charset=X-UNKNOWN

charset=X-UNKNOWN

[Karl Eichwalder]

Pine users have the habit to send messages with such header lines:

    MIME-Version: 1.0
    Content-Type: text/plain; charset=X-UNKNOWN
    Content-Transfer-Encoding: quoted-printable

But if you've a cloaser look at the message body, you'll notice that the
mail actually is written as:

    MIME-Version: 1.0
    Content-Type: text/plain; charset=iso-8859-1
    Content-Transfer-Encoding: 8bit

Unfortunately, Gnus obeys the original haeader lines and displays
special characters (umlauts, accented characters) in the backslashed
notation ;-(

It there an easy way to tell Emacs 20.4 with PGnus 0.95, to be a little
bit smarter (display 8bit as 8bit)?

[I hope, it's clear, what I want so say.]

-- 
                                             work   :        ke@suse.de
Karl Eichwalder                              private: ke@gnu.franken.de

Re: charset=X-UNKNOWN

[Lars Magne Ingebrigtsen]

Karl Eichwalder <ke@gnu.franken.de> writes:

> Pine users have the habit to send messages with such header lines:
> 
>     MIME-Version: 1.0
>     Content-Type: text/plain; charset=X-UNKNOWN
>     Content-Transfer-Encoding: quoted-printable
> 
> But if you've a cloaser look at the message body, you'll notice that the
> mail actually is written as:
> 
>     MIME-Version: 1.0
>     Content-Type: text/plain; charset=iso-8859-1

You could define "x-unknown" as a synonym for "iso-8859-1" in
`mm-charset-synonym-alist'. 

>     Content-Transfer-Encoding: 8bit
> 
> Unfortunately, Gnus obeys the original haeader lines and displays
> special characters (umlauts, accented characters) in the backslashed
> notation ;-(
> 
> It there an easy way to tell Emacs 20.4 with PGnus 0.95, to be a little
> bit smarter (display 8bit as 8bit)?

But it can't possibly guess that a CTE that says that it is qp is
really something else.  

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

Re: charset=X-UNKNOWN

[Karl Eichwalder]

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

|   You could define "x-unknown" as a synonym for "iso-8859-1" in
|   `mm-charset-synonym-alist'. 

This works great so far -- I'm wondering about drawbacks ;) 

|   But it can't possibly guess that a CTE that says that it is qp is
|   really something else.

Yes -- I fear I've to tell them all to use valid Mime headers, sigh...

Thanks for your answer!

-- 
                                             work   :        ke@suse.de
Karl Eichwalder                              private: ke@gnu.franken.de

Re: charset=X-UNKNOWN

[Shenghuo ZHU]

>>>>> "LMI" == Lars Magne Ingebrigtsen <larsi@gnus.org> writes:

LMI> Karl Eichwalder <ke@gnu.franken.de> writes:
>> Pine users have the habit to send messages with such header lines:
>> 
>> MIME-Version: 1.0
>> Content-Type: text/plain; charset=X-UNKNOWN
>> Content-Transfer-Encoding: quoted-printable
>> 
>> But if you've a cloaser look at the message body, you'll notice that the
>> mail actually is written as:
>> 
>> MIME-Version: 1.0
>> Content-Type: text/plain; charset=iso-8859-1

LMI> You could define "x-unknown" as a synonym for "iso-8859-1" in
LMI> `mm-charset-synonym-alist'. 

It would be better to put 'x-unknown in `gnus-newsgroup-ignored-charsets'.  
Let Gnus "guess" the actual charset.

(setq gnus-newsgroup-ignored-charsets (cons 'x-unknown
        gnus-newsgroup-ignored-charsets))

-- 
Shenghuo ZHU

Re: charset=X-UNKNOWN

[Karl Eichwalder]

Shenghuo ZHU <zsh@cs.rochester.edu> writes:

|   It would be better to put 'x-unknown in `gnus-newsgroup-ignored-charsets'.  
|   Let Gnus "guess" the actual charset.
|   
|   (setq gnus-newsgroup-ignored-charsets (cons 'x-unknown
|           gnus-newsgroup-ignored-charsets))

This sounds good -- but when I tried it, Gnus hangs while reading the
active file.  This works:

(eval-after-load "mm-util"
  '(add-to-list 'mm-charset-synonym-alist
               '(x-unknown . iso-8859-1)))

But your example or this one does not:

(eval-after-load "gnus-sum"
  (setq gnus-newsgroup-ignored-charsets
        (cons 'x-unknown gnus-newsgroup-ignored-charsets)))

-- 
                                             work   :        ke@suse.de
Karl Eichwalder                              private: ke@gnu.franken.de

Re: charset=X-UNKNOWN

[Shenghuo ZHU]

>>>>> "KE" == Karl Eichwalder <ke@gnu.franken.de> writes:

 Shenghuo ZHU <zsh@cs.rochester.edu> writes:
 |   It would be better to put 'x-unknown in `gnus-newsgroup-ignored-charsets'.  
 |   Let Gnus "guess" the actual charset.
 |   
 |   (setq gnus-newsgroup-ignored-charsets (cons 'x-unknown
 |           gnus-newsgroup-ignored-charsets))

KE> This sounds good -- but when I tried it, Gnus hangs while reading
KE> the active file.  This works:

This is weird. There is no charset stuff in the active file, I guess.

KE> (eval-after-load "mm-util"
KE>   '(add-to-list 'mm-charset-synonym-alist
KE>                '(x-unknown . iso-8859-1)))

KE> But your example or this one does not:

KE> (eval-after-load "gnus-sum"
KE>   (setq gnus-newsgroup-ignored-charsets
KE>         (cons 'x-unknown gnus-newsgroup-ignored-charsets)))

I use 
        (setq gnus-newsgroup-ignored-charsets '(unknown-8bit x-unknown))
which works for me.

-- 
Shenghuo ZHU

Re: charset=X-UNKNOWN

[Lars Magne Ingebrigtsen]

Shenghuo ZHU <zsh@cs.rochester.edu> writes:

>  |   (setq gnus-newsgroup-ignored-charsets (cons 'x-unknown
>  |           gnus-newsgroup-ignored-charsets))

I've now added x-unknown to the default value of this variable.

> KE> This sounds good -- but when I tried it, Gnus hangs while reading
> KE> the active file.  This works:
> 
> This is weird. There is no charset stuff in the active file, I guess.

I didn't think so either...

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