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
{{ message }}
Repository navigation
fix(web): a disabled Push names the real reason when there are local changes - #36786
Nothing needed changing on this pass, so nothing was pushed and the head is unchanged. PR #36786 is linked to this thread, but I could not stay on it: watch_pull_request failed twice with 'GitHub has no working credential for this host', and gh auth status says the token on this VM is invalid. At handback the PR head was 48483d6, with CI (Build and Typecheck, Lint, the Test Web and Test Server shards) and the Macroscope checks still running or queued. No reviews and no line comments were posted (read over the public REST API without a token). The rest of the watch, meaning failing checks and Macroscope or CodeRabbit findings rated high or above, has to be done by a step with a working GitHub login. Earlier validation of this head still holds: vp run check:changed passed (format, lint, knip, typecheck, tests, react compiler).
Supersedes #29866. The change is @saphid's; their commits are kept.
Clone on feature, in sync with origin/feature, with an uncommitted edit to tracked README.md. Opened Git action options: Commit enabled, Push disabled, Create PR enabled. Hovering Push showed "Commit or stash local changes before pushing." The correct blocker is that there are no local commits to push. 2) Pushed an empty commit to origin/feature from a second clone and fetched it into the first, keeping the edit (status: behind 1, dirty). Reloaded and reopened the menu: the primary action became Commit, Push and Create PR were disabled. Hovering Create PR showed "Branch is behind upstream. Pull/rebase before creating a pull request." Hovering Push still showed "Commit or stash local changes before pushing." 3) Committed a new file locally (ahead 1, behind 1, still dirty): Push was disabled and its tooltip still blamed local changes. 4) Control: stashed, rebased onto origin/feature, popped the stash (ahead 1, behind 0, same uncommitted edit). Push was enabled. Since Push is available with uncommitted edits, local edits are not a Push blocker, and the tooltip in states 1-3 names the wrong reason. The text comes from the hasChanges check at apps/web/src/components/GitActionsControl.tsx:264, which runs before the isBehind and !isAhead checks.
Evidence
Environment: real web client and server, Chromium, 1280×800, light theme, isolated fresh databases, synthetic local bare Git remotes, and free replay provider.
Steps on both commits: create feature tracking origin/feature with an uncommitted tracked README edit; open Git action options and hover Push when in sync, behind 1 after a second clone pushes an empty commit, and ahead 1/behind 1 after committing local.txt. Then stash, rebase onto origin/feature, pop the stash, and verify Push remains enabled with ahead 1/behind 0 and the edit retained. Each state was fully prepared before opening its replay thread to avoid stale status caches.
Proven: base showed “Commit or stash local changes before pushing.” in all three disabled states. Head showed “No local commits to push.” when in sync and “Branch is behind upstream. Pull/rebase before pushing.” when behind or diverged. Push was enabled in the ahead-and-dirty control on both commits.
No new commands this pass, since the code did not change. Earlier on this head: vp test run src/components/GitActionsControl.logic.test.ts (18 passed) and vp run check:changed (all passed).
Not in this change
packages/client-runtime/src/state/gitActions.ts (mobile) keeps 'Commit or stash local changes before pushing.'; it is correct there because mobile's canPush still requires a clean worktree. Whether mobile should match web's push availability is a separate product decision.
packages/client-runtime/src/state/gitActions.ts (mobile) keeps 'Commit or stash local changes before pushing.'; it is correct there because mobile's canPush still requires a clean worktree.
Who did the work
Written by Claude Opus 5.5, in Claude Code
Reviewed by GPT-6.1 Sol, in Codex, and by Claude Opus 5.5, each on its own
Proof captured by GPT-6.1 Sol and checked by Claude Opus 5.5
Macroscope's review found this PR approvable — Disabled Push now explains the actual blocker instead of incorrectly blaming local changes. This is a UI-message correction; action availability and Git execution remain unchanged.
You can add or adjust custom eligibility rules. Learn more.
No actionable comments were generated in the recent review.
Walkthrough
The disabled-action reason helper moved from GitActionsControl.tsx to GitActionsControl.logic.ts. The component now uses the extracted helper, and tests cover push-disabled messages for a branch with no local commits and a branch behind upstream.
