Introduce Node::containingTreeScope(), which asserts that the node's root is a tree scope

Introduce Node::containingTreeScope(), which asserts that the node's root is a tree scope

The difference between Node::containingTreeScope() and Node::treeScope() is that
Node::containingTreeScope() asserts that the node is in a treeScope.
Node::treeScope() does not.

Replace some of the current usage of Node::treeScope() with
Node::containingTreeScope() if containingTreeScope() is better to express their
intention.

This CL also introduces Element::adjustedFocusedElementInTreeScope() to optimize
some code path where we can early exit if the element is not in a tree scope.

BUG=602556
Committed: https://crrev.com/97a40b7bad083e7f9382a92e00adf1bd650db4a5
Cr-Commit-Position: refs/heads/master@{#397065}

[hayato, 2016-04-14 06:51:51]

The CQ bit was checked by hayato@chromium.org to run a CQ dry run

[commit-bot: I haz the power, 2016-04-14 06:51:59]

Dry run: CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1883083002/1
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1883083002/1

[hayato, 2016-04-14 07:15:42]

Description was changed from

==========
Introduce Node::rootTreeScope(), which checks the node's root is a tree scope

BUG=
==========

to

==========
Introduce Node::rootTreeScope(), which asserts that the node's root is a tree
scope

The difference between Node::rootTreeScope() and Node::treeScope() is that
Node::rootTreeScope() asserts that the node is in a treeScope.

Replace some of the current usage of Node::treeScope() with
Node::rootTreeScope()
if rootTreeScope() is better to express the intention.

This CL also optimize some code pass by explicitly checking
Node::isInTreeScope() before calling Node::rootTreeScope(), where we can *early
exit*.

BUG=602556
==========

[hayato, 2016-04-14 07:20:03]

Description was changed from

==========
Introduce Node::rootTreeScope(), which asserts that the node's root is a tree
scope

The difference between Node::rootTreeScope() and Node::treeScope() is that
Node::rootTreeScope() asserts that the node is in a treeScope.

Replace some of the current usage of Node::treeScope() with
Node::rootTreeScope()
if rootTreeScope() is better to express the intention.

This CL also optimize some code pass by explicitly checking
Node::isInTreeScope() before calling Node::rootTreeScope(), where we can *early
exit*.

BUG=602556
==========

to

==========
Introduce Node::rootTreeScope(), which asserts that the node's root is a tree
scope

The difference between Node::rootTreeScope() and Node::treeScope() is that
Node::rootTreeScope() asserts that the node is in a treeScope.

Replace some of the current usage of Node::treeScope() with
Node::rootTreeScope()
if rootTreeScope() is better to express the intention.

This CL also optimize some code path by explicitly checking
Node::isInTreeScope() before calling Node::rootTreeScope(), where we can *early
exit*.

BUG=602556
==========

[hayato, 2016-04-14 07:21:34]

Description was changed from

==========
Introduce Node::rootTreeScope(), which asserts that the node's root is a tree
scope

The difference between Node::rootTreeScope() and Node::treeScope() is that
Node::rootTreeScope() asserts that the node is in a treeScope.

Replace some of the current usage of Node::treeScope() with
Node::rootTreeScope()
if rootTreeScope() is better to express the intention.

This CL also optimize some code path by explicitly checking
Node::isInTreeScope() before calling Node::rootTreeScope(), where we can *early
exit*.

BUG=602556
==========

to

==========
Introduce Node::rootTreeScope(), which asserts that the node's root is a tree
scope

The difference between Node::rootTreeScope() and Node::treeScope() is that
Node::rootTreeScope() asserts that the node is in a treeScope.

Replace some of the current usage of Node::treeScope() with
Node::rootTreeScope()
if rootTreeScope() is better to express the intention.

This CL also optimize some code path by explicitly checking
Node::isInTreeScope() before calling Node::rootTreeScope(),
where we can *early exit* if the node is not in a tree scope.

