bug#12362: GNUS can no longer read UTF-8

bug#12362: GNUS can no longer read UTF-8

[jidanni]

Starting with

emacs-snapshot:
  Installed: 2:20120905-1
  Candidate: 2:20120905-1
  Version table:
 *** 2:20120905-1 0
        500 http://emacs.naquadah.org/ unstable/ Packages
        100 /var/lib/dpkg/status

Even the *Summary* buffers come up garbled.

Try reading even this group,
Newsgroups: gwene.cc.ptt.rss.geography
NNTP-Posting-Host: sea.gmane.org
Content-Type: text/html; charset=utf-8

bug#12362: GNUS can no longer read UTF-8

[Lars Ingebrigtsen]

jidanni@jidanni.org writes:

> Even the *Summary* buffers come up garbled.
>
> Try reading even this group,

Are you sure that the messages are valid?  This is the first subject:

=?utf-8?q?[=e6=96=b0=e8=81=9e]_=e8=bb=8a=e7=ab=99=e6=94=b9=e5=90=8d_=e4=b8=96=e5=8d=9a=e7=ab=99=e2=86=92=e5=8d=83=e7=94=b2=e7=ab=99=e3=80=81=e7=ab=b9=e7=a7=91=e7=ab=99=e2=86=92=e6=96=b0=e8=8e=8a?=

(rfc2047-decode-string
"=?utf-8?q?[=e6=96=b0=e8=81=9e]_=e8=bb=8a=e7=ab=99=e6=94=b9=e5=90=8d_=e4=b8=96=e5=8d=9a=e7=ab=99=e2=86=92=e5=8d=83=e7=94=b2=e7=ab=99=e3=80=81=e7=ab=b9=e7=a7=91=e7=ab=99=e2=86=92=e6=96=b0=e8=8e=8a?=")
=> "[新聾] \350۪\347˙改吭 世孺\347˙→孃甲\347˙、\347˹科\347˙→新讪"

Which contains invalid characters, I think?  Or perhaps not.  I'm not an
expert on ... that.

-- 
(domestic pets only, the antidote for overdose, milk.)
  http://lars.ingebrigtsen.no  *  Lars Magne Ingebrigtsen

bug#12362: GNUS can no longer read UTF-8

[jidanni]

The problem happens in ALL groups.
Reverting to
emacs-snapshot:
  Installed: 2:20120829-1
  Candidate: 2:20120905-1
fixes it.

bug#12362: GNUS can no longer read UTF-8

[Kenichi Handa]

In article <87vcfsf5rb.fsf@jidanni.org>, jidanni@jidanni.org writes:

> The problem happens in ALL groups.
> Reverting to
> emacs-snapshot:
>   Installed: 2:20120829-1
>   Candidate: 2:20120905-1
> fixes it.

That's because of my recent change to
quoted-printable-decode-region for speed up.  In this code,
I misunderstood the meaning of "(case-fold-search t)", and
handled only "[0-9A-Z]" (no lowercase letters).  RFC2045
also says:

 (1)   (General 8bit representation) Any octet, except a CR or
          LF that is part of a CRLF line break of the canonical
          (standard) form of the data being encoded, may be
          represented by an "=" followed by a two digit
          hexadecimal representation of the octet's value.  The
          digits of the hexadecimal alphabet, for this purpose,
          are "0123456789ABCDEF".  Uppercase letters must be
          used; lowercase letters are not allowed.
                ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

Anyway, I've just installed a fix to handle also 'a'..'f'
(in spite of RFC2045's definition).  Could you please try
again?

---
Kenichi Handa
handa@m17n.org

bug#12362: GNUS can no longer read UTF-8

[jidanni]

>>>>> "K" == Kenichi Handa <handa.kenichi@aist.go.jp> writes:
K> Could you please try again?
OK I'll try with next week's snapshot.

Also now when I do
$ emacs -f gnus
but not
$ emacs
I get a mysterious new buffer
*Compile-Log*
that contains this mysterious content

Warning: Use `with-current-buffer' rather than save-excursion+set-buffer
Warning: (lambda (x) ...) quoted with ' rather than with #'

With no indication in the world as to what files in the world it is
talking about!!

So I changed

(add-hook
 'gnus-startup-hook
 '(lambda ()

to

(add-hook
 'gnus-startup-hook
 #(lambda ()

like the stupid advice said and... of course fell for the stupid trick
like a dummy. I reverted it.

You can go ahead and check my http://jidanni.org/comp/configuration/

I don't think the warnings are referring to me.