Rename PaintLayer::enclosingElement -> PaintLayer::enclosingNode

Rename PaintLayer::enclosingElement -> PaintLayer::enclosingNode

This function returns a Node*, not an Element*. The callsites are a
little confusing, currently.

BUG=None

Committed: https://crrev.com/b357b25045f5f8b8080fb26489287fe6af55a7b6
Cr-Commit-Position: refs/heads/master@{#363065}

[Ian Vollick, 2015-12-03 17:49:08]

Description was changed from

==========
Rename PaintLayer::enclosingElement -> PaintLayer::enclosingNode

This function returns a Node*, not an Element*. The callsites are a
little confusing, currently.

BUG=None
==========

to

==========
Rename PaintLayer::enclosingElement -> PaintLayer::enclosingNode

This function returns a Node*, not an Element*. The callsites are a
little confusing, currently.

BUG=None
==========

[Ian Vollick, 2015-12-03 17:49:08]

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

[chrishtr, 2015-12-03 17:57:51]

chrishtr@chromium.org changed reviewers:
+ chrishtr@chromium.org

[chrishtr, 2015-12-03 17:57:52]

https://codereview.chromium.org/1495053002/diff/1/third_party/WebKit/Source/c...
File third_party/WebKit/Source/core/paint/PaintLayer.cpp (right):

https://codereview.chromium.org/1495053002/diff/1/third_party/WebKit/Source/c...
third_party/WebKit/Source/core/paint/PaintLayer.cpp:1549: Node*
PaintLayer::enclosingNode() const
I think paint layers will always be for Elements. Two of the three use cases of
this function also require element, not node (they use toElement()
unconditionally to cast).

[Ian Vollick, 2015-12-03 18:09:56]

On 2015/12/03 17:57:52, chrishtr wrote:
>
https://codereview.chromium.org/1495053002/diff/1/third_party/WebKit/Source/c...
> File third_party/WebKit/Source/core/paint/PaintLayer.cpp (right):
> 
>
https://codereview.chromium.org/1495053002/diff/1/third_party/WebKit/Source/c...
> third_party/WebKit/Source/core/paint/PaintLayer.cpp:1549: Node*
> PaintLayer::enclosingNode() const
> I think paint layers will always be for Elements. Two of the three use cases
of
> this function also require element, not node (they use toElement()
> unconditionally to cast).

PaintLayer::hitTestContents passes the Node* to HitTestResult::setInnerNode.

Some context: see Elliott's last comment on
https://codereview.chromium.org/1405993008/

[chrishtr, 2015-12-03 18:17:31]

On 2015/12/03 at 18:09:56, vollick wrote:
> On 2015/12/03 17:57:52, chrishtr wrote:
> >
https://codereview.chromium.org/1495053002/diff/1/third_party/WebKit/Source/c...
> > File third_party/WebKit/Source/core/paint/PaintLayer.cpp (right):
> > 
> >
https://codereview.chromium.org/1495053002/diff/1/third_party/WebKit/Source/c...
> > third_party/WebKit/Source/core/paint/PaintLayer.cpp:1549: Node*
> > PaintLayer::enclosingNode() const
> > I think paint layers will always be for Elements. Two of the three use cases
of
> > this function also require element, not node (they use toElement()
> > unconditionally to cast).
> 
> PaintLayer::hitTestContents passes the Node* to HitTestResult::setInnerNode.

Right, saw that. But it's ok to pass an Element to setInnerNode. So if it's true
that a PaintLayer is always backed by an Element, then it would be good to
return Element* from this function.

> 
> Some context: see Elliott's last comment on
https://codereview.chromium.org/1405993008/

Thanks for the context.

[esprehn, 2015-12-03 18:58:30]

On 2015/12/03 at 17:57:52, chrishtr wrote:
>
https://codereview.chromium.org/1495053002/diff/1/third_party/WebKit/Source/c...
> File third_party/WebKit/Source/core/paint/PaintLayer.cpp (right):
> 
>
https://codereview.chromium.org/1495053002/diff/1/third_party/WebKit/Source/c...
> third_party/WebKit/Source/core/paint/PaintLayer.cpp:1549: Node*
PaintLayer::enclosingNode() const
> I think paint layers will always be for Elements. Two of the three use cases
of this function also require element, not node (they use toElement()
unconditionally to cast).

