chromeos: More power-related cleanup.

chromeos: More power-related cleanup.

This consolidates SuspendObserver and ResumeObserver in a
PowerEventObserver class within ash/system/chromeos/power
and moves extension-related code into an
ExtensionSystemEventObserver class.

BUG=none

Committed: https://src.chromium.org/viewvc/chrome?view=rev&revision=227855

[Daniel Erat, 2013-10-09 16:01:50]

(It's possible that I'll be able to remove BrightnessObserver completely in this
change; still following up on that.)

[stevenjb, 2013-10-09 18:52:28]

https://codereview.chromium.org/26692002/diff/1/chrome/browser/chromeos/power...
File chrome/browser/chromeos/power/brightness_observer.cc (right):

https://codereview.chromium.org/26692002/diff/1/chrome/browser/chromeos/power...
chrome/browser/chromeos/power/brightness_observer.cc:31:
chrome::AttemptUserExit();
If we do need this, it seems like we should put the logic in PowerEventObserver
(using LoginState::GetLoggedInUserType), and call
Shell::GetInstance()->system_tray_delegate()->SignOut() (which just calls
chrome::AttemptUserExit()).

While you're in there, SystemTrayDelegate::ShutDown() just calls
ash::Shell::GetInstance()->lock_state_controller()->RequestShutdown(), which
seems kind of silly.

[Daniel Erat, 2013-10-09 19:25:29]

https://codereview.chromium.org/26692002/diff/1/chrome/browser/chromeos/power...
File chrome/browser/chromeos/power/brightness_observer.cc (right):

https://codereview.chromium.org/26692002/diff/1/chrome/browser/chromeos/power...
chrome/browser/chromeos/power/brightness_observer.cc:31:
chrome::AttemptUserExit();
On 2013/10/09 18:52:28, stevenjb wrote:
> If we do need this, it seems like we should put the logic in
PowerEventObserver
> (using LoginState::GetLoggedInUserType), and call
> Shell::GetInstance()->system_tray_delegate()->SignOut() (which just calls
> chrome::AttemptUserExit()).

I dunno... I'm still pretty hopeful that this can be deleted now, but if it
can't, it feels like a bit of a stretch to put it in ash. It's very much tied to
public accounts, and I think of those as being a Chrome concept rather than an
Ash concept. 

> While you're in there, SystemTrayDelegate::ShutDown() just calls
> ash::Shell::GetInstance()->lock_state_controller()->RequestShutdown(), which
> seems kind of silly.

Just to make sure I understand: you're proposing that
SystemTrayDelegate::ShutDown() be removed and system/date/tray_date.cc just call
LockStateController::RequestShutdown() instead, right?

[stevenjb, 2013-10-09 20:15:36]

lgtm

https://codereview.chromium.org/26692002/diff/1/chrome/browser/chromeos/power...
File chrome/browser/chromeos/power/brightness_observer.cc (right):

https://codereview.chromium.org/26692002/diff/1/chrome/browser/chromeos/power...
chrome/browser/chromeos/power/brightness_observer.cc:31:
chrome::AttemptUserExit();
On 2013/10/09 19:25:30, Daniel Erat wrote:
> On 2013/10/09 18:52:28, stevenjb wrote:
> > If we do need this, it seems like we should put the logic in
> PowerEventObserver
> > (using LoginState::GetLoggedInUserType), and call
> > Shell::GetInstance()->system_tray_delegate()->SignOut() (which just calls
> > chrome::AttemptUserExit()).
> 
> I dunno... I'm still pretty hopeful that this can be deleted now, but if it
> can't, it feels like a bit of a stretch to put it in ash. It's very much tied
to
> public accounts, and I think of those as being a Chrome concept rather than an
> Ash concept. 
> 
> > While you're in there, SystemTrayDelegate::ShutDown() just calls
> > ash::Shell::GetInstance()->lock_state_controller()->RequestShutdown(), which
> > seems kind of silly.
> 
> Just to make sure I understand: you're proposing that
> SystemTrayDelegate::ShutDown() be removed and system/date/tray_date.cc just
call
> LockStateController::RequestShutdown() instead, right?
1. OK, fair, this is fine here then if we can't remove it.

2. Correct. I'd like to reduce Ash -> Chrome -> Ash calls when there is no
Chrome specific behavior in between. (I'm guessing that this wasn't the case
when first implemented).

[Daniel Erat, 2013-10-09 20:36:39]

https://codereview.chromium.org/26692002/diff/1/chrome/browser/chromeos/power...
File chrome/browser/chromeos/power/brightness_observer.cc (right):

https://codereview.chromium.org/26692002/diff/1/chrome/browser/chromeos/power...
chrome/browser/chromeos/power/brightness_observer.cc:31:
chrome::AttemptUserExit();
On 2013/10/09 20:15:36, stevenjb wrote:
> On 2013/10/09 19:25:30, Daniel Erat wrote:
> > On 2013/10/09 18:52:28, stevenjb wrote:
> > > If we do need this, it seems like we should put the logic in
> > PowerEventObserver
> > > (using LoginState::GetLoggedInUserType), and call
> > > Shell::GetInstance()->system_tray_delegate()->SignOut() (which just calls
> > > chrome::AttemptUserExit()).
> > 
> > I dunno... I'm still pretty hopeful that this can be deleted now, but if it
> > can't, it feels like a bit of a stretch to put it in ash. It's very much
tied
> to
> > public accounts, and I think of those as being a Chrome concept rather than
an
> > Ash concept. 
> > 
> > > While you're in there, SystemTrayDelegate::ShutDown() just calls
> > > ash::Shell::GetInstance()->lock_state_controller()->RequestShutdown(),
which
> > > seems kind of silly.
> > 
> > Just to make sure I understand: you're proposing that
> > SystemTrayDelegate::ShutDown() be removed and system/date/tray_date.cc just
> call
> > LockStateController::RequestShutdown() instead, right?
> 1. OK, fair, this is fine here then if we can't remove it.
> 
> 2. Correct. I'd like to reduce Ash -> Chrome -> Ash calls when there is no
> Chrome specific behavior in between. (I'm guessing that this wasn't the case
> when first implemented).

(Discussed in person: I'm going to clean this up in a following change.)

[commit-bot: I haz the power, 2013-10-09 20:38:32]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/derat@chromium.org/26692002/1

[commit-bot: I haz the power, 2013-10-10 00:53:07]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/derat@chromium.org/26692002/1

[commit-bot: I haz the power, 2013-10-10 01:40:30]

Change committed as 227855