Don't activate a window if screen is locked.

* Don't activate a window if screen is locked.
* StopEventsPropagation should check window visibility.
* Changed IsScreenLocked to use the same logic as 
  CanFocus.

BUG=none
TEST=new test case is added to window_unittests.cc

Committed: http://src.chromium.org/viewvc/chrome?view=rev&revision=117600

[sky, 2012-01-12 01:05:14]

Ben ran into a similar issue with trying to do a fade animation when you close a
window. The problem he had is that when you close a window, we delete everything
right away, which effectively stops the animation. After discussing with him
he's going to look into adding the ability to detach a layer from an
aura::Window. For Ben's particular case the views::Widget and aura::Window would
be deleted when you close the window, but the ui::Layer would not get deleted
until the animation is done.

I think this would fix the bug you're running into as well. The lock screen
animation code would change to delete the aura::Window right away, but the
animation would continue on. Presumably then the windows used for the lock
screen wouldn't mess with the event propagation code and focus/activation would
work.

WDYT?

http://codereview.chromium.org/9181012/diff/9003/chrome/browser/chromeos/logi...
File chrome/browser/chromeos/login/base_login_display_host.cc (right):

http://codereview.chromium.org/9181012/diff/9003/chrome/browser/chromeos/logi...
chrome/browser/chromeos/login/base_login_display_host.cc:202:
GetWidget()->GetNativeView()->set_stops_event_propagation(false);
Add a comment as to why this is needed.

[oshima, 2012-01-12 20:46:10]

Reduced the scope of this CL just to fix window activation issue when screen is
locked and a few minor changes.

PTAL

[sky, 2012-01-12 21:11:36]

http://codereview.chromium.org/9181012/diff/11046/ash/wm/activation_controlle...
File ash/wm/activation_controller.cc (right):

http://codereview.chromium.org/9181012/diff/11046/ash/wm/activation_controlle...
ash/wm/activation_controller.cc:43: window->CanReceiveEvents() &&
One downside of this is that if the lock screen is up, this will return false,
which will prevent the window from coming to the front so that when the lock
screen goes away it is deactivated. Maybe we should move the window to the front
but not activate it.

http://codereview.chromium.org/9181012/diff/11046/ui/aura/window.cc
File ui/aura/window.cc (right):

http://codereview.chromium.org/9181012/diff/11046/ui/aura/window.cc#newcode362
ui/aura/window.cc:362: bool Window::CanReceiveEvents() const {
Do we really need this? Can we instead use CanFocus?

http://codereview.chromium.org/9181012/diff/11046/ui/aura/window.h
File ui/aura/window.h (right):

http://codereview.chromium.org/9181012/diff/11046/ui/aura/window.h#newcode256
ui/aura/window.h:256: // This window is currently stopping event propagation for
any windows behind
Returns true if this window ...

http://codereview.chromium.org/9181012/diff/11046/ui/aura/window.h#newcode304
ui/aura/window.h:304: // Returns true if the Window is behind a window that
stops event propagation.
'the Window' -> 'this Window'

[oshima, 2012-01-12 21:32:50]

http://codereview.chromium.org/9181012/diff/11046/ash/wm/activation_controlle...
File ash/wm/activation_controller.cc (right):

http://codereview.chromium.org/9181012/diff/11046/ash/wm/activation_controlle...
ash/wm/activation_controller.cc:43: window->CanReceiveEvents() &&
On 2012/01/12 21:11:36, sky wrote:
> One downside of this is that if the lock screen is up, this will return false,
> which will prevent the window from coming to the front so that when the lock
> screen goes away it is deactivated. Maybe we should move the window to the
front
> but not activate it.
good point. done.

http://codereview.chromium.org/9181012/diff/11046/ui/aura/window.cc
File ui/aura/window.cc (right):

http://codereview.chromium.org/9181012/diff/11046/ui/aura/window.cc#newcode362
ui/aura/window.cc:362: bool Window::CanReceiveEvents() const {
On 2012/01/12 21:11:36, sky wrote:
> Do we really need this? Can we instead use CanFocus?
I needed separate method because non focusable window (which is checked by
delegate_->CanFocus()), can be activated.

http://codereview.chromium.org/9181012/diff/11046/ui/aura/window.h
File ui/aura/window.h (right):

http://codereview.chromium.org/9181012/diff/11046/ui/aura/window.h#newcode256
ui/aura/window.h:256: // This window is currently stopping event propagation for
any windows behind
On 2012/01/12 21:11:36, sky wrote:
> Returns true if this window ...

Done.

http://codereview.chromium.org/9181012/diff/11046/ui/aura/window.h#newcode304
ui/aura/window.h:304: // Returns true if the Window is behind a window that
stops event propagation.
On 2012/01/12 21:11:36, sky wrote:
> 'the Window' -> 'this Window'

Done.

[sky, 2012-01-12 22:06:44]

LGTM

[commit-bot: I haz the power, 2012-01-13 01:29:02]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/oshima@chromium.org/9181012/10014

[commit-bot: I haz the power, 2012-01-13 03:15:44]

Change committed as 117600