The following two tests assumed that HEAD would always be there in the ‘rad’ remote. It turns out, however, that this isn’t necessarily the case, for example on Debian releases (as consistently noticed in multiple CI runs).
This change fixes this issue by explicitly setting the remote head, where this matters:
- crates/radicle-cli/examples/rad-patch-fetch-2.md
- crates/radicle-cli/examples/rad-patch-via-push.md
The following two tests assumed that HEAD would always be there in the ‘rad’ remote. It turns out, however, that this isn’t necessarily the case, for example on Debian releases (as consistently noticed in multiple CI runs).
This change fixes this issue by explicitly setting the remote head, where this matters:
- crates/radicle-cli/examples/rad-patch-fetch-2.md
- crates/radicle-cli/examples/rad-patch-via-push.md
Fix the test failures related to remote default branch (HEAD)
The following two tests assumed that HEAD would always be there in the ‘rad’ remote. It turns out, however, that this isn’t necessarily the case, depending on the git version that’s used.
Git versions older that v2.48.0 do not touch ‘refs/remotes/
Git v2.48.0 and newer will update ‘refs/remotes/
Therefore, to ensure that test results are consistent across pre-v2.48.0 git and v2.48.0 or newer git, the fixture of test repositories now include setting ‘remote.rad.followRemoteHEAD=never’, and adjusting the outputs that are affected.
These files are affected:
- crates/radicle/src/test/fixtures.rs
- crates/radicle-cli/examples/rad-patch-fetch-2.md
- crates/radicle-cli/examples/rad-patch-via-push.md
Rewording
Rebase