Error with latest Gnus

Error with latest Gnus

[Joerg Jaspert]

Hi

I now always get the following error if i look at a message (doesnt
matter in which group/which backend).
The message displays fine then.

Debugger entered--Lisp error: (invalid-function (macro lambda (group) "Find the real name of a foreign newsgroup." (\` (let ((gname (\, group))) (if (string-match "^[^:]+:" gname) (substring gname (match-end 0)) gname)))))
  gnus-group-real-name("nnimap+mail:INBOX.Mailinglisten.LWN")
  #[nil "ÂÃÄ\b!\"‰\x11‡" [gnus-newsgroup-name in-debian-devel-announce-group-p string-match "debian.devel.announce" gnus-group-real-name] 4]()
  run-hooks(gnus-select-article-hook)
  apply(run-hooks gnus-select-article-hook)
  (save-current-buffer (apply (quote run-hooks) funcs))
  gnus-run-hooks(gnus-select-article-hook)
  gnus-summary-display-article(216 nil)
  gnus-summary-select-article(nil nil pseudo)
  gnus-summary-scroll-up(1)
  call-interactively(gnus-summary-scroll-up)

Some hints please where i can search. I updated to latest Emacs and
Gnus (CVS). I tried older Emacs and Gnus (one month ago CVS, but it
still has an error, so i think its something on my system. But i dont
know where to look :() ).

Thx.
-- 
bye Joerg
<Sahneschnitter> Aquariophile: welches debian/ welche xfree version?
<Aquariophile> woody
<Aquariophile> Xfree version 86

Re: Error with latest Gnus

[Kevin Greiner]

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain; charset=utf-16be, Size: 3343 bytes --]

