[pulseaudio-discuss] Detecting when data source disappears?

[Steven Wawryk]

On 28/05/17 21:30, Georg Chini wrote:
> On 23.05.2017 03:44, Steven Wawryk wrote:
>> Sorry about reposting.  I botched up the subject line, messing up the
>> threading.
>>
>>>>>> I've been reading up and experimenting on both the simple and async
>>>>>> APIs.  In one experiment, I used a CLI file that sets up a set of
>>>>>> module-sine modules with output remapped by module-remap-sink modules
>>>>>> to a stream fed to a module-null-sink module.
>>> Could you please include the command sequence you are using? Maybe I can
>>> reproduce the problem here.
>> I've attached the CLI file and the shell script I use to run it. The
>> sequence is:
>>
>> ./restart-pulse.sh dist/data/sine-2x6.pa
>> parec --device="test_sink0.monitor" --format=ulaw --rate=8000
>> --channels=6 --channel-map="aux0,aux1,aux2,aux3,aux4,aux5" parec0.raw
>>
>> Wait 10 seconds, then in another terminal:
>>
>> pactl unload-module module-remap-sink # or pactl unload-module
>> module-sine
>>
>> Wait another 10 seconds, then kill parec with ctrl-c.  I then look at
>> parec0.raw by importing it in audacity (u-law, 6channels, 8000 smp/sec).
> I tested it here. To me it looks like a bug that occurs because of
> re-winding and using
> a monitor source, but I can't put my finger on it yet.
> With your setup, you are running with 2 seconds latency, which is
> probably not what
> you intend. Add --latency-msec=10 or similar to your parec command. This
> should
> mostly hide the bug.

Thanks, Georg.  I set the latency to 10msec as you said and the length 
of the transitional signal flipping has dropped to less than 200msec.