[Chromecast] Implement new volume control API

[Chromecast] Implement new volume control API

BUG= internal b/34454613

Review-Url: https://codereview.chromium.org/2738873002
Cr-Commit-Position: refs/heads/master@{#456937}
Committed: https://chromium.googlesource.com/chromium/src/+/3801f1a0449592bdb25e7155ac1eabf50610f207

[kmackay, 2017-03-08 05:31:33]

kmackay@chromium.org changed reviewers:
+ bshaya@google.com, gfhuang@chromium.org, halliwell@chromium.org,
jyw@chromium.org, slan@chromium.org

[kmackay, 2017-03-10 21:37:13]

The CQ bit was checked by kmackay@chromium.org to run a CQ dry run

[commit-bot: I haz the power, 2017-03-10 21:37:39]

Dry run: CQ is trying da patch. Follow status at
 
https://chromium-cq-status.appspot.com/v2/patch-status/codereview.chromium.or...

[commit-bot: I haz the power, 2017-03-10 22:36:33]

The CQ bit was unchecked by commit-bot@chromium.org

[commit-bot: I haz the power, 2017-03-10 22:36:34]

Dry run: This issue passed the CQ dry run.

[kmackay, 2017-03-13 16:59:29]

Ping for review

[slan, 2017-03-13 22:09:51]

just nits.

https://codereview.chromium.org/2738873002/diff/20001/chromecast/chromecast.gni
File chromecast/chromecast.gni (right):

https://codereview.chromium.org/2738873002/diff/20001/chromecast/chromecast.g...
chromecast/chromecast.gni:32: enable_assistant = false
The assert from internal should be copied at the end of this file:

# Assistant shall only be enabled on audio-only builds.
assert(!enable_assistant || is_cast_audio_only,
       "Assistant is only available on audio devices.")

https://codereview.chromium.org/2738873002/diff/20001/chromecast/media/cma/ba...
File chromecast/media/cma/backend/alsa/stream_mixer_alsa.cc (right):

https://codereview.chromium.org/2738873002/diff/20001/chromecast/media/cma/ba...
chromecast/media/cma/backend/alsa/stream_mixer_alsa.cc:179: for (auto type :
{AudioContentType::kMedia, AudioContentType::kAlarm,
Perhaps we should use a static assert in here to make sure that we don't miss
this when we add another stream type?

Or, it may just be easier to use a map and a new struct:

struct VolumeConfig {
 public:  
  VolumeConfig() : vol(0.0), limit(0.0), muted(false) {}
  float vol, limit;
  bool muted;
}

std::unordered_map<AudioConentType, VolumeConfig> volume_map_;

That way you get this initialization for free whenever you access it:
volume_map_[type].vol = vol;

https://codereview.chromium.org/2738873002/diff/20001/chromecast/media/cma/ba...
chromecast/media/cma/backend/alsa/stream_mixer_alsa.cc:928: float
effective_volume = std::min(volume_limit_[type], level);
Regarding my earlier comment about using a VolumeConfig struct, you could avoid
repeating this line everywhere by implementing it in a GetEffectiveVolume()
method (ex. line 612 above, 962 below)

https://codereview.chromium.org/2738873002/diff/20001/chromecast/media/cma/ba...
File chromecast/media/cma/backend/alsa/stream_mixer_alsa.h (right):

https://codereview.chromium.org/2738873002/diff/20001/chromecast/media/cma/ba...
chromecast/media/cma/backend/alsa/stream_mixer_alsa.h:275:
std::map<AudioContentType, float> volume_;
Why not std::unordered_map?

I think that Chrome actually prefers a vector of pairs over small maps for
speed, but I like the syntax of maps way better.

https://codereview.chromium.org/2738873002/diff/20001/chromecast/media/cma/ba...
File chromecast/media/cma/backend/alsa/stream_mixer_alsa_unittest.cc (right):

https://codereview.chromium.org/2738873002/diff/20001/chromecast/media/cma/ba...
chromecast/media/cma/backend/alsa/stream_mixer_alsa_unittest.cc:178: void
SetMuted(bool muted) override {}
Why not mock?

