Fix AudioPipeReader to handle EOF properly.

Fix AudioPipeReader to handle EOF properly.

PulseAudio is started on demand. When PulseAudio is not started read() alaways
returns 0 (i.e. EOF). Normally pulseaudio is started on demand by libpulse so that
it can be restarted after it's stoped or crashed. Previously AudioPipeReader::DoCapture()
didn't handle properly the case when read() returns 0, which would cause it to get into 
indefinite loop when pulseaudio is not started.

BUG=167801
Committed: https://src.chromium.org/viewvc/chrome?view=rev&revision=174913

[Lambros, 2013-01-02 22:51:28]

lgtm

https://codereview.chromium.org/11744009/diff/1/remoting/host/linux/audio_pip...
File remoting/host/linux/audio_pipe_reader.cc (right):

https://codereview.chromium.org/11744009/diff/1/remoting/host/linux/audio_pip...
remoting/host/linux/audio_pipe_reader.cc:152: PLOG(ERROR) << "read";
optional: PLOG_IF(ERROR, condition) <<..

[Sergey Ulanov, 2013-01-02 23:02:20]

https://codereview.chromium.org/11744009/diff/1/remoting/host/linux/audio_pip...
File remoting/host/linux/audio_pipe_reader.cc (right):

https://codereview.chromium.org/11744009/diff/1/remoting/host/linux/audio_pip...
remoting/host/linux/audio_pipe_reader.cc:152: PLOG(ERROR) << "read";
On 2013/01/02 22:51:28, Lambros wrote:
> optional: PLOG_IF(ERROR, condition) <<..

I don't think PLOG_IF() would make this code more readable or provide any other
benefits.

[commit-bot: I haz the power, 2013-01-02 23:10:11]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/sergeyu@chromium.org/11744009/1

[commit-bot: I haz the power, 2013-01-03 02:11:23]

Change committed as 174913