Plan step 38 review: de-flaking the reduced-motion test moved the click
out of the retried assertScript, which was right, but also dropped the
arrival read in the click's own task, so a smooth scroll finishing
within assertScript's retries would now pass a test named "scrolls
instantly". The click script records at(1) at once and the assertion
requires it. It is read against the slot size measured before the
click: the row re-measures a frame later (222 against 217.7px at
1280px), which is why a later read in the same evaluation fails.
Chromium: CarouselTest 10 of 10, the test itself three runs green.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01Qwx5USif3wFFmxtHg5U1g9
Found running the full browser suite for plan step 38's last batch,
unrelated to its content changes (confirmed by reverting every
touched file and reproducing both failures against the pre-batch
commit): CarouselTest never pinned a window size, so the multi-browse
row's keyline math only lined "scroll by 2 slots" up with the open
keyline at some widths — the host's own default (here, wider than the
carousel was ever exercised at) left a remainder large enough to fail
four assertions. carouselShowcase() now resizes to 1280x900, matching
the width other browser tests already treat as "large desktop".
One test still failed after that: it put a `.click()` inside the
script assertScript asserts, and assertScript retries its whole
expression on a false result (this file's own comment says so) — a
retry re-fires the click, drifting the carousel one item further each
time and never recovering, unlike every other test here, which clicks
through a separate call first. Split the click out the same way.
Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01Qwx5USif3wFFmxtHg5U1g9
Plan step 38 (last batch), plan step 42's target for tests/: every
Tailwind utility class in a tests/Browser/*.php probe's Blade string
or Livewire component render() (~60 class attributes across 14 files)
becomes an inline style built from --md-sys-color-*/--md-sys-shape-*/
--md-sys-measurement-* tokens or a literal px value for an arbitrary
demo size — PickingTest's clip box, every probe's <body
class="bg-surface">, the grid/stack/row wrapper divs, the carousel
item's coloured filler. No test's assertions, selectors or expected
text change; these are layout containers around the components under
test, not anything a test reads.
Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01Qwx5USif3wFFmxtHg5U1g9
Plan step 36 review of a149970d. Reduced motion writes a zero inset for
every item, so isMasked() never called one partly shown and a press on
an item the row cut off did nothing. It now reads the item's box against
the scrollport there, as it already did for multi-aspect, and the
browser test the rewrite added to document the gap asserts the reveal
instead (plan step 32's owed test).
A focused item's inset ring is 3px in again: Tailwind's
-outline-offset-3 is 3px, not the 12px the rewrite wrote.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01Qwx5USif3wFFmxtHg5U1g9
Plan step 36 (containment group): <x-carousel>'s and <x-carousel-item>'s
class lists move into resources/css/components/carousel.css and
carousel-item.css, keyed on data-md-carousel (its value is the layout)
and the parts' data-md-carousel-* hooks (-probe, -scroller, -controls
with "auto"/"always", -previous/-next, -item, -surface, -content,
-label/-label-text). Every selector uses a `>` combinator rather than a
bare descendant one, because a carousel item can itself hold a nested
carousel whose own root would otherwise match its parent's layout
rules too (list.css already solves the same problem for segmented
list rows).
Behaviour is unchanged: resources/js/carousel.js (the keyline maths,
C-05's reduced-motion fix, C-11's vertical full-screen layout, C-12's
padding, C-18's item-as-tab-stop, the multi-aspect layout) is touched
only where it reads or writes the renamed hooks and dataset
properties; every inline custom property it writes
(--material-carousel-*) is untouched. The item renders the shared
md-focus-ring class (foundation/interaction.css) instead of a
hand-rolled ring, refined to an inset offset since an outward one
would draw under the neighbouring item. The previous/next buttons
mirror whole in RTL from carousel.css rather than through
<x-icon mirror-rtl>, which <x-button icon> has no prop to reach (a
component outside this batch); the technique matches how the
Tailwind-era markup already mirrored the whole button. The overlay
label's literal white ink over the scrim (C-25) is kept, with the
same reasoning as before.
Hooks renamed data-material-carousel(-item/-surface/-content/-label)
-> data-md-carousel(-item/-surface/-content/-label), data-padding(-end)
-> data-md-padding(-end), data-centered -> data-md-centered, updated in
the same commit: resources/js/carousel.js, tests/Feature/Components/
CarouselTest.php (rewritten on data-md-* and ComponentStylesheet) and
tests/Browser/CarouselTest.php. Browser tests owed by
docs/plans/material-3-browser-tests.md, added but not run: the
multi-aspect carousel's previous/next, arrow keys, Home and End
(scoped by data-md-carousel="multi-aspect" rather than a position in
the showcase, so reordering its examples cannot silently mis-target
the wrong carousel); a reduced-motion click on an item cut off only by
the row's own edge, which documents rather than fixes a real gap —
isMasked()'s inset check is always false once C-05 zeroes every
item's inset, so the click-to-reveal affordance does not fire there
(found by the Chromium baseline, step 32; fixing it is outside a hook
rename).
