Use RTCAudioSessionDelegate in AudioDeviceIOS.

Use RTCAudioSessionDelegateAdapter in AudioDeviceIOS.

Part 3 of refactor. Also:
- better weak pointer delegate storage + tests
- we now ignore route changes when we're interrupted
- fixed bug where preferred sample rate wasn't set if audio session
   wasn't active

BUG=

Committed: https://crrev.com/e54467f73e2630bbaf751a4db4326d1278f704e5
Cr-Commit-Position: refs/heads/master@{#12007}

[tkchin_webrtc, 2016-03-15 03:34:49]

Patchset #1 (id:1) has been deleted

[tkchin_webrtc, 2016-03-15 03:34:55]

Patchset #1 (id:20001) has been deleted

[tkchin_webrtc, 2016-03-15 03:35:28]

Description was changed from

==========
Use RTCAudioSessionDelegate in AudioDeviceIOS.

BUG=
==========

to

==========
Use RTCAudioSessionDelegate in AudioDeviceIOS.

Updated the delegate storage to use weak pointers better.

BUG=
==========

[tkchin_webrtc, 2016-03-15 03:58:19]

Description was changed from

==========
Use RTCAudioSessionDelegate in AudioDeviceIOS.

Updated the delegate storage to use weak pointers better.

BUG=
==========

to

==========
Use RTCAudioSessionDelegate in AudioDeviceIOS.

Part 3 of refactor. Also:
- better weak pointer delegate storage + tests
- we now ignore route changes when we're interrupted
- fixed bug where preferred sample rate wasn't set if audio session 
   wasn't active

BUG=
==========

[tkchin_webrtc, 2016-03-15 03:58:19]

tkchin@webrtc.org changed reviewers:
+ haysc@webrtc.org, henrika@webrtc.org

[tkchin_webrtc, 2016-03-15 03:59:11]

Patchset #1 (id:40001) has been deleted

[tkchin_webrtc, 2016-03-15 04:00:45]

Description was changed from

==========
Use RTCAudioSessionDelegate in AudioDeviceIOS.

Part 3 of refactor. Also:
- better weak pointer delegate storage + tests
- we now ignore route changes when we're interrupted
- fixed bug where preferred sample rate wasn't set if audio session 
   wasn't active

BUG=
==========

to

==========
Use RTCAudioSessionDelegateAdapter in AudioDeviceIOS.

Part 3 of refactor. Also:
- better weak pointer delegate storage + tests
- we now ignore route changes when we're interrupted
- fixed bug where preferred sample rate wasn't set if audio session 
   wasn't active

BUG=
==========

[henrika_webrtc, 2016-03-15 08:53:45]

Some minor comments. 

LGTM

Unable to comment on all of the most fancy Objective-C parts in this CL.

https://codereview.webrtc.org/1796983004/diff/60001/webrtc/modules/audio_devi...
File webrtc/modules/audio_device/ios/audio_device_ios.h (right):

https://codereview.webrtc.org/1796983004/diff/60001/webrtc/modules/audio_devi...
webrtc/modules/audio_device/ios/audio_device_ios.h:157: // These methods should
be called in response to audio events.
Any restrictions on the calling thread?

https://codereview.webrtc.org/1796983004/diff/60001/webrtc/modules/audio_devi...
File webrtc/modules/audio_device/ios/audio_device_ios.mm (right):

https://codereview.webrtc.org/1796983004/diff/60001/webrtc/modules/audio_devi...
webrtc/modules/audio_device/ios/audio_device_ios.mm:122: audio_session_observer_
=
Nice!

https://codereview.webrtc.org/1796983004/diff/60001/webrtc/modules/audio_devi...
webrtc/modules/audio_device/ios/audio_device_ios.mm:369: RTCLog("Session sample
rate: %f", session_sample_rate);
Thanks for adding these logs. Very useful when external users report issues.

https://codereview.webrtc.org/1796983004/diff/60001/webrtc/modules/audio_devi...
File webrtc/modules/audio_device/ios/objc/RTCAudioSessionDelegateAdapter.h
(right):

https://codereview.webrtc.org/1796983004/diff/60001/webrtc/modules/audio_devi...
webrtc/modules/audio_device/ios/objc/RTCAudioSessionDelegateAdapter.h:29: /**
Audio device and callback thread are raw pointers and should be kept alive
Are there any restrictions on the thread here in our design? Could it be any?
Should prio be high, low etc.?
Just curious if any details should be added as comment here.

https://codereview.webrtc.org/1796983004/diff/60001/webrtc/modules/audio_devi...
File webrtc/modules/audio_device/ios/objc/RTCAudioSessionTest.mm (right):

https://codereview.webrtc.org/1796983004/diff/60001/webrtc/modules/audio_devi...
webrtc/modules/audio_device/ios/objc/RTCAudioSessionTest.mm:122: -
(void)testZeroingWeakDelegate {
Please add a comment here. Not trivial to figure out what you test here.

https://codereview.webrtc.org/1796983004/diff/60001/webrtc/modules/audio_devi...
webrtc/modules/audio_device/ios/objc/RTCAudioSessionTest.mm:156:
TEST_F(AudioSessionTest, AddAndRemovekDelegates) {
Did you mean "Removek"?

