Latest CVS Gnus just doesn't wanna play fair.

Latest CVS Gnus just doesn't wanna play fair.

[Steve Youngs]

OK this time I'm certain that the problem _isn't_ my ~/.gnus :-)

When composing a mail and trying to move between the mail headers with
C-c C-f whatever:

Symbol's function definition is void: comint-extract-current-pathname


The backtrace:

  # bind (standard-output stack-trace-on-signal debug-on-signal stack-trace-on-error debug-on-error)
  comint-extract-current-pathname()
  # bind (end beg multi-line)
  comint-find-source-code(nil)
  # bind (command-debug-status)
  call-interactively(comint-find-source-code)
  # (condition-case ... . error)
  # (catch top-level ...)


And when trying to send the mail with C-c C-c

C-c C-c not defined


The backtrace:

  # bind (standard-output stack-trace-on-signal debug-on-signal stack-trace-on-error debug-on-error)
  # (condition-case ... . error)
  # (catch top-level ...)



-- 
|---<Steve Youngs>---------------<GnuPG KeyID: 9E7E2820>---|
|            XEmacs - It's not just an editor.             |
|                    It's a way of life.                   |
|------------------------------------<youngs@xemacs.org>---|

Re: Latest CVS Gnus just doesn't wanna play fair.

[Kai Großjohann]

On Sun, 29 Jul 2001, Steve Youngs wrote:

> OK this time I'm certain that the problem _isn't_ my ~/.gnus :-)
> 
> When composing a mail and trying to move between the mail headers
> with C-c C-f whatever:
> 
> Symbol's function definition is void:
> comint-extract-current-pathname

Very curious.  What's the major mode in effect when you do that?  What
does C-h c C-c C-f <foo> print?

What does a backtrace produced via M-x set-variable RET debug-on-error
RET t RET look like?  The other backtraces you show look really
strange, and I'm not sure whether it's because something strange was
going on, or because I'm not familiar with the XEmacs backtraces you
show. 

kai
-- 
~/.signature: No such file or directory

Re: Latest CVS Gnus just doesn't wanna play fair.

[Steve Youngs]

|--==> "SY" == Steve Youngs <youngs@xemacs.org> writes:

  SY> OK this time I'm certain that the problem _isn't_ my ~/.gnus :-)
  SY> When composing a mail and trying to move between the mail headers with
  SY> C-c C-f whatever:

  SY> Symbol's function definition is void: comint-extract-current-pathname

  SY> And when trying to send the mail with C-c C-c

  SY> C-c C-c not defined

I broke this because I had created an 'auto-autoloads.el' and
'custom-load.el' for my Gnus.  Remove them, and goodbye problems. 

I've been building waaay too many XEmacs packages, and got a little
carried away. :-(

-- 
|---<Steve Youngs>---------------<GnuPG KeyID: 9E7E2820>---|
|            XEmacs - It's not just an editor.             |
|                    It's a way of life.                   |
|------------------------------------<youngs@xemacs.org>---|

Re: Latest CVS Gnus just doesn't wanna play fair.

[Steve Youngs]

|--==> "KG" == Kai Großjohann <Kai.Grossjohann@CS.Uni-Dortmund.DE> writes:

  KG> On Sun, 29 Jul 2001, Steve Youngs wrote:
  >>OK this time I'm certain that the problem _isn't_ my ~/.gnus :-)
  >>
  >>When composing a mail and trying to move between the mail headers
  >>with C-c C-f whatever:
  >>
  >>Symbol's function definition is void:
  >>comint-extract-current-pathname

  KG> Very curious.  What's the major mode in effect when you do that?  What
  KG> does C-h c C-c C-f <foo> print?

  KG> What does a backtrace produced via M-x set-variable RET debug-on-error
  KG> RET t RET look like?  The other backtraces you show look really
  KG> strange, and I'm not sure whether it's because something strange was
  KG> going on, or because I'm not familiar with the XEmacs backtraces you
  KG> show. 

Hi Kai

I fixed this (although I don't think properly), see my followup to my
original post.

When I get a chance I'll look into it at more depth and fix it
properly. 


-- 
|---<Steve Youngs>---------------<GnuPG KeyID: 9E7E2820>---|
|            XEmacs - It's not just an editor.             |
|                    It's a way of life.                   |
|------------------------------------<youngs@xemacs.org>---|

Re: Latest CVS Gnus just doesn't wanna play fair.

[Didier Verna]

Steve Youngs <youngs@xemacs.org> wrote:

>   SY> C-c C-c not defined
>
> I broke this because I had created an 'auto-autoloads.el' and
> 'custom-load.el' for my Gnus.  Remove them, and goodbye problems.
>
> I've been building waaay too many XEmacs packages, and got a little
> carried away. :-(

	Steve, I'm experiencing the same problems even after the bug fix in
XEmacs's autoload.el I've just sent (and applied) to xemacs-patches. Something
*must* have changed in Gnus recently because I've been installing it as an
XEmacs package and making auto-autoloads and custom-loads for it for ages
before.

