dpwas: Add display_override tentative tests

Adding manual WPT tests for new display-override field.
PR to add new field to manifest spec is here: w3c/manifest#932

Bug: 1149993
Change-Id: I670c69737e337a4c13e35672daa93e703f97b84e
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/3166014
Commit-Queue: Mike Jackson <[email protected]>
Reviewed-by: Daniel Murphy <[email protected]>
Cr-Commit-Position: refs/heads/main@{#922341}

Add additional structured clone tests

The following behaviors mandated by the spec weren't tested:

- Serializing a non-serializable platform object, or transferring a
  non-transferable platform object, should fail.
- Platform objects must deserialize even if their interface has been
  deleted from the realm's global.
- Objects which have serializable or transferable interfaces in their
  inheritance hierarchy must deserialize as instances of the closest
  such interface.
- Transferring again a detached object will fail.

This change also passes the test object as an additional parameter to
the `f` function of structured clone tests, to make it possible to test
that a promise rejects.

Test that float values are correctly produced while interpolating box…

…-shadow and text-shadow values, added for WebKit bug https://bugs.webkit.org/show_bug.cgi?id=230347. (web-platform-tests#30823)

Add missing import.

Sync interfaces/ with @webref/idl 2.4.1 (web-platform-tests#30818)

[Re-Factor Client Hints] (5) Remove (unlaunched) `lang` client hint {…

…FOLLOWUP}

https://chromium-review.googlesource.com/c/chromium/src/+/3160286
introduced upstream flakiness because several uses of
lang were not removed. This fixes that issue.

Change-Id: Ie9fd542667f1636b51ebb5078ff651f2b63a1ee2
bug: 1250096
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/3163688
Reviewed-by: Yoav Weiss <[email protected]>
Commit-Queue: Yoav Weiss <[email protected]>
Auto-Submit: Ari Chivukula <[email protected]>
Cr-Commit-Position: refs/heads/main@{#922014}

Fix incorrect tree construction by the AAA

Prior to this CL, the following code:
  <code some-attribute>
    <div>
      <code>
        <code>
          <code>
            <code>
            </code>
          </code>
        </code>
      </code>
parsed to this:
  <code some-attribute></code>
  <div>
    <code some-attribute>
      <code>
        <code>
          <code>
            <code>
            </code>
          </code>
        </code>
      </code>
    </code>
  </div>

The adoption agency algorithm reparented the div tag though code tags
were properly nested. A step was added to the spec[1] in order to fix
this, but not implemented in Chromium. This CL implements it.

[1] https://www.w3.org/Bugs/Public/show_bug.cgi?id=24833

Fixed: 1217523
Change-Id: I2ce1ada69d37305041468b9c10f59e6238e9a209
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/3162497
Commit-Queue: Mason Freed <[email protected]>
Reviewed-by: Mason Freed <[email protected]>
Cr-Commit-Position: refs/heads/main@{#922189}

Add BigInt objects to structured cloning test

BigInts were missing on the structured clone test although it was defined in the HTML specification.

Use WebTransportError for errors during session establishment

This is for w3c/webtransport#344.
Test coverage is not great, but we are adding more web platform tests.

Bug: 1011392
Change-Id: Ie84fce58936931bf14ba3e8c9d096cdf73e12d3b
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/3162458
Commit-Queue: Yutaka Hirano <[email protected]>
Reviewed-by: Yoichi Osato <[email protected]>
Reviewed-by: Adam Rice <[email protected]>
Reviewed-by: Kenichi Ishibashi <[email protected]>
Reviewed-by: Nidhi Jaju <[email protected]>
Cr-Commit-Position: refs/heads/main@{#922011}

Pre-paint: Avoid inlines that aren't present in a fragmentainer.

If the parent block fragment isn't an inline formatting context, we
generally don't need to enter inlines, except that we need to clear the
paint flags. Therefore, only enter inlines if the parent is at the last
fragment. Entering at any other fragment may cause us to update the
paint offset, which might set paint invalidation flags for no good
reason.

In the test included, there's a tall float in an inline formatting
context, taking up three columns. There's a line in the first column
with just a BR element. There's also a wide inline-block, but it doesn't
fit beside the float, so it gets pushed to the third column, where we
create a line for it below the float. When a float in an inline
formatting context breaks, it's resumed as a regular block. An inline
formatting context will only be created if there's actual inline content
there as well. So, the second column doesn't get any inline content at
all (i.e. no fragment items).

Therefore we'd visit the BR element in the first column (correct - this
is where the element is). We'd then re-enter it in the second column
(because we had no inline formatting context to search, and just enter
everything - this is bad), and update the paint offset (and possibly
other things), and set invalidation flags. In the third column we'll not
enter it, because there we have an inline formatting context to search
(and the BR will not be found). So we'd end up with paint invalidation
flags not cleared.

Bug: 1249769
Change-Id: Icaa583c0bb1a034a7a2690ab6bc27cbcb7010f75
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/3162102
Reviewed-by: Koji Ishii <[email protected]>
Commit-Queue: Morten Stenshorne <[email protected]>
Cr-Commit-Position: refs/heads/main@{#922120}