Open jebofponderworthy opened 6 years ago
Thanks for the report!
TBH, I've never used multiple JACK daemons before. I've just tried it and I can reproduce the problem you are reporting.
Sadly, I have no idea what's going wrong here ... do you have any?
Not much at all -- I was honestly hoping you would find something obvious that I had missed in the test code! I might be dimly remembering (or imagining) a two-year-old linux-audio-user conversation in which it was said something like this would occur. If the latter is not an imagination, there's something with the standard JACK primitives which gets in the way of one process attaching itself to multiple JACK processes. For my needs right now, I have worked around it, by creating a new JACK client every time I need to check status. Working workaround is in wait_for_jackport() (not wait_for_jack()) in the below.
https://github.com/jebofponderworthy/the-box-of-no-return/blob/master/jpctrl.py
The following code snippet produces the same list of ports, the correct list for the first ('hard', hardware-connected) JACK server, all four times it does the get_ports() print. The .name print comes out correctly though. Any ideas why this is happening?