[ale] Stupid socket tricks

Fulton Green ale at FultonGreen.com
Mon Dec 18 15:25:04 EST 2000


Though you might have to handle users that have UDP firewalled out of their
corporate environment, thereby necessitating an alternate TCP approach for
those users.

On Mon, Dec 18, 2000 at 03:19:03PM -0500, Glenn C. Lasher Jr. wrote:
> 
> Now THIS looks like the real winner.
> 
> Thanks
> 
> On Mon, 18 Dec 2000 hirsch at zapmedia.com wrote:
> 
> > Glenn C. Lasher Jr. writes:
> >  > 
> >  > I'm looking for a solution to a problem, and I stumbled upon the named
> >  > pipe as a possible solution, but it doesn't behave in the needed manner.
> >  > 
> >  > What I am trying to do is cause data from one process to be distributed to
> >  > multiple processes, some of which may or may not be there, but the data is
> >  > real-time, and any data that was there before a receiving process started
> >  > is not to be delivered to that process.  (the application is an MP3
> >  > streamer)
> > 
> > [snip]
> >  > 
> >  > So what do I need to do differently?  I want to have the MP3's generated
> >  > in realtime and anything for which there is no receiver needs to go to the
> >  > bit bucket in the sky.  
> > 
> > You might try a upd multicast socket.  One process can send to the
> > multicast socket and other processes can listen to it.  If they get
> > the packet they can deal with it, and otherwise it just disappears
> > into the aether.  It worked well for an conferencing internet
> > telephone I once worked on, and it has the added advantage that the
--
To unsubscribe: mail majordomo at ale.org with "unsubscribe ale" in message body.





More information about the Ale mailing list