Gnus in Emacs bzr revision 101985 acting weird

Gnus in Emacs bzr revision 101985 acting weird

[Ashish SHUKLA]

[-- Attachment #1: Type: text/plain, Size: 2982 bytes --]

Hi everyone,

I use Gnus as my MUA, and I've noticed that it is acting weird for
me. Everytime I read a mail with Gnus, in the "Summary" buffer, I don't get an
"R" in the left column of the "Summary" buffer as displayed in the image at
the following URL:

http://www.flickr.com/photos/wahjava/5088753249/in/photostream/

To verify the problem, I cloned Gnus repo, restarted my Emacs, and before
loading Gnus with M-x gnus, I eval'd following code in *scratch* buffer:

#v+
(add-to-list 'load-path "~/.emacs.d/elisp/gnus/lisp")
(require 'gnus-load)
#v-

Now in "No Gnus", I can see the "R" in the left column in the "Summary"
buffer, every time I read a mail, as displayed in the image at the following
URL:

http://www.flickr.com/photos/wahjava/5088753247/

During all this, I've not modified my "~/.gnus" or "~/.emacs.d/init.el". I'm
wondering if I should report this as bug or already a known issue ?

Also recently I'm noticing that every time I visit a group in *Group* buffer,
I get following message:

"Key sequence $ t starts with non-prefix key $"

A back trace is pasted below:

#v+
Debugger entered--Lisp error: (error "Key sequence $ t starts with non-prefix key $")
  define-key((keymap #^[nil nil keymap #...) "$t" nnmairix-search-thread-this-article)
  nnmairix-summary-mode-hook()
  run-hooks(gnus-summary-mode-hook)
  apply(run-hooks gnus-summary-mode-hook)
  run-mode-hooks(gnus-summary-mode-hook)
  apply(run-mode-hooks gnus-summary-mode-hook)
  gnus-run-mode-hooks(gnus-summary-mode-hook)
  gnus-summary-mode("nnmaildir:mailing-lists.emacs-devel")
  gnus-summary-setup-buffer("nnmaildir:mailing-lists.emacs-devel")
  gnus-summary-read-group-1("nnmaildir:mailing-lists.emacs-devel" nil t nil nil nil)
  gnus-summary-read-group("nnmaildir:mailing-lists.emacs-devel" nil t nil nil nil nil)
  gnus-group-read-group(nil t)
  gnus-group-select-group(nil)
  gnus-topic-select-group(nil)
  call-interactively(gnus-topic-select-group nil nil)
#v-

Please note that I've removed the keymap from the define-key line above as
it's quite big.

Following is an excerpt from the lisp/gnus/nnmairix.el where this error
occurs:

#v+
;; Summary mode
(defun nnmairix-summary-mode-hook ()
  "Nnmairix summary mode keymap."
  (define-key gnus-summary-mode-map
    (kbd "$ t") 'nnmairix-search-thread-this-article)
  (define-key gnus-summary-mode-map
    (kbd "$ f") 'nnmairix-search-from-this-article)
  (define-key gnus-summary-mode-map
    (kbd "$ m") 'nnmairix-widget-search-from-this-article)
  (define-key gnus-summary-mode-map
    (kbd "$ g") 'nnmairix-create-search-group-from-message)
  (define-key gnus-summary-mode-map
    (kbd "$ o") 'nnmairix-goto-original-article)
  (define-key gnus-summary-mode-map
    (kbd "$ u") 'nnmairix-remove-tick-mark-original-article))
#v-

Thanks
-- 
Ashish SHUKLA

“You don't have to distrust the government to want to use
cryptography.” (Phil Zimmermann)

[-- Attachment #2: Type: application/pgp-signature, Size: 835 bytes --]

Re: Gnus in Emacs bzr revision 101985 acting weird

[Lars Magne Ingebrigtsen]

wahjava.ml@gmail.com (Ashish SHUKLA) writes:

> I use Gnus as my MUA, and I've noticed that it is acting weird for
> me. Everytime I read a mail with Gnus, in the "Summary" buffer, I don't get an
> "R" in the left column of the "Summary" buffer as displayed in the image at
> the following URL:
>
> http://www.flickr.com/photos/wahjava/5088753249/in/photostream/

I'm unable to reproduce this bug.  And Gnus in Emacs bzr and Gnus git
are virtually identical, so this is rather puzzling.