BUG=602556
==========

[hayato, 2016-04-14 07:21:55]

Description was changed from

==========
Introduce Node::rootTreeScope(), which asserts that the node's root is a tree
scope

The difference between Node::rootTreeScope() and Node::treeScope() is that
Node::rootTreeScope() asserts that the node is in a treeScope.

Replace some of the current usage of Node::treeScope() with
Node::rootTreeScope()
if rootTreeScope() is better to express the intention.

This CL also optimize some code path by explicitly checking
Node::isInTreeScope() before calling Node::rootTreeScope(),
where we can *early exit* if the node is not in a tree scope.

BUG=602556
==========

to

==========
Introduce Node::rootTreeScope(), which asserts that the node's root is a tree
scope

The difference between Node::rootTreeScope() and Node::treeScope() is
that Node::rootTreeScope() asserts that the node is in a treeScope.

Replace some of the current usage of Node::treeScope() with
Node::rootTreeScope()
if rootTreeScope() is better to express the intention.

This CL also optimize some code path by explicitly checking
Node::isInTreeScope() before calling Node::rootTreeScope(),
where we can *early exit* if the node is not in a tree scope.

BUG=602556
==========

[hayato, 2016-04-14 07:22:13]

Description was changed from

==========
Introduce Node::rootTreeScope(), which asserts that the node's root is a tree
scope

The difference between Node::rootTreeScope() and Node::treeScope() is
that Node::rootTreeScope() asserts that the node is in a treeScope.

Replace some of the current usage of Node::treeScope() with
Node::rootTreeScope()
if rootTreeScope() is better to express the intention.

This CL also optimize some code path by explicitly checking
Node::isInTreeScope() before calling Node::rootTreeScope(),
where we can *early exit* if the node is not in a tree scope.

BUG=602556
==========

to

==========
Introduce Node::rootTreeScope(), which asserts that the node's root is a tree
scope

The difference between Node::rootTreeScope() and Node::treeScope() is
that Node::rootTreeScope() asserts that the node is in a treeScope.

Replace some of the current usage of Node::treeScope() with
Node::rootTreeScope()
if rootTreeScope() is better to express the intention.

This CL also optimize some code path by explicitly checking
Node::isInTreeScope() 
before calling Node::rootTreeScope(),
where we can *early exit* if the node is not in a tree scope.

BUG=602556
==========

[hayato, 2016-04-14 07:24:07]

The CQ bit was checked by hayato@chromium.org to run a CQ dry run

[hayato, 2016-04-14 07:24:11]

wip

[commit-bot: I haz the power, 2016-04-14 07:24:19]

Dry run: CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1883083002/20001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1883083002/20001

[hayato, 2016-04-14 07:29:57]

hayato@chromium.org changed reviewers:
+ esprehn@chromium.org, rune@opera.com

[hayato, 2016-04-14 07:29:58]

Elliot, Rune, could you give me early feedback about this CL?

It looks there are other places where we can use Node::rootTreeScope(), instead
of Node::treeScope(), but I would like to know whether this CL is a right
direction or not.

https://codereview.chromium.org/1883083002/diff/20001/third_party/WebKit/Sour...
File third_party/WebKit/Source/core/dom/SelectorQuery.cpp (left):

https://codereview.chromium.org/1883083002/diff/20001/third_party/WebKit/Sour...
third_party/WebKit/Source/core/dom/SelectorQuery.cpp:264: if (selector->match()
== CSSSelector::Id &&
!rootNode.document().containsMultipleElementsWithId(selector->value())) {
Using document() here looks a bug to me.

[hayato, 2016-04-14 07:37:30]

Description was changed from

==========
Introduce Node::rootTreeScope(), which asserts that the node's root is a tree
scope

The difference between Node::rootTreeScope() and Node::treeScope() is
that Node::rootTreeScope() asserts that the node is in a treeScope.

Replace some of the current usage of Node::treeScope() with
Node::rootTreeScope()
if rootTreeScope() is better to express the intention.

This CL also optimize some code path by explicitly checking
Node::isInTreeScope() 
before calling Node::rootTreeScope(),
where we can *early exit* if the node is not in a tree scope.

