[nas] pulseaudio integration

[Paul Fox]

i wrote:
 > i'm wondering if the nasd server could be run on top of the
 > pulseaudio shim (i.e., "padsp nasd", as in the previous link). 
 > this would let both nas clients (connecting via nasd) and
 > pulseaudio clients (like firefox, connecting via pulseaudio) both
 > get at local hardware, with the mixing being done in the
 > pulseaudio server.

this seems to work, at least at a gross level.  the devil is in
the details.

when i have nasd running wrapped in padsp, audio that's sent to
the nasd server is clipped -- i lose the last utterance of every
sound file.  i've tried the "keep device open" config option,
thinking that the close of /dev/dsp was causing it, but that
didn't seem to help.  i need to play with that some more, and be
sure i really had the correct version of the config file in place
during the test.  maybe there's a PA config option that can help.

i couldn't get the nas mixer to work, but that may have been
related to other (unrelated) issues that i've since discovered --
i.e. my volume control keyboard hotkeys weren't working after i
stopped testing, either.  so maybe everything was fine.  what i
did notice, though was, that adjusting the volume via the
pulseaudio server directly, bypassing nasd (e.g., with "alsamixer
-Dpulse") worked, but wasn't sticky -- nasd would set it back at
the next access.  i have all the "keep", "don't reset" options in
my nas config, so again, i'm not sure what's going on.


as dave has said a couple of times on the list, i think it's
worth trying to bridge the nas and pulseaudio worlds.  the fact
that pulseaudio supports all alsa-based applications via an alsa
plugin, and has a /dev/dsp wrapper that's clearly more currently
maintained than audiooss, makes it a no-brainer.  i'm hoping that
i can get the bugs out of running nasd on top of pulseaudio, and
end up with a zero-coding transition solution.

paul
=---------------------
 paul fox, pgf at foxharp.boston.ma.us (arlington, ma, where it's 13.1 degrees)