And looking over the code, I can't really immediately see how Gnus would
work otherwise, but not mark articles as read.

Do you have any local code that alters Gnus' behaviour in your load path?

> Also recently I'm noticing that every time I visit a group in *Group* buffer,
> I get following message:
>
> "Key sequence $ t starts with non-prefix key $"

`$' was recently added as a key stroke in gnus-spam, and that introduced
a keymap collision.  I think I'll just back out that change...

-- 
(domestic pets only, the antidote for overdose, milk.)
  larsi@gnus.org * Lars Magne Ingebrigtsen

Re: Gnus in Emacs bzr revision 101985 acting weird

[Ashish SHUKLA]

[-- Attachment #1: Type: text/plain, Size: 2514 bytes --]

Lars Magne Ingebrigtsen writes:
> wahjava.ml@gmail.com (Ashish SHUKLA) writes:

>> I use Gnus as my MUA, and I've noticed that it is acting weird for
>> me. Everytime I read a mail with Gnus, in the "Summary" buffer, I don't get an
>> "R" in the left column of the "Summary" buffer as displayed in the image at
>> the following URL:
>> 
>> http://www.flickr.com/photos/wahjava/5088753249/in/photostream/

> I'm unable to reproduce this bug.  And Gnus in Emacs bzr and Gnus git
> are virtually identical, so this is rather puzzling.

> And looking over the code, I can't really immediately see how Gnus would
> work otherwise, but not mark articles as read.

It is just that the "R" mark is not visible in the "Summary" buffer. The mail
gets marked as in when I leave the "Summary" buffer and return to *Group*
buffer I can see the unread count of mails decreased.

I also noticed that it's reproduces in Gnus git HEAD code as well, but not
always, whereas it's always reproducible with Gnus code (in Emacs bzr).

I'm using Emacs on GNU/Linux, compiled with GCC 4.5.1, with following
value of `system-configuration-options':

'--prefix=/usr' '--localstatedir=/var/lib/emacs' '--libexecdir=/usr/lib/emacs' '--with-xpm' '--with-jpeg' '--with-tiff' '--with-gif' '--with-xim' '--with-png' '--with-x-toolkit=gtk' '--with-m17n-flt' '--with-rsvg' '--with-xft' '--with-libotf' '--with-gpm' '--with-dbus' '--with-sound' '--with-makeinfo' 'CFLAGS=-march=x86-64 -mtune=native -O2 -pipe -g -fno-optimize-sibling-calls' 'LDFLAGS=-Wl,--hash-style=gnu -Wl,--as-needed'

> Do you have any local code that alters Gnus' behaviour in your load path?

And there is nothing I can found in my load path which alters Gnus behavior,
other than BBDB, as shown below:

#v+
;; bbdb
(eval-after-load "bbdb"
  '(progn
	 (bbdb-initialize 'gnus 'sc 'message 'sendmail)
	 (setq bbdb-dwim-net-address-allow-redundancy t)
	 (add-hook 'gnus-startup-hook 'bbdb-insinuate-gnus)
	 (eval-after-load "sendmail"
	   (add-hook 'mail-setup-hook 'bbdb-insinuate-sendmail))))
(require 'bbdb-autoloads)
#v-

>> Also recently I'm noticing that every time I visit a group in *Group* buffer,
>> I get following message:
>> 
>> "Key sequence $ t starts with non-prefix key $"

> `$' was recently added as a key stroke in gnus-spam, and that introduced
> a keymap collision.  I think I'll just back out that change...

Thanks
-- 
Ashish SHUKLA

“With no power, comes no responsibility.” (Kick-Ass, 2010)

[-- Attachment #2: Type: application/pgp-signature, Size: 835 bytes --]

Re: Gnus in Emacs bzr revision 101985 acting weird

[Lars Magne Ingebrigtsen]

wahjava.ml@gmail.com (Ashish SHUKLA) writes:

> It is just that the "R" mark is not visible in the "Summary" buffer. The mail
> gets marked as in when I leave the "Summary" buffer and return to *Group*
> buffer I can see the unread count of mails decreased.