BUG=602556
==========

to

==========
Introduce Node::rootTreeScope(), which asserts that the node's root is a tree
scope

The difference between Node::rootTreeScope() and Node::treeScope() is that
Node::rootTreeScope() asserts that the node is in a treeScope. Node::treeScope()
does not.

Replace some of the current usage of Node::treeScope() with
Node::rootTreeScope()
if rootTreeScope() is better to express the intention.

This CL also optimize some code pass by explicitly checking
Node::isInTreeScope() before calling Node::rootTreeScope(), where we can *early
exit*.

BUG=602556
==========

[hayato, 2016-04-14 07:37:30]

hayato@chromium.org changed reviewers:
- esprehn@chromium.org, rune@opera.com

[hayato, 2016-04-14 07:38:45]

Description was changed from

==========
Introduce Node::rootTreeScope(), which asserts that the node's root is a tree
scope

The difference between Node::rootTreeScope() and Node::treeScope() is that
Node::rootTreeScope() asserts that the node is in a treeScope. Node::treeScope()
does not.

Replace some of the current usage of Node::treeScope() with
Node::rootTreeScope()
if rootTreeScope() is better to express the intention.

This CL also optimize some code pass by explicitly checking
Node::isInTreeScope() before calling Node::rootTreeScope(), where we can *early
exit*.

BUG=602556
==========

to

==========
Introduce Node::rootTreeScope(), which asserts that the node's root is a tree
scope

The difference between Node::rootTreeScope() and Node::treeScope() is that 
Node::rootTreeScope() asserts that the node is in a treeScope.
Node::treeScope() does not.

Replace some of the current usage of Node::treeScope() with
Node::rootTreeScope()
if rootTreeScope() is better to express the intention.

This CL also optimize some code path by explicitly checking
Node::isInTreeScope() before calling Node::rootTreeScope(), where we can *early
exit* if the node is not in a tree scope.

BUG=602556
==========

[hayato, 2016-04-14 07:39:00]

Description was changed from

==========
Introduce Node::rootTreeScope(), which asserts that the node's root is a tree
scope

The difference between Node::rootTreeScope() and Node::treeScope() is that 
Node::rootTreeScope() asserts that the node is in a treeScope.
Node::treeScope() does not.

Replace some of the current usage of Node::treeScope() with
Node::rootTreeScope()
if rootTreeScope() is better to express the intention.

This CL also optimize some code path by explicitly checking
Node::isInTreeScope() before calling Node::rootTreeScope(), where we can *early
exit* if the node is not in a tree scope.

BUG=602556
==========

to

==========
Introduce Node::rootTreeScope(), which asserts that the node's root is a tree
scope

The difference between Node::rootTreeScope() and Node::treeScope() is that 
Node::rootTreeScope() asserts that the node is in a treeScope.
Node::treeScope() does not.

Replace some of the current usage of Node::treeScope() with
Node::rootTreeScope()
if rootTreeScope() is better to express the intention.

This CL also optimize some code path by explicitly checking
Node::isInTreeScope()
before calling Node::rootTreeScope(), where we can *early exit*
if the node is not in a tree scope.

BUG=602556
==========

[rune, 2016-04-14 08:48:00]

rune@opera.com changed reviewers:
+ rune@opera.com

[rune, 2016-04-14 08:48:01]

I think I would prefer to have the DCHECK in treeScope() instead and add a
treeScopeOrDocument() in the cases where that's what we expect.

treeScopeOrDocument() could do:

DCHECK(isInTreeScope() || treeScope() == document())

https://codereview.chromium.org/1883083002/diff/20001/third_party/WebKit/Sour...
File third_party/WebKit/Source/core/dom/SelectorQuery.cpp (left):

https://codereview.chromium.org/1883083002/diff/20001/third_party/WebKit/Sour...
third_party/WebKit/Source/core/dom/SelectorQuery.cpp:264: if (selector->match()
== CSSSelector::Id &&
!rootNode.document().containsMultipleElementsWithId(selector->value())) {
On 2016/04/14 07:29:58, hayato wrote:
> Using document() here looks a bug to me.