þÿ\0J\0o\0e\0r\0g\0 \0J\0a\0s\0p\0e\0r\0t\0 \0<\0j\0o\0e\0r\0g\0@\0g\0a\0n\0n\0e\0f\0f\0.\0d\0e\0>\0 \0w\0r\0i\0t\0e\0s\0:\0
\0
\0>\0 \0H\0i\0
\0>\0
\0>\0 \0I\0 \0n\0o\0w\0 \0a\0l\0w\0a\0y\0s\0 \0g\0e\0t\0 \0t\0h\0e\0 \0f\0o\0l\0l\0o\0w\0i\0n\0g\0 \0e\0r\0r\0o\0r\0 \0i\0f\0 \0i\0 \0l\0o\0o\0k\0 \0a\0t\0 \0a\0 \0m\0e\0s\0s\0a\0g\0e\0 \0(\0d\0o\0e\0s\0n\0t\0
\0>\0 \0m\0a\0t\0t\0e\0r\0 \0i\0n\0 \0w\0h\0i\0c\0h\0 \0g\0r\0o\0u\0p\0/\0w\0h\0i\0c\0h\0 \0b\0a\0c\0k\0e\0n\0d\0)\0.\0
\0>\0 \0T\0h\0e\0 \0m\0e\0s\0s\0a\0g\0e\0 \0d\0i\0s\0p\0l\0a\0y\0s\0 \0f\0i\0n\0e\0 \0t\0h\0e\0n\0.\0
\0>\0
\0>\0 \0D\0e\0b\0u\0g\0g\0e\0r\0 \0e\0n\0t\0e\0r\0e\0d\0-\0-\0L\0i\0s\0p\0 \0e\0r\0r\0o\0r\0:\0 \0(\0i\0n\0v\0a\0l\0i\0d\0-\0f\0u\0n\0c\0t\0i\0o\0n\0 \0(\0m\0a\0c\0r\0o\0 \0l\0a\0m\0b\0d\0a\0 \0(\0g\0r\0o\0u\0p\0)\0 \0"\0F\0i\0n\0d\0 \0t\0h\0e\0 \0r\0e\0a\0l\0 \0n\0a\0m\0e\0 \0o\0f\0 \0a\0 \0f\0o\0r\0e\0i\0g\0n\0 \0n\0e\0w\0s\0g\0r\0o\0u\0p\0.\0"\0 \0(\0\\0`\0 \0(\0l\0e\0t\0 \0(\0(\0g\0n\0a\0m\0e\0 \0(\0\\0,\0 \0g\0r\0o\0u\0p\0)\0)\0)\0 \0(\0i\0f\0 \0(\0s\0t\0r\0i\0n\0g\0-\0m\0a\0t\0c\0h\0 \0"\0^\0[\0^\0:\0]\0+\0:\0"\0 \0g\0n\0a\0m\0e\0)\0 \0(\0s\0u\0b\0s\0t\0r\0i\0n\0g\0 \0g\0n\0a\0m\0e\0 \0(\0m\0a\0t\0c\0h\0-\0e\0n\0d\0 \00\0)\0)\0 \0g\0n\0a\0m\0e\0)\0)\0)\0)\0)\0
\0>\0 \0 \0 \0g\0n\0u\0s\0-\0g\0r\0o\0u\0p\0-\0r\0e\0a\0l\0-\0n\0a\0m\0e\0(\0"\0n\0n\0i\0m\0a\0p\0+\0m\0a\0i\0l\0:\0I\0N\0B\0O\0X\0.\0M\0a\0i\0l\0i\0n\0g\0l\0i\0s\0t\0e\0n\0.\0L\0W\0N\0"\0)\0
\0>\0 \0 \0 \0#\0[\0n\0i\0l\0 \0"\0Â\0Ã\0Ä\0\b\0!\0\\0"\0‰\0\x11\0‡\0"\0 \0[\0g\0n\0u\0s\0-\0n\0e\0w\0s\0g\0r\0o\0u\0p\0-\0n\0a\0m\0e\0 \0i\0n\0-\0d\0e\0b\0i\0a\0n\0-\0d\0e\0v\0e\0l\0-\0a\0n\0n\0o\0u\0n\0c\0e\0-\0g\0r\0o\0u\0p\0-\0p\0 \0s\0t\0r\0i\0n\0g\0-\0m\0a\0t\0c\0h\0 \0"\0d\0e\0b\0i\0a\0n\0.\0d\0e\0v\0e\0l\0.\0a\0n\0n\0o\0u\0n\0c\0e\0"\0 \0g\0n\0u\0s\0-\0g\0r\0o\0u\0p\0-\0r\0e\0a\0l\0-\0n\0a\0m\0e\0]\0 \04\0]\0(\0)\0
\0>\0 \0 \0 \0r\0u\0n\0-\0h\0o\0o\0k\0s\0(\0g\0n\0u\0s\0-\0s\0e\0l\0e\0c\0t\0-\0a\0r\0t\0i\0c\0l\0e\0-\0h\0o\0o\0k\0)\0
\0>\0 \0 \0 \0a\0p\0p\0l\0y\0(\0r\0u\0n\0-\0h\0o\0o\0k\0s\0 \0g\0n\0u\0s\0-\0s\0e\0l\0e\0c\0t\0-\0a\0r\0t\0i\0c\0l\0e\0-\0h\0o\0o\0k\0)\0
\0>\0 \0 \0 \0(\0s\0a\0v\0e\0-\0c\0u\0r\0r\0e\0n\0t\0-\0b\0u\0f\0f\0e\0r\0 \0(\0a\0p\0p\0l\0y\0 \0(\0q\0u\0o\0t\0e\0 \0r\0u\0n\0-\0h\0o\0o\0k\0s\0)\0 \0f\0u\0n\0c\0s\0)\0)\0
\0>\0 \0 \0 \0g\0n\0u\0s\0-\0r\0u\0n\0-\0h\0o\0o\0k\0s\0(\0g\0n\0u\0s\0-\0s\0e\0l\0e\0c\0t\0-\0a\0r\0t\0i\0c\0l\0e\0-\0h\0o\0o\0k\0)\0
