From xemacs-m  Tue Apr 15 21:36:08 1997
Received: from CNRI.Reston.VA.US (CNRI.Reston.VA.US [132.151.1.1])
	by xemacs.org (8.8.5/8.8.5) with SMTP id VAA05760
	for <xemacs-beta@xemacs.org>; Tue, 15 Apr 1997 21:36:07 -0500 (CDT)
Received: from newcnri.cnri.reston.va.us by CNRI.Reston.VA.US id aa26220;
          15 Apr 97 22:39 EDT
Received: from anthem.CNRI.Reston.Va.US by newcnri.CNRI.Reston.Va.US (SMI-8.6/SMI-SVR4)
	id WAA00351; Tue, 15 Apr 1997 22:39:13 -0400
Received: by anthem.CNRI.Reston.Va.US (SMI-8.6/SMI-SVR4)
	id WAA07448; Tue, 15 Apr 1997 22:39:13 -0400
Date: Tue, 15 Apr 1997 22:39:13 -0400
Message-Id: <199704160239.WAA07448@anthem.CNRI.Reston.Va.US>
From: "Barry A. Warsaw" <bwarsaw@CNRI.Reston.VA.US>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
To: Kyle Jones <kyle_jones@wonderworks.com>
Cc: xemacs-beta@xemacs.org
Subject: Re: pending-del.el citizenship
References: <199704152320.TAA07319@anthem.CNRI.Reston.Va.US>
	<QQclln01124.199704160147@crystal.WonderWorks.COM>
X-Mailer: VM 6.23 under 19.15 XEmacs Lucid
Reply-To: bwarsaw@python.org
X-Attribution: BAW
X-Oblique-Strategy: Distorting time
X-Url: http://www.python.org/~bwarsaw


>>>>> "KJ" == Kyle Jones <kyle_jones@wonderworks.com> writes:

>>>>> "BAW" == Barry Warsaw <bwarsaw@python.org> writes:

    BAW>  > [...] As a meta note, I wonder if pre- and
    BAW> post-command-hook shouldn't do something like this instead of
    BAW> the current error-wipes-all behavior?

    KJ> Catching errors is great, but a warning should not emitted
    KJ> unless the function is removed from the hook list.  Otherwise
    KJ> you might get a warning at every command event.  I could deal
    KJ> with that, but some other user might be inclined to consider
    KJ> XEmacs hosed beyod repair, ditch the editing session, and hate
    KJ> us.

The current zap-the-hook behavior is really bogus because the ordinary
user will have no idea what to do to repair this damage.  She'll just
notice that after this error occurs, paren mode stops de-highlighting
(leaving face turds), or pending-del stops working, etc.

I think if you're getting warnings after every command event, then
things are really hosed (as opposed to transiently error-prone).  When
I see this lossage, it's usually not very often.  I wouldn't be
totally opposed to silently ignoring the error, but I still think it's
better to warn and continue.

-Barry