[kmackay, 2017-03-14 00:21:00]

https://codereview.chromium.org/2738873002/diff/20001/chromecast/chromecast.gni
File chromecast/chromecast.gni (right):

https://codereview.chromium.org/2738873002/diff/20001/chromecast/chromecast.g...
chromecast/chromecast.gni:32: enable_assistant = false
On 2017/03/13 22:09:51, slan wrote:
> The assert from internal should be copied at the end of this file:
> 
> # Assistant shall only be enabled on audio-only builds.
> assert(!enable_assistant || is_cast_audio_only,
>        "Assistant is only available on audio devices.")

Done.

https://codereview.chromium.org/2738873002/diff/20001/chromecast/media/cma/ba...
File chromecast/media/cma/backend/alsa/stream_mixer_alsa.cc (right):

https://codereview.chromium.org/2738873002/diff/20001/chromecast/media/cma/ba...
chromecast/media/cma/backend/alsa/stream_mixer_alsa.cc:179: for (auto type :
{AudioContentType::kMedia, AudioContentType::kAlarm,
On 2017/03/13 22:09:51, slan wrote:
> Perhaps we should use a static assert in here to make sure that we don't miss
> this when we add another stream type?
> 
> Or, it may just be easier to use a map and a new struct:
> 
> struct VolumeConfig {
>  public:  
>   VolumeConfig() : vol(0.0), limit(0.0), muted(false) {}
>   float vol, limit;
>   bool muted;
> }
> 
> std::unordered_map<AudioConentType, VolumeConfig> volume_map_;
> 
> That way you get this initialization for free whenever you access it:
> volume_map_[type].vol = vol;

Done.

https://codereview.chromium.org/2738873002/diff/20001/chromecast/media/cma/ba...
chromecast/media/cma/backend/alsa/stream_mixer_alsa.cc:928: float
effective_volume = std::min(volume_limit_[type], level);
On 2017/03/13 22:09:51, slan wrote:
> Regarding my earlier comment about using a VolumeConfig struct, you could
avoid
> repeating this line everywhere by implementing it in a GetEffectiveVolume()
> method (ex. line 612 above, 962 below)

Done.

https://codereview.chromium.org/2738873002/diff/20001/chromecast/media/cma/ba...
File chromecast/media/cma/backend/alsa/stream_mixer_alsa.h (right):

https://codereview.chromium.org/2738873002/diff/20001/chromecast/media/cma/ba...
chromecast/media/cma/backend/alsa/stream_mixer_alsa.h:275:
std::map<AudioContentType, float> volume_;
On 2017/03/13 22:09:51, slan wrote:
> Why not std::unordered_map?
> 
> I think that Chrome actually prefers a vector of pairs over small maps for
> speed, but I like the syntax of maps way better.

The comment in
https://cs.chromium.org/chromium/src/base/containers/small_map.h?dr=C&l=27
indicates that std::map should be the default. I don't think it matters for
performance reasons since there are only 3 entries, and the callsites are not
performance-critical. Also, to use unordered_map, I would need to define some
hash function for AudioContentType, which is a bit of a pain.

https://codereview.chromium.org/2738873002/diff/20001/chromecast/media/cma/ba...
File chromecast/media/cma/backend/alsa/stream_mixer_alsa_unittest.cc (right):

https://codereview.chromium.org/2738873002/diff/20001/chromecast/media/cma/ba...
chromecast/media/cma/backend/alsa/stream_mixer_alsa_unittest.cc:178: void
SetMuted(bool muted) override {}
On 2017/03/13 22:09:51, slan wrote:
> Why not mock?

We use a StrictMock everywhere, and I didn't feel like adding EXPECT_CALL for
these methods everywhere. Maybe there is a better way.