Changes: Git action disabled reasons
Layer / File(s)
Summary
Extract and validate disabled-action reasons apps/web/src/components/GitActionsControl.logic.ts, apps/web/src/components/GitActionsControl.tsx, apps/web/src/components/GitActionsControl.logic.test.ts
The logic module now provides explanations for disabled commit, push, and change-request actions. The component uses the helper, and tests check two push-disabled reasons.
This change shows a more accurate reason when Push is disabled on a branch with local changes. Push behavior itself is unchanged. The change is low risk and ready to merge.
Pre-merge checks: 3 · 1
Failed checks: 1 warning
Check name
Status
Explanation
Resolution
Description check
The description contains substantial, relevant evidence: a clear reproduction on main, a problem statement (Push tooltip blames local changes), before/after screenshots, interaction recordings, and te…
1) Restructure the description to match the template: add '## Problem', '## Change', '## Scope', and '## Verification' headings. 2) In Scope, link the issue that shows the bug, or explain why a small obvious fix needs no issue. Link PR #367…
Passed checks: 3 passed
Check name
Status
Explanation
Title check
The title names the change: a disabled Push now gives the real reason when local changes exist. It follows the conventional commit form and matches the PR's stated objective and the logic change in Gi…
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: Title check
Explanation
The title names the change: a disabled Push now gives the real reason when local changes exist. It follows the conventional commit form and matches the PR's stated objective and the logic change in GitActionsControl.logic.ts and GitActionsControl.tsx.
Full details: Description check
Explanation
The description contains substantial, relevant evidence: a clear reproduction on main, a problem statement (Push tooltip blames local changes), before/after screenshots, interaction recordings, and test results. However, it fails the template's core structure. It has no '## Problem', '## Change', '## Scope', or '## Verification' headings. The Scope section is missing: there is no linked issue that shows the bug, and no statement on why no issue is needed. Instead, the author links PR #36786 and says it supersedes #29866. The PR also says the head is unchanged and that CI and Macroscope checks were still running at handback, so the verification status is unclear. The description also has an unclear or duplicated bullet under 'Not in this change'. It also omits the conventional commit title. The author's own note says the watch was not completed. The 'Tests' section lists earlier runs on the same head, which is acceptable, but it does not match the template's requirement to describe the focused checks and observed results for the changed behavior. The description also includes a process narrative about a VM and GitHub credentials that does not belong in a PR description.
Resolution
Restructure the description to match the template: add '## Problem', '## Change', '## Scope', and '## Verification' headings. 2) In Scope, link the issue that shows the bug, or explain why a small obvious fix needs no issue. Link PR #36786 only if a maintainer requested it. 3) Use a conventional commit title in plain language, for example 'fix(web): disabled Push shows the correct reason'. 4) In Verification, state the focused test commands and results for the changed behavior. Remove the notes about the failed PR watch and VM credentials, or move them out of the PR description. Clearly state which checks are still pending. 5) Remove the duplicated 'Not in this change' bullet. 6) Keep the before/after screenshots and recordings, and confirm the links resolve. 7) Keep the model and harness attribution at the end, as the template asks.
Finishing Touches
Generate unit tests (beta)
Commit to this branch
Create a new PR
Autofix · Keep fixing CodeRabbit findings and required CI, and resolving merge conflicts
Included review availability: This review used your included allowance. Your plan provides up to 10 included reviews per hour; 4 remain after this review.
…5 (Checking that every fork seam survived)
Failing check: Checking that every fork seam survived
The seam check failed because `apps/web/src/components/GitActionsControl.tsx` had 21 `fork: source-control-cockpit` markers, and `docs/fork/seams.tsv` expects 22.
The missing marker had moved, not disappeared. Upstream commit 732198d (pingdotgg#36786) moved `getMenuActionDisabledReason` out of `GitActionsControl.tsx` and into `GitActionsControl.logic.ts`. The merge had already carried the seam over with its marker: the `gitOperationLabel(gitStatus)` early return, which shows an in-progress git operation as the reason an action is disabled. The `gitOperationLabel` import in the logic file is marked too. Only the seams.tsv counts were out of date: that file now has 9 markers where its row said 8, and `GitActionsControl.tsx` has 21 where its row said 22.