\0>\0 \0 \0 \0g\0n\0u\0s\0-\0s\0u\0m\0m\0a\0r\0y\0-\0d\0i\0s\0p\0l\0a\0y\0-\0a\0r\0t\0i\0c\0l\0e\0(\02\01\06\0 \0n\0i\0l\0)\0
\0>\0 \0 \0 \0g\0n\0u\0s\0-\0s\0u\0m\0m\0a\0r\0y\0-\0s\0e\0l\0e\0c\0t\0-\0a\0r\0t\0i\0c\0l\0e\0(\0n\0i\0l\0 \0n\0i\0l\0 \0p\0s\0e\0u\0d\0o\0)\0
\0>\0 \0 \0 \0g\0n\0u\0s\0-\0s\0u\0m\0m\0a\0r\0y\0-\0s\0c\0r\0o\0l\0l\0-\0u\0p\0(\01\0)\0
\0>\0 \0 \0 \0c\0a\0l\0l\0-\0i\0n\0t\0e\0r\0a\0c\0t\0i\0v\0e\0l\0y\0(\0g\0n\0u\0s\0-\0s\0u\0m\0m\0a\0r\0y\0-\0s\0c\0r\0o\0l\0l\0-\0u\0p\0)\0
\0>\0
\0>\0 \0S\0o\0m\0e\0 \0h\0i\0n\0t\0s\0 \0p\0l\0e\0a\0s\0e\0 \0w\0h\0e\0r\0e\0 \0i\0 \0c\0a\0n\0 \0s\0e\0a\0r\0c\0h\0.\0 \0I\0 \0u\0p\0d\0a\0t\0e\0d\0 \0t\0o\0 \0l\0a\0t\0e\0s\0t\0 \0E\0m\0a\0c\0s\0 \0a\0n\0d\0
\0>\0 \0G\0n\0u\0s\0 \0(\0C\0V\0S\0)\0.\0 \0I\0 \0t\0r\0i\0e\0d\0 \0o\0l\0d\0e\0r\0 \0E\0m\0a\0c\0s\0 \0a\0n\0d\0 \0G\0n\0u\0s\0 \0(\0o\0n\0e\0 \0m\0o\0n\0t\0h\0 \0a\0g\0o\0 \0C\0V\0S\0,\0 \0b\0u\0t\0 \0i\0t\0
\0>\0 \0s\0t\0i\0l\0l\0 \0h\0a\0s\0 \0a\0n\0 \0e\0r\0r\0o\0r\0,\0 \0s\0o\0 \0i\0 \0t\0h\0i\0n\0k\0 \0i\0t\0s\0 \0s\0o\0m\0e\0t\0h\0i\0n\0g\0 \0o\0n\0 \0m\0y\0 \0s\0y\0s\0t\0e\0m\0.\0 \0B\0u\0t\0 \0i\0 \0d\0o\0n\0t\0
\0>\0 \0k\0n\0o\0w\0 \0w\0h\0e\0r\0e\0 \0t\0o\0 \0l\0o\0o\0k\0 \0:\0(\0)\0 \0)\0.\0
\0>\0
\0
\0W\0e\0l\0l\0,\0 \0g\0n\0u\0s\0-\0g\0r\0o\0u\0p\0-\0r\0e\0a\0l\0-\0n\0a\0m\0e\0 \0w\0o\0r\0k\0e\0d\0 \0h\0e\0r\0e\0 \0u\0s\0i\0n\0g\0 \0t\0h\0e\0 \0g\0r\0o\0u\0p\0 \0t\0h\0a\0t\0 \0y\0o\0u\0
\0p\0r\0o\0v\0i\0d\0e\0d\0.\0 \0 \0T\0h\0e\0 \0d\0e\0b\0u\0g\0 \0m\0e\0s\0s\0a\0g\0e\0 \0a\0l\0s\0o\0 \0l\0o\0o\0k\0s\0 \0r\0a\0t\0h\0e\0r\0 \0b\0i\0z\0a\0r\0r\0e\0 \0s\0o\0 \0l\0e\0t\0'\0s\0 \0a\0s\0s\0u\0m\0e\0
\0t\0h\0a\0t\0 \0t\0h\0e\0 \0p\0r\0o\0b\0l\0e\0m\0 \0i\0s\0 \0i\0n\0 \0t\0h\0e\0 \0b\0y\0t\0e\0-\0c\0o\0m\0p\0i\0l\0e\0d\0 \0f\0i\0l\0e\0.\0 \0 \0T\0r\0y\0 \0d\0e\0l\0e\0t\0i\0n\0g\0,\0 \0o\0r\0 \0a\0t\0
\0l\0e\0a\0s\0t\0 \0r\0e\0n\0a\0m\0i\0n\0g\0,\0 \0t\0h\0e\0 \0g\0n\0u\0s\0-\0u\0t\0i\0l\0.\0e\0l\0c\0 \0f\0i\0l\0e\0 \0s\0o\0 \0t\0h\0a\0t\0 \0t\0h\0e\0 \0g\0n\0u\0s\0-\0u\0t\0i\0l\0.\0e\0l\0 \0f\0i\0l\0e\0
\0w\0i\0l\0l\0 \0b\0e\0 \0l\0o\0a\0d\0e\0d\0.\0 \0 \0T\0h\0a\0t\0 \0s\0h\0o\0u\0l\0d\0 \0e\0i\0t\0h\0e\0r\0 \0g\0i\0v\0e\0 \0y\0o\0u\0 \0a\0 \0b\0e\0t\0t\0e\0r\0 \0e\0r\0r\0o\0r\0 \0m\0e\0s\0s\0a\0g\0e\0 \0o\0r\0
\0f\0i\0x\0 \0t\0h\0e\0 \0p\0r\0o\0b\0l\0e\0m\0.\0 \0 \0I\0f\0 \0i\0t\0 \0i\0s\0 \0t\0h\0e\0 \0l\0a\0t\0e\0r\0,\0 \0y\0o\0u\0'\0l\0l\0 \0w\0a\0n\0t\0 \0t\0o\0 \0r\0e\0c\0o\0m\0p\0i\0l\0e\0
\0g\0n\0u\0s\0-\0u\0t\0i\0l\0.\0e\0l\0.\0
\0
\0K\0e\0v\0i\0n\0

