[DGD] call_out question

[Jay Shaffstall]

> > 1) If I use call_out with a delay of 0, the docs say that the function "is
> > to be called as soon as possible after termination of the current
> > thread".  If input from users has been waiting, will it all be processed
> > before the call_out?  Can a call_out be starved if users continue to send
> > input?
>
>I do not know, but the existence of the second question suggests
>to me that the answer to the first is "no".

Another possibility is that user input and call_outs are all queued, and 
happen in the order in which they are received.  So the answer to the first 
could be yes and still avoid call_out starvation.  Given what I've seen of 
DGD so far, I'm assuming that's the case, but I wanted to make certain 
before basing any designs on it.

>More generally, when one is using a chain of call_outs for some
>sort of background maintenance or monitoring, and the need for
>it goes away (other than through the destruction of the object
>in question) then one would remove the call_out to stop the
>chain.

Ah, I get it.  You remove the call_out to *stop* it from 
happening.  Removing a call_out that has already happened would be 
meaningless, then.

Thanks!

Jay


_________________________________________________________________
List config page:  http://list.imaginary.com/mailman/listinfo/dgd