Re: Message - which charset?

[Dmitry Yaitskov <dimas@home.com>]

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

> Dmitry Yaitskov <dimas@home.com> writes:
> 
> > 1) the charset of a message, if it is a reply or a followup, would be
> > the same as the charset of the original message (somehow I supposed
> > that this would be the default behavior but I don't think it is on my
> > system).
> 
> If you are using a Mule-enabled Emacs, then that should be the case.

Thanks, but no, I am using a non-Mule emacs. I don't want to pay the
performance penalty for the very limited multilingual support that I
really need. Basically, besides English I only *sometimes* need
Russian support in news and email - which would work fine out of the
box if it were not for the 2 different coding sysems used for Russian.
Even with that, most problems are solved by a lisp module enabling to
select the coding system. And with some simple elisp, I can now
determine which of the 2 coding systems is used in a particular
incoming message. My only problem left is - I'd like to be able to
specify the charset "of my choice" for the message that I compose (for
the whole message or for the 1st part of a mime one), based on the
headers/body of the (1st part of the 1st) message being replied
to/followed up. (For new messages, the default Gnus charset works
fine.) Is it possible to set up a hook like that to tell Gnus which
charset to specify?

> > 2) do something in the message buffer depending on the charset of the
> > message. I wanted to use the message-setup-hook for this but don't
> > know how to make sure that I have the charset available in the message
> > headers when the message setup hook is called. Even if I reply to a
> > message with a specified charset, I don't see it in the reply's
> > headers although I have message-generate-headers-first set to t.
> 
> Well, there may be several charsets in an article, so there's no easy
> way to do that.
> 
> (mm-find-mime-charset-region (point-min) (point-max))
> 
> will return a list of charsets in the current buffer.

Thanks, I will try to use it. I am not sure though this would help as
I'd like the list of charsets in the message(s) being replied to, not
the current (new?) one. In other words: while in the original message
buffer, I can determine (using mm-find-mime-charset-region or
otherwise) what the charset in the reply should be. Now, I hit
reply/followup/whatever, and in the new message buffer need to tell
Gnus somehow to change its "default" charset for this message only to
the one that I just found out in step 1. Is there a way to achieve
this? Thanks.

-- 
Cheers,
-Dima.