Repository navigation
✨ Show ordered preparation rail without changing pipeline stages - #15910
Conversation
Signed-off-by: Andy 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 |
Open PR overlap checkBase branch: Conflicts are checked with every PR merged onto the current tip of
|
There was a problem hiding this comment.
Correctness review @ aa6594a — no blockers. Ran stage_tracker_preparation_visual.test.js against the head sources: 15/15 pass. Verified the pill-suppression path end to end: stageTrackerPreparationVisualSync runs before rows render (tracker-filter-render.js:94), rows re-render when their HTML changes (tracker-motion.js:719-722), nested PRs carry head/base_ref (stage_tracker.go LinkedPR DTO, tracker-row-layout.js:60-66), and static/index.html matches static-src.
One advisory (low, in-scope):
Rail is shown on every deployment before the first scan, even where base preparation is disabled. tracker-filter-render.js:65 calls stageTrackerPreparationVisualSync(null, …) for the initializing/no-items branch, and tracker-preparation.js:100-103 then unhides the region with "Preparation queue unavailable · waiting for a current tracker snapshot." Go only emits base_preparation for repos with RequireCurrentBaseFor (stage_tracker_preparation.go:42), so installs that never use preparation get a labelled "Preparation queues" region during load that then disappears. Previously nothing preparation-related was shown there. If that is not intended, hide the rail unless a prior payload actually carried queues:
if (!data || !Array.isArray(data.items)) {
const had = _stageTrackerPreparationShown.size > 0 || (el.innerHTML && el.innerHTML.indexOf('ham-preparation-group') >= 0);
el.hidden = !had;
if (had) el.innerHTML = '<p class="ham-preparation-note">Preparation queue unavailable · waiting for a current tracker snapshot.</p>';
return;
}
(adjust the inactive policy hides the rail; unavailable/loading state is explicit test accordingly.)
— hive: agent=adjudicator-2 backend=copilot model=claude-fable-5.1 copilot=1.0.88
|
Thank you for your contribution! Your PR has been merged. We'd love to hear how your experience was: share feedback |
Verified source delivery
Proofs: v2 CI, v2 Tests, complete image-build matrix. User-visible resultThe Activity Monitor has a directly visible train-style preparation reference rail: confirmed owner first, then durable first-admission FIFO waiters. It distinguishes filtered/off-page membership, inactive tickets and unverified recorded heads without promising merge order, approval or ready/native queue membership. Pipeline row stages, sorting, filters, pages and linked issue/PR trees remain unchanged. Redundant preparation/behind-owner pills disappear only for exact-current PRs actually represented in that rail; fallback pills and unknown/CI/Review/held warnings remain. Preparation no longer repeats in the gold header chip/popover, and the zero-ready-train chip is hidden; genuine ready trains remain separate. No DTO/store/scheduler/native backend changes, extra polling, production deployment, runtime configuration or activation were performed. CI/source delivery is not a runtime rollout claim. |
Summary
Validation
make dashboard-gen.Scope
Reuses the existing cached
base_preparationDTO and train presentation primitives. No engine/store/scheduler/DTO changes, new polling, future-integration dependency, deployment, configuration change or activation.Owner-requested follow-up to #15887. Includes the explicit requests to remove repeated waiting-behind pills and the redundant
0 train PRs · owner · waitingheader entry. Source-only merge is authorized after complete exact-head applicable CI succeeds.