sticky agenda and clock persistence interaction

sticky agenda and clock persistence interaction

[Brian van den Broek]

[-- Attachment #1: Type: text/plain, Size: 1205 bytes --]

Hi all,

GNU Emacs 23.2.1 (x86_64-pc-linux-gnu, GTK+ Version 2.20.1) of
2010-12-11 on brahms, modified by Debian
Org-mode version 7.9 (org-7.9-3-ga986d3 @ /home/brian/.emacsd/site-lisp/)

I just found that if I have

(setq org-agenda-sticky t)
(org-clock-persistence-insinuate)

in my .emacs---or rather in a file that my .emacs invokes with
load-library---I get

Warning (initialization): An error occurred while loading `/home/brian/.emacs':

Symbol's function definition is void: org-toggle-sticky-agenda

To ensure normal operation, you should investigate and remove the
cause of the error in your initialization file.  Start Emacs with
the `--debug-init' option to view a complete error backtrace.

in a *Warnings* buffer, whereas if I have

(org-clock-persistence-insinuate)
(setq org-agenda-sticky t)

in my .emacs, emacs launches without complaint.

The backtrace from running with --debug-init is attached. (gmail's web
interface barfs on the control codes and won't let me paste more than
the first three lines into the message pane.)

This seems odd and smells of a bug or at least a documentation lapse.

If that's not enough to go on, I'm happy to provide more information.

Best,

Brian vdB

[-- Attachment #2: backtrace --]
[-- Type: application/octet-stream, Size: 2566 bytes --]

Re: sticky agenda and clock persistence interaction

[Bastien]

Hi Brian,

Brian van den Broek <brian.van.den.broek@gmail.com> writes:

> I just found that if I have
>
> (setq org-agenda-sticky t)
> (org-clock-persistence-insinuate)
>
> in my .emacs---or rather in a file that my .emacs invokes with
> load-library---I get
>
> Warning (initialization): An error occurred while loading `/home/brian/.emacs':
>
> Symbol's function definition is void: org-toggle-sticky-agenda

If you are not requiring Org anyhow, org-agenda-sticky will not be
known.

What if you do

(require 'org-install)
(setq org-agenda-sticky t)
(org-clock-persistence-insinuate)

?

> in a *Warnings* buffer, whereas if I have
>
> (org-clock-persistence-insinuate)
> (setq org-agenda-sticky t)
>
> in my .emacs, emacs launches without complaint.

Because org-clock-persistence-insinuate is an autoloaded function.
So calling it will load org-clock.el which will require other Org 
packages as well.

> The backtrace from running with --debug-init is attached.

(Btw, there is a suspicious ~/.emacsd/ here -- not ~/.emacs.d/.  
Looks weird but maybe that's intentional.)

-- 
 Bastien

Re: sticky agenda and clock persistence interaction

[Brian van den Broek]

On 21 September 2012 09:37, Bastien <bzg@altern.org> wrote:
> Hi Brian,
>
> Brian van den Broek <brian.van.den.broek@gmail.com> writes:
>
>> I just found that if I have
>>
>> (setq org-agenda-sticky t)
>> (org-clock-persistence-insinuate)
>>
>> in my .emacs---or rather in a file that my .emacs invokes with
>> load-library---I get
>>
>> Warning (initialization): An error occurred while loading `/home/brian/.emacs':
>>
>> Symbol's function definition is void: org-toggle-sticky-agenda
>
> If you are not requiring Org anyhow, org-agenda-sticky will not be
> known.
>
> What if you do
>
> (require 'org-install)
> (setq org-agenda-sticky t)
> (org-clock-persistence-insinuate)
>
> ?

Hi Bastien,

(After the bad website report, I'm pleased to see I've not done
anything so silly this time :-)

I have a file, ogrconf.el that gets loaded by my .emacs. It starts
with (require 'org-install). So, the error I reported emerged from
what you suggest, save that I have a few hundred lines of config
between the require and the sticky and persistence lines.

I just tested, and if I start out my orgconf.el with

(require 'org-install)
(setq org-agenda-sticky t)
(org-clock-persistence-insinuate)

(thus, putting the relevant lines before any of my other org
configuration) I get the same warning I reported in the original post.
I've not done the backtrace assuming that the same warning with pretty
much the same cause would have the same backtrace; I'm happy to
provide if it is wanted.

>> The backtrace from running with --debug-init is attached.
>
> (Btw, there is a suspicious ~/.emacsd/ here -- not ~/.emacs.d/.
> Looks weird but maybe that's intentional.)

It's intentional. At some point, my .emacs became unwieldy. I
separated my config into a bunch of files which .emacs loads and put
them into a user-created dir ~/.emacsd,  leaving ~/.emacs.d for emacs
to have its way with. I prefer to enforce separation between files I
administer and those under emacs's control.

Best,

Brian

Re: sticky agenda and clock persistence interaction

[Bastien]

Hi Brian,

Can you compare the output of 

C-h f org-agenda RET

when you start Emacs normally and when you start emacs -Q?

I suspect you're another victim of a ghost install...

-- 
 Bastien

Re: sticky agenda and clock persistence interaction

[Achim Gratz]

Brian van den Broek writes:
> (setq org-agenda-sticky t)

That's a defcustom and you should not try to setq it unless you really
know when and how defcustom does its thing (I don't).


Regards,
Achim.
-- 
+<[Q+ Matrix-12 WAVE#46+305 Neuron microQkb Andromeda XTk Blofeld]>+

SD adaptation for Waldorf Blofeld V1.15B11:
http://Synth.Stromeko.net/Downloads.html#WaldorfSDada

Re: sticky agenda and clock persistence interaction

[Bastien]

Achim Gratz <Stromeko@nexgo.de> writes:

> Brian van den Broek writes:
>> (setq org-agenda-sticky t)
>
> That's a defcustom and you should not try to setq it unless you really
> know when and how defcustom does its thing (I don't).

What's wrong with setq'ing a variable defined with defcustom?

AFAIK that's the usual way to add customization in your init file,
independantly of the customization you add through the customize
interface.  And that's the way advertized in the Emacs manual too.

-- 
 Bastien