[Chuck, 2016-03-15 14:33:21]

https://codereview.webrtc.org/1796983004/diff/60001/webrtc/modules/audio_devi...
File webrtc/modules/audio_device/ios/audio_device_ios.mm (right):

https://codereview.webrtc.org/1796983004/diff/60001/webrtc/modules/audio_devi...
webrtc/modules/audio_device/ios/audio_device_ios.mm:352:
LOG_IF_ERROR(AudioOutputUnitStart(vpio_unit_),
Do you want a similar log line to 344 here?

https://codereview.webrtc.org/1796983004/diff/60001/webrtc/modules/audio_devi...
webrtc/modules/audio_device/ios/audio_device_ios.mm:369: RTCLog("Session sample
rate: %f", session_sample_rate);
Is this descriptive enough to know what happened from the log?

https://codereview.webrtc.org/1796983004/diff/60001/webrtc/modules/audio_devi...
File webrtc/modules/audio_device/ios/objc/RTCAudioSession.mm (right):

https://codereview.webrtc.org/1796983004/diff/60001/webrtc/modules/audio_devi...
webrtc/modules/audio_device/ios/objc/RTCAudioSession.mm:37: @synthesize
delegates = _delegates;
Where does _delegates get initialized?

https://codereview.webrtc.org/1796983004/diff/60001/webrtc/modules/audio_devi...
webrtc/modules/audio_device/ios/objc/RTCAudioSession.mm:474: -
(std::vector<__weak id<RTCAudioSessionDelegate> >)delegates {
Does the set for this property need to be synchronized as well?

[tkchin_webrtc, 2016-03-15 20:14:59]

PTAL

https://codereview.webrtc.org/1796983004/diff/60001/webrtc/modules/audio_devi...
File webrtc/modules/audio_device/ios/audio_device_ios.h (right):

https://codereview.webrtc.org/1796983004/diff/60001/webrtc/modules/audio_devi...
webrtc/modules/audio_device/ios/audio_device_ios.h:157: // These methods should
be called in response to audio events.
On 2016/03/15 08:53:44, henrika_webrtc wrote:
> Any restrictions on the calling thread?

Done.

https://codereview.webrtc.org/1796983004/diff/60001/webrtc/modules/audio_devi...
File webrtc/modules/audio_device/ios/audio_device_ios.mm (right):

https://codereview.webrtc.org/1796983004/diff/60001/webrtc/modules/audio_devi...
webrtc/modules/audio_device/ios/audio_device_ios.mm:122: audio_session_observer_
=
On 2016/03/15 08:53:44, henrika_webrtc wrote:
> Nice!

Acknowledged.

https://codereview.webrtc.org/1796983004/diff/60001/webrtc/modules/audio_devi...
webrtc/modules/audio_device/ios/audio_device_ios.mm:352:
LOG_IF_ERROR(AudioOutputUnitStart(vpio_unit_),
On 2016/03/15 14:33:21, Chuck wrote:
> Do you want a similar log line to 344 here?

Done.

https://codereview.webrtc.org/1796983004/diff/60001/webrtc/modules/audio_devi...
webrtc/modules/audio_device/ios/audio_device_ios.mm:369: RTCLog("Session sample
rate: %f", session_sample_rate);
On 2016/03/15 14:33:21, Chuck wrote:
> Is this descriptive enough to know what happened from the log?

This was there before already actually :) Made it more descriptive.

https://codereview.webrtc.org/1796983004/diff/60001/webrtc/modules/audio_devi...
File webrtc/modules/audio_device/ios/objc/RTCAudioSession.mm (right):

https://codereview.webrtc.org/1796983004/diff/60001/webrtc/modules/audio_devi...
webrtc/modules/audio_device/ios/objc/RTCAudioSession.mm:37: @synthesize
delegates = _delegates;
On 2016/03/15 14:33:21, Chuck wrote:
> Where does _delegates get initialized?

It's on the stack so it's ctor gets called automatically.

https://codereview.webrtc.org/1796983004/diff/60001/webrtc/modules/audio_devi...
webrtc/modules/audio_device/ios/objc/RTCAudioSession.mm:474: -
(std::vector<__weak id<RTCAudioSessionDelegate> >)delegates {
On 2016/03/15 14:33:21, Chuck wrote:
> Does the set for this property need to be synchronized as well?

