Move layouet/LayoutView::setSelection() to editing/LayoutSelection

Move layout/LayoutView::setSelection() to editing/LayoutSelection

This moves layout marking functionality from LayoutView to LayoutSelection
 to prepare for LayoutNG.
This is refatoring but difference of each lifetime makes different 
 LayoutObject member clearing timing(see FrameSelection::ContextDestroyed).

Detail:
LayoutView::
 -SetSelection()
 -SelectionBounds()
 -InvalidatePaintForSelection()
 -selection_start_, selection_end_
 -selection_start_pos_, selection_end_pos_
-> move to LayoutSelection. Leave function definition in LayoutView and will move it
 to LayoutSelection later.

 ClearSelection(), SelectionStartEnd()
  They are used from few LayoutObject classes so move function body to 
 LayoutSelection and make LayoutView and FrameSelection proxies.

LayoutSelection::
 +SetSelection()
 +SelectionBounds()
 +InvalidatePaintForSelection()
 +selection_start_, selection_end_
 +selection_start_pos_, selection_end_pos_
 +OnDocumentShutdown(); 
  This just clear above 4 members because LayoutSelection lives longer than 
 LayoutView.

LayoutViewItem::
 -SelectionBounds()
 -InvalidatePaintForSelection()
  They are only used from FrameSelection. Since FrameSelection has LayoutSelection,
 just use it directly.

BUG=708453
Review-Url: https://codereview.chromium.org/2800813006
Cr-Commit-Position: refs/heads/master@{#463978}
Committed: https://chromium.googlesource.com/chromium/src/+/b33836726c84241997ec990495acd9f9616008d0

[yoichio, 2017-04-06 08:28:10]

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

[commit-bot: I haz the power, 2017-04-06 08:28:23]

Dry run: CQ is trying da patch. Follow status at
 
https://chromium-cq-status.appspot.com/v2/patch-status/codereview.chromium.or...

[commit-bot: I haz the power, 2017-04-06 09:13:41]

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

[commit-bot: I haz the power, 2017-04-06 09:13:42]

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

[yoichio, 2017-04-10 06:53:22]

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

[commit-bot: I haz the power, 2017-04-10 06:53:32]

Dry run: CQ is trying da patch. Follow status at
 
https://chromium-cq-status.appspot.com/v2/patch-status/codereview.chromium.or...

[yoichio, 2017-04-10 08:16:17]

Patchset #1 (id:1) has been deleted

[yoichio, 2017-04-10 08:16:25]

Patchset #1 (id:20001) has been deleted

[yoichio, 2017-04-10 08:16:31]

Patchset #1 (id:40001) has been deleted

[yoichio, 2017-04-10 08:36:16]

Description was changed from

==========
Move LayoutView::*selection to editing/LayoutSelection

BUG=708453
==========

to

==========
Move layout/LayoutView::setSelection() to editing/LayoutSelection

This moves layout marking functionality from LayoutView to LayoutSelection
 to prepare for LayoutNG.
This is refatoring but difference of each lifetime makes different 
 LayoutObject member clearing timing(see FrameSelection::ContextDestroyed).

Detail:
LayoutView::
 -SetSelection()
 -SelectionBounds()
 -InvalidatePaintForSelection()
 -selection_start_, selection_end_
 -selection_start_pos_, selection_end_pos_
-> move to LayoutSelection. Leave function definition in LayoutView and will
move it
 to LayoutSelection later.

 ClearSelection(), SelectionStartEnd()
  They are used from few LayoutObject classes so move function body to 
 LayoutSelection and make LayoutView and FrameSelection proxies.

LayoutSelection::
 +SetSelection()
 +SelectionBounds()
 +InvalidatePaintForSelection()
 +selection_start_, selection_end_
 +selection_start_pos_, selection_end_pos_
 +OnDocumentShutdown(); 
  This just clear above 4 members because LayoutSelection lives longer than 
 LayoutView.

LayoutViewItem::
 -SelectionBounds()
 -InvalidatePaintForSelection()
  They are only used from FrameSelection. Since FrameSelection has
LayoutSelection,
 just use it directly.

BUG=708453
==========

[yoichio, 2017-04-10 08:36:16]

yoichio@chromium.org changed reviewers:
+ kojii@chromium.org, yosin@chromium.org

[yosin_UTC9, 2017-04-10 08:56:48]

lgtm

Thanks!

[commit-bot: I haz the power, 2017-04-10 09:09:31]

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

[commit-bot: I haz the power, 2017-04-10 09:09:31]

Dry run: This issue passed the CQ dry run.

[kojii, 2017-04-10 15:16:22]

Wish to help, but I'm afraid I don't know much about LayoutView, when its
frame_view_ is nullptr, and such.

Could you try to find more appropriate reviewers for LayoutView? If that doesn't
work, I can rs.

[yoichio, 2017-04-11 01:39:05]

yoichio@chromium.org changed reviewers:
+ eae@chromium.org

[yoichio, 2017-04-11 01:39:06]

Add Emil for layout/ reviewer. Could you take a look?

