Path: utzoo!utgpu!jarvis.csri.toronto.edu!mailrus!nrl-cmf!ukma!rutgers!bpa!cbmvax!uunet!mcvax!cernvax!ethz!ceb
From: ceb@ethz.UUCP (Charles Buckley)
Newsgroups: comp.emacs
Subject: Re: ARGH!  Ahem.  Umm, answering prompts & abbreviating
Message-ID: <781@ethz.UUCP>
Date: 13 Feb 89 19:18:54 GMT
References: <17901@vax5.CIT.CORNELL.EDU> <89Feb6.111423est.38107@neat.ai.toronto.edu> <780@ethz.UUCP>
Distribution: comp
Organization: ETH Zuerich
Lines: 19
In-reply-to: ceb@ethz.UUCP's message of 13 Feb 89 18:57:09 GMT


A followon to my previous posting about yes-or-no . . . (If things run
where you are the way they do here, you'll probably see this posting
first, so patience) anyway, here it is.

One obvious improvement one might make to what I posted (I bet it
comes from the Yale people first; it seems their style) would be, in
the context of those functions I posted, to check the frame stack and
see what function was pending, y-or-n-p or yes-or-no-p, and to do the
right thing accordingly.

I would do it, but I can't go rooting around in the EMACS guts right
now to find out how to get at the frame-stack if at all.  If not
possible, it would also be doable by defining yet another global state
variable, but I have learned to shy away from this kind of thing.  If
you did define the global variable, you could also fix it so the
functions had no effect when neither one was pending.

That's it.