Yes. Perhaps fix that in a separate CL.

Reported https://crbug.com/603431

[esprehn, 2016-04-14 10:19:12]

On 2016/04/14 at 08:48:01, rune wrote:
> I think I would prefer to have the DCHECK in treeScope() instead and add a
treeScopeOrDocument() in the cases where that's what we expect.
> 
> treeScopeOrDocument() could do:
> 
> DCHECK(isInTreeScope() || treeScope() == document())

I don't think any method that has a name like that makes sense, all tree scopes
are documents. I'd rather we didn't mess with the existing treeScope() method at
all. This patch seems fine and the naming of rootTreeScope seems okay too, it
matches rootNode which was added recently.

> 
>
https://codereview.chromium.org/1883083002/diff/20001/third_party/WebKit/Sour...
> File third_party/WebKit/Source/core/dom/SelectorQuery.cpp (left):
> 
>
https://codereview.chromium.org/1883083002/diff/20001/third_party/WebKit/Sour...
> third_party/WebKit/Source/core/dom/SelectorQuery.cpp:264: if
(selector->match() == CSSSelector::Id &&
!rootNode.document().containsMultipleElementsWithId(selector->value())) {
> On 2016/04/14 07:29:58, hayato wrote:
> > Using document() here looks a bug to me.
> 
> Yes. Perhaps fix that in a separate CL.
> 
> Reported https://crbug.com/603431

Yeah let's fix the bug separately.

[hayato, 2016-04-14 10:25:33]

Thanks. My current plan is:

- Introduce Node::rootTreeScope().
- Do not touch Node::treeScope(). Let it be as is.

That would not upset blink developers, I hope.

Let me refine this CL.

[commit-bot: I haz the power, 2016-04-14 10:26:49]

The CQ bit was unchecked by commit-bot@chromium.org

[commit-bot: I haz the power, 2016-04-14 10:26:50]

Dry run: This issue passed the CQ dry run.

[esprehn, 2016-04-14 10:43:12]

I've been thinking about this from the bug and the reason we have document
and treeScope is to model the global in JS. For example in the blink C++
you write document().createElement(...) or whatever it might be. That's
pretty but confusing since document is also a method and you can write
node.document().foo(...), and we have this inDocument() bit that doesn't
relate.

One path could be to rename the existing method to globalDocument(). I
suspect we don't need globalTreeScope at all, since anyone using that can
just use globalDocument.

Then we rename treeScope() and document() to something like rootTreeScope()
with the assert like you have here and connectedDocument() that similarly
has the inDocument() assert, I think the spec is using the language
connected now too?

That's more verbose but makes each call site clear about whether it means
the global document or the connected one.

-- 
You received this message because you are subscribed to the Google Groups "Blink
Reviews" group.
To unsubscribe from this group and stop receiving emails from it, send an email
to blink-reviews+unsubscribe@chromium.org.

[esprehn, 2016-04-14 10:43:13]

I've been thinking about this from the bug and the reason we have document
and treeScope is to model the global in JS. For example in the blink C++
you write document().createElement(...) or whatever it might be. That's
pretty but confusing since document is also a method and you can write
node.document().foo(...), and we have this inDocument() bit that doesn't
relate.

One path could be to rename the existing method to globalDocument(). I
suspect we don't need globalTreeScope at all, since anyone using that can
just use globalDocument.

Then we rename treeScope() and document() to something like rootTreeScope()
with the assert like you have here and connectedDocument() that similarly
has the inDocument() assert, I think the spec is using the language
connected now too?

That's more verbose but makes each call site clear about whether it means
the global document or the connected one.

-- 
You received this message because you are subscribed to the Google Groups
"Chromium-reviews" group.
To unsubscribe from this group and stop receiving emails from it, send an email
to chromium-reviews+unsubscribe@chromium.org.

