merge_pr_55870

Add test for the version part of the XML prolog.

Bug: 441911594
Change-Id: Ie0662f30fa390077578fe8d0deff43e26db7d8c3
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/7118800
Commit-Queue: Dominik Röttsches drott@chromium.org
Reviewed-by: Dominik Röttsches drott@chromium.org
Cr-Commit-Position: refs/heads/main@{#1540530}

merge_pr_55868

[Masonry] Update masonry-grid-placement-named-lines-002.html

This test has dense packing set, but the expectation was failing to
fill in earlier holes in all cases. Update the test to match the
expected dense packing results.

There is one remaining bug where we are failing to place items into
earlier positions in implicit tracks. As such, this test is left as
failing for further investigation, and the comment was updated to
better reflect why the test is still failing.

Bug: 343257585
Change-Id: Iea6a01a5fa4f6489e2681ec7cf09be4bc590eabd
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/7122078
Reviewed-by: Celeste Pan celestepan@microsoft.com
Commit-Queue: Alison Maher almaher@microsoft.com
Reviewed-by: Kurt Catti-Schmidt kschmi@microsoft.com
Cr-Commit-Position: refs/heads/main@{#1540406}

merge_pr_55856

[SVG] Fix precision loss in non-scaling-stroke with large coordinate
translations

When SVG paths use vector-effect="non-scaling-stroke" with large
viewBox coordinate translations (e.g. viewBox="2137678 0 96 50"),
vertical lines would disappear due to floating-point precision loss
in the transform calculations.

The issue occurred because ComputeNonScalingStrokeTransform() was
zeroing out translation components (SetE(0), SetF(0)) under the
assumption that "stroke width is independent of translation."
While this is true for stroke width calculation, removing
translation can lead to large coordinate values, increasing the
risk of precision loss as seen in the bug.

In this CL, we preserve translation components in the
non-scaling-stroke transform. This mitigates the precision loss
caused by large offset but we only preserve this translation for
non-scaling-stroke-path and not in general.

Bug: 40718465
Change-Id: Id92acf5c5798eab264df33d8a4af4759376483f6
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/7106499
Reviewed-by: Fredrik Söderquist fs@opera.com
Reviewed-by: Vinay Singh vinaysingh@microsoft.com
Commit-Queue: Divyansh Mangal dmangal@microsoft.com
Cr-Commit-Position: refs/heads/main@{#1540470}

merge_pr_55855

Move H2ConnectionGuard lock to be connection-specific

This helps avoid #51981, where we frequently saw timeouts from the H2
server, by avoiding holding a global lock across all connections to
the H2 server.

merge_pr_55864

WebKit export of https://bugs.webkit.org/show_bug.cgi?id=301728

merge_pr_55863

[anchor] Switch position-try-order to use containing-block writing-mode.

It was resolved in:
w3c/csswg-drafts#12869 (comment)

That position-try-order should work in the containing-block writing-mode
and not the candidate writing-mode.

Fixed: 452327359
Change-Id: I60984e57cda1dd57dfcb70e10cbd14a272d972c2
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/7120318
Commit-Queue: David Grogan dgrogan@chromium.org
Auto-Submit: Ian Kilpatrick ikilpatrick@chromium.org
Reviewed-by: David Grogan dgrogan@chromium.org
Commit-Queue: Ian Kilpatrick ikilpatrick@chromium.org
Cr-Commit-Position: refs/heads/main@{#1540244}

merge_pr_55862

[gap-decorations] Make row-rule-outset into shorthand using new props

This CL makes column-rule-outset into a shorthand using the new
longhands w3c/csswg-drafts#12603.

We also implement the paint behavior so that these work for row,
moreover, we add new tests for this property, since the existing ones
are using the row-rule-outset shorthand.

We also update some existing parsing tests, and uncomment parts of tests
that now work since we introduced the shorthand again.

Recently it was resolved that the outset properties will be renamed to
inset w3c/csswg-drafts#12848, so they will
need to be renamed in follow up CLs, but since the asymmetric behavior
resolved first, we are implementing that first. It will also allow us to
move to the new inset behavior incrementally.