Then it sounds like either `gnus-read-mark' has been set to the space
character, or something is altering your `gnus-summary-line-format'.

> And there is nothing I can found in my load path which alters Gnus behavior,
> other than BBDB, as shown below:

Well, perhaps it's BBDB that's interfering, then?

-- 
(domestic pets only, the antidote for overdose, milk.)
  larsi@gnus.org * Lars Magne Ingebrigtsen

Re: Gnus in Emacs bzr revision 101985 acting weird

[Ashish SHUKLA]

[-- Attachment #1: Type: text/plain, Size: 2463 bytes --]

Lars Magne Ingebrigtsen writes:
> wahjava.ml@gmail.com (Ashish SHUKLA) writes:

>> It is just that the "R" mark is not visible in the "Summary" buffer. The mail
>> gets marked as in when I leave the "Summary" buffer and return to *Group*
>> buffer I can see the unread count of mails decreased.

> Then it sounds like either `gnus-read-mark' has been set to the space
> character, or something is altering your `gnus-summary-line-format'.

I'm currently running Gnus (shipped with Emacs bzr revision 101988) in FreeBSD
(amd64).

Value of `gnus-read-mark' is:

82 (#o122, #x52)

Value of `gnus-summary-line-format' is:

"%U%R%z %d %B%(%[%-25,25f%]%) %s
"

which is same as what I've set in my .gnus:

(setq gnus-summary-line-format "%U%R%z %d %B%(%[%-25,25f%]%) %s\n")

In case if it helps, I tried to dump all buffer-local-variables[1] in the
"Summary" buffer by eval-ing following:

#v+
(let ((l (buffer-local-variables)) 
      (buf (generate-new-buffer "debug-output-testing")))
 (set-buffer buf)
 (dolist (v l) (insert (format "%s = %s\n" (car v) (cdr v)))))
#v-

Also I noticed that in drafts group, I can't edit existing drafts, as I used
to do using 'e' key, I looked up Gnus manual, and found mention of 'D e' to
edit draft, but that performs the same function as 'e' key. I did 'C-h c D'
and that resulted in following:

D runs the command gnus-summary-mark-as-read-backward

To edit the draft of this message, I've to use the M-x gnus-draft-edit-message.

I'm wondering if something major has changed between revisions 101606 and
current, which I didn't noticed :(. I'll go through log.

>> And there is nothing I can found in my load path which alters Gnus behavior,
>> other than BBDB, as shown below:

> Well, perhaps it's BBDB that's interfering, then?

I disabled BBDB as well, and nothing changed.

> -- 
> (domestic pets only, the antidote for overdose, milk.)
>   larsi@gnus.org * Lars Magne Ingebrigtsen

Let me know if you need my .gnus or any other configuration or debug output.

References:
[1]  http://people.freebsd.org/~ashish/gnus-debug-out-20101019.txt.gz

Thanks
-- 
Ashish SHUKLA

“I am free, no matter what rules surround me. If I find them
tolerable, I tolerate them; if I find them too obnoxious, I break
them. I am free because I know that I alone am morally responsible for
everything I do.” (Robert A. Heinlein, "The Moon Is a Harsh Mistress",
1966)

[-- Attachment #2: Type: application/pgp-signature, Size: 833 bytes --]

Re: Gnus in Emacs bzr revision 101985 acting weird

[Ashish SHUKLA]

[-- Attachment #1: Type: text/plain, Size: 2266 bytes --]

Hi,

I'm wondering if this has something to do with `face' not getting set
correctly in the "Summary" buffer for mails, which I'm currently reading and
thus getting marked as Read implicitly.

