Repository navigation
fix(minimald): leave HEAD alone when a push targets another branch - #2319
norrietaylor merged 5 commits into
Conversation
The post-receive hook checked out every pushed branch, so pushing a branch other than the workspace's current one moved HEAD and swapped the worktree to it. Capture the checked-out branch and HEAD's born/unborn state up front; a push to a non-checked-out branch now only moves the ref (receive-pack already did that), while the still-unborn first push keeps populating the worktree as before. Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
|
No actionable comments were generated in the recent review. 🎉 ℹ️ Recent review info
📝 Walkthrough
Merge Risk: ⚪ Minimal · up to
|
| Check name | Status | Explanation |
|---|---|---|
| Title check |  |
The title clearly and concisely describes the primary change: preventing pushes to another branch from moving the current HEAD. |
| Description check | |
The description explains the behavior change and provides detailed verification commands and results. It does not include the template's Checklist section, but the required summary and testing informa… |
| Docstring Coverage | |
Docstring coverage is 100.00% which is sufficient. The required threshold is 80.00%. Docstring coverage is scoped to functions touched by this diff. Analyzed 8 functions across 2 files. (1 skipped: 1 … |
| Linked Issues check | |
Check skipped because no linked issues were found for this pull request. |
| Out of Scope Changes check | |
Check skipped because no linked issues were found for this pull request. |
Full details: Description check
-
Explanation
The description explains the behavior change and provides detailed verification commands and results. It does not include the template's Checklist section, but the required summary and testing information are present.
Full details: Docstring Coverage
-
Explanation
Docstring coverage is 100.00% which is sufficient. The required threshold is 80.00%. Docstring coverage is scoped to functions touched by this diff. Analyzed 8 functions across 2 files. (1 skipped: 1 unsupported.)
✨ Finishing Touches
-
📝 Generate docstrings
-
- Commit to this branch
- Create a new PR
🧪 Generate unit tests (beta)
-
- Commit to this branch
- Create a new PR
-
- Autofix · Keep fixing CodeRabbit findings and required CI, and resolving merge conflicts
Warning
Some tools did not complete. Review the errors below.
🔧 Vale (3.23.0)
-
{
"Line": 0,
"Path": "",
"Text": "E100 [loadStyles] Runtime error\n\nstyle 'ste' does not exist on StylesPath\n\nExecution stopped with code 1.",
"Code": "E100",
"Span": 0
}
A rabbit watches branches hop,
The first eligible branch takes its spot.
Other refs advance along,
The worktree stays where it belongs.
Detached HEAD gets a notice from the burrow.
Comment @coderabbitai help to get the list of available commands.
A first push carrying several branches checked each one out in turn, so HEAD ended on the last. Record the checkout so later refs in the same push only move their refs, and cover a multi-ref push that includes the current branch.
…e born flag A first push into a fresh workspace that carries two branches checks out only the first, and every multi-ref test now checks the other ref lands at the pushed tip. The hook's HEAD flag is a string (born=yes) instead of a raw exit status, and its header comment now matches the first-checkout-wins behaviour.
|
/review |
A detached HEAD in the session worktree matches no branch, so the post-receive hook moves the pushed refs and never checks anything out. That is intended, and matches plain git, but it happened silently. The hook now prints one stderr line, which the pusher sees as a `remote:` line, naming the detached HEAD and the `git switch` that re-attaches it. A new git_remote test detaches the workspace HEAD, pushes, and checks the message, the still-detached HEAD and the moved ref. The cli-min push paragraph covers the detached case and corrects which branch the first push into an empty workspace checks out.
|
/review |
There was a problem hiding this comment.
Approved on ARCH PASS at 6b3f7a1.
…2319) The post-receive hook checked out every pushed branch, so pushing a branch other than the workspace's current one moved HEAD and swapped the worktree to it. A push to a non-checked-out branch now only moves the ref. A first push into a fresh workspace checks out the first branch and leaves the rest as refs, and a push onto a detached HEAD says so on a remote: line with the git switch that re-attaches it. The cli-min push paragraph describes all three. Backport of #2319. Applied without adaptation. (cherry picked from commit 0a297d0)
…2319) The post-receive hook checked out every pushed branch, so pushing a branch other than the workspace's current one moved HEAD and swapped the worktree to it. A push to a non-checked-out branch now only moves the ref. A first push into a fresh workspace checks out the first branch and leaves the rest as refs, and a push onto a detached HEAD says so on a remote: line with the git switch that re-attaches it. The cli-min push paragraph describes all three. Backport of #2319. Applied without adaptation. (cherry picked from commit 0a297d0)
Routing-Key: inbox-route/I_kwDOSUhdos8AAAABX3veMQ
git push (redacted) <branch>checked out every pushed branch inthe session workspace, so pushing any branch other than the current one
silently moved the session's HEAD and swapped the worktree to it.
The post-receive hook in
handle_git_receivenow records thechecked-out branch and whether HEAD is born before the ref loop; a
push to a non-checked-out branch only moves the ref (receive-pack
already did that), while the still-unborn first push keeps populating
the worktree as before (informed by #347).
Verification
cargo fmt --all --checkcargo clippy --workspace --all-targets --locked -- -D warningscargo test --locked -p minimal --test git_remote— 2/2 passcargo test --locked -p minimald --lib exec::— 50/50 passcargo test --locked -p minimald -p minimalfull suites show onlyfailures identical on pristine
mainin this sandbox(net::switch/answerer/connection/rpc + one legacy-network-spelling
unit test) — pre-existing environment limits, untouched by this diff
with the fix
Summary by CodeRabbit
HEADbranch when it is pushed, or the first pushed branch otherwise.HEADis detached. In the latter case, a notice explains how to switch to a branch.HEADare handled.