Auto rotate on lid rotation changes.

Auto rotate on lid rotation changes.

Based on patch to read the accelerometer https://codereview.chromium.org/200643005, this listens to accelerometer updates and rotates the screen.

BUG=342907
TEST=MaximizeModeControllerTest.AccelerometerRotation, MaximizeModeControllerTest.AccelerometerIgnoresLowAngles, MaximizeModeControllerTest.AccelerometerSticky, MaximizeModeControllerTest.RotationOnlyInMaximizeMode
TEST=Rotate device, ash should rotate to match device orientation.

Committed: https://src.chromium.org/viewvc/chrome?view=rev&revision=261692

[flackr, 2014-04-01 00:02:36]

Hi, this is dependent on https://codereview.chromium.org/200643005/ which has
not yet landed, however the difference is easily separable. You can ignore
everything except for ash/display/* and the two lines in ash/shell.cc which add
and remove DisplayManager as an observer. Mind taking a look so that when the
other lands this can land sooner? Thanks!

[oshima, 2014-04-01 00:57:02]

FYI: You can upload diffs between commits using git cl upload.
(like this https://codereview.chromium.org/218493010)

https://codereview.chromium.org/196413017/diff/30001/ash/accelerometer/accele...
File ash/accelerometer/accelerometer_controller.cc (right):

https://codereview.chromium.org/196413017/diff/30001/ash/accelerometer/accele...
ash/accelerometer/accelerometer_controller.cc:13: #if defined(OS_CHROMEOS)
What is the reason why we don't want to make this chromeos only?
(Do you want to run test on win ash too or do we have a plan to support this on
win ash?)

https://codereview.chromium.org/196413017/diff/30001/ash/accelerometer/accele...
File ash/accelerometer/accelerometer_observer.h (right):

https://codereview.chromium.org/196413017/diff/30001/ash/accelerometer/accele...
ash/accelerometer/accelerometer_observer.h:16: public:
virtual dtor

https://codereview.chromium.org/196413017/diff/30001/ash/display/display_mana...
File ash/display/display_manager.h (right):

https://codereview.chromium.org/196413017/diff/30001/ash/display/display_mana...
ash/display/display_manager.h:286: const gfx::Vector3dF& lid) OVERRIDE;
Hmm, I don't think this should belong here. (I should have DPES here. it's my
fault)

This class is pretty much independent of most of ash classes and want to keep it
that way.

[flackr, 2014-04-01 01:46:31]

I'm not sure what you used to upload that diff. Can you send me the command?

https://codereview.chromium.org/196413017/diff/30001/ash/accelerometer/accele...
File ash/accelerometer/accelerometer_controller.cc (right):

https://codereview.chromium.org/196413017/diff/30001/ash/accelerometer/accele...
ash/accelerometer/accelerometer_controller.cc:13: #if defined(OS_CHROMEOS)
On 2014/04/01 00:57:03, oshima wrote:
> What is the reason why we don't want to make this chromeos only?
> (Do you want to run test on win ash too or do we have a plan to support this
on
> win ash?)

The code to read the accelerometer is chromeos specific for now. We can easily
update this later to support win ash as well.

https://codereview.chromium.org/196413017/diff/30001/ash/accelerometer/accele...
File ash/accelerometer/accelerometer_observer.h (right):

https://codereview.chromium.org/196413017/diff/30001/ash/accelerometer/accele...
ash/accelerometer/accelerometer_observer.h:16: public:
On 2014/04/01 00:57:03, oshima wrote:
> virtual dtor

Done in original CL.

https://codereview.chromium.org/196413017/diff/30001/ash/display/display_mana...
File ash/display/display_manager.h (right):

https://codereview.chromium.org/196413017/diff/30001/ash/display/display_mana...
ash/display/display_manager.h:286: const gfx::Vector3dF& lid) OVERRIDE;
On 2014/04/01 00:57:03, oshima wrote:
> Hmm, I don't think this should belong here. (I should have DPES here. it's my
> fault)
> 
> This class is pretty much independent of most of ash classes and want to keep
it
> that way.
> 

Okay, so something else to observe accelerometer updates and call into
DisplayManager?

