Add a new audio constraint, 'googHotword', that requests the hotword input

Add a new audio constraint, 'googHotword', that requests the hotword input
stream where available.

BUG=403138
Committed: https://crrev.com/ee0205db24387b76fda082b8506f73e70288b320
Cr-Commit-Position: refs/heads/master@{#310171}

[Anand Mistry (off Chromium), 2015-01-02 00:46:25]

amistry@chromium.org changed reviewers:
+ dalecurtis@chromium.org, rlp@chromium.org

[DaleCurtis, 2015-01-05 22:57:53]

https://codereview.chromium.org/816193003/diff/20001/content/browser/renderer...
File content/browser/renderer_host/media/media_stream_manager.cc (right):

https://codereview.chromium.org/816193003/diff/20001/content/browser/renderer...
content/browser/renderer_host/media/media_stream_manager.cc:146:
chromeos::CrasAudioHandler::Get()->GetAudioDevices(&devices);
Is this something you want to check every time? Device enumeration can be slow.

[rpetterson, 2015-01-06 00:14:48]

I defer to Dale on this CL but have no issues with it.

[Anand Mistry (off Chromium), 2015-01-06 02:02:40]

https://codereview.chromium.org/816193003/diff/20001/content/browser/renderer...
File content/browser/renderer_host/media/media_stream_manager.cc (right):

https://codereview.chromium.org/816193003/diff/20001/content/browser/renderer...
content/browser/renderer_host/media/media_stream_manager.cc:146:
chromeos::CrasAudioHandler::Get()->GetAudioDevices(&devices);
On 2015/01/05 22:57:52, DaleCurtis wrote:
> Is this something you want to check every time? Device enumeration can be
slow.

As far as I can tell, this is a non-blocking operation (just a map iteration).
It should only be an issue if the list of devices is extremely large.

[DaleCurtis, 2015-01-06 02:37:18]

Great, I see that now. Can you add a unittest to MediaStreamConstraintsUtilTest
as well as any other that might be applicable?

[Anand Mistry (off Chromium), 2015-01-06 06:35:38]

On 2015/01/06 02:37:18, DaleCurtis wrote:
> Great, I see that now. Can you add a unittest to
MediaStreamConstraintsUtilTest
> as well as any other that might be applicable?

I found one relevant test and extended it, but I couldn't find anything similar.
None of the other options are unit tested to a great extent.

[DaleCurtis, 2015-01-06 19:04:26]

lgtm

[Anand Mistry (off Chromium), 2015-01-06 22:42:39]

The CQ bit was checked by amistry@chromium.org

[commit-bot: I haz the power, 2015-01-06 22:43:58]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/816193003/60001

[commit-bot: I haz the power, 2015-01-06 23:36:16]

Committed patchset #4 (id:60001)

[commit-bot: I haz the power, 2015-01-06 23:36:58]

Patchset 4 (id:??) landed as
https://crrev.com/ee0205db24387b76fda082b8506f73e70288b320
Cr-Commit-Position: refs/heads/master@{#310171}