I had to change the mido.open_xy(port_name) port name from nanoKONTROL2 MIDI 1 to nanoKONTROL2:nanoKONTROL2 nanoKONTROL2 _ CTR 32:0, which i obtained from running mido.get_input_names() in a python3 interpreter.
Otherwise it would crash on startup.
My mind immediatly went to a hacky string operation to dynamically grab the port name on runtime from that get_input_names() output, but i don't know whether that is appropriate.

Traceback (most recent call last):
  File "./nanoKontroller", line 307, in <module>
    inport = mido.open_input('nanoKONTROL2 MIDI 1')
  File "/home/sparr/.local/lib/python2.7/site-packages/mido/backends/backend.py", line 91, in open_input
    return self.module.Input(name, **self._add_api(kwargs))
  File "/home/sparr/.local/lib/python2.7/site-packages/mido/ports.py", line 161, in __init__
    BasePort.__init__(self, name, **kwargs)
  File "/home/sparr/.local/lib/python2.7/site-packages/mido/ports.py", line 86, in __init__
    self._open(**kwargs)
  File "/home/sparr/.local/lib/python2.7/site-packages/mido/backends/rtmidi.py", line 118, in _open
    rtapi = _get_api_id(api)
  File "/home/sparr/.local/lib/python2.7/site-packages/mido/backends/rtmidi.py", line 34, in _get_api_id
    return rtmidi.API_UNSPECIFIED
AttributeError: 'module' object has no attribute 'API_UNSPECIFIED'

not sure where to go from here to collect more info

I spent a bit of time trying to get application-specific volume controls to work with this library, but ran into a roadblock I'm not sure if I can solve, or is even solvable :)

I augmented the get_audio_devices function to also get the sink_input_list from pulsectl and then pipe it through the rest of the flow, including having it inside the config file.
It actually works too! The only issue is the way sink inputs work, possibly. In my case, I wanted it for Google Chrome specifically. Changing Chrome's volume or muting it instead of muting the entire output device. The problem is that the pulsectl library is returning the media.name property as the name from Pulse, which in the case of Chrome is "Playback". Like I mentioned, this does work successfully, however if there are multiple sink inputs with "Playback" as the media.name then it fails. This includes having more than one tab in Chrome playing audio as they're both sink inputs with the name of "Playback".

There may be an easier way to handle this, and I could share my changes if that helps point someone in the right direction for this, or if it's an interesting enough problem to solve for my use case. It definitely seems like the bigger issue is the way the pulsectl library is handling retrieval of PulseObjects as well as how sink inputs are treated, but I would love to collaborate on this, or discuss more if anyone would like to work on it :)

Thanks!

Hi Thomas,

I'm considering picking this device up with a Raspberry Pi. I know it's been a while, but could you give more details on this line?

LED control will not work until the magic control sequence has been sent to the device. Use the Windows control software to switch to this mode.

What do you mean with the magic control sequence for the led control?