Re: nndiary documentation bug

Re: nndiary documentation bug

[Reiner Steib]

On Thu, May 03 2007, Holger Sparr wrote:

> From (info "(gnus)Running NNDiary") : 
>
> ,----
> |    In order to use `nndiary' in autonomous mode, you have several
> | things to do:
> | 
> |    * Allow `nndiary' to retrieve new mail by itself.  Put the following
> |      line in your `gnusrc' file:
> | 
> |           (setq nndiary-get-new-mail t)
> `----
>
> This seems to be no longer true.
>
> By now, `nndiary-get-new-mail' is actually designed as a server variable
> and should therefor be used like this:
>
> (setq gnus-secondary-select-methods '((nndiary "" 
>                                                (nndiary-get-new-mail t))))

I think Holger is right.  Didier, what do you think?

BTW, is there any reason _not_ to include (info "(gnus)Running
NNDiary") in the stable branch (v5-10)?

Bye, Reiner.
-- 
       ,,,
      (o o)
---ooO-(_)-Ooo---  |  PGP key available  |  http://rsteib.home.pages.de/

Re: nndiary documentation bug

[Didier Verna]

Reiner Steib <reinersteib+from-uce@imap.cc> wrote:

> On Thu, May 03 2007, Holger Sparr wrote:
>
>> From (info "(gnus)Running NNDiary") : 
>>
>> ,----
>> |    In order to use `nndiary' in autonomous mode, you have several
>> | things to do:
>> | 
>> |    * Allow `nndiary' to retrieve new mail by itself.  Put the following
>> |      line in your `gnusrc' file:
>> | 
>> |           (setq nndiary-get-new-mail t)
>> `----
>>
>> This seems to be no longer true.
>>
>> By now, `nndiary-get-new-mail' is actually designed as a server
>> variable and should therefore be used like this:
>>
>> (setq gnus-secondary-select-methods '((nndiary "" 
>>                                                (nndiary-get-new-mail t))))
>
> I think Holger is right.  Didier, what do you think?


  I don't quite understand the "This seems to be no longer true." part,
as AFAICR, the design of nndiary-get-new-mail has never changed. I'm
actually reluctant to advertise this variable as a server one (although
technically, I guess it's possible to use it that way).

As explained in the doc, nndiary is not a traditionnal mail backend (you
would not use it for retrieving your emails), so if you want to use it
in its "autonomous" mode, you probably also have to customize
nndiary-mail-sources and nndiary-split-methods as well. This is not like
setting up one *and only one* mail backend as the preferred method for
getting mails.



> BTW, is there any reason _not_ to include (info "(gnus)Running
> NNDiary") in the stable branch (v5-10)?

  If nndiary is present in the stable branch, the whole "Email Based
Diary" node should be included in the doc. If it is not, I don't know
why.

-- 
Read the Jazz Blog !! http://jazzblog.didierverna.com

Didier Verna, didier@lrde.epita.fr, http://www.lrde.epita.fr/~didier

EPITA / LRDE, 14-16 rue Voltaire   Tel.+33 (1) 44 08 01 85
94276 Le Kremlin-Bicêtre, France   Fax.+33 (1) 53 14 59 22   didier@xemacs.org

Re: nndiary documentation bug

[Holger Sparr]

Didier Verna wrote:

> Reiner Steib <reinersteib+from-uce@imap.cc> wrote:
>
>> On Thu, May 03 2007, Holger Sparr wrote:
>>
>>> From (info "(gnus)Running NNDiary") : 
>>>
>>> ,----
>>> |    In order to use `nndiary' in autonomous mode, you have several
>>> | things to do:
>>> | 
>>> |    * Allow `nndiary' to retrieve new mail by itself.  Put the following
>>> |      line in your `gnusrc' file:
>>> | 
>>> |           (setq nndiary-get-new-mail t)
>>> `----
>>>
>>> This seems to be no longer true.
>>>
>>> By now, `nndiary-get-new-mail' is actually designed as a server
>>> variable and should therefore be used like this:
>>>
>>> (setq gnus-secondary-select-methods '((nndiary "" 
>>>                                                (nndiary-get-new-mail t))))
>>
>> I think Holger is right.  Didier, what do you think?
>
>
>   I don't quite understand the "This seems to be no longer true." part,
> as AFAICR, the design of nndiary-get-new-mail has never changed. I'm
> actually reluctant to advertise this variable as a server one (although
> technically, I guess it's possible to use it that way).
>
> As explained in the doc, nndiary is not a traditionnal mail backend (you
> would not use it for retrieving your emails), so if you want to use it
> in its "autonomous" mode, you probably also have to customize
> nndiary-mail-sources and nndiary-split-methods as well. This is not like
> setting up one *and only one* mail backend as the preferred method for
> getting mails.

I cannot really say when getting new mail for the nndiary backend stopped. 

I only guess that it happened after `gnus-server-regenerate-server' for
my nnml backend. 

This is what I get now:

,----[ C-h v nndiary-get-new-mail RET ]
| nndiary-get-new-mail is a variable defined in `nndiary.el'.
| Its value is nil
| 
| 
| Documentation:
| Whether nndiary gets new mail and split it.
| Contrary to traditional mail back ends, this variable can be set to t
| even if your primary mail back end also retreives mail. In such a case,
| NDiary uses its own mail-sources and split-methods.
| 
| This is a Gnus server variable.  See Info node `(gnus)Select Methods'.
| ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
| [back]
`----

and I assumed something has been changed. Since

,----
| (require 'gnus-diary)
| (setq nndiary-get-new-mail t)
`----

in my .gnus.el has no effect anymore (meaning the default of nil is not changed).