[hayato, 2016-04-15 10:17:59]

The CQ bit was checked by hayato@chromium.org to run a CQ dry run

[hayato, 2016-04-15 10:18:03]

rebased

[commit-bot: I haz the power, 2016-04-15 10:18:16]

Dry run: CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1883083002/40001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1883083002/40001

[hayato, 2016-04-15 10:29:00]

I see. Thanks. Thus, the long time goal would be:

1).
Document& Node::globalDocument() {
  return ... // The behavior is same to the current Node::document().
}

2).
Document& Node::connectedDocument() {
  DCHECK(inShadowIncludingDocument());
  return globalDocument();
}

3).
Document& Node::treeScope() {
  DCHECK(inTreeScope());
  return m_treeScope;
}

4).
TreeScope& Node::treeScopeOrGlobalDocument() {   // I am not sure whether we
still need this
  return m_treeScope;
}

That would makes sense.


A bug of querySelectorAll was fixed in another CL:
https://codereview.chromium.org/1892743002

I rebased this CL on top on it. In any case, this CL might be okay because if we
introduce Node::rootTreeScope() here, we can rename it to 3) mechanically later.

[commit-bot: I haz the power, 2016-04-15 11:53:23]

The CQ bit was unchecked by commit-bot@chromium.org

[commit-bot: I haz the power, 2016-04-15 11:53:24]

Dry run: This issue passed the CQ dry run.

[hayato, 2016-04-19 06:20:09]

PTAL

[rune, 2016-04-19 07:36:29]

lgtm, but please wait for esprehn.

[hayato, 2016-04-22 07:48:46]

Elliott, could you take a look?

[esprehn, 2016-04-23 02:11:01]

On 2016/04/22 at 07:48:46, hayato wrote:
> Elliott, could you take a look?

Yeah I'll look soon, sorry for the delay.

[esprehn, 2016-04-27 07:52:23]

https://codereview.chromium.org/1883083002/diff/40001/third_party/WebKit/Sour...
File third_party/WebKit/Source/core/dom/Element.cpp (right):

https://codereview.chromium.org/1883083002/diff/40001/third_party/WebKit/Sour...
third_party/WebKit/Source/core/dom/Element.cpp:2212: if (isInTreeScope() &&
rootTreeScope().adjustedFocusedElement() == this) {
Why do we need to add the check here? We do adjustedFocusedElement() == this in
many places, if that's too expensive then we should add a method to Element that
checks isInTreeScope for you.

[esprehn, 2016-05-18 00:36:35]

I think this got tackled some other way? Do we still want to do this? What's the
status? :)

[hayato, 2016-05-18 02:27:09]

On 2016/05/18 at 00:36:35, esprehn wrote:
> I think this got tackled some other way? Do we still want to do this? What's
the status? :)

No update. :)

I guess I have to rebase this CL. This CL is not a urgent task for Blink.
Let me remove you from reviewers, tentatively. If this CL becomes ready again, I
might send another review request.

[hayato, 2016-05-18 02:27:21]

hayato@chromium.org changed reviewers:
- esprehn@chromium.org, rune@opera.com

[hayato, 2016-05-23 05:21:20]

The CQ bit was checked by hayato@chromium.org to run a CQ dry run

[hayato, 2016-05-23 05:21:25]

Add Element::adjustedFocusElementinTreeScope

[commit-bot: I haz the power, 2016-05-23 05:21:33]

Dry run: CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1883083002/60001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1883083002/60001

[hayato, 2016-05-23 05:24:27]

Description was changed from

==========
Introduce Node::rootTreeScope(), which asserts that the node's root is a tree
scope

The difference between Node::rootTreeScope() and Node::treeScope() is that 
Node::rootTreeScope() asserts that the node is in a treeScope.
Node::treeScope() does not.

Replace some of the current usage of Node::treeScope() with
Node::rootTreeScope()
if rootTreeScope() is better to express the intention.

This CL also optimize some code path by explicitly checking
Node::isInTreeScope()
before calling Node::rootTreeScope(), where we can *early exit*
if the node is not in a tree scope.

BUG=602556
==========

to

==========
Introduce Node::rootTreeScope(), which asserts that the node's root is a tree
scope

The difference between Node::rootTreeScope() and Node::treeScope() is that 
Node::rootTreeScope() asserts that the node is in a treeScope.
Node::treeScope() does not.