Re: Error with latest Gnus

[Joerg Jaspert]

Kevin Greiner <kgreiner@xpediantsolutions.com> writes:

>> Debugger entered--Lisp error: (invalid-function (macro lambda (group) "Find the real name of a foreign newsgroup." (\` (let ((gname (\, group))) (if (string-match "^[^:]+:" gname) (substring gname (match-end 0)) gname)))))
>>   gnus-group-real-name("nnimap+mail:INBOX.Mailinglisten.LWN")
>>   #[nil "ÂÃÄ\b!\"‰\x11‡" [gnus-newsgroup-name in-debian-devel-announce-group-p string-match "debian.devel.announce" gnus-group-real-name] 4]()
>>   run-hooks(gnus-select-article-hook)
>>   apply(run-hooks gnus-select-article-hook)
>>   (save-current-buffer (apply (quote run-hooks) funcs))
>>   gnus-run-hooks(gnus-select-article-hook)
>>   gnus-summary-display-article(216 nil)
>>   gnus-summary-select-article(nil nil pseudo)
>>   gnus-summary-scroll-up(1)
>>   call-interactively(gnus-summary-scroll-up)
> Well, gnus-group-real-name worked here using the group that you
> provided.

It doesnt matter in which group, its everywhere.

> The debug message also looks rather bizarre so let's assume that the
> problem is in the byte-compiled file.  Try deleting, or at least
> renaming, the gnus-util.elc file so that the gnus-util.el file will be
> loaded.  That should either give you a better error message or fix the
> problem.  If it is the later, you'll want to recompile gnus-util.el.

Sorry, i should have written that. This is without byte-compile. With a
byte-compiled file you cant really read/understand the error message.

I already tried it without any .elc from Gnus. No difference. :()

-- 
bye Joerg
Die dümmsten Hähne haben die dicksten Eier.

Re: Error with latest Gnus

[Robert Epprecht]

Kevin Greiner <kgreiner@xpediantsolutions.com> writes:

> þÿ

[ snip ]

> 

Kevin, I can't read your message with CVS Gnus from a few minutes ago.

It looks like there is a garbage byte between all the letters.
Like this:  ^@K^@e^@v^@i^@n^@  (where ^@ is *one* character)

Robert

Re: Error with latest Gnus

[Kevin Greiner]

Robert Epprecht <epprecht@solnet.ch> writes:

> Kevin Greiner <kgreiner@xpediantsolutions.com> writes:
>
>> þÿ
>
> [ snip ]
>
>> 
>
> Kevin, I can't read your message with CVS Gnus from a few minutes ago.
>
> It looks like there is a garbage byte between all the letters.
> Like this:  ^@K^@e^@v^@i^@n^@  (where ^@ is *one* character)
>
> Robert

np, Joerg's original post contained control characters that I then
included in the reply.  They apparently trashed the charset selection.

Of course, if THIS message doesn't appear correctly, I have a problem.

