Your comments
Because the changes for the feature (not the ones for the existing part of the code base) may cause conflicts. It can be a lot of manual work to sort that out.
I have a long running branch to develop a complex new feature. Part of that effort is extending some existing parts of the code base. To reduce the size of the feature branch and to test the extensions separately, I cherry-pick them into a separate branch. But those extensions evolved over time and their changes are spread over several commits. To cherry-pick just and all of the extension changes, I perform the actions given in my request.
Improvement over the existing behavior: Either you have to perform the actions given in my request. Or else you have to cherry-pick the commits one by one and discard the unwanted changes.
Although I hope my request is understandable, I should have been more precise:
- create a new temporary local branch based on the later of the two selected commits
- enter the new branch
- squash the commits between the earlier commit (excluding) and the later commit (including)
- cherry-pick the selected files as it is currently implemented for single commits
- change into the original branch
- delete the temporary branch
Customer support service by UserEcho
The extensions of the existing code base usually do not cause conflicts. That's the reason to cherry-pick them into a branch of their own and to test them separately.