[kojii, 2017-04-11 05:43:14]

kojii@chromium.org changed reviewers:
+ dcheng@chromium.org

[kojii, 2017-04-11 05:43:15]

dcheng@, are you familiar with LayoutView and its life cycles?

[dcheng, 2017-04-11 06:29:15]

https://codereview.chromium.org/2800813006/diff/60001/third_party/WebKit/Sour...
File third_party/WebKit/Source/core/editing/LayoutSelection.cpp (right):

https://codereview.chromium.org/2800813006/diff/60001/third_party/WebKit/Sour...
third_party/WebKit/Source/core/editing/LayoutSelection.cpp:34:
LayoutSelection::LayoutSelection(FrameSelection& frameSelection)
Nit: mind fixing the name to be frame_selection?

https://codereview.chromium.org/2800813006/diff/60001/third_party/WebKit/Sour...
File third_party/WebKit/Source/core/layout/LayoutView.cpp (left):

https://codereview.chromium.org/2800813006/diff/60001/third_party/WebKit/Sour...
third_party/WebKit/Source/core/layout/LayoutView.cpp:840: if (!frame_view_)
I'm not actually sure if there is any time this will ever be true:
- LayoutView is created during Document::Initialize()
- Document::Initialize() is called by LocalDOMWindow::InstallNewDocument()
- LocalDOMWindow::InstallNewDocument() is called when we are committing a
provisional load:
https://cs.chromium.org/chromium/src/third_party/WebKit/Source/core/loader/Do...
- But before we install a new document, we ensure the FrameView is created in
https://cs.chromium.org/chromium/src/third_party/WebKit/Source/core/loader/Fr...

So probably all these null checks are pointless.

https://codereview.chromium.org/2800813006/diff/60001/third_party/WebKit/Sour...
File third_party/WebKit/Source/core/layout/LayoutView.cpp (right):

https://codereview.chromium.org/2800813006/diff/60001/third_party/WebKit/Sour...
third_party/WebKit/Source/core/layout/LayoutView.cpp:840: if
(!frame_selection_->GetDocument().GetLayoutView()->GetFrameView())
I would probably try DCHECKing this. It looks like this is called during layout,
and I wouldn't expect to be updating layout on an inactive Document.

[yoichio, 2017-04-11 07:20:53]

Patchset #2 (id:80001) has been deleted

[yoichio, 2017-04-11 07:21:11]

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

[commit-bot: I haz the power, 2017-04-11 07:21:31]

Dry run: CQ is trying da patch. Follow status at
 
https://chromium-cq-status.appspot.com/v2/patch-status/codereview.chromium.or...

[yoichio, 2017-04-11 07:23:58]

https://codereview.chromium.org/2800813006/diff/60001/third_party/WebKit/Sour...
File third_party/WebKit/Source/core/editing/LayoutSelection.cpp (right):

https://codereview.chromium.org/2800813006/diff/60001/third_party/WebKit/Sour...
third_party/WebKit/Source/core/editing/LayoutSelection.cpp:34:
LayoutSelection::LayoutSelection(FrameSelection& frameSelection)
On 2017/04/11 06:29:14, dcheng wrote:
> Nit: mind fixing the name to be frame_selection?

Done.

https://codereview.chromium.org/2800813006/diff/60001/third_party/WebKit/Sour...
File third_party/WebKit/Source/core/layout/LayoutView.cpp (right):

https://codereview.chromium.org/2800813006/diff/60001/third_party/WebKit/Sour...
third_party/WebKit/Source/core/layout/LayoutView.cpp:840: if
(!frame_selection_->GetDocument().GetLayoutView()->GetFrameView())
On 2017/04/11 06:29:15, dcheng wrote:
> I would probably try DCHECKing this. It looks like this is called during
layout,
> and I wouldn't expect to be updating layout on an inactive Document.
I agree. I would DCHECK in another patch.

[commit-bot: I haz the power, 2017-04-11 08:53:10]

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

[commit-bot: I haz the power, 2017-04-11 08:53:11]

Dry run: This issue passed the CQ dry run.

[yoichio, 2017-04-12 01:49:09]

PTAL.

[dcheng, 2017-04-12 07:20:50]

LGTM

https://codereview.chromium.org/2800813006/diff/100001/third_party/WebKit/Sou...
File third_party/WebKit/Source/core/layout/LayoutView.cpp (right):

https://codereview.chromium.org/2800813006/diff/100001/third_party/WebKit/Sou...
third_party/WebKit/Source/core/layout/LayoutView.cpp:840: if
(!frame_selection_->GetDocument().GetLayoutView()->GetFrameView())
Nit: add a TODO to try DCHECKing this, just in case this gets pre-empted by
other work =)

[eae, 2017-04-12 07:35:31]

LGTM

Lifecycle looks fine, thanks for checking!

[yoichio, 2017-04-12 08:12:19]

Thanks all for reviewing!

https://codereview.chromium.org/2800813006/diff/100001/third_party/WebKit/Sou...
File third_party/WebKit/Source/core/layout/LayoutView.cpp (right):

https://codereview.chromium.org/2800813006/diff/100001/third_party/WebKit/Sou...
third_party/WebKit/Source/core/layout/LayoutView.cpp:840: if
(!frame_selection_->GetDocument().GetLayoutView()->GetFrameView())
On 2017/04/12 07:20:50, dcheng wrote:
> Nit: add a TODO to try DCHECKing this, just in case this gets pre-empted by
> other work =)

Done.

