Path: utzoo!utgpu!jarvis.csri.toronto.edu!mailrus!uwm.edu!gem.mps.ohio-state.edu!ginosko!uunet!cbmvax!vu-vlsi!swatsun!jackiw
From: jackiw@cs.swarthmore.edu (Nick Jackiw)
Newsgroups: comp.sys.mac.programmer
Subject: Re: VBL or patch traps?
Message-ID: <3178@carthage.cs.swarthmore.edu>
Date: 17 Oct 89 15:02:24 GMT
References: <189@uhura.cs.wisc.edu>
Reply-To: jackiw@carthage (Nick Jackiw)
Distribution: usa
Organization: Visual Geometry Project, Swarthmore College, PA
Lines: 32

In article <189@uhura.cs.wisc.edu> qureshi@uhura (Mohammad Qureshi) writes:
> I have been considering writing an INIT that would, after a set time, present
> a dialog to the user.  This would be a sort of alarm which the user would
> be able to set (using a cdev interface).  My question is, should I implement
> this alarm as a VBL task, checking each time if one of the alarm times has
> been reached, or should I patch WaitNextEvent/GetNextEvent and/or appropriate
> traps to check whether or not the alarm should go off?
> 
> Any assistance will be greatly appreciated.

I'd definitely go with the trap patch, not the VBL.  That way, while you
may take up more or less processor cycles than a VBL (depending on the
frequency of calls to WNE), you're guaranteed to be doing it at a time 
when the foreground application is "looking for something else to do",
as opposed to doing it during the middle of somebody else's processor-
intensive computations.

Furthermore, you've got a *lot* more flexibility in memory access and 
toolbox use in an event-loop trap than you do running at interrupt level.

Lastly, you can take as much time as you like if it's not in a VBL---they
didn't name it "WAITnextEvent" for nothing :-)!

-Nick



-- 
     _  _|\____    Nick Jackiw | Visual Geometry Project | Math Department
   / /_/   O>  \   ------------+-------------------------+ Swarthmore College
   |       O>   |  215-328-8225| jackiw@cs.swarthmore.edu| Swarthmore PA 19081
    \_Guernica_/   ------------+-------------------------+                 USA