[slan, 2017-03-14 17:21:28]

lgtm

https://codereview.chromium.org/2738873002/diff/20001/chromecast/media/cma/ba...
File chromecast/media/cma/backend/alsa/stream_mixer_alsa.h (right):

https://codereview.chromium.org/2738873002/diff/20001/chromecast/media/cma/ba...
chromecast/media/cma/backend/alsa/stream_mixer_alsa.h:275:
std::map<AudioContentType, float> volume_;
On 2017/03/14 00:21:00, kmackay wrote:
> On 2017/03/13 22:09:51, slan wrote:
> > Why not std::unordered_map?
> > 
> > I think that Chrome actually prefers a vector of pairs over small maps for
> > speed, but I like the syntax of maps way better.
> 
> The comment in
> https://cs.chromium.org/chromium/src/base/containers/small_map.h?dr=C&l=27
> indicates that std::map should be the default. I don't think it matters for
> performance reasons since there are only 3 entries, and the callsites are not
> performance-critical. Also, to use unordered_map, I would need to define some
> hash function for AudioContentType, which is a bit of a pain.

Acknowledged.

https://codereview.chromium.org/2738873002/diff/20001/chromecast/media/cma/ba...
File chromecast/media/cma/backend/alsa/stream_mixer_alsa_unittest.cc (right):

https://codereview.chromium.org/2738873002/diff/20001/chromecast/media/cma/ba...
chromecast/media/cma/backend/alsa/stream_mixer_alsa_unittest.cc:178: void
SetMuted(bool muted) override {}
On 2017/03/14 00:21:00, kmackay wrote:
> On 2017/03/13 22:09:51, slan wrote:
> > Why not mock?
> 
> We use a StrictMock everywhere, and I didn't feel like adding EXPECT_CALL for
> these methods everywhere. Maybe there is a better way.

Eh, gmock doesn't have good support for this case. We probably shouldn't be
using StrictMock if we don't care about some of the API calls, but oh well.

[kmackay, 2017-03-14 17:31:20]

The CQ bit was checked by kmackay@chromium.org

[commit-bot: I haz the power, 2017-03-14 17:32:39]

The CQ bit was unchecked by commit-bot@chromium.org

[commit-bot: I haz the power, 2017-03-14 17:32:40]

This CL has an open dependency (Issue 2712883006 Patch 200001). Please resolve
the dependency and try again.
If you are sure that there is no real dependency, please use one of the options
listed in https://goo.gl/9Es4OR to land the CL.

[kmackay, 2017-03-14 22:05:10]

The CQ bit was checked by kmackay@chromium.org

[commit-bot: I haz the power, 2017-03-14 22:06:07]

CQ is trying da patch. Follow status at
 
https://chromium-cq-status.appspot.com/v2/patch-status/codereview.chromium.or...

[kmackay, 2017-03-14 23:34:46]

The CQ bit was unchecked by kmackay@chromium.org

[kmackay, 2017-03-14 23:34:47]

The CQ bit was checked by kmackay@chromium.org

[commit-bot: I haz the power, 2017-03-14 23:35:25]

CQ is trying da patch. Follow status at
 
https://chromium-cq-status.appspot.com/v2/patch-status/codereview.chromium.or...

[commit-bot: I haz the power, 2017-03-15 01:24:36]

CQ is committing da patch.

Bot data: {"patchset_id": 40001, "attempt_start_ts": 1489534487186300,
"parent_rev": "3bddfe5d8dd5ca4f1431cbef44d3bbaf4dced3e3", "commit_rev":
"3801f1a0449592bdb25e7155ac1eabf50610f207"}

[commit-bot: I haz the power, 2017-03-15 01:25:24]

Description was changed from

==========
[Chromecast] Implement new volume control API

BUG= internal b/34454613
==========

to

==========
[Chromecast] Implement new volume control API

BUG= internal b/34454613

Review-Url: https://codereview.chromium.org/2738873002
Cr-Commit-Position: refs/heads/master@{#456937}
Committed:
https://chromium.googlesource.com/chromium/src/+/3801f1a0449592bdb25e7155ac1e...
==========

[commit-bot: I haz the power, 2017-03-15 01:25:26]

Committed patchset #3 (id:40001) as
https://chromium.googlesource.com/chromium/src/+/3801f1a0449592bdb25e7155ac1e...