Repository navigation
π fix: keep the issue row's merge-gate pill to one line - #15102
Conversation
The merge-gate pill (#15084) read an issue row's lane reason β 'open PR #15090 covers it (PR at 060-reviewed)' β as a gate reason and drew it as the pill's detail, a two-line clipped label on every issue whose PR sat at the gate. The pill is now the stage word plus a gate fact only; the issue row reads the gate from the nested PR it mirrors so it agrees with the PR sub-row, and the cover reason leads the glyph tooltip instead. Fixes #15101 Signed-off-by: Andrew Anderson <andy@clubanderson.com>
|
[APPROVALNOTIFIER] This PR is NOT APPROVED This pull-request has been approved by: The full list of commands accepted by this bot can be found here. DetailsNeeds approval from an approver in each of these files:Approvers can indicate their approval by writing |
|
@clubanderson is auto-promoted for contributor-lane merging on this PR: the last 5 merged PRs by this author in this hive's repositories have no revert or sentinel-alert evidence. CI, reviewer verdict, diff fences, soak, holds and branch protection still apply. |
Open PR overlap checkBase branch: Conflicts are checked with every PR merged onto the current tip of
|
There was a problem hiding this comment.
Read the tree, not just the diff. Looks correct to me.
Why the gate read is safe. The only producer of open PR #N covers it (PR at <stage>) is src/pkg/stageengine/derive_issue.go:160, and it returns mirrored=true on the same path β so whenever STAGE_TRACKER_COVER_REASON_RE matches, the issue row's stage is the covering PR's stage, and stageTrackerCoveringPR (static/index.html:14236) resolves to the same PR the sub-row draws. The server already fills linked_prs[].merge_gate / soak_remaining_seconds from the PR item (stage_tracker.go:2558, pre-existing on v5), so no API change was needed.
Partial cover. The addresses part of it (β¦; PR at <stage>) reason (derive_issue.go:158) doesn't match the regex, but the prs.find(p => p.stage === item.stage) fallback still picks the mirrored PR and laneReason is '' for every kind:'issue' row, so that shape no longer leaks into the pill either β it just loses the tooltip prefix. Fine.
PR rows unchanged. laneReason falls back to item.reason exactly as before for kind:'pr' (a PR row's reason never matches the cover regex), so stageTrackerMergeGateInfoFor is behaviour-preserving there.
Couldn't run the node-backed TestStageTrackerMergeGate* here (go test is disabled on agent hosts); CI is the verdict for those.
β hive: agent=reviewer backend=copilot model=claude-fable-5.1 copilot=1.0.88
There was a problem hiding this comment.
Read the tree at ce9dcaf, not just the diff. Looks correct to me.
- The regex
STAGE_TRACKER_COVER_REASON_RE(src/pkg/dashboard/static/index.html:14227) matches the producer byte-for-byte:stageengine/derive_issue.go:160emits"open PR #" + N + " covers it (PR at " + stage + ")". The older bare tokenopen PR covers it(derive_issue.go:25) does not match and so is neither a cover text nor β sincelaneReasonis''forkind === 'issue'atindex.html:14267β a gate reason; the test covers that case. stageTrackerCoveringPRreadslinked_prs[].stage/.merge_gate, which the server does populate (stage_tracker.go:2558mirrorspr.MergeGateonto the nested entry;StageTrackerLinkedPR.Stageat:476). Fallback to the issue's ownstagewhen the nested PR has nomerge_gatekeeps older payloads at the right word.- All three callers (
index.html:14297, 14306, 14607) go throughstageTrackerMergeGateInfo, so the pill, the glyph tooltip and the row strip all take the new path; PR rows are untouched becausestageTrackerCoveringPRreturnsnullforkind === 'pr'.
I could not run the node test here (go test is blocked on agent hosts), but test (dashboard 1-4/4) is green on this head.
β hive: agent=reviewer-2 backend=copilot model=claude-fable-5.1 copilot=1.0.88
|
merged by Hive trusted-author auto-merge: author @clubanderson holds role owner and GitHub permission admin on repo; CI green at ce9dcaf. |
|
Thank you for your contribution! Your PR has been merged. We'd love to hear how your experience was: share feedback |
Fixes #15101
What
In the Activity Monitor, an issue row whose covering PR sits in the merge gate rendered a two-line, clipped pill after #15084 rolled:
(#15088β#15090, #15023β#15029, #15083β#15086, hotshot#316β#317.)
Root cause
stageTrackerMergeGateInfobuilt the gate-pending detail frommg.reason || item.reason. A PR row hasmerge_gate.reason(a gate token). An issue row has nomerge_gateof its own and itsreasonis the issue lane'sopen PR #N covers it (PR at <stage>), which fell throughstageTrackerGateReasonTextunchanged and was concatenated into the pill label.Fix
stageTrackerCoverReasonTextrecognises the issue lane's cover reason; it is never used as a gate reason, so the pill is one line: the stage word plus a gate fact only.stageTrackerCoveringPRpicks the nestedlinked_prs[]entry the reason names (else the one at the mirrored stage); the issue row readsmerge_gatefrom it, so the issue row and the PR sub-row always agree on word / detail / cue.Open PR #15090 covers it Β· at 060-reviewed β β¦β no label name in parentheses (Activity Monitor: operator-readability findings from a live walkthrough (12:11β16:59 ET, 0470256 β aa9b530)Β #15096 iteration-6 note).TestStageTrackerMergeGateIssueRowPillNode(node, stdin fixture) asserts the pill has nocovers it/(PR attext and the tooltip does, that both rows agree, and the older-payload / no-linked-PR / two-linked-PR cases.docs/dashboard-sections.mdand achangelog.d/fixed-*fragment.090 vs 060 on the same rows
The label timeline shows #15086 / #15029 really wore
stage/090-ready-for-human-mergebefore the roll and were relabelledstage/060-reviewedat 21:17Z; both displays were faithful to the label at the time. The 090 label was the wrong one (N7 in #15096: hive-authored PRs flagged untrusted) β a stage-derivation issue, not display; this PR is display-only.