Replace some of the current usage of Node::treeScope() with
Node::rootTreeScope()
if rootTreeScope() is better to express their intention.

This CL also introduces Element::adjustedFocusedElementInTreeScope() to optimize
some code path where we can early exit if the element is not in a tree scope.

BUG=602556
==========

[hayato, 2016-05-23 05:24:27]

hayato@chromium.org changed reviewers:
+ rune@opera.com

[hayato, 2016-05-23 05:24:47]

Description was changed from

==========
Introduce Node::rootTreeScope(), which asserts that the node's root is a tree
scope

The difference between Node::rootTreeScope() and Node::treeScope() is that 
Node::rootTreeScope() asserts that the node is in a treeScope.
Node::treeScope() does not.

Replace some of the current usage of Node::treeScope() with
Node::rootTreeScope()
if rootTreeScope() is better to express their intention.

This CL also introduces Element::adjustedFocusedElementInTreeScope() to optimize
some code path where we can early exit if the element is not in a tree scope.

BUG=602556
==========

to

==========
Introduce Node::rootTreeScope(), which asserts that the node's root is a tree
scope

The difference between Node::rootTreeScope() and Node::treeScope() is that 
Node::rootTreeScope() asserts that the node is in a treeScope.
Node::treeScope() does not.

Replace some of the current usage of Node::treeScope() with
Node::rootTreeScope()
if rootTreeScope() is better to express their intention.

This CL also introduces Element::adjustedFocusedElementInTreeScope() to optimize
some code path where we can early exit if the element is not in a tree scope.

BUG=602556
==========

[hayato, 2016-05-23 06:22:09]

hayato@chromium.org changed reviewers:
+ esprehn@chromium.org

[hayato, 2016-05-23 06:22:10]

PTAL again. I have rebased this.

https://codereview.chromium.org/1883083002/diff/40001/third_party/WebKit/Sour...
File third_party/WebKit/Source/core/dom/Element.cpp (right):

https://codereview.chromium.org/1883083002/diff/40001/third_party/WebKit/Sour...
third_party/WebKit/Source/core/dom/Element.cpp:2212: if (isInTreeScope() &&
rootTreeScope().adjustedFocusedElement() == this) {
Done.

Yeah, TreeScope::adjustedFocusedElement() could be expensive.
I introduced Element::adjustedFocusedElementInTreeScope(), and updated callers
of TreeScope::adjustedFocusedElement() to use that. That would be a small
performance gain.

[commit-bot: I haz the power, 2016-05-23 08:21:22]

The CQ bit was unchecked by commit-bot@chromium.org

[commit-bot: I haz the power, 2016-05-23 08:21:23]

Dry run: Try jobs failed on following builders:
  linux_chromium_rel_ng on tryserver.chromium.linux (JOB_FAILED,
http://build.chromium.org/p/tryserver.chromium.linux/builders/linux_chromium_...)

[hayato, 2016-05-26 04:54:03]

Elliott, could you take a look?

[esprehn, 2016-05-26 05:00:31]

Sure I'll look tomorrow, sorry for the delay.

-- 
You received this message because you are subscribed to the Google Groups
"Chromium-reviews" group.
To unsubscribe from this group and stop receiving emails from it, send an email
to chromium-reviews+unsubscribe@chromium.org.

[esprehn, 2016-05-26 05:00:32]

Sure I'll look tomorrow, sorry for the delay.

-- 
You received this message because you are subscribed to the Google Groups "Blink
Reviews" group.
To unsubscribe from this group and stop receiving emails from it, send an email
to blink-reviews+unsubscribe@chromium.org.

[esprehn, 2016-05-31 19:40:33]

Hmm, the name rootTreeScope is confusing since this isn't root of the tree of
scopes, maybe that's okay. I might have gone with containingTreeScope like
containingShadowRoot?

[esprehn, 2016-05-31 19:41:08]

lgtm

[hayato, 2016-06-01 00:36:21]

On 2016/05/31 at 19:40:33, esprehn wrote:
> Hmm, the name rootTreeScope is confusing since this isn't root of the tree of
scopes, maybe that's okay. I might have gone with containingTreeScope like
containingShadowRoot?

