merge_pr_49808

[wdspec] Fix WebDriver classic and BiDi drag and drop position tests.

Differential Revision: https://phabricator.services.mozilla.com/D232725

bugzilla-url: https://bugzilla.mozilla.org/show_bug.cgi?id=1871346
gecko-commit: d13caad2c6161773f359021d3ebe132880d93bf6
gecko-reviewers: webdriver-reviewers, Sasha

merge_pr_49800

layout: Fix intrinsic contributions of anonymous blocks

In order to compute the inline min-content and max-content contributions
of an anonymous block, we were finding its min-content and max-content
inline size with a SizeConstraint coming from the block size of the box.

However, anonymous blocks do not establish a containing block for their
contents, so this patch uses a SizeConstraint from the block size of the
containing block.

Signed-off-by: Oriol Brufau [email protected]

merge_pr_49814

FLEDGE: Make B&A WPT tests more hermetic

...by making sure to always explicitly use a .test coordinator.

Bug: 383600441
Change-Id: Ib163b5042ebfdb04d906c9bd549f323cbed45768
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/6094505
Reviewed-by: Mason Freed [email protected]
Reviewed-by: Russ Hamilton [email protected]
Commit-Queue: Maks Orlovich [email protected]
Cr-Commit-Position: refs/heads/main@{#1399474}

merge_pr_49807

Block-in-inline shouldn't keep an IFC from being non-contiguous.

A block-in-inline may only contain fragmented parallel flows, for all we
know, so it cannot unconditionally count as "in-flow", since we might
fail to mark a container as potentially non-contiguous that way.

In the test included, there's a text node "x xx", where the first word
should be in the first column, and the second word should be in the
fourth. There's no room for such a wide piece of text ("xx") before
we're past the tall float.

Bug: 346876226
Change-Id: I5bcf8840a1c71783dfa0bdce442a92eb5ddbf7b4
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/6108448
Reviewed-by: Ian Kilpatrick [email protected]
Commit-Queue: Morten Stenshorne [email protected]
Cr-Commit-Position: refs/heads/main@{#1399542}

merge_pr_49811

webnn: add wpt for bidirectional lstm

Fixes some minor bugs with CoreML bidrection implementation.

Also adjusted ULP for this complex op to 3, this is same as gru.

Bug: 360052663
Change-Id: Ie15ab15a359be70e219fd61ee0d3a3fea21f3405
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/6103076
Commit-Queue: Phillis Tang [email protected]
Reviewed-by: Reilly Grant [email protected]
Cr-Commit-Position: refs/heads/main@{#1399398}

merge_pr_49810

DOM: Test 'slotchange' event firing during move

See whatwg/dom#1307 (comment).

R=[email protected]

Bug: 40150299
Change-Id: I9529941e05591620da5b3c11635693a15f0ce866
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/6111212
Reviewed-by: Noam Rosenthal [email protected]
Commit-Queue: Dominic Farolino [email protected]
Cr-Commit-Position: refs/heads/main@{#1399350}

merge_pr_49809

Update Wasm tests (#49809)

• Update Wasm tests

• Ignore linting errors in automatically imported Wasm tests

---

Co-authored-by: wpt-pr-bot [email protected]
Co-authored-by: Panos Astithas [email protected]

merge_pr_49804

Only update first appended element

Since we are only planning to support one element in
, this patch only performs a clone into the first in tree order. This is in response to feedback here: https://github.com/whatwg/html/pull/10633#discussion_r1834376985 This patch also removes the logic which clears elements when they are removed/disconnected. Change-Id: I1580ec9f12df463d1f5134905e3e527cfefa694d Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/6043186 Reviewed-by: David Baron [email protected] Commit-Queue: Joey Arhar [email protected] Cr-Commit-Position: refs/heads/main@{#1399301}

merge_pr_49803

[carousel] Implement targeted scroll-marker selection for ::column

::column pseudos[1] may create ::scroll-markers, but do not exist in the
DOM tree as parents of the elements which are flowed into them.
They also do not appear to have LayoutObjects and are not found[2]
in the layout tree. So, when searching for the appropriate scroll-marker
to select for a targeted scroll we need to consider whether the
scroll-marker-generating element (marker-generating-element) found in
the layout tree is also flowed into the same scroll-marker-generating
::column (marker-generating-column) as the target of the scrollIntoView
operation. If the target and marker-generating-element are both in
marker-generating-column, marker-generating-element is the preferred[3]
choice. The only means we have of checking whether an element is flowed
into a ::column is by looking at the rect associated with that ::column.

[1] https://drafts.csswg.org/css-multicol-2/#column-pseudo
[2] https://source.chromium.org/chromium/chromium/src/+/165f5ddad25b289f886a248ca810d075b1e7bb87:third_party/blink/renderer/core/page/scrolling/snap_coordinator.cc;l=254-255
[3] w3c/csswg-drafts#10738 (comment)

Bug: 380062280
Change-Id: Ie2ad58c3beca13a4945325d9aa9c4aec6a5fc332
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/6088285
Commit-Queue: David Awogbemila [email protected]
Reviewed-by: Steve Kobes [email protected]
Cr-Commit-Position: refs/heads/main@{#1399198}

merge_pr_49802

[carousel] Pin the selected scroll-marker for targeted scrolls

The CSS working group resolved[1] that when a scroll operation is aimed
at an element, i.e. Element.scrollIntoView, the associated
scroll-marker-group should select the active scroll-marker based on the
element which the operation is intending to scroll to. In such cases,
the scroll-marker that should be selected is the scroll-marker
associated with the first scroll target (a scroll target is an element
which generates a scroll-marker) found through a search starting from
the target of the scrollIntoView itself and backwards in tree-order.
As soon as some other type of scroll occurs, e.g. Element.scrollTo, or
a user gesture scroll, the scroll-marker-group should no longer
consider its active marker pinned, i.e. it should be based on the
scroll position.

This patch implements this for elements in general, but not for
::column pseudo elements which may also act as scroll targets since
::column::scroll-marker is allowed. The ::column case needs to be
handled specially as ::column pseudos are not parents of the elements
which are flowed into them in the DOM tree. This will be done in a
follow-up patch.

[1] w3c/csswg-drafts#10738 (comment)

Bug: 380062280
Change-Id: I363e0f055f9791ead0b35f4bbe037db91f299624
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/6089232
Reviewed-by: Steve Kobes [email protected]
Commit-Queue: David Awogbemila [email protected]
Cr-Commit-Position: refs/heads/main@{#1399197}