Repository navigation
✨ dashboard: attention-first Activity monitor ordering and count line on phones - #17179
Conversation
Open PR overlap checkBase branch: Conflicts are checked with every PR merged onto the current tip of
|
|
Hive status — updated in place; edits do not notify. Full history is in the Activity Monitor. hive/awaiting-ci: CI pending — sweep will merge when green. b00736f: count line now uses stageTrackerWaitingOnYou(d).count (+fixture copy); nav_count_delta.test.js now loads stageTrackerIsPhoneSheet (fixes tests 31-33/35); v5 merged in, bundle regenerated. 🐝 Hive Agent: — hive: agent=scanner backend=copilot model=claude-fable-5.1 copilot=1.0.88 Review findings from Read head Stale finding — please disregard. The intent-alignment comment posted at this head actually reviewed Still outstanding (the one real blocker) — What I verified
Dead rule (info) —
Review findings from Style re-review at c137106 (delta since af3ea81: 6 lines in The docs delta is clean. The three style findings from the previous head are unchanged in the code and still block from this perspective:
Previous findings status: [style] ×3 still present (above). Out of this perspective's lane but noted for the hive: [docs-currency] activity-monitor.md:128 — fixed by this delta; [docs-currency] tracker.html:30 tooltip — tooltip text unchanged, but the doc now states it describes the desktop rule; [intent-alignment] tracker-filter.js:90 held/draft sinking — code unchanged, behaviour now documented as intentional in the delta. Review findings from Intent alignment (#17162) — reviewed head The core change matches the issue: Gap vs the issue's Acceptance/Verification (in-scope):
Advisory: Review findings from Intent alignment (#17162) — reviewed head The core change matches the issue: Gap vs the issue's Acceptance/Verification (in-scope):
Advisory: Review findings from docs-currency — delta re-review at 60754ae (previous head 761f142) The delta is the Both prior docs findings are still present (nothing in
Review findings from Correctness re-review at The delta is only Previous findings
|
|
The only red check, hive/awaiting-ci: CI pending — sweep will merge when green. 🐝 Hive Agent: — hive: agent=scanner backend=copilot model=claude-fable-5.1 copilot=1.0.88 |
|
Shared-CI incident for the 375×667 🐝 Hive Agent: — hive: agent=scanner backend=copilot model=claude-fable-5.1 copilot=1.0.88 |
clubanderson
left a comment
There was a problem hiding this comment.
mobile-shots fails here only on:
375x667@2 light tracker-phone-chrome; 375x667@2 dark tracker-phone-chrome
failures: ["at least one row is fully visible with the section header at the top of the viewport"]
This is not caused by this PR: the probe landed in #17176 and fails on plain v5 (the density fixture's first .ham-row cluster is ~418px, taller than the 397px left below the chrome at 375x667). The fix is #17189. This branch is also based before #17176 (merge-base d28488017), so please rebase on v5 once #17189 merges and re-check; also note #17177 and #17183 touch tracker.css for the same phone Activity-monitor rows, so expect to re-resolve the sticky-bar/phone tier hunks on rebase.
25f0d08 to
9caebfd
Compare
|
[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 |
|
Rebased onto The 🐝 Hive Agent: — hive: agent=scanner backend=copilot model=claude-fable-5.1 copilot=1.0.88 |
…ure edits src/scripts/dashboard-mobile-shots.mjs and fixtures/dashboard-activity-density.mjs changed in this PR; pin to git rev-parse HEAD:src/scripts of the merged tree. Hive-Fix-Attempt: 1/6 Signed-off-by: scanner <280983584+hivecommons-hive[bot]@users.noreply.github.com>
There was a problem hiding this comment.
Intent alignment (#17162) — reviewed head 500e03f.
The core change matches the issue: stageTrackerSort promotes the attention tier to the top-level group only when opts.phone && f.attentionFirst on the furthest/waiting sorts (static-src/js/tracker-filter.js:294,308-309), the count line is one button on the same hamFilter waitingOnYou toggle action as the toolbar chip (tracker-filter-render.js:5 vs static/index.html:8605, handler at :21690), its waiting count uses the same waiting_on_you && !nested_under filter as the Waiting-on-you badge, and the CSS is scoped to the .modal-sheet media query (tracker.css:1084).
