Skip to content

fix(ui): end the list search dock at the glass pane, not 4px below it - #14995

Merged
iamwhatever merged 1 commit into
mainfrom
fix/glass-dock-seam
Sep 29, 2026
Merged

iamwhatever merged 1 commit into
mainfrom
fix/glass-dock-seam

Conversation

@CrysisDeu

Copy link
Copy Markdown
Contributor

Problem / Motivation

Goal: A row scrolling under the list panels' glass search field emerges at the glass, not through a strip of a third tone below it.

After #14896 the dock ended 4px under the pane: the search row kept its pb-1, and with nothing on the shelf the shelf still painted a 4px scrim fade. Scrolled under, a row passed through that band — half-clipped text on a pale strip, then a hard step into the row's own colour — and the maintainer read it as a leftover box behind the pane.

Why it matters

The seam sits on the two surfaces every user scrolls first, and it undoes the point of the material: the glass should be the only thing between the list and the chrome.

Not a goal

  • No change to the material, the field's radius, or the rest position of the first row (still 4px below the dock).
  • No change to how rows look under the glass itself (blurred through the pane, sharp below it — as under the composer).

What changed (motivation → approach → change)

The pane's bottom edge is the dock's edge. SearchFilterBar's wrapper drops its pb-1 (both mounts sit in a dock, whose business the gap now is); the shelf is empty:hidden, so a bare field's dock ends at the glass (measured: dock 40px = pane bottom on both lists, was 44px). The roster's star-error wrapper mounts only while there is an error, so an empty wrapper no longer holds the shelf — and its 4px scrim — open.

The rest gap moves into the measured variable. ListDock publishes --list-dock-h as its height plus 4px and --list-dock-pin-inset as 4px, so LIST_BODY_CLS still pads and scroll-pads the lane to dock + 4 (first row clear at rest, deep-linked rows land clear) while a pinned folder header, offset by the inset, sits exactly on the dock's bottom edge (measured 40.14 = 40.14). clearPinnedHeaders subtracts the inset too, so the keyboard-reveal band ends where the header stack ends instead of 4px past it (the overshoot Opus noted on #14896).

Backwards compatibility

Compatible: same components, ids and names; SearchFilterBar's default wrapper loses 4px of bottom padding and its only two mounts are the docks this fixes.

Tests

  • src/test/ListDock.test.tsx: --list-dock-h is the measured height plus the gap (48px for a 44px dock, 88 after growth) and --list-dock-pin-inset is 4px.
  • playwright/sidebar-sticky-folder-header.spec.ts (unchanged) still measures pinned headers against the dock's bottom edge, which is where they sit.
  • src/test/ChatSidebar.arrowNav.test.tsx (unchanged) covers the band with no dock var; the inset subtraction is exercised by the sticky spec's arrow-key claim in the browser gate.
  • Full suite not run locally by policy; CI is the runner.

Manual verification

website/scripts/capture-glass-search-dock.mjs now asserts each scroller's padding equals dock height + 4 and the pinned header's top equals the dock's bottom, on both lists, dark and light, in all seven scenes.

Screenshots / video

Sessions sidebar, dark — rows scrolled under; the row now emerges at the pane's edge (compare #14896's frame, where a pale 4px band sat between):

Sessions scrolled under the glass, dark
Pinned folder header on the pane's edge, dark

Rest, roster, light

Sessions at rest, dark
Crew Members scrolled, dark
Sessions scrolled, light

Related Issues

Follow-up to #14896.

Pattern harvest

Not generalizable: a padding left on a row that became the edge of a floating surface; the dock now owns its own edge.

Checklist

  • At most two commits (one is the norm), with a Conventional Commits title
  • 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

The dock shipped in #14896 ended 4px under the pane: the search row kept
its pb-1, and a bare field's shelf still painted a 4px scrim fade. A row
scrolling under the dock therefore emerged through a strip of a third
tone between the blurred glass and the sharp list, and that strip read as
a leftover box behind the pane.

The pane's bottom edge is now the dock's edge. The search row drops its
bottom padding, the shelf hides when empty (the roster's star-error
wrapper mounts only while there is an error, so it does not hold the
shelf open), and the 4px rest gap between dock and first row moves into
the measured variable: ListDock publishes --list-dock-h as its height
plus the gap and --list-dock-pin-inset as the gap, so the lane pads and
scroll-pads to dock + 4 while a pinned folder header still sits exactly
on the dock's bottom edge. The keyboard-reveal band subtracts the inset
too, so it ends where the header stack ends instead of 4px past it.
@CrysisDeu
CrysisDeu requested a review from a team September 29, 2026 09:04
@CrysisDeu
CrysisDeu requested a review from a team as a code owner September 29, 2026 09:04
@github-actions github-actions Bot added the readiness: checking Automated validation is still running label Sep 29, 2026
@github-actions

Copy link
Copy Markdown
Contributor

First Principles Review (Fable 5) — ✅ PASS

Premise-level review of f33b496265d341f73805ab45febd8045db3fb18f — 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/CONCERNS are advisory.

