This CL improves the testing of template cloning with Parts, testing
these four cases:

  1. Main document parsing
  2. Template (content fragment) parsing
  3. Template/fragment cloning
  4. Declarative Shadow DOM parsing and cloning

This CL fixes the behavior for #3 above, but leaves #4 broken. The
following changes in behavior are made:

1. Part::MoveToRoot() can be used to change the root(), including
   to set it to nullptr. This happens when a Node tree is removed
   from the DOM, and it contains Parts that refer to the old root.
2. IsDocumentPartRoot() is now virtual, because during a tree move,
   the root() for a Part can be made nullptr even when it's a
   ChildNodePart.
3. Part::disconnected_ is added to keep track of whether the
   Part has been disconnected, since root() can now be nullptr.
4. (This is a bug fix) When using ChildNodePart::setNextSibling(),
   the new sibling node wasn't having its Part registered with
   NodeRareData, which caused a CHECK failure when trying to
   subsequently clone that Part. This is caught in the new test
   which clones declaratively-built templates containing Parts.

Bug: 1453291
Change-Id: Ic1c1475431cf6bd658f191db78003204412ef78f
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/4713668
Reviewed-by: David Baron <[email protected]>
Auto-Submit: Mason Freed <[email protected]>
Commit-Queue: Mason Freed <[email protected]>
Cr-Commit-Position: refs/heads/main@{#1175782}

Following the discussion on issue #3 [1], this CL adds support to soft
navigations triggered by keyboard shortcuts, by adding unfocused keydown
events to the events that can trigger the soft navigation heuristic.

[1] WICG/soft-navigations#3

Bug: 1478772
Change-Id: Ib423a3cfc09eaf4dd9a2221b3494ab1016fa8668
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/4839506
Commit-Queue: Yoav Weiss <[email protected]>
Reviewed-by: Ian Clelland <[email protected]>
Cr-Commit-Position: refs/heads/main@{#1193004}

…on triggers"

This reverts commit 6efe71286a014d3d3872bc990e3ea2d08dd46dba.

Reason for revert: One check added in this CL causes crash. see
crbug.com/1480047

Original change's description:
> [soft navigations] Enable keyboard shortcuts as soft navigation triggers
>
> Following the discussion on issue #3 [1], this CL adds support to soft
> navigations triggered by keyboard shortcuts, by adding unfocused keydown
> events to the events that can trigger the soft navigation heuristic.
>
> [1] WICG/soft-navigations#3
>
> Bug: 1478772
> Change-Id: Ib423a3cfc09eaf4dd9a2221b3494ab1016fa8668
> Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/4839506
> Commit-Queue: Yoav Weiss <[email protected]>
> Reviewed-by: Ian Clelland <[email protected]>
> Cr-Commit-Position: refs/heads/main@{#1193004}

Bug: 1478772
Change-Id: I3a518c165e6b19239a6bf7900e94c1ef9c3e5a5a
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/4859802
Reviewed-by: Ian Clelland <[email protected]>
Commit-Queue: Hao Liu <[email protected]>
Owners-Override: Daniel Cheng <[email protected]>
Bot-Commit: Rubber Stamper <[email protected]>
Cr-Commit-Position: refs/heads/main@{#1196100}

…attempt #3

This converts IDL-exposed promises in ReadableStream,
ReadableStreamBYOBReader, ReadableStreamDefaultReader, and
ReadableStreamGenericReader to use typed ScriptPromiseResolver
instead of StreamPromiseResolver and to return typed
ScriptPromises.

Bug: 329702363
Change-Id: I8ad1af1a7c9c909d711881ce7621c6c9fac58931
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/5429731
Reviewed-by: Adam Rice <[email protected]>
Reviewed-by: Nidhi Jaju <[email protected]>
Commit-Queue: Nate Chapin <[email protected]>
Cr-Commit-Position: refs/heads/main@{#1289397}

Fixing these tests involved the following updates:
1. Update all grid id refs to masonry.

2. In the row test, update the item.style.expectedRows to
item.style.gridRow. I suspect this was a bad AI suggestion I didn't
catch originally.

3. The min-content/max-content contribution in the block direction
(i.e. row direction) is always the same (which is the block content
contribution). This means that the row test expectations needed to
be updated for all min-content entries to match that of max-content.

The reason the grid test for #3 is different is that test sets both
column and row constraints, which impacts what the rows are sized to.
In this case, we are only constraining the row sizes, so we can end
up with different expectations as a result.

Bug: 343257585
Change-Id: I81f19a30d332eab5e18b7638d98467fbb81cde02
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/7038202
Commit-Queue: Alison Maher <[email protected]>
Reviewed-by: Kurt Catti-Schmidt <[email protected]>
Cr-Commit-Position: refs/heads/main@{#1532680}

To "prepare the script element" [1] exits early when the element is disconnected.
In this test, the script element represented by frag_script_2 which logs "inline script #3"
is disconnected in the preceeding script (frag_script_1 / "inline script #2").
Since post-connection steps [2] are executed in the tree order, by the time we are to
"prepare the script element" on frag_script_2 / "inline script #3", the script element is
already disconnected and the script will not be executed.

This test failing in Chrome and Safari but passing in Firefox. After the change, the test
passes in Chrome and Safari and fails in Firefox.

[1] https://html.spec.whatwg.org/multipage/scripting.html#prepare-the-script-element
[2] https://dom.spec.whatwg.org/#concept-node-post-connection-ext

…s#58514)

To "prepare the script element" [1] exits early when the element is disconnected.
In this test, the script element represented by frag_script_2 which logs "inline script #3"
is disconnected in the preceeding script (frag_script_1 / "inline script #2").
Since post-connection steps [2] are executed in the tree order, by the time we are to
"prepare the script element" on frag_script_2 / "inline script #3", the script element is
already disconnected and the script will not be executed.

This test failing in Chrome and Safari but passing in Firefox. After the change, the test
passes in Chrome and Safari and fails in Firefox.

[1] https://html.spec.whatwg.org/multipage/scripting.html#prepare-the-script-element
[2] https://dom.spec.whatwg.org/#concept-node-post-connection-ext