Aura: Switch window mode when screen resolution changes

Aura: Switch window mode when screen resolution changes

If the user plugs in a high-resolution external monitor, dynamically switch out of compact window mode into overlapping window mode.  Force re-layout of all containers to ensure launcher, status area, etc. show up in proper locations.

BUG=110174
TEST=aura_shell_unittests ShellTest, try plugging and unplugging a monitor with resolution > 1400 pixels wide


Committed: http://src.chromium.org/viewvc/chrome?view=rev&revision=120692

[James Cook, 2012-02-03 18:44:22]

Scott, PTAL at ash/*

Oshima, PTAL at opaque_browser_frame_view.cc.

[oshima, 2012-02-03 19:00:30]

On 2012/02/03 18:44:22, James Cook (Chromium) wrote:
> Scott, PTAL at ash/*
> 
> Oshima, PTAL at opaque_browser_frame_view.cc.

LGTM on opaque_browser_frame_view.cc

[sky, 2012-02-03 21:47:42]

https://chromiumcodereview.appspot.com/9320076/diff/1/ash/shell.cc
File ash/shell.cc (right):

https://chromiumcodereview.appspot.com/9320076/diff/1/ash/shell.cc#newcode425
ash/shell.cc:425: for (Window::Windows::const_iterator it =
root_window->children().begin();
Don't the children handle this themselves?

[James Cook, 2012-02-03 22:02:43]

http://codereview.chromium.org/9320076/diff/1/ash/shell.cc
File ash/shell.cc (right):

http://codereview.chromium.org/9320076/diff/1/ash/shell.cc#newcode425
ash/shell.cc:425: for (Window::Windows::const_iterator it =
root_window->children().begin();
On 2012/02/03 21:47:42, sky wrote:
> Don't the children handle this themselves?

They do, but too early for it to be useful for me.  By the time I get the
OnRootWindowResized() call in Shell the children have already had their bounds
set and OnWindowResized called.  Do you think the window-mode-selection code
should be triggered by RootWindowLayoutManager::OnWindowResized() so it can
happen before child bounds are set?

[James Cook, 2012-02-06 17:34:53]

Scott, PTAL - I moved the code to trigger the window mode change into
RootWindowLayoutManager, which means Shell doesn't have to be a
RootWindowObserver and we don't have to explicitly trigger a re-layout of the
windows.  Does this look better?

[sky, 2012-02-06 17:54:05]

LGTM

[commit-bot: I haz the power, 2012-02-06 20:33:25]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/jamescook@chromium.org/9320076/11003

[commit-bot: I haz the power, 2012-02-06 22:50:24]

The commit queue went berserk retrying too often for a
seemingly flaky test. Builder is mac_rel, revision is 120625, job name
was 9320076-11003 (previous was lost) (retry) (previous was lost) (previous was
lost).

[commit-bot: I haz the power, 2012-02-07 00:59:11]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/jamescook@chromium.org/9320076/11003

[commit-bot: I haz the power, 2012-02-07 02:41:18]

Change committed as 120692