Skip to content

fix(dashboard): scroll the list when "More below" is clicked - #31800

Merged
bolichen97 merged 1 commit into
mainfrom
fix/21588-more-below-scrolls
Nov 7, 2026
Merged

bolichen97 merged 1 commit into
mainfrom
fix/21588-more-below-scrolls

Conversation

@bolichen97

@bolichen97 bolichen97 commented Nov 6, 2026 •

Copy link
Copy Markdown
Contributor

Problem / Motivation

Goal: Clicking the "More below" strip under a clipped Settings list scrolls that list down, so the strip does what it looks like it does.

A first-time user in the nightly GUI test opened Settings on a short window, saw "More below" in accent text under the side rail, clicked it, and nothing happened. They had to scroll the list by hand to find Developer. The same strip sits under the two-pane section list inside a Settings tab.

Why it matters

The strip reads as a control (accent colour, a chevron, its own bar), so a click that does nothing reads as broken. The rows it points at (Developer, Releases, About) are the ones a new user is looking for on a laptop-height window. The GUI lane has reported this three times.

Not a goal

  • Changing when the strip shows or hides, its wording, or its look beyond a pointer cursor.
  • Moving keyboard focus into the list on click: the rows stay where they are in the tab order.
  • Changing other scroll cues (the MORE_BELOW_MASK fades, JumpToBottomButton).

What changed (motivation → approach → change)

The strip was a plain label with no click behaviour, while its styling invites a click.

MoreBelowCue is now a real <button> that takes the list it sits under (scrollerRef). A click scrolls that list down by 80% of its visible height, so the rows that were at the bottom stay in view; it scrolls smoothly, or jumps when the user prefers reduced motion. Both hosts, the Settings side rail (SidePanelLayout) and the capped two-pane section list (SettingsSubNav), keep a ref to their scrolling box next to the existing edge-measuring ref and pass it in. This is the same shape as the chat suggestion strip's scroll arrows in FollowUpBar.

At the end of the list the strip hides its label as before, which also takes the button out of the tab order, so it is never a stop that does nothing. If the button has keyboard focus when it hides, focus moves to the last row of the list instead of falling back to the page. website/docs/page-layout.md and the translator note for components.sidePanelLayout.more_below in en.context.json now describe it as a button.

Backwards compatibility

Compatible: the strip shows and hides exactly as before; the only change is that a click on it scrolls the list it sits under.

Removes nothing: the label, its visibility rules and every row stay; the new prop is required at both call sites, and both pass it.

Tests

  • SidePanelLayout.tabListScrollCue.test.tsx: the rail's strip is the button named "More below"; a click calls scrollBy on the tab list with 80% of its height, smooth, and with behavior: 'auto' under reduced motion; the rows are still real buttons; a focused strip that hides at the end of the list hands focus to the last row.
  • SettingsSubNav.railPin.test.tsx: the capped section list's strip is that button too, and a click scrolls that list's box the same way.
  • Both fail on origin/main's components (2 failed, 8 passed) and pass with the fix (10 passed). Determinism: 20 serial repeats of both files green, and a shuffled run (--sequence.shuffle --sequence.seed=19969) green.

Proof: not provable -- frontend test (prove.py runs Python only)

Manual verification

Ran the built dashboard gateway-free (stubbed API) at 1280x600 on /settings, dark and light: the rail is clipped and shows "More below"; one click scrolls the rows up with Shortcuts and Skills still in view; a second click reaches Developer, Releases and About, and the label hides at the end.

Screenshots / video

Settings rail at 1280x600, cropped to the rail. Before the click, after one click, and at the end of the list:

Before After one click End of list
Dark dark before dark after dark end
Light light before light after light end

Recording of the two clicks (dark):

clicking More below

Related Issues

Closes #21588

Sibling sweep

Swept: git grep -n "more_below|MoreBelowCue|MORE_BELOW_MASK" -- website/src
Sibling: website/src/components/SidePanelLayout.tsx -- changed here
Sibling: website/src/components/SettingsSubNav.tsx -- changed here
Sibling: website/src/components/ModelEffortDropdown.tsx -- not affected -- MORE_BELOW_MASK is a fade with no label, nothing invites a click

Pattern harvest

Not generalizable: one shared component styled like a control with no action; both of its hosts are fixed here.