        The weird stuff actually happens for me when compiling the autoloads:

didier(pts/3)% $src/dxemacs/src/xemacs -batch -q -f batch-byte-compile auto-autoloads.el custom-load.el
Compiling /usr/local/share/xemacs/site-packages/lisp/gnus/auto-autoloads.el...
While compiling message-mode in file /usr/local/share/xemacs/site-packages/lisp/gnus/auto-autoloads.el:
  ** reference to free variable message-face-alist
  ** reference to free variable message-mode-menu
  ** reference to free variable message-mode-field-menu
  ** reference to free variable message-mail-alias-type
While compiling toplevel forms:
  ** multiple autoloads for smiley-region
While compiling the end of the data:
  ** The following functions are not known to be defined:
    message-setup-fill-variables, message-setup-toolbar,
    message-tool-bar-map, mail-abbrevs-setup,
    message-set-auto-save-file-name, mm-enable-multibyte, mml-mode
Wrote /usr/local/share/xemacs/site-packages/lisp/gnus/auto-autoloads.elc
Compiling /usr/local/share/xemacs/site-packages/lisp/gnus/custom-load.el...
Wrote /usr/local/share/xemacs/site-packages/lisp/gnus/custom-load.elc
Done

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

EPITA / LRDE, 14-16 rue Voltaire   Tel.+33 (1) 53 14 59 47
94276 Le Kremlin-Bicêtre, France   Fax.+33 (1) 44 08 01 99   didier@xemacs.org

Re: Latest CVS Gnus just doesn't wanna play fair.

[Katsumi Yamaoka]

Hi,

>>>>> In <muxbslqgbyc.fsf@uzeb.lrde.epita.fr> 
>>>>>	Didier Verna <didier@lrde.epita.fr> wrote:

drv> Steve Youngs <youngs@xemacs.org> wrote:
drv>>   SY> C-c C-c not defined

[...]

drv> The weird stuff actually happens for me when compiling the autoloads:

This problem has been fixed in latest Gnus:

>>>>> In the list bugs@gnus.org, xemacs-beta@xemacs.org,
>>>>> Message-ID: <yosuae249sz8.fsf@jpl.org>
>>>>>	Katsumi Yamaoka <yamaoka@ga.sony.co.jp> wrote:

[...]

> XEmacs' autoload does not handle `define-derived-mode'.

> Here is a patch for the problem.  But it is no more than an ad-
> hoc solution, it should be fixed in XEmacs, I think.

> 2001-07-17  Katsumi Yamaoka  <yamaoka@jpl.org>

> 	* dgnushack.el (dgnushack-make-auto-load): Advise `make-autoload'
> 	to handle `define-derived-mode'.

Re: Latest CVS Gnus just doesn't wanna play fair.

[Didier Verna]

Katsumi Yamaoka <yamaoka@jpl.org> wrote:

> This problem has been fixed in latest Gnus:

	Nope. I was talking about a real `auto-autoloads.el' file, that you
make with "xemacs -batch -q -l autoload -f batch-update-directory". But
you're right, this should be fixed in XEmacs, and the recent patch to
message.el to use defined-derived-mode is at the origin of my problem. I'd
missed that one.

Thanks.

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

EPITA / LRDE, 14-16 rue Voltaire   Tel.+33 (1) 53 14 59 47
94276 Le Kremlin-Bicêtre, France   Fax.+33 (1) 44 08 01 99   didier@xemacs.org

Re: Latest CVS Gnus just doesn't wanna play fair.

[Didier Verna]

I wrote:

> Katsumi Yamaoka <yamaoka@jpl.org> wrote:
>
>> This problem has been fixed in latest Gnus:
>
> 	Nope. I was talking about a real `auto-autoloads.el' file, that you
> make with "xemacs -batch -q -l autoload -f batch-update-directory". But
> you're right, this should be fixed in XEmacs, and the recent patch to
> message.el to use defined-derived-mode is at the origin of my problem. I'd
> missed that one.

        Fixed in CVS (XEmacs 21.5).

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

EPITA / LRDE, 14-16 rue Voltaire   Tel.+33 (1) 53 14 59 47
94276 Le Kremlin-Bicêtre, France   Fax.+33 (1) 44 08 01 99   didier@xemacs.org

Re: Latest CVS Gnus just doesn't wanna play fair.

[Steve Youngs]

|--==> "drv" == Didier Verna <didier@lrde.epita.fr> writes:

  drv> Steve Youngs <youngs@xemacs.org> wrote:
  SY> C-c C-c not defined
  >>
  >>I broke this because I had created an 'auto-autoloads.el' and
  >>'custom-load.el' for my Gnus.  Remove them, and goodbye problems.
  >>
  >>I've been building waaay too many XEmacs packages, and got a little
  >>carried away. :-(

  drv> Steve, I'm experiencing the same problems even after the bug fix in
  drv> XEmacs's autoload.el I've just sent (and applied) to xemacs-patches. Something
  drv> *must* have changed in Gnus recently because I've been installing it as an
  drv> XEmacs package and making auto-autoloads and custom-loads for it for ages
  drv> before.

Just to let you know, I had no problems with current CVS.

-- 
|---<Steve Youngs>---------------<GnuPG KeyID: 9E7E2820>---|
|            XEmacs - It's not just an editor.             |
|                    It's a way of life.                   |
|------------------------------------<youngs@xemacs.org>---|