realtime design

Bjorn Pettersen BPettersen at NAREX.com
Tue Oct 15 02:20:36 CEST 2002


> From: Aahz [mailto:aahz at pythoncraft.com] 
> 
> In article <cb035744.0210140629.43767d40 at posting.google.com>,
> Will Stuyvesant <hwlgw at hotmail.com> wrote:
> >
> >I am trying to come up with a method to run a function in such a way 
> >that the whole thing will take at most x milliseconds, with x a 
> >reasonable value like 200.  If the function is not finished by that 
> >time, a default value should be used.  If the function *does* finish
in 
> >time, the return value of the function should be used.
> 
> That last part isn't really possible, not if there are 
> statements that depend on external connectivity like I/O.  
> But it's certainly possible to make your main loop work the 
> way you want: have your function put its result on a 
> Queue.Queue.  In your main function, time.sleep() for however 
> long it's supposed to, then do a non-blocking get() on the 
> queue.  If the get() fails, do whatever's appropriate.  This 
> gets trickier if your main program needs to be responsive 
> during the 200ms, but it's not impossible.

You should be able to re-work the "Futures" receipe
(http://aspn.activestate.com/ASPN/Cookbook/Python/Recipe/84317) into
doing what you want... (basically set a timestamp in __init__ and
conditionally return a default value in __call__).

-- bjorn




More information about the Python-list mailing list