Add ash SessionStateDelegate

Add ash SessionStateDelegate

This CL refactors the ShellDelegate by adding a SessionStateDelegate to
which methods dealing with the session state can be moved. This cleans up
the huge ShellDelegate interface and paves the way for further Chrome OS
multiprofile work which will need to add several new methods related to
the session state.

This CL is only the first step. Several other methods should also move to
SessionStateDelegate but I do not want to overburden a single CL.

BUG=None
TEST=Manual and browser/unit tests

TBR=sky (for c/b/idle_chromeos.cc and c/chrome_browser_ui.gypi)

Committed: https://src.chromium.org/viewvc/chrome?view=rev&revision=194578

[bartfab (slow), 2013-04-16 17:18:52]

Hi Daniel,

Could you review the ash/ and chrome/browser/ui/ash/ parts of this CL?

Hi Nikita,

Could you review the rest of this CL?

Steven, FYI.

[Daniel Erat, 2013-04-16 17:34:15]

lgtm

https://codereview.chromium.org/14295008/diff/5001/ash/ash.gyp
File ash/ash.gyp (right):

https://codereview.chromium.org/14295008/diff/5001/ash/ash.gyp#newcode653
ash/ash.gyp:653: 'session_state_delegate_stub.cc',
nit: fix alphabetization

https://codereview.chromium.org/14295008/diff/5001/ash/ash.gyp#newcode787
ash/ash.gyp:787: 'session_state_delegate_stub.cc',
here too

https://codereview.chromium.org/14295008/diff/5001/ash/session_state_delegate.h
File ash/session_state_delegate.h (right):

https://codereview.chromium.org/14295008/diff/5001/ash/session_state_delegate...
ash/session_state_delegate.h:32: // Locks the screen.
nit: mention that is possibly asynchronous

https://codereview.chromium.org/14295008/diff/5001/ash/test/test_session_stat...
File ash/test/test_session_state_delegate.cc (right):

https://codereview.chromium.org/14295008/diff/5001/ash/test/test_session_stat...
ash/test/test_session_state_delegate.cc:37: screen_locked_ = true;
does it make sense to CHECK(can_lock_screen_) here?

[stevenjb, 2013-04-16 18:42:43]

lgtm

[Nikita (slow), 2013-04-16 20:00:41]

lgtm

[bartfab (slow), 2013-04-17 08:14:40]

https://chromiumcodereview.appspot.com/14295008/diff/5001/ash/ash.gyp
File ash/ash.gyp (right):

https://chromiumcodereview.appspot.com/14295008/diff/5001/ash/ash.gyp#newcode653
ash/ash.gyp:653: 'session_state_delegate_stub.cc',
On 2013/04/16 17:34:15, Daniel Erat wrote:
> nit: fix alphabetization

Done.

https://chromiumcodereview.appspot.com/14295008/diff/5001/ash/ash.gyp#newcode787
ash/ash.gyp:787: 'session_state_delegate_stub.cc',
On 2013/04/16 17:34:15, Daniel Erat wrote:
> here too

Done.

https://chromiumcodereview.appspot.com/14295008/diff/5001/ash/session_state_d...
File ash/session_state_delegate.h (right):

https://chromiumcodereview.appspot.com/14295008/diff/5001/ash/session_state_d...
ash/session_state_delegate.h:32: // Locks the screen.
On 2013/04/16 17:34:15, Daniel Erat wrote:
> nit: mention that is possibly asynchronous

Done.

https://chromiumcodereview.appspot.com/14295008/diff/5001/ash/test/test_sessi...
File ash/test/test_session_state_delegate.cc (right):

https://chromiumcodereview.appspot.com/14295008/diff/5001/ash/test/test_sessi...
ash/test/test_session_state_delegate.cc:37: screen_locked_ = true;
On 2013/04/16 17:34:15, Daniel Erat wrote:
> does it make sense to CHECK(can_lock_screen_) here?

It's not an error to request a screen lock when screen lock is not possible. The
code should not CHECK - it should just ignore the request. I modified it to this
effect.

[bartfab (slow), 2013-04-17 12:22:06]

Committed patchset #6 manually as r194578 (presubmit successful).