[InputEvent] Add 'beforeinput' logic in |will*()| and guarded by a flag (3/11)

[InputEvent] Add 'beforeinput' logic in |will*()| and guarded by a flag (3/11)

This CL added 'beforeinput' logic in:
1. |CompositeEditCommand::willApplyEditing()|
2. |EditCommandComposition::willUnapply()|
3. |EditCommandComposition::willReapply()|

They are disabled by default, and following CLs will move 'beforeinput' logic one by
one.

This CL doesn't have behavior change.

Please see bug and the design doc below for the big picture:
https://docs.google.com/a/chromium.org/document/d/1M9bQaLKdRpR1qJ9i7zZKN-hKwd4ZkgJs5GPde2ZsDy0/edit?usp=sharing

BUG=678795

[chongz, 2017-01-05 20:30:50]

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

[commit-bot: I haz the power, 2017-01-05 20:31:23]

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, 2017-01-05 22:33:46]

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

[commit-bot: I haz the power, 2017-01-05 22:33:46]

Dry run: This issue passed the CQ dry run.

[chongz, 2017-01-05 23:05:12]

Description was changed from

==========
Implemented 'beforeinput' logic in will*

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

to

==========
[InputEvent] Add 'beforeinput' logic in |will*()| and guarded by a flag (3/11)

This CL added 'beforeinput' logic in:
1. |CompositeEditCommand::willApplyEditing()|
2. |EditCommandComposition::willUnapply()|
3. |EditCommandComposition::willReapply()|

They are disabled by default, and following CLs will move 'beforeinput' logic
one by
one.

This CL doesn't have behavior change.

Please see bug and the design doc below for the big picture:
https://docs.google.com/a/chromium.org/document/d/1M9bQaLKdRpR1qJ9i7zZKN-hKwd...

BUG=678795
==========

[chongz, 2017-01-05 23:40:27]

chongz@chromium.org changed reviewers:
+ xiaochengh@chromium.org

[chongz, 2017-01-05 23:40:28]

PTAL, thanks!

[Xiaocheng, 2017-01-06 02:24:52]

xiaochengh@chromium.org changed reviewers:
+ yosin@chromium.org

[Xiaocheng, 2017-01-06 02:24:53]

+yosin

https://codereview.chromium.org/2618583003/diff/1/third_party/WebKit/Source/c...
File third_party/WebKit/Source/core/editing/EditingUtilities.cpp (right):

https://codereview.chromium.org/2618583003/diff/1/third_party/WebKit/Source/c...
third_party/WebKit/Source/core/editing/EditingUtilities.cpp:2200:
!target->isConnected())
We currently don't check this.

Why do we need to cancel subsequent processing when the target node is removed?

[chongz, 2017-01-09 23:26:42]

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

[commit-bot: I haz the power, 2017-01-09 23:27:28]

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

[chongz, 2017-01-09 23:28:07]

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

[commit-bot: I haz the power, 2017-01-09 23:28:34]

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

[chongz, 2017-01-09 23:42:48]

xiaocheng@ Thanks for the review! I've changed event target to
|ensureComposition()->startingRootEditableElement()|, PTAL again, thanks!

https://codereview.chromium.org/2618583003/diff/1/third_party/WebKit/Source/c...
File third_party/WebKit/Source/core/editing/EditingUtilities.cpp (right):

https://codereview.chromium.org/2618583003/diff/1/third_party/WebKit/Source/c...
third_party/WebKit/Source/core/editing/EditingUtilities.cpp:2200:
!target->isConnected())
On 2017/01/06 02:24:53, Xiaocheng wrote:
> We currently don't check this.
> 
> Why do we need to cancel subsequent processing when the target node is
removed?

We currently check |isConnected()| in
|Editor::deleteSelectionAfterDraggingWithEvents()| and
|Editor::replaceSelectionAfterDraggingWithEvents()| but not other places, but I
think we should do the same check for all 'beforeinput' to keep consistence.

See discussion with yosin@:
https://codereview.chromium.org/2288913002/#msg21

[Xiaocheng, 2017-01-10 03:05:12]

lgtm

https://codereview.chromium.org/2618583003/diff/1/third_party/WebKit/Source/c...
File third_party/WebKit/Source/core/editing/EditingUtilities.cpp (right):

https://codereview.chromium.org/2618583003/diff/1/third_party/WebKit/Source/c...
third_party/WebKit/Source/core/editing/EditingUtilities.cpp:2200:
!target->isConnected())
On 2017/01/09 at 23:42:48, chongz wrote:
> On 2017/01/06 02:24:53, Xiaocheng wrote:
> > We currently don't check this.
> > 
> > Why do we need to cancel subsequent processing when the target node is
removed?
> 
> We currently check |isConnected()| in
|Editor::deleteSelectionAfterDraggingWithEvents()| and
|Editor::replaceSelectionAfterDraggingWithEvents()| but not other places, but I
think we should do the same check for all 'beforeinput' to keep consistence.
> 
> See discussion with yosin@:
> https://codereview.chromium.org/2288913002/#msg21

Seems reasonable... As the event target is the editing host, if it gets removed,
no more operation can be done.

[commit-bot: I haz the power, 2017-01-10 04:29:07]

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

[commit-bot: I haz the power, 2017-01-10 04:29:08]

Dry run: This issue passed the CQ dry run.