I moved the seam's row in `docs/fork/seams.tsv` to follow the code: `GitActionsControl.logic.ts` goes from 8 to 9 and `GitActionsControl.tsx` from 22 to 21. I didn't change any code. I reran `check_seams` and it passes.
Pipyakas
pushed a commit
to Pipyakas/t3code
that referenced
this pull request
Nov 15, 2026
…changes (pingdotgg#36786)
Co-authored-by: Alex Southwell <4596216+saphid@users.noreply.github.com>
Co-authored-by: github-actions[bot] <41898282+github-actions[bot]@users.noreply.github.com>
(cherry picked from commit 732198d)
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
size:L100-499 changed lines (additions + deletions).vouch:trustedAuthor is trusted by repo permissions or the VOUCHED list.
2 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
github-actions
Bot
added
vouch:trusted
![macroscopeapp[bot]](https://avatars.githubusercontent.com/in/900172?s=60&v=4){kind=small}
Nothing needed changing on this pass, so nothing was pushed and the head is unchanged. PR #36786 is linked to this thread, but I could not stay on it: watch_pull_request failed twice with 'GitHub has no working credential for this host', and gh auth status says the token on this VM is invalid. At handback the PR head was 48483d6, with CI (Build and Typecheck, Lint, the Test Web and Test Server shards) and the Macroscope checks still running or queued. No reviews and no line comments were posted (read over the public REST API without a token). The rest of the watch, meaning failing checks and Macroscope or CodeRabbit findings rated high or above, has to be done by a step with a working GitHub login. Earlier validation of this head still holds: vp run check:changed passed (format, lint, knip, typecheck, tests, react compiler).
Supersedes #29866. The change is @saphid's; their commits are kept.
Reproduction on main
Reproduced on main at cebdf39:
feature, in sync with origin/feature, with an uncommitted edit to tracked README.md. Opened Git action options: Commit enabled, Push disabled, Create PR enabled. Hovering Push showed "Commit or stash local changes before pushing." The correct blocker is that there are no local commits to push. 2) Pushed an empty commit to origin/feature from a second clone and fetched it into the first, keeping the edit (status: behind 1, dirty). Reloaded and reopened the menu: the primary action became Commit, Push and Create PR were disabled. Hovering Create PR showed "Branch is behind upstream. Pull/rebase before creating a pull request." Hovering Push still showed "Commit or stash local changes before pushing." 3) Committed a new file locally (ahead 1, behind 1, still dirty): Push was disabled and its tooltip still blamed local changes. 4) Control: stashed, rebased onto origin/feature, popped the stash (ahead 1, behind 0, same uncommitted edit). Push was enabled. Since Push is available with uncommitted edits, local edits are not a Push blocker, and the tooltip in states 1-3 names the wrong reason. The text comes from the hasChanges check at apps/web/src/components/GitActionsControl.tsx:264, which runs before the isBehind and !isAhead checks.Evidence
Environment: real web client and server, Chromium, 1280×800, light theme, isolated fresh databases, synthetic local bare Git remotes, and free replay provider.
Steps on both commits: create feature tracking origin/feature with an uncommitted tracked README edit; open Git action options and hover Push when in sync, behind 1 after a second clone pushes an empty commit, and ahead 1/behind 1 after committing local.txt. Then stash, rebase onto origin/feature, pop the stash, and verify Push remains enabled with ahead 1/behind 0 and the edit retained. Each state was fully prepared before opening its replay thread to avoid stale status caches.
Proven: base showed “Commit or stash local changes before pushing.” in all three disabled states. Head showed “No local commits to push.” when in sync and “Branch is behind upstream. Pull/rebase before pushing.” when behind or diverged. Push was enabled in the ahead-and-dirty control on both commits.
Scenario and seed driver
Tests
No new commands this pass, since the code did not change. Earlier on this head: vp test run src/components/GitActionsControl.logic.test.ts (18 passed) and vp run check:changed (all passed).
Not in this change
Who did the work
Run by the t3code factory in T3 Code.