Kevin

Re: Error with latest Gnus

[Kevin Greiner]

Joerg Jaspert <joerg@ganneff.de> writes:

> Kevin Greiner <kgreiner@xpediantsolutions.com> writes:
>
>>> Debugger entered--Lisp error: (invalid-function (macro lambda (group) "Find the real name of a foreign newsgroup." (\` (let ((gname (\, group))) (if (string-match "^[^:]+:" gname) (substring gname (match-end 0)) gname)))))
>>>   gnus-group-real-name("nnimap+mail:INBOX.Mailinglisten.LWN")
>>>   #[nil "..." [gnus-newsgroup-name in-debian-devel-announce-group-p string-match "debian.devel.announce" gnus-group-real-name] 4]()
>>>   run-hooks(gnus-select-article-hook)
>>>   apply(run-hooks gnus-select-article-hook)
>>>   (save-current-buffer (apply (quote run-hooks) funcs))
>>>   gnus-run-hooks(gnus-select-article-hook)
>>>   gnus-summary-display-article(216 nil)
>>>   gnus-summary-select-article(nil nil pseudo)
>>>   gnus-summary-scroll-up(1)
>>>   call-interactively(gnus-summary-scroll-up)
>> Well, gnus-group-real-name worked here using the group that you
>> provided.
>
> It doesnt matter in which group, its everywhere.
>
>> The debug message also looks rather bizarre so let's assume that the
>> problem is in the byte-compiled file.  Try deleting, or at least
>> renaming, the gnus-util.elc file so that the gnus-util.el file will be
>> loaded.  That should either give you a better error message or fix the
>> problem.  If it is the later, you'll want to recompile gnus-util.el.
>
> Sorry, i should have written that. This is without byte-compile. With a
> byte-compiled file you cant really read/understand the error message.
>
> I already tried it without any .elc from Gnus. No difference. :()

What does (macroexpand '(gnus-group-real-name
"nnimap+mail:INBOX.Mailinglisten.LWN")) return?

You might also try (progn (debug) (gnus-group-real-name
"nnimap+mail:INBOX.Mailinglisten.LWN")) so that you can step into the
macro expansion.  That can be fairly ugly but this may be an expansion
error rather than an evaluation error.

Kevin

Re: Error with latest Gnus

[Reiner Steib]

On Tue, May 06 2003, Kevin Greiner wrote:

> Robert Epprecht <epprecht@solnet.ch> writes:
[...]
>> Kevin, I can't read your message with CVS Gnus from a few minutes ago.
>>
>> It looks like there is a garbage byte between all the letters.
>> Like this:  ^@K^@e^@v^@i^@n^@  (where ^@ is *one* character)
[...]
> np, Joerg's original post contained control characters that I then
> included in the reply.  They apparently trashed the charset selection.

It's a known Bug in Emacs 21.3.  Ken'ichi HANDA fixed it[1] today in the
RC branch (which will probably be released as Emacs 21.4).

A simple workaround is, to replace the b0rked function
`sort-coding-systems' by a correct one, see [2].

Bye, Reiner.

[1] <URL:http://news.gmane.org/onethread.php?group=gmane.emacs.devel&
root=%3C200305060708.QAA21428@etlken.m17n.org%3E>

[2] <URL:http://theotp1.physik.uni-ulm.de/~ste/comp/emacs/
rs-sort-coding-systems-21-3-fix.el>
-- 
       ,,,
      (o o)
---ooO-(_)-Ooo--- PGP key available via WWW   http://rsteib.home.pages.de/

Re: Error with latest Gnus

[Joerg Jaspert]

Kevin Greiner <kgreiner@xpediantsolutions.com> writes:

>> Sorry, i should have written that. This is without byte-compile. With a
>> byte-compiled file you cant really read/understand the error message.
>> I already tried it without any .elc from Gnus. No difference. :()
> What does (macroexpand '(gnus-group-real-name
> "nnimap+mail:INBOX.Mailinglisten.LWN")) return?

(let ((gname "nnimap+mail:INBOX.Mailinglisten.LWN")) (if (string-match
"^[^:]+:" gname) (substring gname (match-end 0)) gname))

No beep, no error.

> You might also try (progn (debug) (gnus-group-real-name
> "nnimap+mail:INBOX.Mailinglisten.LWN")) so that you can step into the
> macro expansion.  That can be fairly ugly but this may be an expansion
> error rather than an evaluation error.

