You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Unreleased pairs of commits where the first commit would cause a version change and the second commit is a revert of the first do not cause any changes in the version.
Unreleased pairs of commits should be ignored.
If the original commit was released, would you expect the revert commit to trigger a second release? If so, what level of version bump?
If the original commit was released, would you expect the revert commit to trigger a second release? If so, what level of version bump?
The behavior of the JS implementation is to trigger a patch release for a revert commit that's already in a release.
I'm not sure whether that's the right level of bump for all situations. An argument can be made that any reverted released commits that triggered a release (with the exception of perf) are a breaking change since code written between the commit and the reverted commits can break if it's depending on a bug fix, feature introduction or breaking change.
@cpcloud wired place to ask, but I can't find a way to reach out for you otherwise: do you have an other package recommendation which handles semantic-releases better?
@cpcloud wired place to ask, but I can't find a way to reach out for you otherwise: do you have an other package recommendation which handles semantic-releases better?
It has been 90 days since the last update on this confirmed issue. @python-semantic-release/team can you provide an update on the status of this issue?
It has been 90 days since the last update on this confirmed issue. @python-semantic-release/team can you provide an update on the status of this issue?
Activity
cpcloud commentedon Jan 4, 2022
It looks like the logic to do this in the javascript implementation is here https://github.com/conventional-changelog/conventional-changelog/blob/21b73f974adc4490a318d9c892055655cb82c538/packages/conventional-commits-filter/index.js#L14-L57
danth commentedon Jan 4, 2022
Unreleased pairs of commits should be ignored.
If the original commit was released, would you expect the revert commit to trigger a second release? If so, what level of version bump?
cpcloud commentedon Jan 4, 2022
Indeed, but they are not at the moment.
The behavior of the JS implementation is to trigger a patch release for a revert commit that's already in a release.
I'm not sure whether that's the right level of bump for all situations. An argument can be made that any reverted released commits that triggered a release (with the exception of
perf) are a breaking change since code written between the commit and the reverted commits can break if it's depending on a bug fix, feature introduction or breaking change.cpcloud commentedon Jan 23, 2022
I have since stopped using python-semantic-release so this is no longer an issue for me.
danth commentedon Jan 23, 2022
Perhaps this should be left open, since it could be an important feature for other users?
jacksbox commentedon Feb 8, 2022
@cpcloud wired place to ask, but I can't find a way to reach out for you otherwise: do you have an other package recommendation which handles semantic-releases better?
cpcloud commentedon Feb 8, 2022
@jacksbox
You can use the
semantic-release/execplugin to execute shell scripts in the different stages.Here's one of my projects that does this: https://github.com/cpcloud/protoletariat
The scripts are here: https://github.com/cpcloud/protoletariat/tree/main/ci/release
Config is here: https://github.com/cpcloud/protoletariat/blob/main/.releaserc.json
21 remaining items
github-actions commentedon Dec 13, 2025
It has been 90 days since the last update on this confirmed issue. @python-semantic-release/team can you provide an update on the status of this issue?
codejedi365 commentedon Dec 14, 2025
No update yet, still in backlog
github-actions commentedon Mar 15, 2026
It has been 90 days since the last update on this confirmed issue. @python-semantic-release/team can you provide an update on the status of this issue?