Checklist

  • At most two commits (one is the norm), with a Conventional Commits title (feat|fix|docs|style|refactor|perf|test|chore|ci|build|revert: ...)
  • Existing tests pass and new tests added for new functionality
  • Self-review completed; code follows project style guidelines
  • Documentation updated (if applicable)
  • No secrets, credentials, or internal references in the diff

Contribution License Agreement

Contributed under the project's Apache-2.0 license (see CONTRIBUTING.md).

@bolichen97
bolichen97 requested review from a team and KiroCrewCaptain as code owners November 6, 2026 20:56
@bolichen97
bolichen97 enabled auto-merge (squash) November 6, 2026 21:13
@github-actions

github-actions Bot commented Nov 6, 2026 •

Copy link
Copy Markdown
Contributor

First Principles Review (Opus 5.5) — ✅ PASS

Premise-level review of b6087520871b7895ee6cb49acabfbc649c9f077f — why this exists and whether the shipped surface is the smallest honest version. Updated in place on each push. A BLOCK verdict blocks PR readiness; PASS does not.

First-Principles-Verdict: PASS

Nothing blocks. The deleted "a label, not a control: decorative" pin is countered by linked issue #21588 and three GUI-lane reports of users clicking a strip that did nothing.

What this change ships

Inventory (5 items) — 5 justified

Intent: let someone who sees "More below" under a clipped Settings list click it and reach the hidden rows (FIX).

  1. Clicking the rail's "More below" strip scrolls the Settings side list down by most of its height — justified
  2. The same click behaviour on the capped two-pane section list inside a Settings tab — justified
  3. The strip becomes a focusable button with a pointer cursor and is announced to screen readers; it was previously aria-hidden — justified
  4. A strip that hides while focused now moves keyboard focus to the last row — justified
  5. Under reduced motion the scroll jumps instead of animating — justified

[FIRST-PRINCIPLES-REVIEWED] b608752

Description read

[DESCRIPTION-READ] b02682c82f3d1b3e2286a35044bef5d2fabd89d75435c73e6cf603ebe3535ba6

That sha256 names the PR description this verdict read. Recompute it with .github/scripts/pr-description-capture.sh; if it differs, the description changed after this verdict was formed and any finding drawn from the description is unproven.

@github-actions

github-actions Bot commented Nov 6, 2026 •

Copy link
Copy Markdown
Contributor

GPT 6.1 Review — ✅ no blocking findings

GPT 6.1 completed its review of b6087520871b7895ee6cb49acabfbc649c9f077f and found no blocking issues.

This comment is updated in place on each push.

Review details

No findings.
[GPT-REVIEWED] b608752

False positive or not applicable? A repository writer can comment:
/ai-review override gpt b6087520871b7895ee6cb49acabfbc649c9f077f false-positive: <reason, 20 to 500 characters>

@github-actions

github-actions Bot commented Nov 6, 2026 •

Copy link
Copy Markdown
Contributor

Design Review (Opus 5.5) — ✅ PASS

Design-level review of b6087520871b7895ee6cb49acabfbc649c9f077f — updated in place on each push. A BLOCK verdict blocks PR readiness; PASS does not.

Design-Verdict: PASS

Nothing blocks: the click now does what the strip's styling suggests, the fix is the same at both hosts, and screenshots of this revision are attached.

Notes

  • This deletes the old "decorative label, not a control" pin. The description gives no reason that pin's rationale ("every row it points at is already in the tab order") no longer holds, so add one line saying why.

[DESIGN-REVIEWED] b608752

Description read

[DESCRIPTION-READ] cec395a3bbde66f7b0ac470d0eaa92b3359547f10bf3739ef16248570e552a47

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 .github/scripts/pr-description-capture.sh over the same inputs; if it differs, the description or that evidence changed after this verdict was formed and any finding drawn from either is unproven.

@github-actions

github-actions Bot commented Nov 6, 2026 •

Copy link
Copy Markdown
Contributor

Opus 5.5 Review — ✅ no blocking findings

Reviewed b6087520871b7895ee6cb49acabfbc649c9f077f — this comment is updated in place on each push.

