud team.
What Robert said. There isn’t any more discussion because there’s nothing to add to this [!important] request.
I wonder if Alastair's comment was in itself an auto-updating pull request that could’ve used human intervention to make a proper decision here.
FYI, we’ve moved to github because of this issue not being addressed. Sadly having to drop HG for git.
Closing it would seem to indicate that it will never be fixed rather than just that it isn’t top priority, let’s just say that then? It seems like the issue has been known and ignored for ~4 years, and now the rationale that it is not worth fixing is that it has been known for ~4 years. Disappointing, as it is still a major issue, and the suggestion of auto-making a PR branch is relatively straightforward.
The lack of discussion, is because I believe that everyone know what needs
to be done. There is nothing else that warrants discussion.
A wontfix status makes me very sad.
Hi everyone,
Given the relative lack of discussion on this issue in recent years, and since the team is currently focused on higher priority items, I am closing this issue to indicate that we don't plan to address this in the near future.
(FYI – comments are not deleted when updates are pushed; they become outdated and can be viewed by clicking on the "
prior comments" button on a file's diff header)
Thanks,
Alastair
Bitbucket PM
This (along with the auto-deletion of comments during updates) makes the built-in pull request reviews almost unusable. This is also an issue if integrations are enabled. For example, if Code Collaborator has an update pushed every time a branch is updated, you'll end up with a TON of extra commits in the review. The normal dev cycle is push, push, push, squash, put into review, push, push, squash, push, squash, put into review again, so you can't trigger on a push or a squash, you need something else.
+1 on this. Thinking about switching to GitHub if nothing happens soon