Gnus 5.6.38 and Emacs 20.3

Gnus 5.6.38 and Emacs 20.3

[Bill White]

Gnus v5.6.38, GNU Emacs 20.3.1 (i586-unicent-linux-gnu, X toolkit) of
Thu Aug 20 1998 on banshee

gnus-group-suspend works fine so far with 20.3.1, but C-c C-c (aka M-x
message-send-and-exit) returns the following message in the minibuffer
and doesn't sent the message:

----------------------------------------
Attempt to set a constant symbol: :-pos
----------------------------------------

Is there something more I could report to help debug things?

Thanks  -

Bill
-- 
Bill White
billw@wolfram.com

Re: Gnus 5.6.38 and Emacs 20.3

[Lars Magne Ingebrigtsen]

Bill White <billw@wolfram.com> writes:

> gnus-group-suspend works fine so far with 20.3.1, but C-c C-c (aka M-x
> message-send-and-exit) returns the following message in the minibuffer
> and doesn't sent the message:
> 
> ----------------------------------------
> Attempt to set a constant symbol: :-pos
> ----------------------------------------

`(setq debug-on-error t)', repeat the bug and mail me the resulting
backtrace.

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

Re: Gnus 5.6.38 and Emacs 20.3

[Harald Meland]

[Bill White]

> Gnus v5.6.38, GNU Emacs 20.3.1 (i586-unicent-linux-gnu, X toolkit) of
> Thu Aug 20 1998 on banshee
> 
> gnus-group-suspend works fine so far with 20.3.1, but C-c C-c (aka M-x
> message-send-and-exit) returns the following message in the minibuffer
> and doesn't sent the message:
> 
> ----------------------------------------
> Attempt to set a constant symbol: :-pos
> ----------------------------------------

I haven't seen this problem (yet, this is my second message using
20.3/5.6.38 :), but on the subject of Emacs 20.3: I had to add this
snippet to my .emacs in order for Gnus not to misunderstand the names
of my 8-bit-character-containing-mail-group-names:

;; And I don't want my .newsrc.eld MULEified, as this messes up mail
;; group names with 8-bit characters:
(setq file-coding-system-alist
      (nconc '(("\\.newsrc\\.eld\\'" no-conversion . no-conversion))
	     file-coding-system-alist))

Could Gnus maybe include somthing like that, or bind
`coding-system-for-read' appropriately before it reads our
"~/.newsrc.eld"s?

-- 
Harald

Re: Gnus 5.6.38 and Emacs 20.3

[Simon Josefsson]

Harald Meland <Harald.Meland@usit.uio.no> writes:

> ;; And I don't want my .newsrc.eld MULEified, as this messes up mail
> ;; group names with 8-bit characters:
> (setq file-coding-system-alist
>       (nconc '(("\\.newsrc\\.eld\\'" no-conversion . no-conversion))
> 	     file-coding-system-alist))
> 
> Could Gnus maybe include somthing like that, or bind
> `coding-system-for-read' appropriately before it reads our
> "~/.newsrc.eld"s?

I have this in my ~/.gnus:

;; I'm using ÅÄÖ in my .newsrc.eld
(setq gnus-startup-file-coding-system 'iso-8859-1)

/s

Re: Gnus 5.6.38 and Emacs 20.3

[Lars Magne Ingebrigtsen]

Harald Meland <Harald.Meland@usit.uio.no> writes:

> I haven't seen this problem (yet, this is my second message using
> 20.3/5.6.38 :), but on the subject of Emacs 20.3: I had to add this
> snippet to my .emacs in order for Gnus not to misunderstand the names
> of my 8-bit-character-containing-mail-group-names:

Gnus binds `coding-system-for-read' before reading the file(s) with
the value of `gnus-startup-file-coding-system'.  Which is `binary' by
default.  Should it be `no-conversion' instead?

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

Re: Gnus 5.6.38 and Emacs 20.3

[François Pinard]

Bill White <billw@wolfram.com> writes:

> Attempt to set a constant symbol: :-pos

> Is there something more I could report to help debug things?

