Partial fix for keyboard occlusion.

Partial fix for keyboard occlusion.

This fixes three aspects of keyboard occlusion.  When the keyboard is
showing:
- full-screen windows are resized to still be 100% visible
- panels are moved above the keyboard
- the shelf is hidden

This CL does not address:
- moving focused text fields into view in web pages

BUG=235157
Committed: https://src.chromium.org/viewvc/chrome?view=rev&revision=197129

[sadrul, 2013-04-24 22:56:24]

Does this take care of dialog boxes with text fields that may be hidden by the
keyboard (or if the omnibox is in a place that gets hidden by the keyboard)?

[bryeung, 2013-04-25 14:49:55]

No.  If I understand the question, both of those cases would require moving the
window/dialog into the work area.  If Ash doesn't do that already, this CL won't
help.

[sadrul, 2013-04-26 02:54:16]

> No.  If I understand the question, both of those cases would require moving
the
> window/dialog into the work area.  If Ash doesn't do that already, this CL
won't
> help.

I don't think ash does that at the moment. We might need to do something about
it.

LGTM with nits:

https://codereview.chromium.org/14477010/diff/1/ash/shelf/shelf_layout_manage...
File ash/shelf/shelf_layout_manager.cc (right):

https://codereview.chromium.org/14477010/diff/1/ash/shelf/shelf_layout_manage...
ash/shelf/shelf_layout_manager.cc:706: keyboard_bounds_.height(),
If you set bottom to:

  keyboard_bounds_.height() + work_area_insets.bottom();

I suppose that would end up showing the shelf right on top of the keyboard?

https://codereview.chromium.org/14477010/diff/1/ui/keyboard/keyboard_controll...
File ui/keyboard/keyboard_controller.cc (right):

https://codereview.chromium.org/14477010/diff/1/ui/keyboard/keyboard_controll...
ui/keyboard/keyboard_controller.cc:188: gfx::Rect(0, 0, 0, 0)));
just gfx::Rect()

[bryeung, 2013-04-26 13:44:50]

Thanks Sadrul.

+ben for OWNERS

https://codereview.chromium.org/14477010/diff/1/ash/shelf/shelf_layout_manage...
File ash/shelf/shelf_layout_manager.cc (right):

https://codereview.chromium.org/14477010/diff/1/ash/shelf/shelf_layout_manage...
ash/shelf/shelf_layout_manager.cc:706: keyboard_bounds_.height(),
On 2013/04/26 02:54:16, sadrul wrote:
> If you set bottom to:
> 
>   keyboard_bounds_.height() + work_area_insets.bottom();
> 
> I suppose that would end up showing the shelf right on top of the keyboard?

When the keyboard is showing, work_area_inserts.bottom() == 0.

Either:
- the shelf is horizontal, and we would have set shelf_height to zero above
- we selected a value with 0 for bottom above

That said, it seems a bit safer and more consistent to include the
work_area_insets.bottom() value.

https://codereview.chromium.org/14477010/diff/1/ui/keyboard/keyboard_controll...
File ui/keyboard/keyboard_controller.cc (right):

https://codereview.chromium.org/14477010/diff/1/ui/keyboard/keyboard_controll...
ui/keyboard/keyboard_controller.cc:188: gfx::Rect(0, 0, 0, 0)));
On 2013/04/26 02:54:16, sadrul wrote:
> just gfx::Rect()

Oh, that's nicer.  Thanks!

[bryeung, 2013-04-29 15:53:11]

ping

[Ben Goodger (Google), 2013-04-29 16:16:38]

lgtm

https://codereview.chromium.org/14477010/diff/12001/ui/keyboard/keyboard_cont...
File ui/keyboard/keyboard_controller.h (right):

https://codereview.chromium.org/14477010/diff/12001/ui/keyboard/keyboard_cont...
ui/keyboard/keyboard_controller.h:36: class Observer {
avoid nested classes, they can't be forward-declared

[bryeung, 2013-04-29 18:31:43]

Thanks Ben: moved the Observer out.

https://codereview.chromium.org/14477010/diff/12001/ui/keyboard/keyboard_cont...
File ui/keyboard/keyboard_controller.h (right):

https://codereview.chromium.org/14477010/diff/12001/ui/keyboard/keyboard_cont...
ui/keyboard/keyboard_controller.h:36: class Observer {
On 2013/04/29 16:16:38, Ben Goodger (Google) wrote:
> avoid nested classes, they can't be forward-declared

Got it.  Fixed.

[commit-bot: I haz the power, 2013-04-29 18:31:54]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/bryeung@chromium.org/14477010/17001

[commit-bot: I haz the power, 2013-04-29 18:37:03]

Failed to trigger a try job on linux_chromeos_clang
HTTP Error 400: Bad Request

[commit-bot: I haz the power, 2013-04-29 18:37:32]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/bryeung@chromium.org/14477010/18014

[commit-bot: I haz the power, 2013-04-29 20:52:58]

Change committed as 197129