I do not have a strong opinion on the name. Thus, let's rename it to
containingTreeScope.
I'll land this patch after renaming.  Thanks!

[hayato, 2016-06-01 01:08:39]

Description was changed from

==========
Introduce Node::rootTreeScope(), which asserts that the node's root is a tree
scope

The difference between Node::rootTreeScope() and Node::treeScope() is that 
Node::rootTreeScope() asserts that the node is in a treeScope.
Node::treeScope() does not.

Replace some of the current usage of Node::treeScope() with
Node::rootTreeScope()
if rootTreeScope() is better to express their intention.

This CL also introduces Element::adjustedFocusedElementInTreeScope() to optimize
some code path where we can early exit if the element is not in a tree scope.

BUG=602556
==========

to

==========
Introduce Node::containingTreeScope(), which asserts that the node's root is a
tree scope

The difference between Node::containingTreeScope() and Node::treeScope() is that

Node::containingTreeScope() asserts that the node is in a treeScope.
Node::treeScope() does not.

Replace some of the current usage of Node::treeScope() with 
Node::containingTreeScope() if containingTreeScope() is better to express their
intention.

This CL also introduces Element::adjustedFocusedElementInTreeScope() to optimize
some code path where we can early exit if the element is not in a tree scope.

BUG=602556
==========

[hayato, 2016-06-01 01:09:57]

The CQ bit was checked by hayato@chromium.org to run a CQ dry run

[hayato, 2016-06-01 01:10:04]

renamed

[commit-bot: I haz the power, 2016-06-01 01:10:50]

Dry run: CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1883083002/80001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1883083002/80001

[hayato, 2016-06-01 01:15:40]

The CQ bit was unchecked by hayato@chromium.org

[hayato, 2016-06-01 01:15:47]

The CQ bit was checked by hayato@chromium.org

[hayato, 2016-06-01 01:15:48]

The patchset sent to the CQ was uploaded after l-g-t-m from rune@opera.com,
esprehn@chromium.org
Link to the patchset: https://codereview.chromium.org/1883083002/#ps80001
(title: "renamed")

[commit-bot: I haz the power, 2016-06-01 01:16:27]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1883083002/80001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1883083002/80001

[commit-bot: I haz the power, 2016-06-01 05:32:42]

Committed patchset #5 (id:80001)

[commit-bot: I haz the power, 2016-06-01 05:34:33]

Description was changed from

==========
Introduce Node::containingTreeScope(), which asserts that the node's root is a
tree scope

The difference between Node::containingTreeScope() and Node::treeScope() is that

Node::containingTreeScope() asserts that the node is in a treeScope.
Node::treeScope() does not.

Replace some of the current usage of Node::treeScope() with 
Node::containingTreeScope() if containingTreeScope() is better to express their
intention.

This CL also introduces Element::adjustedFocusedElementInTreeScope() to optimize
some code path where we can early exit if the element is not in a tree scope.

BUG=602556
==========

to

==========
Introduce Node::containingTreeScope(), which asserts that the node's root is a
tree scope

The difference between Node::containingTreeScope() and Node::treeScope() is that
Node::containingTreeScope() asserts that the node is in a treeScope.
Node::treeScope() does not.

Replace some of the current usage of Node::treeScope() with
Node::containingTreeScope() if containingTreeScope() is better to express their
intention.

This CL also introduces Element::adjustedFocusedElementInTreeScope() to optimize
some code path where we can early exit if the element is not in a tree scope.

BUG=602556

Committed: https://crrev.com/97a40b7bad083e7f9382a92e00adf1bd650db4a5
Cr-Commit-Position: refs/heads/master@{#397065}
==========

[commit-bot: I haz the power, 2016-06-01 05:34:35]

Patchset 5 (id:??) landed as
https://crrev.com/97a40b7bad083e7f9382a92e00adf1bd650db4a5
Cr-Commit-Position: refs/heads/master@{#397065}