Grrr!  I'm sure I've seen this problem before, a while ago in fact, and
I remember that the mere solution was to rename a few variables all over
some Emacs LISP source.  The sad thing is that I do not remember which,
nor why.  It was something like changing `...:-pos' to `...:-position',
maybe.  I likely reported the problem, whatever it was.  The trick Lars
gave you, that is "(setq debug-on-error t)", should help you identifying
where the problem occurred, at least.

-- 
François Pinard                            mailto:pinard@iro.umontreal.ca
Join the free Translation Project!    http://www.iro.umontreal.ca/~pinard

Re: Gnus 5.6.38 and Emacs 20.3

[Bill White]

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain, Size: 1197 bytes --]

Lars came up with the right answer. I had just upgraded to Emacs 20.3
(my first emacs upgrade) and just copied all the .el and .elc files
from 20.2's site-lisp dir to 20.3's. All I had to do was delete the
.elc files & byte-compile with 20.3. Problem solved!

Cheers -

bw
-- 
Bill White
billw@wolfram.com
http://www.wolfram.com/~billw


François Pinard <pinard@iro.umontreal.ca> writes:

> Bill White <billw@wolfram.com> writes:
> 
> > Attempt to set a constant symbol: :-pos
> 
> > Is there something more I could report to help debug things?
> 
> Grrr!  I'm sure I've seen this problem before, a while ago in fact, and
> I remember that the mere solution was to rename a few variables all over
> some Emacs LISP source.  The sad thing is that I do not remember which,
> nor why.  It was something like changing `...:-pos' to `...:-position',
> maybe.  I likely reported the problem, whatever it was.  The trick Lars
> gave you, that is "(setq debug-on-error t)", should help you identifying
> where the problem occurred, at least.
> 
> -- 
> François Pinard                            mailto:pinard@iro.umontreal.ca
> Join the free Translation Project!    http://www.iro.umontreal.ca/~pinard

Re: Gnus 5.6.38 and Emacs 20.3

[Hrvoje Niksic]

François Pinard <pinard@iro.umontreal.ca> writes:

> Bill White <billw@wolfram.com> writes:
> 
> > Attempt to set a constant symbol: :-pos
> 
> > Is there something more I could report to help debug things?
> 
> Grrr!  I'm sure I've seen this problem before, a while ago in fact,
> and I remember that the mere solution was to rename a few variables
> all over some Emacs LISP source.  The sad thing is that I do not
> remember which, nor why.  It was something like changing `...:-pos'
> to `...:-position', maybe.

No, it's `:-pos' to `colon-pos'.  The file in question is
`mail/mail-extr.el', and the problem occurs because Emacs 20 supports
unquoted keywords, and signals an error if you try to assign a
different value to them.

-- 
Hrvoje Niksic <hniksic@srce.hr> | Student at FER Zagreb, Croatia
--------------------------------+--------------------------------
I'm a Lisp variable -- bind me!

Re: Gnus 5.6.38 and Emacs 20.3

[SL Baur]

Lars Magne Ingebrigtsen <larsi@gnus.org> writes in ding@gnus.org:

> Gnus binds `coding-system-for-read' before reading the file(s) with
> the value of `gnus-startup-file-coding-system'.  Which is `binary' by
> default.  Should it be `no-conversion' instead?

`binary' should be correct.  `no-conversion' can change LFs to CR-LFs
and vice-versa.

Re: Gnus 5.6.38 and Emacs 20.3

[Karl Kleinpaste]

SL Baur <steve@xemacs.org> writes:
> `no-conversion' can change LFs to CR-LFs and vice-versa.

Let me see if I have this straight: `no-conversion' does a conversion?

Re: Gnus 5.6.38 and Emacs 20.3

[Hrvoje Niksic]

Karl Kleinpaste <karl@jprc.com> writes:

> SL Baur <steve@xemacs.org> writes:
> > `no-conversion' can change LFs to CR-LFs and vice-versa.
> 
> Let me see if I have this straight: `no-conversion' does a
> conversion?

Yup.  Splendid, isn't it?  Call it "hysterical reasons"!  :-(

-- 
Hrvoje Niksic <hniksic@srce.hr> | Student at FER Zagreb, Croatia
--------------------------------+--------------------------------
Sauron is alive in Argentina!

