merge_pr_54727

Fix SVG text scaling when using font-size-adjust

SVG text was not properly scaled when font-size-adjust was applied
due to two issues. First, the adjusted font size was not set before
retrieving the font. When font-size-adjust is used, the adjusted size—
not the computed size—should be referenced during font selection.
Second, the aspect ratio of the font was calculated incorrectly in some
conditions. The ratio was derived from the computed size in
FontDescription, which may not match the intrinsic size of the reference
font. This change corrects both issues to ensure proper SVG text scaling
with font-size-adjust.

This change introduces a failure in font-size-adjust-009.html on
Windows. The test assumes that the Noto font has a cap-height of
1462/2048 units. However, DirectWrite on Windows reports a cap-height of
1472/2048 units. This mismatch prevents the test from passing on
Windows. Besides, this change affects the Windows result of
font-size-adjust-system-fallback.html. The Windows expectation was based
on an incorrect aspect-ratio calculation, so we need to update it accordingly.

Test:
external/wpt/svg/fonts/font-size-adjust-scale-text.svg (new)

Bug: 364926904
Change-Id: I6eb7cbde52aed72bdd2df75cf99ce99b56e89a43
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/6473830
Reviewed-by: Dominik Röttsches [email protected]
Commit-Queue: ChangSeok Oh [email protected]
Cr-Commit-Position: refs/heads/main@{#1511868}

merge_pr_54726

[webaudio-testharness] Migrate simple-input-output.html

Convert third_party/blink/web_tests/external/wpt/webaudio
/the-audio-api/the-scriptprocessornode-interface
/simple-input-output.html from the legacy audit.js runner to pure
testharness.js

Bug: 396477778
Change-Id: Ia7317b1bc0c4825c22d02d4da8d28f83fb5e74ae
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/6905430
Reviewed-by: Hongchan Choi [email protected]
Reviewed-by: Michael Wilson [email protected]
Commit-Queue: Saqlain [email protected]
Cr-Commit-Position: refs/heads/main@{#1511834}

merge_pr_54725

WPT: Snapshot internal observers before Subscriber iterates over them

This behavior was already implemented in
https://crrev.com/c/6311209, which was a follow-up bug fix to
https://crrev.com/c/6221901. It was only recently spec'ed in
WICG/observable#214, and this CL adds a WPT for
it.

This behavior was obliquely tested by:
https://source.chromium.org/chromium/chromium/src/+/main:third_party/blink/web_tests/external/wpt/dom/observable/tentative/observable-from.any.js;l=675-740;drc=5f8643cf150e97b694f2214ee092c34cc0ccf5fe,
as discovered in
keithamus/observable-polyfill#38, however this
CL adds a simpler, more direct test for this behavior.

R=jarhar

Bug: 40282760
Change-Id: I43748a09a2f5a235c25ffdbd807e630c187ef2b3
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/6915657
Reviewed-by: Joey Arhar [email protected]
Commit-Queue: Dominic Farolino [email protected]
Cr-Commit-Position: refs/heads/main@{#1511774}

merge_pr_54724

Basic main-thread implementation of EventTrigger

Bug: chromium:441908430

Change-Id: I58e2c0b2ef88f0b49783a971cbf08124fe0795f4
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/6896549
Reviewed-by: Philip Rogers [email protected]
Commit-Queue: Stefan Zager [email protected]
Cr-Commit-Position: refs/heads/main@{#1511747}

merge_pr_54722

Update HSTS top-level navigation test to set HSTS for same origin hostname.

This change adds an explicit step to pin the same origin hostname via
HSTS before testing top-level navigation upgrades. This ensures that the
general use case for HSTS to be taken in to account by all the browsers
is tested .

Change-Id: Iae8f48740240d3f67824e14eca88a7c4b10726ef
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/6846440
Reviewed-by: Mike Taylor [email protected]
Commit-Queue: Mike Taylor [email protected]
Reviewed-by: Andrew Williams [email protected]
Cr-Commit-Position: refs/heads/main@{#1511662}

merge_pr_54720

Update css-view-trransition idl test (#54720)

The idl file has changed name, so update the test.

merge_pr_54719

VT: Remap old child transform based on new parent transform.

If we only have old snapshots, then we don't update the transforms in
response to parent transform changing (in nested vt).

This patch addresses the problem to make sure we do the remapping to
preserve visual continuity if target transform changes.

Bug: 443266269
R=[email protected], [email protected]

Change-Id: I59ab5db56c7f18f012905f7148cdcb568b46798c
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/6917755
Reviewed-by: Steve Kobes [email protected]
Commit-Queue: Vladimir Levin [email protected]
Cr-Commit-Position: refs/heads/main@{#1511669}

merge_pr_54717

Re-add border to customizable select in-page

There was a discussion about this to re-add the borders here:
w3c/csswg-drafts#12510 (comment)

Bug: 357649033
Change-Id: I971f5e8268ca2b247842cad37f71fa2f7372b785
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/6916262
Reviewed-by: Dominic Farolino [email protected]
Commit-Queue: Joey Arhar [email protected]
Cr-Commit-Position: refs/heads/main@{#1511807}

merge_pr_54714

Fix a crash with empty @contents blocks.

Also fixed a typo in the test while we were on it.

Fixed: 442735990
Change-Id: I4c448272bd9bfeef04ba0be60d9ccef751516d85
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/6919119
Reviewed-by: Anders Hartvoll Ruud [email protected]
Auto-Submit: Steinar H Gunderson [email protected]
Commit-Queue: Steinar H Gunderson [email protected]
Cr-Commit-Position: refs/heads/main@{#1511504}

merge_pr_54705

Increase window min-width to keep the window control buttons visible.

When the profiles menu is visible, the window control buttons can be
pushed offscreen on windows and linux.

It looks like 450px was set as max-width in 2020 in bug 1610497, despite
some unhappy comments in bug 897160. It seems like enough time has
passed to bump up the min-width, especially as both mobile viewports and
desktop viewports have increased in resolution in the five years since
the min-width was last increased.

It would be nice to avoid using magic numbers at all, but this change
fixes the current bug for now.

Differential Revision: https://phabricator.services.mozilla.com/D258989

bugzilla-url: https://bugzilla.mozilla.org/show_bug.cgi?id=1960002
gecko-commit: c41e657c71acabb81588d518da846070d84320eb
gecko-reviewers: dao, desktop-theme-reviewers, extension-reviewers, rpl, webdriver-reviewers, jdescottes, devtools-reviewers