GNU bug report logs - #79066
30.1; emacsclient.desktop may start a new instance of the daemon

Previous Next

To reply to this bug, email your comments to 79066 AT debbugs.gnu.org.

Toggle the display of automated, internal messages from the tracker.

Message #5 received at submit <at> debbugs.gnu.org (full text, mbox):

From: Ulrich Müller <ulm <at> gentoo.org>
To: bug-gnu-emacs <at> gnu.org
Subject: 30.1; emacsclient.desktop may start a new instance of the daemon
Date: Mon, 21 Jul 2025 08:44:18 +0200

The desktop file calls emacsclient with the --alternate-editor= option
(with empty option argument). I find this problematic for two reasons:

1. The user may run the Emacs daemon unter the control of the service
manager (like OpenRC or systemd). So in the regular case the option is
not needed because a daemon would always be running.

However, if for some reason the daemon failed to start or if the Emacs
process crashed, then it is very surprising behaviour for emacsclient to
start a "rogue" daemon that isn't under the service manager's control.

2. The user can get the present behaviour (or any other behaviour that
they prefer) by defining the ALTERNATE_EDITOR environment variable in
their startup files. The desktop file is the wrong place to override
the user's configuration with the --alternate-editor= option, therefore
taking away control from the user.

Message #8 received at 79066 <at> debbugs.gnu.org (full text, mbox):

From: Eli Zaretskii <eliz <at> gnu.org>
To: Ulrich Müller <ulm <at> gentoo.org>
Cc: 79066 <at> debbugs.gnu.org
Subject: Re: bug#79066: 30.1;
 emacsclient.desktop may start a new instance of the daemon
Date: Mon, 21 Jul 2025 15:00:20 +0300

> From: Ulrich Müller <ulm <at> gentoo.org>
> Date: Mon, 21 Jul 2025 08:44:18 +0200
> 
> The desktop file calls emacsclient with the --alternate-editor= option
> (with empty option argument). I find this problematic for two reasons:
> 
> 1. The user may run the Emacs daemon unter the control of the service
> manager (like OpenRC or systemd). So in the regular case the option is
> not needed because a daemon would always be running.
> 
> However, if for some reason the daemon failed to start or if the Emacs
> process crashed, then it is very surprising behaviour for emacsclient to
> start a "rogue" daemon that isn't under the service manager's control.
> 
> 2. The user can get the present behaviour (or any other behaviour that
> they prefer) by defining the ALTERNATE_EDITOR environment variable in
> their startup files. The desktop file is the wrong place to override
> the user's configuration with the --alternate-editor= option, therefore
> taking away control from the user.

<rant>
I can only reiterate my deep unhappiness that we agreed to add these
files to the Emacs distribution, and thus agreed to maintain it.  I
think this should be the job of the downstream distributions: they
know much better how the Emacs server is started, and thus can do a
much better job with the file.

One of these days I will simply remove these desktop files from the
tree.
</rant>

Message #11 received at 79066 <at> debbugs.gnu.org (full text, mbox):

From: Ulrich Müller <ulm <at> gentoo.org>
To: Eli Zaretskii <eliz <at> gnu.org>
Cc: Ulrich Müller <ulm <at> gentoo.org>, 79066 <at> debbugs.gnu.org
Subject: Re: bug#79066: 30.1; emacsclient.desktop may start a new instance
 of the daemon
Date: Mon, 21 Jul 2025 14:18:18 +0200

>>>>> On Mon, 21 Jul 2025, Eli Zaretskii wrote:

> One of these days I will simply remove these desktop files from the
> tree.
> </rant>

No objections. :/

In fact, Gentoo packages its own versions of emacs.desktop and
emacsclient.desktop since a long time (since 2007). So we're not
directly affected, but I still tend to file upstream bugs for any
problems that we see.

Similarly, we loosely sync our files to the ones in the Emacs tree.
(And the last change that we copied was support for org-protocol,
which caused bug #79068 ...)

Previous Next

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