Following is the output of C-u C-x = when my point is in the "Summary" buffer
on my reply (which I last sent to this list, and is currently opened in
*Article* buffer):

#v+
        character: SPC (32, #o40, #x20)
preferred charset: ascii (ASCII (ISO646 IRV))
       code point: 0x20
           syntax:   	which means: whitespace
         category: .:Base, a:ASCII, l:Latin
      buffer code: #x20
        file code: #x20 (encoded by coding system utf-8)
          display: by this font (glyph code)
    xft:-unknown-Droid Sans Mono-normal-normal-normal-*-14-*-*-*-m-0-iso10646-1 (#x03)

Character code properties: customize what to show
  name: SPACE
  general-category: Zs (Separator, Space)

There are text properties here:
  face                 gnus-summary-high-unread
  gnus-number          5400

[back]
#v-

After this, I exited the group and back in *Group* buffer, and did 'g' to
check for new messages. After this I visited
"nnmaildir:mailing-lists.emacs-devel" group again, did '/ o' to list old
messages and moved my point to the same reply, and did "C-u C-x =" again, and
following is the output:

#v+
        character: n (110, #o156, #x6e)
preferred charset: ascii (ASCII (ISO646 IRV))
       code point: 0x6E
           syntax: w 	which means: word
         category: .:Base, a:ASCII, l:Latin, r:Roman
      buffer code: #x6E
        file code: #x6E (encoded by coding system utf-8)
          display: by this font (glyph code)
    xft:-unknown-Droid Sans Mono-normal-normal-normal-*-14-*-*-*-m-0-iso10646-1 (#x51)

Character code properties: customize what to show
  name: LATIN SMALL LETTER N
  general-category: Ll (Letter, Lowercase)

There are text properties here:
  face                 gnus-summary-high-ancient
  gnus-number          5400

[back]
#v-

Thanks
-- 
Ashish SHUKLA

“Until the day when God shall deign to reveal the future to man, all
human wisdom is summed up in these two words,-"Wait and hope".”
(Alexandre Dumas, père, "The Count of Monte Cristo", 1845-1846)

[-- Attachment #2: Type: application/pgp-signature, Size: 833 bytes --]

Re: Gnus in Emacs bzr revision 101985 acting weird

[Lars Magne Ingebrigtsen]

wahjava.ml@gmail.com (Ashish SHUKLA) writes:

> Also I noticed that in drafts group, I can't edit existing drafts, as I used
> to do using 'e' key, I looked up Gnus manual, and found mention of 'D e' to
> edit draft, but that performs the same function as 'e' key. I did 'C-h c D'
> and that resulted in following:
>
> D runs the command gnus-summary-mark-as-read-backward

The `D e' binding is set up on group entry in the drafts groups.

It really sounds like your Emacs/Gnus setup is subtly broken, and that
you have different bits of Gnus loading now and then, so my only advice
to you is to try to get to a base Emacs installation without any
customisations.

-- 
(domestic pets only, the antidote for overdose, milk.)
  larsi@gnus.org * Lars Magne Ingebrigtsen

Re: Gnus in Emacs bzr revision 101985 acting weird

[Ashish SHUKLA]

[-- Attachment #1: Type: text/plain, Size: 1082 bytes --]

Hi Lars,

Lars Magne Ingebrigtsen writes:
> wahjava.ml@gmail.com (Ashish SHUKLA) writes:

>> Also I noticed that in drafts group, I can't edit existing drafts, as I used
>> to do using 'e' key, I looked up Gnus manual, and found mention of 'D e' to
>> edit draft, but that performs the same function as 'e' key. I did 'C-h c D'
>> and that resulted in following:
>> 
>> D runs the command gnus-summary-mark-as-read-backward

> The `D e' binding is set up on group entry in the drafts groups.

> It really sounds like your Emacs/Gnus setup is subtly broken, and that
> you have different bits of Gnus loading now and then, so my only advice
> to you is to try to get to a base Emacs installation without any
> customisations.

I upgraded to Emacs bzr revision 102021, and without changing anything in my
.gnus or .emacs.d/init.el, it started working for me again. I guess it had
something to do with the '$' prefix in nnmairix key-binding.

Thanks
-- 
Ashish SHUKLA

“All great truths begin as blasphemies.” (George Bernard Shaw,
"Annajanska", 1919)

[-- Attachment #2: Type: application/pgp-signature, Size: 833 bytes --]