Bug: 357648037, 440054556
Change-Id: I89086f1df91383468d5379680b7aeb038e99023c
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/7107409
Reviewed-by: Alison Maher almaher@microsoft.com
Commit-Queue: Javier Contreras javiercon@microsoft.com
Cr-Commit-Position: refs/heads/main@{#1540297}

merge_pr_55861

Reland "Refactor menu event handling code [2/4]"

This is a reland of commit 95f912cedcfed49b1360d308897f441882b57f13

Original change's description:

Refactor menu event handling code [2/4]

Another refactoring CL for menu element code. Previously, the command
invoker logic was split (and roughly duplicated) between
HTMLButtonElement and HTMLMenuItemElement. Since both can be command
invokers, it makes sense to move the command management stuff to the
common ancestor class, HTMLElement. This CL does that, and de-dupes
the code out of both subclasses. There is now just one base class
method that is overridden in sub-classes that can be command
invokers: CanBeCommandInvoker().

There was also a weird combination of popover activation command
code, which belongs on HTMLElement because any element can be a
popover, and menu activation command code, which belongs on the
HTMLMenuItemElement class, because only that class handles menu
commands.

This touches a few outside classes like ax_object, which get simpler
now that they don't have to deal with each case of a command invoker.

There is one behavior change here: now, if a menuitem is both
checkable and a sub-menu invoker, the sub-menu wins, and the
item does not do checked/unchecked behaviors.

I also added a test that seemed missing: make sure only buttons and
menuitems can be command invokers. Even though isn't
shipped, this test should pass because it explicitly skips ,
and anyway menuitem isn't included in the HTML5_ELEMENTS list.

Bug: 406566432
Change-Id: I281eda0eda4b0dced3ad38bd09a488f2c6b1ddf4
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/7078858
Reviewed-by: Dominic Farolino dom@chromium.org
Commit-Queue: Mason Freed masonf@chromium.org
Cr-Commit-Position: refs/heads/main@{#1538871}

Bug: 406566432
Change-Id: Iec91410726ae22a3673bdf32eb64aad79f672d6a
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/7113161
Commit-Queue: Dominic Farolino dom@chromium.org
Auto-Submit: Mason Freed masonf@chromium.org
Reviewed-by: Dominic Farolino dom@chromium.org
Cr-Commit-Position: refs/heads/main@{#1540144}

merge_pr_55860

[wpt] Remove the align-dialog-* tests.

It was resolved to remove the {align,justify}-self:dialog values in:
w3c/csswg-drafts#10258 (comment)

Tests for auto-margin ignoring were added in:
fantasai@e283b4a

Change-Id: I33dd95cb45e05ca7a2cf485a846cbfef22986429
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/7119879
Commit-Queue: Ian Kilpatrick ikilpatrick@chromium.org
Reviewed-by: Ian Kilpatrick ikilpatrick@chromium.org
Auto-Submit: Ian Kilpatrick ikilpatrick@chromium.org
Reviewed-by: David Grogan dgrogan@chromium.org
Cr-Commit-Position: refs/heads/main@{#1540163}

merge_pr_55859

[gap-decorations] Make column-rule-outset into shorthand using new props

This CL makes column-rule-outset into a shorthand using the new
longhands w3c/csswg-drafts#12603.

We also implement the paint behavior so that these work for column,
moreover, we add new tests for this property, since the existing ones
are using the column-rule-outset shorthand.

We also update some existing parsing tests, and comment out some parts
of tests that are not applicable yet since row-rule-outset and its
longhands will be implemented in following CLs.

We also temporarily remove the rule-outset shorthand, but will be
reintroduced once the above is done.

Recently it was resolved that the outset properties will be renamed to
inset w3c/csswg-drafts#12848, so they will
need to be renamed in follow up CLs, but since the asymmetric behavior
resolved first, we are implementing that first. It will also allow us to
move to the new inset behavior incrementally.

Bug: 357648037, 440054556
Change-Id: I37cdbe50cfabc371344d4bf2aadccd02f1b99c85
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/7107919
Commit-Queue: Javier Contreras javiercon@microsoft.com
Reviewed-by: Alison Maher almaher@microsoft.com
Cr-Commit-Position: refs/heads/main@{#1540149}