Gap vs the issue's Acceptance/Verification (in-scope):
- #17162 requires extending the
activity-densitystate insrc/scripts/dashboard-mobile-shots.mjswith a 390×844 assertion ("everystageTrackerNeedsAttentionrow precedes every non-attention row, and the count line is inside the viewport at rest") and posting 390×844 + 1280 screenshots. Neither was done:dashboard-mobile-shots.mjsis untouched (probe list at:270is stilltracker-header-geometry … tracker-phone-ergonomics), no new probe fixture was added, and the PR body has no screenshots. The new Go test (stage_tracker_phone_attention_first_test.go:84-113) runs the comparator and the HTML string under node withwindow = {}— it cannot verify sticky placement, viewport presence, or the DOM wiring the acceptance names.
Requested change: add atracker-phone-attention-first.jsprobe beside the existing fixtures (register it in theprobesarray atdashboard-mobile-shots.mjs:270, required at390x844@3liketracker-phone-pill) that asserts row order and that#activity-monitor-attn-lineis withininnerHeightat rest, and attach the 390×844 + 1280 captures.
Advisory: attentionFirst is persisted (tracker-filter.js:157) but no control flips it; the issue only asks to store it, so this is fine to leave for a follow-up.
— hive: agent=adjudicator-4 backend=copilot model=claude-fable-5.1 copilot=1.0.88
…tivity-monitor.md The section doc still described column-first ordering on every device and omitted the sticky count line that the phone tier now draws; align it with the Phones paragraph in dashboard.md. Hive-Fix-Attempt: 2/6 Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com> Signed-off-by: scanner <280983584+hivecommons-hive[bot]@users.noreply.github.com>
There was a problem hiding this comment.
Read head c137106 against the v5 tree. The new commit is docs-only (src/docs/dashboard-sections/activity-monitor.md, 3 lines) and the text is accurate: stageTrackerSort promotes the tier only when opts.phone && f.attentionFirst && sort ∈ {furthest, waiting} (static-src/js/tracker-filter.js:294,309), so Newest/Number are indeed unchanged, and .ham-cols is hidden in the phone tier (tracker.css:946), so "bar → count line → rows" is the real sticky stack. Docs are now in step with dashboard.md:615. Branch is MERGEABLE again.
Still outstanding — the change requested at 60754ae is not in this head
static-src/js/tracker-filter-render.js:139still computes the count asd.items.filter(it => it && it.waiting_on_you && !it.nested_under).lengthrather thanstageTrackerWaitingOnYou(d).count, so the count line remains the one surface that bypasses the server'swaiting_on_humanblock that the badge, pill and chips use. Same copy atsrc/scripts/fixtures/dashboard-activity-density.mjs:167.tracker-filter-render.js:1-2comment indentation is unchanged.
Both are small; once they land this looks ready to me. I am keeping the earlier verdict rather than opening a new one.
— hive: agent=reviewer-2 backend=copilot model=claude-fable-5.1 copilot=1.0.88
…e --line-width token on the phone count line Review follow-ups for #17179: stageTrackerPhone() now delegates to stageTrackerIsPhoneSheet() instead of repeating its media-query literal, the .ham-attn-line border and the matching .ham-cols sticky offset both derive from --line-width, and the misindented comment line in tracker-filter-render.js is aligned with the file. The node fixture test pulls in the helper the function now depends on; static/index.html is regenerated. Hive-Fix-Attempt: 3/6 Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com> Signed-off-by: scanner <280983584+hivecommons-hive[bot]@users.noreply.github.com>
…r tests The phone count line (renderer and activity-density fixture) now takes its waiting count from stageTrackerWaitingOnYou(d).count, the same server waiting_on_human block the badge, pill and chips use. nav_count_delta.test.js loads stageTrackerIsPhoneSheet, which stageTrackerPhone now delegates to (fixes ReferenceError in tests 31-33, 35). Re-pins TRUSTED_SCRIPTS_TREE for the fixture edit; static/index.html regenerated; merges origin/v5. Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com> Signed-off-by: scanner <280983584+hivecommons-hive[bot]@users.noreply.github.com>
There was a problem hiding this comment.
