Migrate AudioFocusChangeListener to browser and always send GAIN

Migrate AudioFocusChangeListener to browser and always send GAIN

When each player has its own AudioFocusChangeListener, they must all
request AUDIOFOCUS_GAIN_TRANSIENT_MAY_DUCK and not react to
AUDIOFOCUS_LOSS_TRANSIENT_CAN_DUCK so that they don't affect each
other. However, DUCK is intended to be used for temporary audio focus
and can result in overlapping audio between the player and an external
application.

In order to signal external applications to stop playing, GAIN must be
sent. By migrating the AudioFocusChangeListener to the browser level,
GAIN can be sent on behalf of all players and any type of
AUDIOFOCUS_LOSS can interrupt all players.

BUG=429192
TEST=Run ContentShell instrumentation test:
./build/android/test_runner.py instrumentation \
  --test-apk=ContentShellTest --isolate-file-path \
  content/content_shell_test_apk.isolate -f "AudioFocusManagerTest*"

[demarem, 2015-05-07 21:13:54]

demarem@amazon.com changed reviewers:
+ boliu@chromium.org, qinmin@chromium.org, tedchoc@chromium.org

[demarem, 2015-05-07 21:13:54]

demarem@amazon.com changed required reviewers:
+ qinmin@chromium.org

[demarem, 2015-05-07 21:13:55]

tedchoc, please review as owner of:
- content/browser/android/*
- content/public/android/*
- content/public/test/android/*

qinmin, we communicated via email on this change. Please review as original
architect of AudioFocus and owner of:
- content/browser/media/android/*
- media/base/android/*

boliu, please review as owner of:
- android_webview/* 

Thanks everyone!

[boliu, 2015-05-07 21:19:47]

glanced at some bits I thought might be important for webview

https://codereview.chromium.org/1131793004/diff/1/content/public/android/java...
File
content/public/android/java/src/org/chromium/content/browser/AudioFocusManager.java
(right):

https://codereview.chromium.org/1131793004/diff/1/content/public/android/java...
content/public/android/java/src/org/chromium/content/browser/AudioFocusManager.java:33:
mContext = context;
.getApplicationContext() to avoid leaking Activities in webview

https://codereview.chromium.org/1131793004/diff/1/content/public/test/android...
File
content/public/test/android/javatests/src/org/chromium/content/browser/test/util/DOMUtils.java
(right):

https://codereview.chromium.org/1131793004/diff/1/content/public/test/android...
content/public/test/android/javatests/src/org/chromium/content/browser/test/util/DOMUtils.java:24:
public enum PlaybackState {PLAYING, PAUSED}
prefer static final ints over enums (due to android jvm sucks at enums and uses
more memory)

[demarem, 2015-05-08 00:00:40]

Thanks boliu, I made those changes.

https://codereview.chromium.org/1131793004/diff/1/content/public/android/java...
File
content/public/android/java/src/org/chromium/content/browser/AudioFocusManager.java
(right):

https://codereview.chromium.org/1131793004/diff/1/content/public/android/java...
content/public/android/java/src/org/chromium/content/browser/AudioFocusManager.java:33:
mContext = context;
On 2015/05/07 21:19:47, boliu wrote:
> .getApplicationContext() to avoid leaking Activities in webview

Done.

https://codereview.chromium.org/1131793004/diff/1/content/public/test/android...
File
content/public/test/android/javatests/src/org/chromium/content/browser/test/util/DOMUtils.java
(right):

https://codereview.chromium.org/1131793004/diff/1/content/public/test/android...
content/public/test/android/javatests/src/org/chromium/content/browser/test/util/DOMUtils.java:24:
public enum PlaybackState {PLAYING, PAUSED}
On 2015/05/07 21:19:47, boliu wrote:
> prefer static final ints over enums (due to android jvm sucks at enums and
uses
> more memory)

Done.

[boliu, 2015-05-08 01:16:37]

android_webview rubberstamp lgtm

[qinmin, 2015-05-08 19:05:19]

qinmin@chromium.org changed reviewers:
+ avayvod@chromium.org, mlamouri@chromium.org, philipj@opera.com

[qinmin, 2015-05-08 19:05:20]

Adding a couple folks that are working on MediaFocus

[whywhat, 2015-05-12 20:18:04]

Thanks for your patch! Mounir has just sent an Intent to Implement [1] to
chromium-dev which would fix the issue. Please, see if it fixes the bug (and
perhaps makes this change redundant, sorry). Feel free to provide feedback in
the group.

[1]
https://groups.google.com/a/chromium.org/d/topic/chromium-dev/pB2bdML9ELM/dis...

[philipj_slow, 2015-05-26 09:41:28]

Any progress on this? Needless to say I'm supportive, but I'm not an owner here.

[mlamouri (slow - plz ping), 2015-05-26 09:48:09]

On 2015/05/26 at 09:41:28, philipj wrote:
> Any progress on this? Needless to say I'm supportive, but I'm not an owner
here.

I believe that https://codereview.chromium.org/1110833004 will fix the issue
this CL tries to fix. Is there any reason we should try to land this instead/in
addition of https://codereview.chromium.org/1110833004?

[philipj_slow, 2015-05-26 10:51:27]

On 2015/05/26 09:48:09, Mounir Lamouri wrote:
> On 2015/05/26 at 09:41:28, philipj wrote:
> > Any progress on this? Needless to say I'm supportive, but I'm not an owner
> here.
> 
> I believe that https://codereview.chromium.org/1110833004 will fix the issue
> this CL tries to fix. Is there any reason we should try to land this
instead/in
> addition of https://codereview.chromium.org/1110833004

Oops, I found the wrong CL by searching for your username, didn't pay attention
to which one.

[mlamouri (slow - plz ping), 2015-06-01 11:43:42]

On 2015/05/26 at 10:51:27, philipj wrote:
> On 2015/05/26 09:48:09, Mounir Lamouri wrote:
> > On 2015/05/26 at 09:41:28, philipj wrote:
> > > Any progress on this? Needless to say I'm supportive, but I'm not an owner
> > here.
> > 
> > I believe that https://codereview.chromium.org/1110833004 will fix the issue
> > this CL tries to fix. Is there any reason we should try to land this
instead/in
> > addition of https://codereview.chromium.org/1110833004
> 
> Oops, I found the wrong CL by searching for your username, didn't pay
attention to which one.

I think this CL has been deprecated by 
https://codereview.chromium.org/1110833004

[demarem, 2015-06-01 13:07:19]

On 2015/06/01 at 11:43:42, mlamouri wrote:
> On 2015/05/26 at 10:51:27, philipj wrote:
> > On 2015/05/26 09:48:09, Mounir Lamouri wrote:
> > > On 2015/05/26 at 09:41:28, philipj wrote:
> > > > Any progress on this? Needless to say I'm supportive, but I'm not an
owner
> > > here.
> > > 
> > > I believe that https://codereview.chromium.org/1110833004 will fix the
issue
> > > this CL tries to fix. Is there any reason we should try to land this
instead/in
> > > addition of https://codereview.chromium.org/1110833004
> > 
> > Oops, I found the wrong CL by searching for your username, didn't pay
attention to which one.
> 
> I think this CL has been deprecated by 
https://codereview.chromium.org/1110833004

I agree, now that that one has landed I have closed this one.