Review details

No findings.

[OPUS-REVIEWED] b608752

Verdict parsed from the review's SHA-scoped output markers for commit b6087520871b7895ee6cb49acabfbc649c9f077f.

False positive or not applicable? A repository writer can comment:
/ai-review override fable b6087520871b7895ee6cb49acabfbc649c9f077f false-positive: <reason, 20 to 500 characters>

@github-actions

github-actions Bot commented Nov 6, 2026 •

Copy link
Copy Markdown
Contributor

UX Review (Opus 5.5) — ✅ PASS

UX-level review of b6087520871b7895ee6cb49acabfbc649c9f077f — updated in place on each push. A BLOCK verdict blocks PR readiness; PASS does not.

UX-Verdict: PASS

Nothing blocks. The blind reader guessed correctly that "More below" scrolls the list and said they would click it. The screenshots and the GIF show the strip before a click, after one click and at the end of the list.

Notes

  • No screenshot shows the SettingsSubNav copy of the strip, under the two-pane section list inside a Settings tab. A wide, capped section list showing "More below" would cover it.
  • The blind reader wrote "I can't be sure it's clickable… it could be just a label." The diff adds a pointer cursor but no hover state. A hover state on the MoreBelowCue button (for example an underline) would tell users it is clickable before they try it.

[UX-REVIEWED] b608752

Description read

[DESCRIPTION-READ] fb974862ec93145535e07a7a30a85c65cd787a52ec19c896a2475b4175f478c3

That sha256 names everything this verdict read: the PR description, plus the 10 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 .github/scripts/pr-description-capture.sh over the same inputs; if it differs, the description or that evidence changed after this verdict was formed and any finding drawn from either is unproven.

@github-actions github-actions Bot added readiness: checking Automated validation is still running readiness: action required A blocking check or review needs attention and removed readiness: checking Automated validation is still running labels Nov 6, 2026
@bolichen97
bolichen97 force-pushed the fix/21588-more-below-scrolls branch from 44f213a to 4f350ee Compare November 7, 2026 00:55
@bolichen97

Copy link
Copy Markdown
Contributor Author

Rebased: main moved in website/src/i18n/en.context.json since this head went green (26c8238 -> f71f5dd); CI re-running.

The new base also carries #31803, which fixes the Windows shard's pull_request_target allowlist test that was red on the old base; the other reds on the old head were GitHub API rate-limit read failures.

[operator: bolichen97]

@bolichen97
bolichen97 force-pushed the fix/21588-more-below-scrolls branch from 4f350ee to c8ced73 Compare November 7, 2026 04:54
@bolichen97

bolichen97 commented Nov 7, 2026 •

Copy link
Copy Markdown
Contributor Author

Rebased: main moved in website/src/i18n/en.context.json since this head went green (f71f5dd -> e4611c5); CI re-running.

The patch is unchanged (git range-diff reports it identical). The previous head's reds were all GitHub API rate-limit failures in CI (Dependency License Gate, PR Policy, Sensitive Change Review, the readiness signal, and the AI review lanes), not findings. Locally: the related vitest set (606 files, 8583 tests), eslint on the touched files and lint:i18n pass on 2552c90; tsc -p tsconfig.app.json and the touched test files pass again on e4611c5, whose extra commits touch none of this PR's files.

[operator: bolichen97]

@bolichen97
bolichen97 force-pushed the fix/21588-more-below-scrolls branch from c8ced73 to 0f12e58 Compare November 7, 2026 07:53
@bolichen97

Copy link
Copy Markdown
Contributor Author

Rebased: main moved in website/src/components/SidePanelLayout.tsx and website/src/i18n/en.context.json since this head went green (e4611c5 -> 5f7f8d1); CI re-running.

The patch is unchanged (git range-diff reports it identical); #31471's edit to SidePanelLayout.tsx is in the tab-selection code, a region this PR does not touch. The previous head's reds were GitHub API rate-limit failures in CI (Dependency License Gate, PR Policy's screenshot read, Sensitive Change Review, the readiness signal, the AI review lanes) plus a setup-python step failure in Backend Tests shard 1 that Coverage Gate inherited, not findings. Locally on b65af1a: the related vitest set (611 files, 8635 tests) passes except one ChatPage.responsivePanel case that failed under host load and passes when its file runs alone; tsc -p tsconfig.app.json, eslint on the touched files and lint:i18n pass. On 5f7f8d1, whose extra commits touch only en.context.json among this PR's files, lint:i18n and the touched test files pass again.

