Consider this:
- uses: EndBug/add-and-commit@v9
with:
add: '["file-that-changed.txt"]'
remove: '["file-that-was-removed.txt"]'
author_name: GitHub-Automated
author_email: foo@example.com
default_author: github_actor
message: "Changing files"
pathspec_error_handling: exitImmediately
pull: "--rebase --autostash"
push: false
First the file is removed here
Then if pull is used, the file is also removed here which breaks the action, since the file is already marked as removed.
::error::Error: fatal: pathspec 'file-that-was-removed.txt' did not match any files
::error::Error: Remove command did not match any file:%0A git rm file-that-was-removed.txt
If one wants to rebase it does not make sense that the commit is done after the pull. The commit should be done, then a pull rebase should happen, that would not break the flow.
I guess it's hard to account for these different approaches.
Consider this:
First the file is removed here
Then if
pullis used, the file is also removed here which breaks the action, since the file is already marked as removed.If one wants to rebase it does not make sense that the commit is done after the pull. The commit should be done, then a pull rebase should happen, that would not break the flow.
I guess it's hard to account for these different approaches.