[oshima, 2014-04-01 14:52:50]

> I'm not sure what you used to upload that diff. Can you send me the command?

git cl upload accepts an argument you can pass to git diff.
Here is what I did.

1) apply the 1st patch (git cl patch A)
2) apply the 2nd patch (git cl patch B)
   note: I had to resolve minor conflict
3) git cl upload HEAD^ (or specify commit)

|HEAD^| may vary depending on how you made your local branch though (step 1,2).

https://codereview.chromium.org/196413017/diff/30001/ash/display/display_mana...
File ash/display/display_manager.h (right):

https://codereview.chromium.org/196413017/diff/30001/ash/display/display_mana...
ash/display/display_manager.h:286: const gfx::Vector3dF& lid) OVERRIDE;
On 2014/04/01 01:46:31, flackr wrote:
> On 2014/04/01 00:57:03, oshima wrote:
> > Hmm, I don't think this should belong here. (I should have DPES here. it's
my
> > fault)
> > 
> > This class is pretty much independent of most of ash classes and want to
keep
> it
> > that way.
> > 
> 
> Okay, so something else to observe accelerometer updates and call into
> DisplayManager?

sgtm. I don't have strong preference as to where/which class. May be in a class
that controls TouchView mode?

[flackr, 2014-04-03 02:50:26]

Hey, as suggested I've merged this with the maximize mode detection which I will
eventually rename to touchview to more accurately describe it. I'm trying to
keep the diff minimal to allow for easier merging back. Please take a look,
thanks!

[oshima, 2014-04-03 19:19:48]

lgtm with nits

https://codereview.chromium.org/196413017/diff/70001/ash/wm/maximize_mode/max...
File ash/wm/maximize_mode/maximize_mode_controller.cc (right):

https://codereview.chromium.org/196413017/diff/70001/ash/wm/maximize_mode/max...
ash/wm/maximize_mode/maximize_mode_controller.cc:137: // back to ROTATE_0 if it
was last set by the accelerometer.
SetDisplayRotation will save the setting to the local store.
Can you also add this issue to this TODO (we probably need to store this as
separate pref)

https://codereview.chromium.org/196413017/diff/70001/ash/wm/maximize_mode/max...
File ash/wm/maximize_mode/maximize_mode_controller_unittest.cc (right):

https://codereview.chromium.org/196413017/diff/70001/ash/wm/maximize_mode/max...
ash/wm/maximize_mode/maximize_mode_controller_unittest.cc:34:
GetDisplayAt(0).id());
There is DisplayManagerTestApi::SetFirstDisplayAsInternalDisplay()

[flackr, 2014-04-03 19:43:46]

https://codereview.chromium.org/196413017/diff/70001/ash/wm/maximize_mode/max...
File ash/wm/maximize_mode/maximize_mode_controller.cc (right):

https://codereview.chromium.org/196413017/diff/70001/ash/wm/maximize_mode/max...
ash/wm/maximize_mode/maximize_mode_controller.cc:137: // back to ROTATE_0 if it
was last set by the accelerometer.
On 2014/04/03 19:19:49, oshima wrote:
> SetDisplayRotation will save the setting to the local store.
> Can you also add this issue to this TODO (we probably need to store this as
> separate pref)

Done.

https://codereview.chromium.org/196413017/diff/70001/ash/wm/maximize_mode/max...
File ash/wm/maximize_mode/maximize_mode_controller_unittest.cc (right):

https://codereview.chromium.org/196413017/diff/70001/ash/wm/maximize_mode/max...
ash/wm/maximize_mode/maximize_mode_controller_unittest.cc:34:
GetDisplayAt(0).id());
On 2014/04/03 19:19:49, oshima wrote:
> There is DisplayManagerTestApi::SetFirstDisplayAsInternalDisplay()
> 

Done.

[flackr, 2014-04-04 03:47:19]

The CQ bit was checked by flackr@chromium.org

[commit-bot: I haz the power, 2014-04-04 03:47:28]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/flackr@chromium.org/196413017/130001

[commit-bot: I haz the power, 2014-04-04 06:26:08]

Change committed as 261692