Remove final RenderWidgetHostImpl dependency from OverscrollController

Remove final RenderWidgetHostImpl dependency from OverscrollController

Shift the view bounds calculation from the RWHI to the
OverscrollControllerDelegate. This allows more flexibility both in testing
and the location of OverscrollController ownership.

BUG=306194
Committed: https://src.chromium.org/viewvc/chrome?view=rev&revision=230502

[jdduke (slow), 2013-10-21 19:56:53]

sadrul@: PTAL.  After this I'd like to move this into the input/ folder, and
perhaps split off the unit tests (if possible, I haven't looked closely enough
to see how much they rely on the GEF/Router/RWHI.

https://codereview.chromium.org/30683003/diff/80001/content/browser/renderer_...
File content/browser/renderer_host/overscroll_controller.cc (right):

https://codereview.chromium.org/30683003/diff/80001/content/browser/renderer_...
content/browser/renderer_host/overscroll_controller.cc:145: if (!delegate_)
Are there cases where we want to process overscroll without a valid |delegate_|?
 I would think |delegate_|'s lifetime here is closely associated with the view?
This may be a change in behavior we don't want, but it's hard for me to say
without knowing more.

[sadrul, 2013-10-21 23:28:35]

LGTM

https://codereview.chromium.org/30683003/diff/80001/content/browser/renderer_...
File content/browser/renderer_host/overscroll_controller.cc (right):

https://codereview.chromium.org/30683003/diff/80001/content/browser/renderer_...
content/browser/renderer_host/overscroll_controller.cc:145: if (!delegate_)
On 2013/10/21 19:56:53, jdduke wrote:
> Are there cases where we want to process overscroll without a valid
|delegate_|?
>  I would think |delegate_|'s lifetime here is closely associated with the
view?
> This may be a change in behavior we don't want, but it's hard for me to say
> without knowing more.

There shouldn't be such cases. This can probably be replaced with a
DCHECK(delegate_);

https://codereview.chromium.org/30683003/diff/80001/content/browser/renderer_...
File content/browser/renderer_host/render_widget_host_unittest.cc (right):

https://codereview.chromium.org/30683003/diff/80001/content/browser/renderer_...
content/browser/renderer_host/render_widget_host_unittest.cc:61:
TestOverscrollDelegate(RenderWidgetHostView* view)
explicit

[jdduke (slow), 2013-10-22 15:15:58]

https://codereview.chromium.org/30683003/diff/80001/content/browser/renderer_...
File content/browser/renderer_host/overscroll_controller.cc (right):

https://codereview.chromium.org/30683003/diff/80001/content/browser/renderer_...
content/browser/renderer_host/overscroll_controller.cc:145: if (!delegate_)
On 2013/10/21 23:28:35, sadrul wrote:
> On 2013/10/21 19:56:53, jdduke wrote:
> > Are there cases where we want to process overscroll without a valid
> |delegate_|?
> >  I would think |delegate_|'s lifetime here is closely associated with the
> view?
> > This may be a change in behavior we don't want, but it's hard for me to say
> > without knowing more.
> 
> There shouldn't be such cases. This can probably be replaced with a
> DCHECK(delegate_);

Done.

https://codereview.chromium.org/30683003/diff/80001/content/browser/renderer_...
File content/browser/renderer_host/render_widget_host_unittest.cc (right):

https://codereview.chromium.org/30683003/diff/80001/content/browser/renderer_...
content/browser/renderer_host/render_widget_host_unittest.cc:61:
TestOverscrollDelegate(RenderWidgetHostView* view)
On 2013/10/21 23:28:35, sadrul wrote:
> explicit

Done.

[jdduke (slow), 2013-10-22 15:18:27]

jochen@: owner review for content/browser/web_contents/? Thanks.

[jochen (gone - plz use gerrit), 2013-10-23 14:02:50]

lgtm

[commit-bot: I haz the power, 2013-10-23 14:57:11]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/jdduke@chromium.org/30683003/270001

[commit-bot: I haz the power, 2013-10-23 20:47:13]

Change committed as 230502