Repository navigation
π mobile-shots: phone-chrome probe accepts a tall first fixture row - #17189
Conversation
The tracker-phone-chrome acceptance added in #17176 asserts that a row is fully visible at 375x667 with the section header at the top. The density fixture's first .ham-row is a whole work-item cluster (parent issue, child PR, decision, notes: ~418px) and never fits the 397px left below the chrome, so every dashboard PR fails mobile-shots at 375x667 regardless of its diff. Keep the intent (chrome stays short, a row shows without scrolling) but accept a first row taller than the remaining viewport when at least 120px of it is visible. Ignore the scratch Chrome profile the harness creates. 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 |
|
Changelog: this PR changes code but carries no changelog entry If it is user-visible β a feature, a fix an operator would notice, a This is a reminder, not a gate; it never blocks a merge. |
Open PR overlap checkBase branch: Conflicts are checked with every PR merged onto the current tip of
|
There was a problem hiding this comment.
Decision needed: @clubanderson should decide whether #17189 keeps only its .gitignore hunk and lets #17185 carry the tracker-phone-chrome.js fix, or the reverse β both edit the same assertion and cannot merge together.
This overlaps #17185 on src/pkg/dashboard/webstatic/testdata/tracker-phone-chrome.js β same failure, same lines, different fix. They conflict; one must yield.
What I verified (read both diffs in full against v5):
- Both PRs target the identical
rowsFullyVisible >= 1assertion attracker-phone-chrome.js:55-56(base numbering) and the identical post-#17176 failure at 375Γ667. - #17185 (1 line) credits
measurements.fixtureNoteback in the row-visibility filter:box.bottom - measurements.fixtureNote <= innerHeight + 1. The probe at:49-52already subtracts that note fromchromebecause "real renders do not have" it; #17185 extends that same correction to the row check. The "a row is fully visible" criterion is unchanged β it just stops counting a fixture-only 108px paragraph against it. Your own measurement (418px row, 397px remaining β 21px short) is well inside that credit. - #17189 keeps the assertion but adds an escape: when the first row is taller than the space left, accept
firstRowPeek >= 120px. That is correct as written (firstRowTooTallβ row bottom below viewport; peek math is right), but it relaxes the criterion for every render, not only the fixture β a future layout whose first row genuinely does not fit would still pass with 120px showing, with only the separatechrome <= 160check bounding it.
So I'd lean to #17185 for the probe hunk, because it removes the fixture artefact rather than loosening the acceptance; the header-comment rewrite here then also becomes unnecessary. The .gitignore hunk (.chrome-dashboard-mobile-shots/) is independent and worth keeping β nothing in #17185 covers it.
Not a defect in either PR; this is a which-one-lands call, so flagging rather than judging. I have not closed anything.
β hive: agent=reviewer-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 |
stageTrackerPopSheet called document.querySelector unconditionally, which broke the Go-driven node harness (stage_tracker_cues_test.go) whose document stub has no querySelector, failing TestStageTrackerOdometerNode and the merged-chip tests. Also rebased onto v5 to pick up the phone-chrome probe fix (#17189) for mobile-shots. Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com> Signed-off-by: scanner <280983584+hivecommons-hive[bot]@users.noreply.github.com>
β¦n phones (#17181) * β¨ dashboard: Activity monitor merged and queue lists open as sheets on phones Signed-off-by: scanner <280983584+hivecommons-hive[bot]@users.noreply.github.com> Co-authored-by: clubanderson <407614+clubanderson@users.noreply.github.com> * π test(dashboard): celebrate node bundle carries the pop-sheet helpers stageTrackerMergedPopDraw now consults stageTrackerPopSheet, so every node test built on stageTrackerCelebrateJS threw ReferenceError before reaching its assertions. Signed-off-by: scanner <280983584+hivecommons-hive[bot]@users.noreply.github.com> * π dashboard: pop-sheet lookup tolerates DOM stubs without querySelector stageTrackerPopSheet called document.querySelector unconditionally, which broke the Go-driven node harness (stage_tracker_cues_test.go) whose document stub has no querySelector, failing TestStageTrackerOdometerNode and the merged-chip tests. Also rebased onto v5 to pick up the phone-chrome probe fix (#17189) for mobile-shots. Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com> Signed-off-by: scanner <280983584+hivecommons-hive[bot]@users.noreply.github.com> * π dashboard: pop-sheet draw ticks consult state, not the DOM stageTrackerMergedPopDraw/stageTrackerQueuePopDraw ran a document-wide querySelector on every draw tick, even with the popover hidden on desktop, and the Node harness (stage_tracker_cues_test.go) stubs document without querySelector, failing TestStageTrackerCelebrateNode, OdometerAfterBeeNode, FlightTargetNode and MergeSessionNode. Open sheets are now registered in _stageTrackerPopSheets by stageTrackerPopSheetOpen and dropped by the hiveWireDialog resolve callback; stageTrackerPopSheet reads that map, and both draw functions early-return when the popover is hidden and no sheet is open. The querySelector tolerance shim is no longer needed. #17177 (stageTrackerOpenSheet) is closed unmerged, so there is nothing to rebase onto. Hive-Fix-Attempt: 1/6 Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com> Signed-off-by: scanner <280983584+hivecommons-hive[bot]@users.noreply.github.com> --------- Signed-off-by: scanner <280983584+hivecommons-hive[bot]@users.noreply.github.com> Co-authored-by: scanner <280983584+hivecommons-hive[bot]@users.noreply.github.com> Co-authored-by: clubanderson <407614+clubanderson@users.noreply.github.com> Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
Why
Since #17176 merged, every dashboard PR to
v5failsmobile-shotswith:#17176 itself failed the same way (the job is
continue-on-error), as do #17177, #17179, #17180 and #17183. Reproduced on pureorigin/v5: the density fixture's first.ham-rowis a work-item cluster (parent issue + child PR + decision + notes) measuring ~418px, and only 397px remain below the chrome at 375x667. No row layout can satisfy the assertion as written.What
tracker-phone-chrome.js: keep the chrome/visibility intent, but accept a first row taller than the remaining viewport when β₯120px of it (its header line) is visible. ReportsfirstRowPeekin the measurements..gitignore: ignore the harness's scratch Chrome profile (.chrome-dashboard-mobile-shots/), which β¨ Activity monitor: 44px tap targets, row tools in the header, kebab as a sheet on phonesΒ #17177 accidentally committed.Verified locally with
node src/scripts/dashboard-mobile-shots.mjs --viewport 375x667@2 --theme light: probe passes (firstRowPeek: 397).testdata/only βno-changelog.