The Document has a PaintLayer and is not an Element.

[Ian Vollick, 2015-12-03 19:01:04]

On 2015/12/03 18:58:30, esprehn wrote:
> On 2015/12/03 at 17:57:52, chrishtr wrote:
> >
>
https://codereview.chromium.org/1495053002/diff/1/third_party/WebKit/Source/c...
> > File third_party/WebKit/Source/core/paint/PaintLayer.cpp (right):
> > 
> >
>
https://codereview.chromium.org/1495053002/diff/1/third_party/WebKit/Source/c...
> > third_party/WebKit/Source/core/paint/PaintLayer.cpp:1549: Node*
> PaintLayer::enclosingNode() const
> > I think paint layers will always be for Elements. Two of the three use cases
> of this function also require element, not node (they use toElement()
> unconditionally to cast).
> 
> The Document has a PaintLayer and is not an Element.

Elliott, are you ok with this change, then?

[esprehn, 2015-12-03 19:07:21]

On 2015/12/03 at 19:01:04, vollick wrote:
> On 2015/12/03 18:58:30, esprehn wrote:
> > On 2015/12/03 at 17:57:52, chrishtr wrote:
> > >
> >
https://codereview.chromium.org/1495053002/diff/1/third_party/WebKit/Source/c...
> > > File third_party/WebKit/Source/core/paint/PaintLayer.cpp (right):
> > > 
> > >
> >
https://codereview.chromium.org/1495053002/diff/1/third_party/WebKit/Source/c...
> > > third_party/WebKit/Source/core/paint/PaintLayer.cpp:1549: Node*
> > PaintLayer::enclosingNode() const
> > > I think paint layers will always be for Elements. Two of the three use
cases
> > of this function also require element, not node (they use toElement()
> > unconditionally to cast).
> > 
> > The Document has a PaintLayer and is not an Element.
> 
> Elliott, are you ok with this change, then?

Yeah, lgtm. The unconditional casts are technically safe since we never set
those properties on the Document. The filter code in general just needs clean
up.

[Ian Vollick, 2015-12-03 20:15:51]

The CQ bit was checked by vollick@chromium.org

[commit-bot: I haz the power, 2015-12-03 20:16:53]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1495053002/40001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1495053002/40001

[commit-bot: I haz the power, 2015-12-03 21:37:39]

Description was changed from

==========
Rename PaintLayer::enclosingElement -> PaintLayer::enclosingNode

This function returns a Node*, not an Element*. The callsites are a
little confusing, currently.

BUG=None
==========

to

==========
Rename PaintLayer::enclosingElement -> PaintLayer::enclosingNode

This function returns a Node*, not an Element*. The callsites are a
little confusing, currently.

BUG=None
==========

[commit-bot: I haz the power, 2015-12-03 21:37:40]

Committed patchset #3 (id:40001)

[commit-bot: I haz the power, 2015-12-03 21:39:16]

Description was changed from

==========
Rename PaintLayer::enclosingElement -> PaintLayer::enclosingNode

This function returns a Node*, not an Element*. The callsites are a
little confusing, currently.

BUG=None
==========

to

==========
Rename PaintLayer::enclosingElement -> PaintLayer::enclosingNode

This function returns a Node*, not an Element*. The callsites are a
little confusing, currently.

BUG=None

Committed: https://crrev.com/b357b25045f5f8b8080fb26489287fe6af55a7b6
Cr-Commit-Position: refs/heads/master@{#363065}
==========

[commit-bot: I haz the power, 2015-12-03 21:39:17]

Patchset 3 (id:??) landed as
https://crrev.com/b357b25045f5f8b8080fb26489287fe6af55a7b6
Cr-Commit-Position: refs/heads/master@{#363065}