Content-Transfer-Encoding

Content-Transfer-Encoding

[Shigeki Uno]

Hello.

I write a e-mail message in Japanese and send it as usual,
so "Content-Transfer-Encoding" line in header shows *quoted-printable*.
But some japanese can't read this *quoted-printable* encoded messages.

Following is the example I send to myself:
-----start of headers----- 
To: Shigeki Uno <shigeki@mediawars.ne.jp>
Subject: test 
From: Shigeki Uno <shigeki@mediawars.ne.jp> 
Date: 13 Jan 2000 18:29:33 +0900 
User-Agent: Gnus/5.0804 (Gnus v5.8.4) Emacs/20.5
Message-ID: <86hfgitj42.fsf@mediawars.ne.jp> 
MIME-Version: 1.0
Content-Type: text/plain; charset=iso-2022-jp
Content-Transfer-Encoding: quoted-printable <-------- HERE 

Here is sentences written in japanese...

-----end of headers-----

I want to encode messages "7bit" type automatically. 
How do I do that ?

Thanks.
-- 
Shigeki Uno <mailto:shigeki@mediawars.ne.jp>

Re: Content-Transfer-Encoding

[Katsumi Yamaoka]

>>>>> In <86d7r6tgmc.fsf@mediawars.ne.jp> 
>>>>>	Shigeki Uno <shigeki@mediawars.ne.jp> wrote:

> I write a e-mail message in Japanese and send it as usual,
> so "Content-Transfer-Encoding" line in header shows *quoted-printable*.
> But some japanese can't read this *quoted-printable* encoded messages.

[...]

> I want to encode messages "7bit" type automatically. 
> How do I do that ?

I was troubled with the same problem. :-)
Please try the following code in your .gnus file:

(setq mm-body-charset-encoding-alist
      '((iso-2022-jp . 7bit) (iso-2022-jp-2 . 7bit)))
-- 
~{I=8T?KC@~} <yamaoka@jpl.org>

Re: Content-Transfer-Encoding

[Shigeki Uno]

Katsumi Yamaoka <yamaoka@jpl.org> writes:
 
> (setq mm-body-charset-encoding-alist
>       '((iso-2022-jp . 7bit) (iso-2022-jp-2 . 7bit)))
 
Thanks to Yamaoka-san, It works great. *WOW*.

# This relate to mm-bodies.el e-lisp file, doesn't this ? 
# I couldn't find that. :-\

-------- result --------
To: Shigeki Uno <shigeki@mediawars.ne.jp> 
Subject: test
From: Shigeki Uno <shigeki@mediawars.ne.jp>
Date: 14 Jan 2000 01:03:53 +0900
User-Agent: Gnus/5.0804 (Gnus v5.8.4) Emacs/20.5
Message-ID: <86zouavtzq.fsf@mediawars.ne.jp>
MIME-Version: 1.0
Content-Type: text/plain; charset=iso-2022-jp
Lines: 7
Xref: nana outbox:346
----- end of result -----

BTW, If I'd like to read *quoted-printable* encoded e-mail messages 
someone sent, are there any way to solve this problem (say, decode 
messages) ?  No way ?

Some *quoted-printable* encoded mails are sometimes sent to me. 

Thanks.
-- 
Shigeki Uno <mailto:shigeki@mediawars.ne.jp>

Re: Content-Transfer-Encoding

[Katsumi Yamaoka]

[-- Attachment #1: Type: text/plain, Size: 1239 bytes --]

>>>>> In <86vh4xx5v9.fsf@mediawars.ne.jp> 
>>>>>	Shigeki Uno <shigeki@mediawars.ne.jp> wrote:
 
>> (setq mm-body-charset-encoding-alist
>>       '((iso-2022-jp . 7bit) (iso-2022-jp-2 . 7bit)))
 
> Thanks to Yamaoka-san, It works great. *WOW*.

> # This relate to mm-bodies.el e-lisp file, doesn't this ? 
> # I couldn't find that. :-\

It is defined in mm-bodies.el, but it looks like an internal variable.
We japanese want it to be an user option and the default value is set
as follows:

(defcustom mm-body-charset-encoding-alist
  '((iso-2022-jp . 7bit) (iso-2022-jp-2 . 7bit))
  "Alist of MIME charsets to encodings.
Valid encodings are `7bit', `8bit', `quoted-printable' and `base64'."
  :type '(repeat (cons (symbol :tag "charset")
		       (choice :tag "encoding"
			       (const 7bit)
			       (const 8bit)
			       (const quoted-printable)
			       (const base64))))
  :group 'mime)

> BTW, If I'd like to read *quoted-printable* encoded e-mail messages 
> someone sent, are there any way to solve this problem (say, decode 
> messages) ?  No way ?

> Some *quoted-printable* encoded mails are sometimes sent to me. 

Eh, can't you read such messages with Gnus 5.8.4?  I can do it without
any customizing nor modifications...

[-- Attachment #2: Type: text/plain, Size: 107 bytes --]

うの-ｓａｎ, ｃａｎ'ｔ ｙｏｕ ｒｅａｄ ｔｈｉｓ ｌｉｎｅ?

[-- Attachment #3: Type: text/plain, Size: 35 bytes --]

-- 
~{I=8T?KC@~} <yamaoka@jpl.org>

Re: Content-Transfer-Encoding

[Shigeki Uno]

Katsumi Yamaoka <yamaoka@jpl.org> writes:

> Eh, can't you read such messages with Gnus 5.8.4?  I can do it without
> any customizing nor modifications...

