[BRLTTY] strange behavior udev and systemd

[Sébastien Hinderer]

Dave Mielke (2020/11/26 10:54 -0500):
> >There is no way to automatically determine what the "right" BRLTTY
> >instance would be.  It depends on what the user wants.
> 
> This is something I've been thinking about from time to time. What we really
> need is for a given USB device to be associated with a gien brltty.conf file.
> Perhaps this could be inferred by the driver that's been specified via the udev
> rules (i.e. via $BRLTTY_BRAILLE_DRIVER).

When I read Mario's message, I started to think and wonder whether we
couldn't somehow declare that a given bluetooth address and a USB id do
actually refer to the same device. would it then be possible that a
single brltty instance would keep running when, say, the device is
disconnectedfrom USB and if, later, a bluetooth device is detected whose
adress corresponds to the USB identifier, then the same instance of
brltty would take control of that bluetooth connection?

I fear this owuld make things difficult from a BRLTTY pointof view. From
an application point of view, though, it feels to me it would solve
theproblmees discussed in this thread.

Sébastien.