Imported from the Containment block of components.css.
Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01Qwx5USif3wFFmxtHg5U1g9
Plan step 32, found by the Browser suite. Content padding (16dp by default
since containment.md C-12) shrinks every keyline near the ends, so large
items resting at the start or the end carry a small inset. Bring-into-view
took any inset over half a pixel for "not fully open", so focusing or
pressing a large item at rest scrolled the row to it. The focal keyline's
own inset at the current scroll is the threshold now. The press test also
checks that focusing and pressing an open item moves nothing.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01Qwx5USif3wFFmxtHg5U1g9
Plan step 32. Since containment.md C-12 the carousel defaults to M3's 16dp
of content padding, and Compose's createShiftedKeylineListForContentPadding
shrinks every keyline near the ends by a share of it, so a large item at
rest at the start or the end is masked by 1.33px a side in the showcase.
The tests asked for a zero inset there and failed; `open(i)` asks whether
an item is at the largest size any item has at that position. The helper's
list of showcase carousels gains the multi-aspect one.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01Qwx5USif3wFFmxtHg5U1g9
Plan step 19, containment.md C-18. The row was the focusable `region`
and the items were not focusable at all, which is the thing M3's
accessibility page draws a Don't for: "use Tab to place initial focus on
the first carousel item", "avoid focusing on the carousel container".
Each item is now `tabindex="0"` with the focus ring drawn inside it, the
row is out of the tab order, and from a focused item the arrows move one
item (moving focus with them), Home and End go to the ends, and Space or
Enter opens one that is not fully in view.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01Qwx5USif3wFFmxtHg5U1g9
Plan step 12, containment.md C-05. Only the parallax half of M3's rule
was honoured: the mask was still written every frame, so items kept
growing and shrinking between the keylines. Under reduced motion the
inset, the shift and the label fade are now all zero, so every item
stays at the strategy's large size and the keylines only decide where
the row snaps. `--material-carousel-pin` went with it: with no mask
there is nothing to pin the content to.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01Qwx5USif3wFFmxtHg5U1g9
<x-menu>, <x-fab-menu> and <x-rich-tooltip> gave their popover an id
that is new with every render, and Livewire's morph matches an element
without a wire:key by its id. So every render of the component around
them swapped the popover for a closed copy: an open menu closed, its
listeners stayed behind on the old element (Escape or a press outside
then left aria-expanded="true" on the trigger and focus unreturned), a
keep-open item's wire:click closed its menu when the response came, and
a rich tooltip went on showing the detached bubble, so it never opened
again. <x-carousel>'s row had the same kind of id: after a render it
scrolled without its listeners, and its items stopped re-masking.
The popover, the bubble and the row now carry a wire:key, so the morph
patches them in place and changes the id and anchor name together with
the trigger's, as it already did for everything else. The key goes
through an attribute bag: Livewire compiles a wire:key written in a
template into the key of the loop iteration around it, which would have
given every child component after the menu in a row the same key.
A morph also removes the menu button's ARIA attributes, which only
script writes. menu.js now writes them again after every morph, so the
button of a menu that stays open, and a FAB menu's close look, still
say it is open, and aria-controls names the popover's new id.
A second press on an open menu's button opened it again, render or not:
the popover closes on the press, and the guard against the click that
follows was timed from the toggle event, which is queued and arrives
after that click. It is timed from beforetoggle now.
Browser tests with Livewire probes in Chromium, Firefox and WebKit, and
a render test for the keys and the keys of the child components after.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01RHoXZSHc8gGpZjFmA5fPc2
The showcase is now an overview and one page per section behind a
grouped navigation rail, moved between with wire:navigate, instead of
one long page under a scrolling top bar. The switch exposed a WebKit
bug in the menu: inside a focusable region, WebKit moves focus out of
the closing popover before its toggle event, so Escape no longer
returned focus to the trigger. The menu now reads it on beforetoggle.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01V9NnLxnPp8vaaurb3Z1MFy
Multi-browse, hero, uncontained and full-screen layouts on a native
scroll-snap row, with Compose's keyline maths masking each item as it
scrolls. Completes Phase 5.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01V9NnLxnPp8vaaurb3Z1MFy