Re: Gnus 5.6.38 and Emacs 20.3

[Harold G. Stevenson]

i just got xemacs running and i  need to know how to set up my machine 
so that i can read mail and gnus. any help will be greatly appreciated.

from tucson-:))

harold

On 22 Aug 1998, Karl Kleinpaste wrote:

> SL Baur <steve@xemacs.org> writes:
> > `no-conversion' can change LFs to CR-LFs and vice-versa.
> 
> Let me see if I have this straight: `no-conversion' does a conversion?
> 

Re: Gnus 5.6.38 and Emacs 20.3

[Michael Welsh Duggan]

SL Baur <steve@xemacs.org> writes:

> Lars Magne Ingebrigtsen <larsi@gnus.org> writes in ding@gnus.org:
> 
> > Gnus binds `coding-system-for-read' before reading the file(s) with
> > the value of `gnus-startup-file-coding-system'.  Which is `binary' by
> > default.  Should it be `no-conversion' instead?
> 
> `binary' should be correct.  `no-conversion' can change LFs to CR-LFs
> and vice-versa.

Partially incorrect.  no-conversion doesn't change line endings.
binary is an alias for no-conversion.

-- 
Michael Duggan
(md5i@cs.cmu.edu)
.

Re: Gnus 5.6.38 and Emacs 20.3

[Hrvoje Niksic]

Michael Welsh Duggan <md5i@cs.cmu.edu> writes:

> SL Baur <steve@xemacs.org> writes:
> 
> > Lars Magne Ingebrigtsen <larsi@gnus.org> writes in ding@gnus.org:
> > 
> > > Gnus binds `coding-system-for-read' before reading the file(s) with
> > > the value of `gnus-startup-file-coding-system'.  Which is `binary' by
> > > default.  Should it be `no-conversion' instead?
> > 
> > `binary' should be correct.  `no-conversion' can change LFs to CR-LFs
> > and vice-versa.
> 
> Partially incorrect.  no-conversion doesn't change line endings.
> binary is an alias for no-conversion.

Not so under XEmacs.  I don't guarantee for FSF Emacs/MULE.

-- 
Hrvoje Niksic <hniksic@srce.hr> | Student at FER Zagreb, Croatia
--------------------------------+--------------------------------
A radioactive cat has eighteen half-lives.

Re: Gnus 5.6.38 and Emacs 20.3

[Harald Meland]

[Simon Josefsson]

> Harald Meland <Harald.Meland@usit.uio.no> writes:
> 
> > ;; And I don't want my .newsrc.eld MULEified, as this messes up mail
> > ;; group names with 8-bit characters:
> > (setq file-coding-system-alist
> >       (nconc '(("\\.newsrc\\.eld\\'" no-conversion . no-conversion))
> > 	     file-coding-system-alist))
> > 
> > Could Gnus maybe include somthing like that, or bind
> > `coding-system-for-read' appropriately before it reads our
> > "~/.newsrc.eld"s?
> 
> I have this in my ~/.gnus:
> 
> ;; I'm using ÅÄÖ in my .newsrc.eld
> (setq gnus-startup-file-coding-system 'iso-8859-1)

That doesn't work here -- it converts all instances of e.g. the single
iso-8859-1 character "Å" into the four characters "\305".  It might
work if you had this set up when you created your first
8-bit-character-containing-group-name, but that's not what I want.
-- 
Harald

Re: Gnus 5.6.38 and Emacs 20.3

[Harald Meland]

[Lars Magne Ingebrigtsen]

> Harald Meland <Harald.Meland@usit.uio.no> writes:
> 
> > I haven't seen this problem (yet, this is my second message using
> > 20.3/5.6.38 :), but on the subject of Emacs 20.3: I had to add this
> > snippet to my .emacs in order for Gnus not to misunderstand the names
> > of my 8-bit-character-containing-mail-group-names:
> 
> Gnus binds `coding-system-for-read' before reading the file(s) with
> the value of `gnus-startup-file-coding-system'.

Great.  Dunno how I got to believe my snippet was needed at all, as it
doesn't appear to be needed now.

-- 
Harald