Repository navigation
fix(settings): keep the restart row while the update card has none - #40331
KiroCrewAutoBuilder wants to merge 1 commit into
Conversation
Builder-Run: build-39354-7b990ef73db4
UX Review (Opus 5.5) — ✅ PASS (a BLOCK with no Trigger and no Outcome does not block)UX-level review of UX-Verdict: PASS cannot evaluate: missing screenshots of the standing "Restart gateway" row next to the in-app update flow and next to the installer card that has no command. The PR attaches no images, so the blind read was not performed. Evidence gaps
[UX-REVIEWED] 21071a3 Description read[DESCRIPTION-READ] 9ade2a8d10b25340e0212a868519e7f9b6fc77827eb329ebe452befaa92b0561 That sha256 names everything this verdict read: the PR description, plus the 3 evidence file(s) the workflow collected from it and handed this review. It covers that evidence because the media strip replaces every attachment URL with the same placeholder, so an edit swapping one attachment for another would leave the description's captured text identical while what this review judged changed. Recompute it with |
First Principles Review (Opus 5.5) — ✅ PASSPremise-level review of First-Principles-Verdict: PASS This is a fix aimed at the mechanism: the standing row now hides only when the update card draws its own Restart button. The new test for the in-app flow fails on the base commit, so the defect has checkable provenance. Notes
What this change shipsInventory (5 items) — 4 justifiedIntent: let a dashboard-only user restart the gateway to finish an update they approved. This is a FIX.
[FIRST-PRINCIPLES-REVIEWED] 21071a3 Description read[DESCRIPTION-READ] 883d5c72f6bc1feb3786212de6ec02aac6306d8233f7bd4cf042b511e2bd67fb That sha256 names the PR description this verdict read. Recompute it with |
Design Review (Opus 5.5) — ✅ PASS (a BLOCK with no Trigger and no Outcome does not block)Design-level review of Design-Verdict: PASS cannot evaluate: missing screenshot of Settings → About with the standing Restart row beside the in-app update flow and beside the command-less installer card. The diff changes which controls those states show, and no rendered evidence came with it. BlockersChanged control set in two update states, with no rendered evidence Notes
[DESIGN-REVIEWED] 21071a3 Description read[DESCRIPTION-READ] 8b4f7a2965b2b2f0fab242272a297aa0eeab27c4e78f986915cac63486220302 That sha256 names everything this verdict read: the PR description, plus the 1 evidence file(s) the workflow collected from it and handed this review. It covers that evidence because the media strip replaces every attachment URL with the same placeholder, so an edit swapping one attachment for another would leave the description's captured text identical while what this review judged changed. Recompute it with |
Opus 5.5 Review — ✅ no blocking findingsReviewed Verdict parsed from the review's SHA-scoped output markers for commit False positive or not applicable? A repository writer can comment: |
GPT 6.1 Review — ✅ no blocking findingsGPT 6.1 completed its review of This comment is updated in place on each push. Review detailsNo findings. False positive or not applicable? A repository writer can comment: |
|
This PR and #39370 make the same fix for part 1 of #39354. Both add the same Please keep one of #39370 and #40331 and close the other, carrying over anything only this one has. |
Problem / Motivation
Goal: Settings → About always offers a way to restart the gateway while an update is pending or has been applied.
Why it matters
Not a goal
What changed (motivation → approach → change)
Backwards compatibility
Compatible: the row only appears in more states. The installer card that has a command still has the only Restart button.
Removes nothing: the change only shows an existing control in states where it was hidden.
Tests
Every test red on the patched tree in Verify, in
backend-1,backend-2,backend-3,backend-5,backend-6,backend-10,backend-11,backend-12, fails at the base commit in the Builder's verify run as well (below, Red without this patch too, and Red only after other tests for those that passed when Verify ran each again alone).Proof: not provable -- frontend test (prove.py runs Python only)
Red without this patch too
These tests fail with the patch, and fail at the base commit in the Builder's verify run as well: the draft PR is opened with them red, and GitHub's CI is likely red on them too. Verify lists at most 50 a shard.
Red only after other tests
These tests failed in Verify's run of their shard, passed when Verify ran each again alone on the patched tree, and failed in the same run at the base commit too: they depend on what ran before them, at the base as with the patch. GitHub's CI may be red on them, by the order its runners give them.
Manual verification
Real-machine verification
Screenshots / video
The Builder captured no screenshots or video. This patch touches UI paths, so the screenshot gate stays red until the owner adds them. The UI paths:
Related Issues
Closes #39354
Sibling sweep
Swept: git grep -n "showManualUpdate" -- website/src/
Sibling: website/src/pages/settings/AboutPanelGatewayUpdate.tsx -- changed here (the row and its error notice, both now on cardHasRestart)
Sibling: website/src/pages/settings/DesktopGatewayRestartRow.tsx -- not affected -- the desktop app renders its own restart row, gated by restartControls
Pattern harvest
Not generalizable: a visibility guard keyed on a broader state than the duplicate it meant to avoid; specific to this panel's layout.
Checklist
Contribution License Agreement
Local review blocked
Local review blocked: the Builder's own review of this patch did not pass it (a reviewer still blocked it after the agent's fix pass, or could not be reached, unavailable), and the patch went to verify anyway (
review_blocked). Verify passed it and Check judged it as any patch; the repository's own review lanes judge it here. What each reviewer said, the blocking ones' items in their own words:Draft PR by the Autobuilder (its agent's own words are quoted as code), run
build-39354-7b990ef73db4. The owner marks it ready for review.