Thanks for catching that. No setter.

https://codereview.webrtc.org/1796983004/diff/60001/webrtc/modules/audio_devi...
File webrtc/modules/audio_device/ios/objc/RTCAudioSessionDelegateAdapter.h
(right):

https://codereview.webrtc.org/1796983004/diff/60001/webrtc/modules/audio_devi...
webrtc/modules/audio_device/ios/objc/RTCAudioSessionDelegateAdapter.h:29: /**
Audio device and callback thread are raw pointers and should be kept alive
On 2016/03/15 08:53:44, henrika_webrtc wrote:
> Are there any restrictions on the thread here in our design? Could it be any?
> Should prio be high, low etc.?
> Just curious if any details should be added as comment here.

Good pt. Edited interfaces a bit.

https://codereview.webrtc.org/1796983004/diff/60001/webrtc/modules/audio_devi...
File webrtc/modules/audio_device/ios/objc/RTCAudioSessionTest.mm (right):

https://codereview.webrtc.org/1796983004/diff/60001/webrtc/modules/audio_devi...
webrtc/modules/audio_device/ios/objc/RTCAudioSessionTest.mm:122: -
(void)testZeroingWeakDelegate {
On 2016/03/15 08:53:44, henrika_webrtc wrote:
> Please add a comment here. Not trivial to figure out what you test here.

Done.

https://codereview.webrtc.org/1796983004/diff/60001/webrtc/modules/audio_devi...
webrtc/modules/audio_device/ios/objc/RTCAudioSessionTest.mm:156:
TEST_F(AudioSessionTest, AddAndRemovekDelegates) {
On 2016/03/15 08:53:44, henrika_webrtc wrote:
> Did you mean "Removek"?

Done.

[Chuck, 2016-03-15 20:55:49]

lgtm

https://codereview.webrtc.org/1796983004/diff/100001/webrtc/modules/audio_dev...
File webrtc/modules/audio_device/ios/audio_device_ios.mm (right):

https://codereview.webrtc.org/1796983004/diff/100001/webrtc/modules/audio_dev...
webrtc/modules/audio_device/ios/audio_device_ios.mm:382: 
nit: extra line

[tkchin_webrtc, 2016-03-15 21:10:54]

Made some interface changes to address your threading comment Henrik. Let me
know if you want me to change anything.

https://codereview.webrtc.org/1796983004/diff/100001/webrtc/modules/audio_dev...
File webrtc/modules/audio_device/ios/audio_device_ios.mm (right):

https://codereview.webrtc.org/1796983004/diff/100001/webrtc/modules/audio_dev...
webrtc/modules/audio_device/ios/audio_device_ios.mm:382: 
On 2016/03/15 20:55:49, Chuck wrote:
> nit: extra line

Done.

[tkchin_webrtc, 2016-03-15 21:11:00]

The CQ bit was checked by tkchin@webrtc.org

[tkchin_webrtc, 2016-03-15 21:11:00]

The patchset sent to the CQ was uploaded after l-g-t-m from henrika@webrtc.org,
haysc@webrtc.org
Link to the patchset: https://codereview.webrtc.org/1796983004/#ps120001 (title:
"Some nits.")

[commit-bot: I haz the power, 2016-03-15 21:11:04]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1796983004/120001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1796983004/120001

[commit-bot: I haz the power, 2016-03-15 23:54:06]

Description was changed from

==========
Use RTCAudioSessionDelegateAdapter in AudioDeviceIOS.

Part 3 of refactor. Also:
- better weak pointer delegate storage + tests
- we now ignore route changes when we're interrupted
- fixed bug where preferred sample rate wasn't set if audio session 
   wasn't active

BUG=
==========

to

==========
Use RTCAudioSessionDelegateAdapter in AudioDeviceIOS.

Part 3 of refactor. Also:
- better weak pointer delegate storage + tests
- we now ignore route changes when we're interrupted
- fixed bug where preferred sample rate wasn't set if audio session 
   wasn't active

BUG=
==========

[commit-bot: I haz the power, 2016-03-15 23:54:08]

Committed patchset #4 (id:120001)

[commit-bot: I haz the power, 2016-03-15 23:54:18]

Description was changed from

==========
Use RTCAudioSessionDelegateAdapter in AudioDeviceIOS.

Part 3 of refactor. Also:
- better weak pointer delegate storage + tests
- we now ignore route changes when we're interrupted
- fixed bug where preferred sample rate wasn't set if audio session 
   wasn't active

BUG=
==========

to

==========
Use RTCAudioSessionDelegateAdapter in AudioDeviceIOS.

Part 3 of refactor. Also:
- better weak pointer delegate storage + tests
- we now ignore route changes when we're interrupted
- fixed bug where preferred sample rate wasn't set if audio session
   wasn't active

BUG=

Committed: https://crrev.com/e54467f73e2630bbaf751a4db4326d1278f704e5
Cr-Commit-Position: refs/heads/master@{#12007}
==========

[commit-bot: I haz the power, 2016-03-15 23:54:19]

Patchset 4 (id:??) landed as
https://crrev.com/e54467f73e2630bbaf751a4db4326d1278f704e5
Cr-Commit-Position: refs/heads/master@{#12007}