Path: utzoo!utgpu!jarvis.csri.toronto.edu!cs.utexas.edu!usc!apple!tecot
From: tecot@Apple.COM (Ed Tecot)
Newsgroups: comp.sys.mac.programmer
Subject: Re: Death of jGNEFilter?
Message-ID: <36871@apple.Apple.COM>
Date: 30 Nov 89 03:22:01 GMT
References: <2323@hudson.acc.virginia.edu> <9068@hoptoad.uucp>
Organization: Apple Computer Inc, Cupertino, CA
Lines: 29

In article <9068@hoptoad.uucp> tim@hoptoad.UUCP (Tim Maroney) writes:
>In article <2323@hudson.acc.virginia.edu> cak3g@astsun8.astro.Virginia.EDU
>(Colin Klipsch) writes:
>>I just read a recent article in this group which suggested that
>>the days of jGNEfilter may be numbered.  (This is a low memory
>>global which contains a pointer to an event filter routine.)
>
>The days of all low-memory globals are supposedly numbered, presumably
>to reduce context switching overhead in future releases of the
>operating system.

Or to support (gasp) protected memory in the future without putting the MMU
into a tailspin.

>>If I wish to remain compatible with future Macs (as I do in fact), how
>>does one respond to events without using jGNEFilter?  Will there be some
>>new method?  Is there a more subtle way of responding to events of
>>which I am not aware?
>
>Sure, just tail patch GetNextEvent!  Seriously, the answer to your
>question is that there is no way to do this kind of thing and be
>compatible with the guidelines.  You can either tail patch, which may
>break now or soon, or use jGNEfilter, which may break in 1991.  I'd go
>for the latter, that is, keep doing what you're doing now.

Much better is to head patch SystemEvent, which gets a chance to filter every
event which goes by.

						_emt