merge_pr_50159

Get started cleaning up interesttarget implementation [1/N]

This is the first in a series of patches to clean up and update the
interesttarget implementation to match recent OpenUI discussions.

This CL:

• Gets rid of interestaction. See the large comment posted here:
  openui/open-ui#1064 (comment)
• Adds a connection to InterestLost when elements are de-focused.
• Adds support (tentatively) for dialogs being shown modally.
• Remove keyboard/focus support for interesttarget. This will
  get re-added later in its new form, via a hotkey rather than
  focus.

Bug: 326681249
Change-Id: I26f07a00c4fb1d2b1da92b64d91f330c02a11468
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/6166890
Auto-Submit: Mason Freed [email protected]
Reviewed-by: David Baron [email protected]
Commit-Queue: Mason Freed [email protected]
Cr-Commit-Position: refs/heads/main@{#1408202}

merge_pr_50127

Don't render optgroup label if there is a legend

Fixed: 378601807
Change-Id: I1e7c9f90d862384691d16a323013e7e479feb843
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/6176323
Reviewed-by: Traian Captan [email protected]
Commit-Queue: Joey Arhar [email protected]
Cr-Commit-Position: refs/heads/main@{#1408220}

merge_pr_50073

WebKit export: Implement CSS Values 5 random() (#50073)

https://bugs.webkit.org/show_bug.cgi?id=284344

merge_pr_50157

Migrate 2d.canvas.host.size.attributes.* to the new canvas WPT generator

This test uses different types of newline characters, like \n, \r and
\f. The indent Jinja filter used by the test template 1 was dropping
all of these, replacing them with normal newlines. This looks like a
bug in Jinja: the indent filter could easily be made to preserve
newline types. This CL overrides the default indent filter with a
version doing exactly that.

Bug: 40207206
Change-Id: I5ece8e0ce4199bb0c915e59131c26eed4ce58693
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/6172748
Commit-Queue: Jean-Philippe Gravel [email protected]
Reviewed-by: Yi Xu [email protected]
Cr-Commit-Position: refs/heads/main@{#1408116}

merge_pr_50154

Use shared helper function in B&A WPT.

Use shared BA.testWithMutatedServerResponse() in
server-response.https.window.js instead of a local one. The shared one
was modified from the local one.

Change-Id: I982e0498f76a8c6eb5dec168a3892847f26bb8a5
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/6182327
Reviewed-by: Maks Orlovich [email protected]
Commit-Queue: Qingxin Wu [email protected]
Cr-Commit-Position: refs/heads/main@{#1408091}

merge_pr_50149

Reland: Inherit service worker controller for srcdoc iframe

This is a reland of http://crrev.com/c/6054691 with changes from
http://crrev.com/c/6081248 for fixing the DCHECK/DwC from the original
change.

Same origin srcdoc iframes should be controlled by the same service
worker that controls the parent frame.

To achieve this, expanded ServiceWorkerClient::InheritControllerFrom to
also support srcdoc iframes and used it to setup service worker client
for srcdoc frames when NavigationRequests for them start.

To allow these clients to show up in clients.matchAll, we update same
origin check in GetWindowClients to use GetLastCommittedOrigin instead
of GetLastCommittedURL from the frames.

As the url for srcdoc iframe is about:srcdoc, when we use origin of that
url for access check, it will fail and hit DCHECK/DwC. So, we switch to
use its creator url, which is captured as url for scope match, when
performing access checks.

As the srcdoc frame test in about-blank-replacement.https.html now
passes, updated expectation to reflect that.

Also added wpt tests to ensure that service worker features work as
expected in srcdoc iframes.

The new behavior is behind the ServiceWorkerSrcdocSupport feature which
is enabled by default.

Bug: 41411856,382895651,382895896
Change-Id: I5d7e96b4838c666842315fce6795a9221e19f61b
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/6085871
Reviewed-by: Rakina Zata Amni [email protected]
Commit-Queue: Liang Zhao [email protected]
Reviewed-by: Yoshisato Yanagisawa [email protected]
Cr-Commit-Position: refs/heads/main@{#1408026}

merge_pr_50148

DOM: Fix moveBefore() disconnected iframe CHECK crash

This CL updates the iframe insertion/removal bookkeeping specific to
moveBefore() that was written with the assumption that all moved nodes
are always connected to the DOM. This code was not updated since
changes to moveBefore() preconditions were made. This CL fixes that,
and adds a test.

R=masonf, nrosenthal

Bug: 40150299, 389339248
Change-Id: I31f23d13f78af3eaa87d03ee7d5703ea97f363c5
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/6182440
Reviewed-by: Noam Rosenthal [email protected]
Commit-Queue: Dominic Farolino [email protected]
Cr-Commit-Position: refs/heads/main@{#1408015}

merge_pr_50147

DOM: Fix moveBefore null parent crash

Before this CL, we didn't early-return from Node::moveBefore() when
insertBefore() threw an error due to an invalid node hierarchy. This
wasn't necessary until we introduced the MovedFrom() hook, since
there was no significant code after the insertBefore() call. But once
we introduced that hook, we didn't add the early-return before calling
the hook, which caused this crash.

R=masonf, nrosenthal

Bug: 40150299, 388934346
Change-Id: Ie5f7401ed5097b7f64097daa557f315a9889664f
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/6179644
Commit-Queue: Dominic Farolino [email protected]
Reviewed-by: Noam Rosenthal [email protected]
Cr-Commit-Position: refs/heads/main@{#1407973}

merge_pr_50137

[B&A] Add WPT for B&A forDebuggingOnly.

Change-Id: I74be87eeae3301292a907a03623ff58ca3c082ef
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/6179638
Reviewed-by: Robert Flack [email protected]
Reviewed-by: Maks Orlovich [email protected]
Commit-Queue: Qingxin Wu [email protected]
Cr-Commit-Position: refs/heads/main@{#1407982}

merge_pr_50136

Add test for resolving rlh on the :root element

There was only coverage for testing rlh on the :root element for the
font and line-height properties, not other lengths.

Change-Id: Ib67faf7f78d657c7bdf917a32eb7f5786916db98
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/6169143
Commit-Queue: Rune Lillesveen [email protected]
Reviewed-by: Anders Hartvoll Ruud [email protected]
Cr-Commit-Position: refs/heads/main@{#1407838}