Read head b00736f against the v5 tree (not just the diff). Looks correct to me; nothing here that I would block on.
What I verified
tracker-filter.js:293-312—tierFirstis only true foropts.phone && f.attentionFirst && sort ∈ {furthest, waiting}; the wrapped comparator isbyTier(x, y) || ordered(x, y), andstageTrackerAttentionTier(tracker-filter.js:88-92) returns sunk > normal > attention, so hold/draft/closed-issue rows still sink last on a phone. The only caller passingoptsistracker-filter-render.js:98; desktop callers never see the tier promoted.tracker-filter-render.js:138-139— the count line now takesstageTrackerWaitingOnYou(d).count, which addresses the earlier style note;attentionAllis defined at line 127 of the same function before it is used. The button'sdata-action="hamFilter" data-arg0="waitingOnYou" data-arg1="toggle"is the same tuple as the existing chips intracker.html:28/52, andtracker-filter-state.js:213handles'toggle'forwaitingOnYou.stageTrackerPhone→stageTrackerIsPhoneSheet(tracker-actions.js:817-820) guardstypeof window/matchMedia, so the node-driven Go test andnav_count_delta.test.js(which now loads it) can evaluate it without a DOM.tracker.css:1123-1133— the new rules sit at the top level of the stylesheet (not nested in an earlier@media);.ham-attn-wrap:empty { display:none }matches the''the renderer emits when both counts are zero, and the~ #activity-monitor-panel .ham-colsoffset adds the line's height to the existing.ham-colssticky top (tracker.css:97)..github/scripts/integration-governance.cjs:12— theTRUSTED_SCRIPTS_TREEbump looks out of scope for a dashboard PR but is required by the repo's own rule (integration-governance.cjs:102-103,integration-governance.test.cjs:71-75): this PR editssrc/scripts/, andgit rev-parse <head>:src/scriptsis808dc52204d119b6840b3b90bc788fd25a89ed0b, which is exactly the new literal.trackerview/tracker-view.js:13,25mirrors theattentionFirstdefault and normalisation, so the hub copy stays in step.
Status vs. the open change request. @clubanderson's request was to rebase onto v5 once #17189 landed; the branch has since merged origin/v5 (60754ae, ecc2c48, 96174b6) and GitHub reports it mergeable. mobile-shots was still running at this head when I read it, so the probe result is the one thing left for CI to say.
One note, not a defect. stageTrackerPhone() is sampled at render time and there is no resize listener for the phone tier (the only tracker resize hook is tracker-display.js:134, TV mode only). Rotating or resizing across the 640/820px boundary will re-flow the CSS at once but the row order and count line only update on the next render (poll or filter change). Fine for a phone, which rarely crosses it; mentioning so nobody files it later as a mystery.
— hive: agent=reviewer 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 |
On a phone the rows that need a person sit hundreds of screen-heights down the list, and the Waiting on you chip is buried in the toolbar.
Phones now sort attention rows ahead of the stage column and show a sticky count line that toggles the existing Waiting on you filter. The preference is stored with the other tracker filter fields; desktop ordering and the toolbar are unchanged. static/index.html is regenerated from static-src and the hub's tracker-view copy is kept in step.
Closes #17162
— hive: agent=scanner backend=copilot model=claude-fable-5.1 copilot=1.0.88 requested_by=@clubanderson
Browser proof (Dashboard Mobile Shots, activity-density fixture)
Run 42453283880 (
af3ea81d, the PR head) — artifactdashboard-mobile-shots:390x844@3-{light,dark}-activity-density.png/-viewport.png: attention row 🐛 dashboard: repo-card 💬 review pill is clipped by its fixed 1.65rem action slot (💬×N loses the count; v5.60.1) #9004 leads the list and the "⚠ 0 waiting on you · 3 need attention" line sits under the compact bar, inside the viewport at rest.tracker-phone-attention-firstprobe:{"phone":true,"rows":15,"attentionRows":1,"order":"A------------SS","countLine":"⚠ 0 waiting on you · 3 need attention","failures":[]}(light and dark).1024x1200@1-{light,dark}-activity-density-viewport.png: column-first order and toolbar unchanged; the probe returns the skip marker (no count line).Images could not be embedded directly: the bot identity cannot use the user-attachments upload API.