Why is UTF-8 subject base64 encoded? (ä)

Why is UTF-8 subject base64 encoded? (ä)

[Frank Schmitt]

Hi

I've told Gnus to use UTF-8 whenever I send a message which contains
non-ASCII characters by saying
(setq mm-coding-system-priorities '(mule-utf-8))
in .gnus.

However, there's one problem: Gnus now always uses base64 instead of
quoted printable to encode 8bit characters in the subject and other
headers. Is this a bug or wanted behavior?

-- 
Did you ever realize how much text fits in eighty columns? If you now consider
that a signature usually consists of up to four lines, this gives you enough
space to spread a tremendous amount of information with your messages. So seize
this opportunity and don't waste your signature with bullshit nobody will read.

Re: Why is UTF-8 subject base64 encoded? (ä)

[Katsumi Yamaoka]

Hi,

>>>>> In <u65jqmlff.fsf@mid.gehheimdienst.de>
>>>>>	Frank Schmitt <usereplyto@Frank-Schmitt.net> wrote:

> Hi

> I've told Gnus to use UTF-8 whenever I send a message which contains
> non-ASCII characters by saying
> (setq mm-coding-system-priorities '(mule-utf-8))
> in .gnus.

> However, there's one problem: Gnus now always uses base64 instead of
> quoted printable to encode 8bit characters in the subject and other
> headers. Is this a bug or wanted behavior?

rfc2047.el uses B encoding by default for the UTF-8 charset.
You can alter the behavior as following:

(eval-after-load "rfc2047"
  '(push '(utf-8 . Q) rfc2047-charset-encoding-alist))

I don't know whether that's a right thing, though.
-- 
Katsumi Yamaoka <yamaoka@jpl.org>

Re: Why is UTF-8 subject base64 encoded? (ä)

[Per Abrahamsen]

Katsumi Yamaoka <yamaoka@jpl.org> writes:

> I don't know whether that's a right thing, though.

I guess Q would be preferable for most latin scripts, and B for
everything else.

The best default would probably be to try both, and use whatever is
shortest.  I believe VM does that.

Re: Why is UTF-8 subject base64 encoded? (ä)

[Florian Weimer]

Per Abrahamsen <abraham@dina.kvl.dk> writes:

> The best default would probably be to try both, and use whatever is
> shortest.  I believe VM does that.

Gnus does it for text/* parts, too.

Re: Why is UTF-8 subject base64 encoded? (ä)

[Lars Magne Ingebrigtsen]

Per Abrahamsen <abraham@dina.kvl.dk> writes:

> The best default would probably be to try both, and use whatever is
> shortest.

Yup.  Fix in Gnus v5.10.3 (i. e., CVS).
-- 
(domestic pets only, the antidote for overdose, milk.)
  larsi@gnus.org * Lars Magne Ingebrigtsen