Events not executing until lock/unlock performed about omnidroid HOT 12 OPEN

so you're saying when you plugged in the charger, screen was NOT locked but 
brightness still wasn't set. does this happen if you have only one rule? 

yes. I start the test with the screen unlocked. Plug in the charger. The screen 
brightness is not changed until after i click the lock button on the phone. 
When i do this the screen is turned of, by design, then i click the button on 
the phone again to unlock and the unlock screen comes up at the correct 
brightness. Currently i only have this rule and the reduce brightness rule 
running. 

Confirmed defect using "Power connected"->"Change Screen Brightness".  
Omnidroid said it did it in the logs, but no brightness actually changed.  Once 
I locked the screen though and then went to unlock it, the screen was dimmed.

It looks like we're changing the brightness by changing the system settings 
(from 
http://code.google.com/p/omnidroid/source/browse/trunk/omnidroid/src/edu/nyu/cs/
omnidroid/app/controller/external/actions/OmniActionService.java):

    int brightness = intent.getIntExtra(SetScreenBrightnessAction.PARAM_BRIGHTNESS, 200);
    Settings.System.putInt(getContentResolver(), Settings.System.SCREEN_BRIGHTNESS, brightness);
    ResultProcessor.process(this, intent, ResultProcessor.RESULT_SUCCESS, null);


Which is good, we're specifying it in the settings, but I'm assuming this isn't 
activated immediately, just only once you go to unlock does it check this.

I found an example on the web that might help us do it immediately 
(http://www.tutorialforandroid.com/2009/01/changing-screen-brightness.html):
    try {
      IHardwareService hardware = IHardwareService.Stub.asInterface(
ServiceManager.getService("hardware"));
      if (hardware != null) {
        hardware.setScreenBacklight(brightness);
      }
    } catch (RemoteException doe) {
    } 

Salome points out this thread about it not being an API call and may change at 
Google's whim:

http://markmail.org/message/zpsh67wetmgims5b#query:+page:1+mid:42x4boeiboyoiyfg+
state:results

If it works though, it seems like the way to do it until a proper API call is 
available.

I don't quite understand the context here. I was just wondering, is it because 
one rule killed another. For example, the conditions of two rules are both met, 
then the adaptation is undeterministic.

No.  It has to do with when android updates the screen brightness based on the 
data stored in the system settings.  It doesn't update the hardware 
immediately.  We're setting the system settings (see my first code block), but 
since it doesn't go into effect immediately the result isn't actually visible 
until android decides to update the hardware (after a lock sequeence).  If we 
use the IHardwareService we can issue the hardware to update immediately.

why are you talking to yourself publicly? do you need a hug?

Sure who couldn't use a hug now and then, but "[email protected]" isn't 
actually me. :-)

[deleted comment]

you guys are funny. I am your group member, just interested in this great 
project :-)

:D:D yeah I see now. Actually I was saying the other day that andrew is the 
most comsci name.

Oh, it's a different person, I also thought that Andrew was talking to himself!