merge_pr_49326

[CompositeClipPathAnimation] Fix crash with position: absolute

A new clip path crash was discovered during experimentation,
exacerbated by a recent change that enables clip path animations
when there is a positive delay or clip-path: none.

In this case when the positioning is absolute, the clip path style
property is a nullptr instead of being filled with the value for
none. This breaks assumptiosn in clip path clipper, which assume
(reasonably, based on prior usage of the code) that an element
with a clip path animation or a mask based clip will have a clip path
value in its style.

This change adds conditionals and removes a DCHECK to prevent any
crash in this case.

Bug: 378027650

Change-Id: If7e2814a4de647c1f6cebe6e7f8291419bcac018
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/6001026
Reviewed-by: Philip Rogers pdr@chromium.org
Commit-Queue: Claire Chambers clchambers@microsoft.com
Cr-Commit-Position: refs/heads/main@{#1386634}

merge_pr_49314

Add "remoteAsyncEvents" and "sessionHistoryInParent" keywords in the list of update_properties for wpt-update command.

Differential Revision: https://phabricator.services.mozilla.com/D229771

bugzilla-url: https://bugzilla.mozilla.org/show_bug.cgi?id=1932597
gecko-commit: c1a15d8f1a23b026bb4e33c2adf5871e41d9a776
gecko-reviewers: jgraham

merge_pr_49313

Ensure inline elements inside an empty line has zero block-size and no baseline in scrolled block frames.

Differential Revision: https://phabricator.services.mozilla.com/D229709

bugzilla-url: https://bugzilla.mozilla.org/show_bug.cgi?id=1931466
gecko-commit: 160977ed2e6cd8576c70840d388d1a6bcfc07262
gecko-reviewers: layout-reviewers, dholbert

merge_pr_49320

Fix leak in single-touch.html

This test could trigger leak detection since not waiting on pointerup
after all of the async tests have completed.

Actions().send() does not wait on the dispatched events to be handled.
Each async test marked itself as done once the expected event was
received. Refactored to be a promise driven test.

2 flakes out of 100 runs without the fix.
0 flakes out of 500 runs with the fix.

Bug: 40945204
Change-Id: I5468e3bc67ecdfa5c09e53a705abdcbc22024cde
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/6033293
Commit-Queue: Kevin Ellis kevers@chromium.org
Reviewed-by: Mustaq Ahmed mustaq@chromium.org
Cr-Commit-Position: refs/heads/main@{#1386408}

merge_pr_49319

VT: Deflake nothing captured view transition test.

Instead of relying on a double rAF, use waitForCompositorReady in
an attempt to deflake the test.

Also update the takeScreenshot condition to await the ready
promise instead of a double raf.

R=kevers@chromium.org

Bug: 378536381, 378637954
Change-Id: Icc54a556d7b26d9d74bb9b7fd39a25aef6adccd5
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/6041322
Commit-Queue: Vladimir Levin vmpstr@chromium.org
Reviewed-by: Kevin Ellis kevers@chromium.org
Cr-Commit-Position: refs/heads/main@{#1386391}

merge_pr_49317

DOM: Support disconnected->disconnected moveBefore()

Per discussion in whatwg/dom#1307, we've
decided to support moveBefore() in the disconnected->disconnected
scenario. This CL enables and tests that.

R=nrosenthal@chromium.org

Bug: 40150299
Change-Id: Ia0cf64a1a623c53ed5d9ae01b50b0e04f7028da2
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/6037586
Reviewed-by: Noam Rosenthal nrosenthal@chromium.org
Commit-Queue: Dominic Farolino dom@chromium.org
Cr-Commit-Position: refs/heads/main@{#1386266}

merge_pr_49316

[css-pseudo] Fix marker-text-combine-upright.html

This test was failing for 2 reasons:

1. ::marker had text-align: start and text-align-last: start as
   important UA rules, while the style adjuster only enforces
   text-align: center in the presence of text-combine-upright: all.
   This resulted in text-align: center; text-align-last: start.

   Arguably the style adjuster should also set text-align-last,
   but a smaller fix is changing the important UA rule for ::marker
   to text-align-last: auto. Note text-align-last: auto behaves as
   text-align-last: start when text-align: start, so the change
   has no effect in the great majority of cases.

2. The style adjuster enforces white-space: pre on outside markers,
   even if the test uses all: unset.

   So this changes the test reference to expect white-space: pre,
   and explicitly sets white-space: pre in the test for Firefox.

Bug: 1060007

TEST=external/wpt/css/css-pseudo/marker-text-combine-upright.html

Change-Id: I06e6ac4034abadf956b9d7194b945c9a2c338542
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/6021669
Commit-Queue: Oriol Brufau obrufau@igalia.com
Reviewed-by: Rune Lillesveen futhark@chromium.org
Cr-Commit-Position: refs/heads/main@{#1386262}

merge_pr_49315

WebKit export of https://bugs.webkit.org/show_bug.cgi?id=282962 (#49315)

merge_pr_49312

[@scope] Pass TreeScope separately to SelectorChecker

The scope member of SelectorCheckingContext is currently used
for several closely related (but different) things, causing subtle
bugs and a hard-to-understand model.

To address this, pass the TreeScope to SelectorChecker separately,
and use context.tree_scope for ShadowRoot-dependent decisions
in selector matching.

The context.scope field is now only used for two things:

1. For matching the :scope pseudo-class.
2. For creating the default activation for @scope.

This fixes Issue 377647716, because since we no longer "clobber"
context.scope during scope activation, making ParentElement
traverse to the shadow host.

Fixed: 377647716
Change-Id: I833f1dfb03c092927712bc6fd649a414a45ac7a4
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/6035173
Reviewed-by: Rune Lillesveen futhark@chromium.org
Commit-Queue: Anders Hartvoll Ruud andruud@chromium.org
Cr-Commit-Position: refs/heads/main@{#1386249}

merge_pr_49311

Make ConsumeNumberOrPercent() not crash on unresolvable calc().

Instead of calling GetDoubleValue() on percentage calc expressions,
we just return them directly, and then add a new rule saying that
ComputeNumber() on them returns the percentage value divided by 100.
This avoids crashes when we try to force them during parsing.

This also takes us halfway towards fixing the opacity-valid.html test;
we no longer convert it into just a double (without surrounding calc()),
but since we don't properly simplify comparison expressions containing
percentages, we don't simplify as much as we should either, so the test
is still broken. (And even after that, we'd probably need to convert
percentages to decimals inside calc for such properties.)

Change-Id: I8e0e25c35d3df82da75cfb1cf551d7f25d825696
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/6037437
Reviewed-by: Anders Hartvoll Ruud andruud@chromium.org
Commit-Queue: Steinar H Gunderson sesse@chromium.org
Cr-Commit-Position: refs/heads/main@{#1386209}