First-Principles-Verdict: PASS

The reversed "Never hidden" shelf pin rests on the maintainer's #14896 report — confirm that comment exists before merge; it is the sole evidence.

What this change ships

Inventory (7 items) — 7 justified

Intent: rows scrolling under the glass search field should emerge at the pane's edge, not through a 4px pale band below it — a FIX (follow-up to #14896).

  1. Rows scrolled under a bare search field emerge at the glass; the 4px band is gone — justified
  2. Empty shelf no longer renders, reversing the prior "Never hidden" pin — justified
  3. Roster star-error wrapper mounts only while an error exists — justified
  4. First row's rest position preserved 4px below the dock via --list-dock-h including the gap — justified
  5. Pinned folder headers still sit on the dock's bottom edge via the 4px pin inset — justified
  6. Arrow-key reveal band now ends at the pinned header stack, not 4px past it — justified
  7. Capture script asserts the exact pad (dock + 4) and header seat, both lists, both themes — justified

[FIRST-PRINCIPLES-REVIEWED] f33b496

Description read

[DESCRIPTION-READ] 9ebf2b5ed70c8bd32c3b754f301ec1b343ec61612c0cd4860ac9374e6d564fc9

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

Copy link
Copy Markdown
Contributor

Design Review (Fable 5) — ✅ PASS

Design-level review of f33b496265d341f73805ab45febd8045db3fb18f — updated in place on each push. A BLOCK verdict blocks PR readiness; PASS/CONCERNS are advisory.

Design-Verdict: PASS

Verify the bare-field dock stays pinned closed: a future always-mounted shelf child silently re-opens the 4px scrim, and no current gate catches it.

Suggestions

  • The capture script's assertion is relative (pad = dock + 4), so the regression class this PR fixes — an idle wrapper holding the shelf open — passes it; add one absolute check that the bare-field scene's dock bottom equals the glass pane's bottom.

[DESIGN-REVIEWED] f33b496

Description read

[DESCRIPTION-READ] dde9e3208002def9e5daca1f48fcc3e4a91956cdfea5ef7f234301b28e93e254

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

Copy link
Copy Markdown
Contributor

GPT 5.6 Review — ✅ no blocking findings

GPT 5.6 completed its review of f33b496265d341f73805ab45febd8045db3fb18f and found no blocking issues.

This comment is updated in place on each push.

Review details

No findings.
[GPT-REVIEWED] f33b496

False positive or not applicable? A repository writer can comment:
/ai-review override gpt f33b496265d341f73805ab45febd8045db3fb18f: <one-sentence reason>

@github-actions

Copy link
Copy Markdown
Contributor

UX Review (Fable 5) — ✅ PASS

UX-level review of f33b496265d341f73805ab45febd8045db3fb18f — updated in place on each push. A BLOCK verdict blocks PR readiness; PASS/CONCERNS are advisory.

UX-Verdict: PASS

Verify arrow-key navigation on a grouped list once: the reveal-band inset change (clearPinnedHeaders) is behavioral and no static shot can show it.

[UX-REVIEWED] f33b496

Description read

[DESCRIPTION-READ] 9190ecdeb104b3c7995f8460e8a66295811032114845c552ef28edbb47a82288

That sha256 names everything this verdict read: the PR description, plus the 8 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

Copy link
Copy Markdown
Contributor

Opus 5 Review — ✅ no blocking findings

Reviewed f33b496265d341f73805ab45febd8045db3fb18f — this comment is updated in place on each push.

Review details

No findings.

[OPUS-REVIEWED] f33b496

Verdict parsed from the review's SHA-scoped output markers for commit f33b496265d341f73805ab45febd8045db3fb18f.

False positive or not applicable? A repository writer can comment:
/ai-review override fable f33b496265d341f73805ab45febd8045db3fb18f: <one-sentence reason>

@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 Sep 29, 2026

@iamwhatever iamwhatever left a comment

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

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

Reviewed. ListDock sets --list-dock-h to height+4 and --list-dock-pin-inset to 4px; the pinned header lands at the dock's bottom edge; the shelf gets empty:hidden and SearchFilterBar drops pb-1. Only ChatSidebar and MembersPage mount ListDock and both shelves are conditional, so :empty holds at rest. sessionRowNav reads the inherited property and falls back to 0 with no dock, so no-dock behaviour is unchanged. No security touch.

Non-blocking test gap: the inset subtraction in clearPinnedHeaders (sessionRowNav.ts:103) has no unit test, and nothing in CI exercises empty:hidden; a small arrowNav test with --list-dock-pin-inset set on the scroller would pin it.

@iamwhatever
iamwhatever enabled auto-merge (squash) September 29, 2026 16:32
@iamwhatever
iamwhatever merged commit 046e5c9 into main Sep 29, 2026
96 checks passed
@iamwhatever
iamwhatever deleted the fix/glass-dock-seam branch September 29, 2026 16:38
@github-actions github-actions Bot removed the readiness: passed Eligible automated validation passed for the current revision label Sep 29, 2026
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants