GNU bug report logs - #51360
29.0.50; variable `server-name' initialized too late for use in init file

Previous Next

To add a comment to this bug, you must first unarchive it, by sending
a message to control AT debbugs.gnu.org, with unarchive 51360 in the body.
You can then email your comments to 51360 AT debbugs.gnu.org in the normal way.

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

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

From: Gregor Zattler <grfz <at> gmx.de>
To: bug-gnu-emacs <at> gnu.org
Subject: 29.0.50; variable `server-name' initialized too late for use in
 init file
Date: Sat, 23 Oct 2021 20:39:25 +0200

Dear emacs developers, if emacs is started as a named daemon
like this: `emacs --damon=foo' or like this:
`emacsclient -a "" -s foo [...]' the variable `server-name'
is initialized with the name "foo" but too late to be used
in init.el:



Warning (initialization): An error occurred while loading ´/home/grfz/.config/emacs/init.el´:

Symbol's value as variable is void: server-name



It would be helpful if `server-name' would be initialized
early enough to be used in init.el in order to enable
conditional configuration of the server, depending on it's
name.



Thanks for your attention, Gregor

Message #8 received at 51360 <at> debbugs.gnu.org (full text, mbox):

From: Stefan Kangas <stefan <at> marxist.se>
To: Gregor Zattler <grfz <at> gmx.de>
Cc: 51360 <at> debbugs.gnu.org
Subject: Re: bug#51360: 29.0.50; variable `server-name' initialized too late
 for use in init file
Date: Sun, 24 Oct 2021 05:46:36 -0700

Gregor Zattler <grfz <at> gmx.de> writes:

> Warning (initialization): An error occurred while loading ´/home/grfz/.config/emacs/init.el´:
>
> Symbol's value as variable is void: server-name
>
> It would be helpful if `server-name' would be initialized
> early enough to be used in init.el in order to enable
> conditional configuration of the server, depending on it's
> name.

Doesn't this just depend on when you call `server-start'?

Do you have a recipe for reproducing this?

Message #11 received at 51360 <at> debbugs.gnu.org (full text, mbox):

From: Gregor Zattler <grfz <at> gmx.de>
To: Stefan Kangas <stefan <at> marxist.se>
Cc: 51360 <at> debbugs.gnu.org
Subject: Re: bug#51360: 29.0.50; variable `server-name' initialized too late
 for use in init file
Date: Sun, 24 Oct 2021 17:34:19 +0200

Hi Stefan, emacs devs,
* Stefan Kangas <stefan <at> marxist.se> [2021-10-24; 05:46]:
> Gregor Zattler <grfz <at> gmx.de> writes:
>
>> Warning (initialization): An error occurred while loading ´/home/grfz/.config/emacs/init.el´:
>>
>> Symbol's value as variable is void: server-name
>>
>> It would be helpful if `server-name' would be initialized
>> early enough to be used in init.el in order to enable
>> conditional configuration of the server, depending on it's
>> name.
>
> Doesn't this just depend on when you call `server-start'?

When emacs is started with emacs --daemon=foo (or
emacsclinet -a "" -s foo [...]) one does not use
`server-start'.

> Do you have a recipe for reproducing this?

Have

(when (equal server-name "foo") (message "My name is foo!"))

in .emacs and start emacs with   emacs --daemon=foo



Ciao,
-- 
Gregor

Message #14 received at 51360 <at> debbugs.gnu.org (full text, mbox):

From: Gregory Heytings <gregory <at> heytings.org>
To: Gregor Zattler <grfz <at> gmx.de>
Cc: 51360 <at> debbugs.gnu.org, Stefan Kangas <stefan <at> marxist.se>
Subject: Re: bug#51360: 29.0.50; variable `server-name' initialized too late
 for use in init file
Date: Sun, 24 Oct 2021 15:45:21 +0000

[Message part 1 (text/plain, inline)]

>
> Warning (initialization): An error occurred while loading 
> ´/home/grfz/.config/emacs/init.el´:
>
> Symbol's value as variable is void: server-name
>

You should (require 'server) before referring to server-name.  None of the 
defcustoms in server.el are autoloaded.

Message #17 received at 51360 <at> debbugs.gnu.org (full text, mbox):

From: Eli Zaretskii <eliz <at> gnu.org>
To: Gregor Zattler <grfz <at> gmx.de>
Cc: 51360 <at> debbugs.gnu.org, stefan <at> marxist.se
Subject: Re: bug#51360: 29.0.50;
 variable `server-name' initialized too late for use in init file
Date: Sun, 24 Oct 2021 18:59:06 +0300

> From: Gregor Zattler <grfz <at> gmx.de>
> Date: Sun, 24 Oct 2021 17:34:19 +0200
> Cc: 51360 <at> debbugs.gnu.org
> 
> When emacs is started with emacs --daemon=foo (or
> emacsclinet -a "" -s foo [...]) one does not use
> `server-start'.
> 
> > Do you have a recipe for reproducing this?
> 
> Have
> 
> (when (equal server-name "foo") (message "My name is foo!"))
> 
> in .emacs and start emacs with   emacs --daemon=foo

Whatever you do that you need server-name for, can't you do that in
emacs-startup-hook?  AFAIR, that gets called after the daemon already
finished its initialization and provided its name to Emacs, so
server-name should be already populated.

Message #22 received at 51360 <at> debbugs.gnu.org (full text, mbox):

From: Gregor Zattler <grfz <at> gmx.de>
To: 51360 <at> debbugs.gnu.org
Subject: Re: bug#51360: 29.0.50; variable `server-name' initialized too late
 for use in init file
Date: Sun, 24 Oct 2021 21:42:57 +0200

Hi Gregory, emacs devs,
* Gregory Heytings <gregory <at> heytings.org> [2021-10-24; 15:45]:
>> Warning (initialization): An error occurred while loading 
>> ´/home/grfz/.config/emacs/init.el´:
>>
>> Symbol's value as variable is void: server-name
>
> You should (require 'server) before referring to server-name.  None of the 
> defcustoms in server.el are autoloaded.

thanks.  Requiring server prevents this error, but `server-name'
is not set to the command line value, though:

~$ cat ~/.emacs
(require 'server)
(when (equal server-name "foo") (setq greeting "My name is foo!"))
~$ emacsclient -a "" -s foo --eval '(message greeting)'
emacsclient: Should XDG_RUNTIME_DIR='/run/user/1000' be in the environment?
emacsclient: (Be careful: XDG_RUNTIME_DIR is security-related.)
emacsclient: can't find socket; have you started the server?
emacsclient: To start the server in Emacs, type "M-x server-start".
Starting Emacs daemon.
Emacs daemon should have started, trying to connect again
*ERROR*: Symbol's value as variable is void: greeting

Ciao,
-- 
Gregor

Message #25 received at 51360 <at> debbugs.gnu.org (full text, mbox):

From: Gregor Zattler <grfz <at> gmx.de>
To: 51360 <at> debbugs.gnu.org
Subject: Re: bug#51360: 29.0.50; variable `server-name' initialized too late
 for use in init file
Date: Sun, 24 Oct 2021 22:00:21 +0200

Hi Eli, emacs devs,
* Eli Zaretskii <eliz <at> gnu.org> [2021-10-24; 18:59]:
>> From: Gregor Zattler <grfz <at> gmx.de>
>> Date: Sun, 24 Oct 2021 17:34:19 +0200
>> Cc: 51360 <at> debbugs.gnu.org
>>
>> When emacs is started with emacs --daemon=foo (or
>> emacsclinet -a "" -s foo [...]) one does not use
>> `server-start'.
>>
>> > Do you have a recipe for reproducing this?
>>
>> Have
>>
>> (when (equal server-name "foo") (message "My name is foo!"))
>>
>> in .emacs and start emacs with   emacs --daemon=foo
>
> Whatever you do that you need server-name for, can't you do that in
> emacs-startup-hook?  AFAIR, that gets called after the daemon already
> finished its initialization and provided its name to Emacs, so
> server-name should be already populated.

the idea is to initialise the emacs daemon depending on it's
server-name.  Therefore doing it via emacs-startup-hook is
somewhat counter-intuitive


But you are right that works.  Thanks.


Ciao,
--
Gregor

Message #28 received at 51360 <at> debbugs.gnu.org (full text, mbox):

From: Eli Zaretskii <eliz <at> gnu.org>
To: Gregor Zattler <grfz <at> gmx.de>
Cc: 51360 <at> debbugs.gnu.org
Subject: Re: bug#51360: 29.0.50;
 variable `server-name' initialized too late for use in init file
Date: Mon, 25 Oct 2021 05:27:37 +0300

> From: Gregor Zattler <grfz <at> gmx.de>
> Date: Sun, 24 Oct 2021 22:00:21 +0200
> 
> > Whatever you do that you need server-name for, can't you do that in
> > emacs-startup-hook?  AFAIR, that gets called after the daemon already
> > finished its initialization and provided its name to Emacs, so
> > server-name should be already populated.
> 
> the idea is to initialise the emacs daemon depending on it's
> server-name.  Therefore doing it via emacs-startup-hook is
> somewhat counter-intuitive

Sorry, I don't think I follow: what has intuition got to do with this?
And what do you mean by "initialise the emacs daemon"? -- that happens
out of your control, in the internal Emacs code.

> But you are right that works.  Thanks.

So does this mean your problem is solved?

Message #31 received at 51360 <at> debbugs.gnu.org (full text, mbox):

From: Gregor Zattler <grfz <at> gmx.de>
To: 51360 <at> debbugs.gnu.org
Subject: Re: bug#51360: 29.0.50; variable `server-name' initialized too late
 for use in init file
Date: Mon, 25 Oct 2021 10:02:16 +0200

[sorry for second post, first did not include bug address]
Hi Eli, emacs devs,
* Eli Zaretskii <eliz <at> gnu.org> [2021-10-25; 05:27]:
>> From: Gregor Zattler <grfz <at> gmx.de>
>> Date: Sun, 24 Oct 2021 22:00:21 +0200
>>
>> > Whatever you do that you need server-name for, can't you do that in
>> > emacs-startup-hook?  AFAIR, that gets called after the daemon already
>> > finished its initialization and provided its name to Emacs, so
>> > server-name should be already populated.
>>
>> the idea is to initialise the emacs daemon depending on it's
>> server-name.  Therefore doing it via emacs-startup-hook is
>> somewhat counter-intuitive
>
> Sorry, I don't think I follow: what has intuition got to do with this?
> And what do you mean by "initialise the emacs daemon"? -- that happens
> out of your control, in the internal Emacs code.
>
>> But you are right that works.  Thanks.
>
> So does this mean your problem is solved?

Yes, thanks.  Please close the bug report.

Ciao,
--
Gregor

Message #36 received at 51360-done <at> debbugs.gnu.org (full text, mbox):

From: Eli Zaretskii <eliz <at> gnu.org>
To: Gregor Zattler <grfz <at> gmx.de>
Cc: 51360-done <at> debbugs.gnu.org
Subject: Re: bug#51360: 29.0.50;
 variable `server-name' initialized too late for use in init file
Date: Mon, 25 Oct 2021 15:41:55 +0300

> From: Gregor Zattler <grfz <at> gmx.de>
> Date: Mon, 25 Oct 2021 10:02:16 +0200
> 
> > So does this mean your problem is solved?
> 
> Yes, thanks.  Please close the bug report.

Thanks, closing.

Previous Next

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