Debugger entered: nil
  (progn (debug) (gnus-group-real-name "nnimap+mail:INBOX.Mailinglisten.LWN"))
  eval((progn (debug) (gnus-group-real-name "nnimap+mail:INBOX.Mailinglisten.LWN")))
  eval-last-sexp-1(nil)
  eval-last-sexp(nil)
* call-interactively(eval-last-sexp)

Same as above.

Ive tried some other things, nothing worked. :(
(Deleted and recompiled *every* *.elc i found, deleted .newsrc.eld and
everything in ~/News/* to be sure its not just an error with that files).

My .gnus and .emacs are the same since mid of October, so i dont think
they are important here.

-- 
bye Joerg
2.5 million B.C.: OOG the Open Source Caveman develops the axe and
releases it under the GPL. The axe quickly gains popularity as a means
of crushing moderators heads.

Re: Error with latest Gnus

[Kevin Greiner]

Joerg Jaspert <joerg@ganneff.de> writes:

> Kevin Greiner <kgreiner@xpediantsolutions.com> writes:
>
>>> Sorry, i should have written that. This is without byte-compile. With a
>>> byte-compiled file you cant really read/understand the error message.
>>> I already tried it without any .elc from Gnus. No difference. :()
>> What does (macroexpand '(gnus-group-real-name
>> "nnimap+mail:INBOX.Mailinglisten.LWN")) return?
>
> (let ((gname "nnimap+mail:INBOX.Mailinglisten.LWN")) (if (string-match
> "^[^:]+:" gname) (substring gname (match-end 0)) gname))

What happens when you eval the let form?

Kevin

Re: Error with latest Gnus

[Joerg Jaspert]

Kevin Greiner <kgreiner@xpediantsolutions.com> writes:

>>>> Sorry, i should have written that. This is without byte-compile. With a
>>>> byte-compiled file you cant really read/understand the error message.
>>>> I already tried it without any .elc from Gnus. No difference. :()
>>> What does (macroexpand '(gnus-group-real-name
>>> "nnimap+mail:INBOX.Mailinglisten.LWN")) return?
>> (let ((gname "nnimap+mail:INBOX.Mailinglisten.LWN")) (if (string-match
>> "^[^:]+:" gname) (substring gname (match-end 0)) gname))
> What happens when you eval the let form?

"INBOX.Mailinglisten.LWN"

It works there. It still beeps and gives the error in the Summary
Buffer whenever i open a new mail/news.
Hmm. On Sunday i have enough time to delete my whole Gnus and Emacs
(with all packages).  Maybe it works after a reinstall, maybe i find
the reason then.

-- 
bye Joerg
<maxx> Aqua mach mal man brain....
<Aquariophile> maxx: schon probiert das gibts ned

Re: Error with latest Gnus

[Dave Love]

Joerg Jaspert <joerg@ganneff.de> writes:

> Debugger entered--Lisp error: (invalid-function (macro lambda (group) "Find the real name of a foreign newsgroup." (\` (let ((gname (\, group))) (if (string-match "^[^:]+:" gname) (substring gname (match-end 0)) gname)))))
>   gnus-group-real-name("nnimap+mail:INBOX.Mailinglisten.LWN")
>   #[nil "ÂÃÄ\b!\"\211^Q\207" [gnus-newsgroup-name in-debian-devel-announce-group-p string-match "debian.devel.announce" gnus-group-real-name] 4]()
>   run-hooks(gnus-select-article-hook)
>   apply(run-hooks gnus-select-article-hook)

Something in your hooks is broken, either compiled incorrectly
(without the macro being expanded) or it's trying to apply the macro.
By the look of the backtrace it isn't something from Gnus.

Re: Error with latest Gnus

[Dave Love]

Reiner Steib <4.uce.03.r.s@nurfuerspam.de> writes:

> It's a known Bug in Emacs 21.3.

I disagree.  The problem is that the utf-16 coding systems have a
mime-charset property.  It happens that I just fixed the Gnus charset
detection to cope, and I'm negotiating either to remove the property
from those systems in Emacs or to add another one which says whether
a coding system is appropriate for text parts.

> A simple workaround is, to replace the b0rked function
> `sort-coding-systems' by a correct one, see [2].

It isn't incorrect.  My change is a kludge to account for the fact
that utf-16 normally has higher priority than utf-8 and currently has
a mime-charset defined.  If you remove the latter, the issue goes
away, and that change probably should go now.