In case it has something to do with my setup, where and how can I investigate? 


Holger

Re: nndiary documentation bug

[Didier Verna]

Holger Sparr <sparr@mfkrs1.mw.tu-dresden.de> wrote:

> This is what I get now:
>
> | This is a Gnus server variable.  See Info node `(gnus)Select Methods'.
> | ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

        This is the resut of defining the variable via defvoo, which has
always been the case.


> ,----
> | (require 'gnus-diary)
> | (setq nndiary-get-new-mail t)
> `----
>
> in my .gnus.el has no effect anymore (meaning the default of nil is
> not changed).

        I'm surprised. My setup is identical; I have just
(nndiary "diary") in gnus-secondary-select-methods, and it works fine.


> In case it has something to do with my setup, where and how can I
> investigate?

        I don't really know... Are you sure you don't have a
contradictory line somewhere else in your Emacs initialization files ?

-- 
Read the Jazz Blog !! http://jazzblog.didierverna.com

Didier Verna, didier@lrde.epita.fr, http://www.lrde.epita.fr/~didier

EPITA / LRDE, 14-16 rue Voltaire   Tel.+33 (1) 44 08 01 85
94276 Le Kremlin-Bicêtre, France   Fax.+33 (1) 53 14 59 22   didier@xemacs.org

Re: nndiary documentation bug

[Reiner Steib]

On Mon, May 07 2007, Didier Verna wrote:

> I'm actually reluctant to advertise this variable as a server one
> (although technically, I guess it's possible to use it that way).

Okay.

>> BTW, is there any reason _not_ to include (info "(gnus)Running
>> NNDiary") in the stable branch (v5-10)?
>
>   If nndiary is present in the stable branch, the whole "Email Based
> Diary" node should be included in the doc.

Apart from minor issues (compatibility code) and comments (moved to
gnus.texi), the versions of nndiary.el and gnus-diary.el in trunk and
v5-10 are identical.  So I will sync the lisp files and install (info
"(gnus)Email Based Diary") in v5-10 unless you'd like to do it
yourself.

Bye, Reiner.

--8<---------------cut here---------------start------------->8---
--- v5-10/lisp/nndiary.el	2007-02-04 17:09:38.000000000 +0100
+++ No/lisp/nndiary.el	        2007-01-26 20:41:00.000000000 +0100
@@ -725,7 +606,7 @@
     (nconc rest articles)))
 
 (deffoo nndiary-request-move-article
-    (article group server accept-form &optional last)
+    (article group server accept-form &optional last move-is-internal)
   (let ((buf (get-buffer-create " *nndiary move*"))
 	result)
     (nndiary-possibly-change-directory group server)
@@ -994,7 +875,7 @@
 		  (search-forward id nil t)) ; We find the ID.
 	;; And the id is in the fourth field.
 	(if (not (and (search-backward "\t" nil t 4)
-		      (not (search-backward"\t" (gnus-point-at-bol) t))))
+		      (not (search-backward"\t" (point-at-bol) t))))
 	    (forward-line 1)
 	  (beginning-of-line)
 	  (setq found t)
@@ -1215,9 +1096,7 @@
     (push (list group
 		(cons (or (caar files) (1+ last))
 		      (max last
-			   (or (let ((f files))
-				 (while (cdr f) (setq f (cdr f)))
-				 (caar f))
+			   (or (caar (last files))
 			       0))))
 	  nndiary-group-alist)))
 
--- v5-10/lisp/gnus-diary.el	2007-02-04 17:09:37.000000000 +0100
+++ plain_No/lisp/gnus-diary.el	2007-01-26 20:40:58.000000000 +0100
@@ -402,7 +347,7 @@
 	   (when (re-search-forward (concat "^" header ":") nil t)
 	     (unless (eq (char-after) ? )
 	       (insert " "))
-	     (setq value (buffer-substring (point) (gnus-point-at-eol)))
+	     (setq value (buffer-substring (point) (point-at-eol)))
 	     (and (string-match "[ \t]*\\([^ \t]+\\)[ \t]*" value)
 		  (setq value (match-string 1 value)))
 	     (condition-case ()
--8<---------------cut here---------------end--------------->8---
-- 
       ,,,
      (o o)
---ooO-(_)-Ooo---  |  PGP key available  |  http://rsteib.home.pages.de/

Re: nndiary documentation bug

[Didier Verna]

Reiner Steib <reinersteib+gmane@imap.cc> wrote:

>>> BTW, is there any reason _not_ to include (info "(gnus)Running
>>> NNDiary") in the stable branch (v5-10)?
>>
>>   If nndiary is present in the stable branch, the whole "Email Based
>> Diary" node should be included in the doc.
>
> Apart from minor issues (compatibility code) and comments (moved to
> gnus.texi), the versions of nndiary.el and gnus-diary.el in trunk and
> v5-10 are identical. So I will sync the lisp files and install (info
> "(gnus)Email Based Diary") in v5-10 unless you'd like to do it
> yourself.

        Please do.

-- 
Read the Jazz Blog !! http://jazzblog.didierverna.com

Didier Verna, didier@lrde.epita.fr, http://www.lrde.epita.fr/~didier

EPITA / LRDE, 14-16 rue Voltaire   Tel.+33 (1) 44 08 01 85
94276 Le Kremlin-Bicêtre, France   Fax.+33 (1) 53 14 59 22   didier@xemacs.org

Re: nndiary documentation bug

[Kamen TOMOV]

test