[operator: bolichen97]

@bolichen97

Copy link
Copy Markdown
Contributor Author

[operator: bolichen97] Backend Tests (3.12, 2), Backend Tests (Windows) (2) and the Coverage Gate behind them fail on test/test_security_gate_liveness.py::test_the_package_stays_within_its_line_budget (34855 > 34845 lines). This PR changes only website/; origin/main at 0795ebc carries the same count, so the red is main's. #32183 raises that budget; this PR waits for it and re-runs on top of it. The other red checks on this head are GitHub API rate-limit failures inside CI, not findings.

@bolichen97
bolichen97 force-pushed the fix/21588-more-below-scrolls branch from 0f12e58 to 2fd9c7a Compare November 7, 2026 14:12
@bolichen97

Copy link
Copy Markdown
Contributor Author

Rebased: main moved in website/src/i18n/en.context.json since this head went green (5f7f8d1 -> 959bc04); CI re-running.

The new base carries #32252 and #32209, which fix the two main reds (the security line-budget test on Backend Tests shard 2, Windows shard 2 and Coverage Gate, and the Windows ignore-list test on shard 7). The patch is unchanged (git range-diff reports it identical). The previous head's First Principles, Design and other red lanes posted no verdict: they failed on GitHub API rate limits, not on findings. Locally on 2fd9c7a: the 42 test files that reference the touched components (640 tests), tsc -p tsconfig.app.json, eslint on the touched files and lint:i18n pass; local GPT and Opus reviews returned PASS.

[operator: bolichen97]

@bolichen97
bolichen97 force-pushed the fix/21588-more-below-scrolls branch from 2fd9c7a to b608752 Compare November 7, 2026 14:36
@github-actions github-actions Bot added readiness: checking Automated validation is still running and removed readiness: action required A blocking check or review needs attention labels Nov 7, 2026
@bolichen97

bolichen97 commented Nov 7, 2026 •

Copy link
Copy Markdown
Contributor Author

self-added: yes
mechanism: focus hand-off when the More below button hides
resolution: narrowed: one layout effect in MoreBelowCue, no new prop or host change; the earlier blocked pushes were rate-limit no-verdict reruns, not findings

  • span=9b54fe009fc6 MoreBelowCue hides while it holds keyboard focus

fixed in b608752: when visible turns false while the button is document.activeElement, a layout effect moves focus to the last focusable row inside scrollerRef, before the browser paints the hidden button. Both hosts (SidePanelLayout, SettingsSubNav) pass that ref.
Test: SidePanelLayout.tabListScrollCue.test.tsx focuses the strip, scrolls to the end, and expects focus on the last row; it fails on the previous head and passes on this one.

[operator: bolichen97]

@github-actions github-actions Bot added readiness: passed Eligible automated validation passed for the current revision and removed readiness: checking Automated validation is still running labels Nov 7, 2026

@KiroCrewCaptain KiroCrewCaptain left a comment

Copy link
Copy Markdown
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Tier 1 auto-approve: fix (7 files). Criteria: no conflict, no requested changes, security path denylist clean, design-doc gate clean, security checklist all-NO, no reviewer-lane findings, readiness signal fresh for this head SHA, declared category matches the diff. Category: the 'More below' cue becomes a button that scrolls its list down; with updated tests.

@bolichen97
bolichen97 merged commit b0ffbf6 into main Nov 7, 2026
98 of 99 checks passed
@bolichen97
bolichen97 deleted the fix/21588-more-below-scrolls branch November 7, 2026 16:10
buluoray pushed a commit that referenced this pull request Nov 8, 2026
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

readiness: passed Eligible automated validation passed for the current revision

Projects

None yet

Development

Successfully merging this pull request may close these issues.

ux(members): Clicking "More below" did nothing visible, so I had to scroll the list myself to find Dev…

2 participants