[yoichio, 2017-04-12 08:12:24]

The CQ bit was checked by yoichio@chromium.org

[yoichio, 2017-04-12 08:12:24]

The patchset sent to the CQ was uploaded after l-g-t-m from yosin@chromium.org,
eae@chromium.org, dcheng@chromium.org
Link to the patchset: https://codereview.chromium.org/2800813006/#ps120001
(title: "nit")

[commit-bot: I haz the power, 2017-04-12 08:12:38]

CQ is trying da patch. Follow status at
 
https://chromium-cq-status.appspot.com/v2/patch-status/codereview.chromium.or...

[commit-bot: I haz the power, 2017-04-12 10:00:27]

CQ is committing da patch.

Bot data: {"patchset_id": 120001, "attempt_start_ts": 1491984744027020,
"parent_rev": "1e2b7d3bcefad78b6062382c40f05c96da1bd364", "commit_rev":
"b33836726c84241997ec990495acd9f9616008d0"}

[commit-bot: I haz the power, 2017-04-12 10:01:22]

Description was changed from

==========
Move layout/LayoutView::setSelection() to editing/LayoutSelection

This moves layout marking functionality from LayoutView to LayoutSelection
 to prepare for LayoutNG.
This is refatoring but difference of each lifetime makes different 
 LayoutObject member clearing timing(see FrameSelection::ContextDestroyed).

Detail:
LayoutView::
 -SetSelection()
 -SelectionBounds()
 -InvalidatePaintForSelection()
 -selection_start_, selection_end_
 -selection_start_pos_, selection_end_pos_
-> move to LayoutSelection. Leave function definition in LayoutView and will
move it
 to LayoutSelection later.

 ClearSelection(), SelectionStartEnd()
  They are used from few LayoutObject classes so move function body to 
 LayoutSelection and make LayoutView and FrameSelection proxies.

LayoutSelection::
 +SetSelection()
 +SelectionBounds()
 +InvalidatePaintForSelection()
 +selection_start_, selection_end_
 +selection_start_pos_, selection_end_pos_
 +OnDocumentShutdown(); 
  This just clear above 4 members because LayoutSelection lives longer than 
 LayoutView.

LayoutViewItem::
 -SelectionBounds()
 -InvalidatePaintForSelection()
  They are only used from FrameSelection. Since FrameSelection has
LayoutSelection,
 just use it directly.

BUG=708453
==========

to

==========
Move layout/LayoutView::setSelection() to editing/LayoutSelection

This moves layout marking functionality from LayoutView to LayoutSelection
 to prepare for LayoutNG.
This is refatoring but difference of each lifetime makes different 
 LayoutObject member clearing timing(see FrameSelection::ContextDestroyed).

Detail:
LayoutView::
 -SetSelection()
 -SelectionBounds()
 -InvalidatePaintForSelection()
 -selection_start_, selection_end_
 -selection_start_pos_, selection_end_pos_
-> move to LayoutSelection. Leave function definition in LayoutView and will
move it
 to LayoutSelection later.

 ClearSelection(), SelectionStartEnd()
  They are used from few LayoutObject classes so move function body to 
 LayoutSelection and make LayoutView and FrameSelection proxies.

LayoutSelection::
 +SetSelection()
 +SelectionBounds()
 +InvalidatePaintForSelection()
 +selection_start_, selection_end_
 +selection_start_pos_, selection_end_pos_
 +OnDocumentShutdown(); 
  This just clear above 4 members because LayoutSelection lives longer than 
 LayoutView.

LayoutViewItem::
 -SelectionBounds()
 -InvalidatePaintForSelection()
  They are only used from FrameSelection. Since FrameSelection has
LayoutSelection,
 just use it directly.

BUG=708453

Review-Url: https://codereview.chromium.org/2800813006
Cr-Commit-Position: refs/heads/master@{#463978}
Committed:
https://chromium.googlesource.com/chromium/src/+/b33836726c84241997ec990495ac...
==========

[commit-bot: I haz the power, 2017-04-12 10:01:23]

Committed patchset #3 (id:120001) as
https://chromium.googlesource.com/chromium/src/+/b33836726c84241997ec990495ac...

[hugoh_UTC2, 2017-04-13 06:53:03]

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

[hugoh_UTC2, 2017-04-13 06:53:04]

https://codereview.chromium.org/2800813006/diff/120001/third_party/WebKit/Sou...
File third_party/WebKit/Source/core/editing/LayoutSelection.h (right):

https://codereview.chromium.org/2800813006/diff/120001/third_party/WebKit/Sou...
third_party/WebKit/Source/core/editing/LayoutSelection.h:78: //
|m_selectionStartPos| and |m_selectionEndPos| are only valid for
Post-merge nit: we should update the comment, now that m_selectionStartPos is
renamed to selection_start_pos_ etc.