[nas] another patch for auplay

[Paul Fox]

 > > 
 > > these patches were driven by my home voice-response menuing system. 
 > 
 > 	Sounds interesting... what do you use for voice recognition?  

ah, sorry.  didn't mean to mislead.  the input is via IR remote, using
the lirc package, and a couple of random remote controllers.

 > > the other option, "-s", didn't go in quite as cleanly, i'm afraid, but
 > > i found it was equally important in my application.  in order for new
 > 
 > 	This one does too, but will be a little tricky... IMHO, I think
 > auplay should always stop the flow immediately when it's interrupted.  
 > The issue now is how to amend the API in such a way that you can get
 > access to the flowid from AuSoundPlaySynchronousFromFile() (and it's
 > asynchronous cousin.)

yes, that was my problem.  :-)

 >   Or perhaps auplay should just be enhanced to always
 > use it's own playFile routine using low-level calls.  This would also make
 > it easier to add another often requested feature - the ability to detect
 > and play audio from stdin...  Thoughts?

doesn't it do this now?  the code certainly tries -- if there are no
filename args, it tries to play the file called "-".  i assumed the next
lower level did the right thing, but i didn't actually go and look.  :-)

btw, as i was putting together that patch, i noticed that the signal
handler should probably, as is done when falling out of main(), call
AuCloseServer() before calling exit.  but doing that broke my setup
badly, and i'm not quite sure why.  somehow actually telling the server
that you're "going away" causes something to misbehave in a way that's
different than if the client simply drops its end of the connection.  i
confess i didn't investigate as well as i should have.

paul
=---------------------
  paul fox, pgf at foxharp.boston.ma.us (arlington, ma, where it's 30.0 degrees)