Well, some e-mails sent to one Mailing List causes unintelligible se-
quence of characters, so I looked into their headers and found that
_Content-Transfer-Encoding is Quoted-Printable_. These messages are
sent by one person and I can't look into which MUA he uses. Because
no X-Mailer info or something like that was shown in their headers. :-(
# Perhaps he uses MUA programmed by himself.

*That's why* I said I can't read such messages. Only his messages.

> うの-ｓａｎ, ｃａｎ'ｔ ｙｏｕ ｒｅａｄ ｔｈｉｓ ｌｉｎｅ?
Yes, I can read the above line without any problems. :-)

Later,
-- 
Shigeki Uno <mailto:shigeki@mediawars.ne.jp>
ICQ UIN   43741937

Re: Content-Transfer-Encoding

[Lars Magne Ingebrigtsen]

Katsumi Yamaoka <yamaoka@jpl.org> writes:

> It is defined in mm-bodies.el, but it looks like an internal variable.
> We japanese want it to be an user option and the default value is set
> as follows:
> 
> (defcustom mm-body-charset-encoding-alist
>   '((iso-2022-jp . 7bit) (iso-2022-jp-2 . 7bit))

I've now installed this in Gnus 5.8.5.

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

Re: Content-Transfer-Encoding

[Pavel Janik ml.]

   From: Lars Magne Ingebrigtsen <larsi@gnus.org>
   Date: 21 Apr 2000 15:22:01 +0200

Hi,

   > > (defcustom mm-body-charset-encoding-alist
   > >   '((iso-2022-jp . 7bit) (iso-2022-jp-2 . 7bit))
   > 
   > I've now installed this in Gnus 5.8.5.

hmm, can you please add (iso-8859-2 . 8bit)?
-- 
Pavel Janík ml.
Pavel.Janik@inet.cz

Re: Content-Transfer-Encoding

[Lars Magne Ingebrigtsen]

Pavel.Janik@inet.cz (Pavel Janik ml.) writes:

>    > > (defcustom mm-body-charset-encoding-alist
>    > >   '((iso-2022-jp . 7bit) (iso-2022-jp-2 . 7bit))
>    > 
>    > I've now installed this in Gnus 5.8.5.
> 
> hmm, can you please add (iso-8859-2 . 8bit)?

Is that the preferred encoding, though?  Why not qp?

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

Re: Content-Transfer-Encoding

[Pavel Janik ml.]

   From: Lars Magne Ingebrigtsen <larsi@gnus.org>
   Date: 22 Apr 2000 02:16:28 +0200

Hi,

   > > hmm, can you please add (iso-8859-2 . 8bit)?
   > 
   > Is that the preferred encoding, though?  Why not qp?

there are at least two good reasons:

        - there are too many people with not-MIME compliant browsers and they
          can not read =3D like stuff at all. 8bit is readable for
          them. Remember that there are crazy people who do not use Gnus and
          even Emacs...
        - if you send something with qp, the first well configured server does
          recode the stuff to 8bit :-) so it is better to directly send with
          8bit and you will have easily grep'able mail archive. I do not think
          that Gnus or anything in Emacs is good for grepping through my 100MB+
          mail archive...

The last reason can be that =3D is three bytes long, but = is only one byte. I
think you can find more resons for this...
-- 
Pavel Janík ml.
Pavel.Janik@inet.cz

Re: Content-Transfer-Encoding

[Lars Magne Ingebrigtsen]

Pavel.Janik@inet.cz (Pavel Janik ml.) writes:

>    > > hmm, can you please add (iso-8859-2 . 8bit)?
>    > 
>    > Is that the preferred encoding, though?  Why not qp?
> 
> there are at least two good reasons:
> 
>         - there are too many people with not-MIME compliant browsers and they
>           can not read =3D like stuff at all. 8bit is readable for
>           them. Remember that there are crazy people who do not use Gnus and
>           even Emacs...

Mail is a MIME world these days.  I don't think there are any
widely-used mail readers that don't grok QP.

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

content-transfer-encoding

[vvv]

[-- Attachment #1: Type: text/plain, Size: 1384 bytes --]

Hi,

rfc2045 has the following requirement:

 RFC> It should also be noted that, by definition, if a composite entity                                                           
 RFC> has a transfer-encoding value such as "7bit", but one of the enclosed                                                        
 RFC> entities has a less restrictive value such as "8bit", then either the                                                        
 RFC> outer "7bit" labelling is in error, because 8bit data are included,                                                          
 RFC> or the inner "8bit" labelling placed an unnecessarily high demand on                                                         
 RFC> the transport system because the actual included data were actually                                                          
 RFC> 7bit-safe.                                                                                                                   

but when gnus sends a multipart messages, it does not set the
content-transfer-encoding=8bit of the whole message when one of the
parts has 8bit (or content-transfer-encoding=binary when one of the
parts has binary CTE).

by default, the CTE, if omitted, is assumed to be 7bit (as stated in
the same RFC).

This message has a part with 8bit CTE, but gnus does not set the 8bit
CTE for the whole message.

Best,
v.

[-- Attachment #2: Type: text/plain, Size: 37 bytes --]

this part has a 8bit CTE:
проба пера