GNU bug report logs - #37826
Very annoying autoraise client/server behavior with -t option

Previous Next

Full log

Message #124 received at 37826 <at> debbugs.gnu.org (full text, mbox):

From: Carlos Pita <carlosjosepita <at> gmail.com>
To: Eli Zaretskii <eliz <at> gnu.org>
Cc: Juanma Barranquero <lekktu <at> gmail.com>, 37826 <at> debbugs.gnu.org
Subject: Re: bug#37826: Very annoying autoraise client/server behavior with -t
 option
Date: Sun, 27 Oct 2019 14:19:51 -0300

> We want a solution that avoids raising the wrong frame for some
> unimportant messages, but still does raise some frame for displaying
> important messages, such as errors.  If that doesn't happen in some
> scenario, then we cannot use this idea.

Temporarily redirecting the output of message was a solution that
fitted that description:

* Questions are still asked.
* Errors still raise some frame with a message.

And then, at the end, you get a dump of all temporarily silenced
messages in the right frame.

You dislike the instrumenting/monkey-patching approach, be it by
advicing or by directly accessing the function slot of the symbol. I'm
not sure why, message is a function with a simple and well defined
interface that I'm honouring except for its side effect, and
everything is carefully done inside an unwind-protect clause. But, in
any case, we could add some global flag or something that message
could check in order to change its output destination. That wouldn't
be an instrumentation. But I find that considerably more complex and
it affects the overall behavior of message instead of just mocking it
for a while.

Previous Next

GNU bug tracking system
Copyright (C) 1999 Darren O. Benham, 1997,2003 nCipher Corporation Ltd, 1994-97 Ian Jackson.