UTF-8 still a problem

UTF-8 still a problem

[Ichimusai]

Emacs 21.2.1
Gnus v0.15 (Oort)

Following the help from Kai in
<84smre8zw7.fsf@lucy.is.informatik.uni-duisburg.de> I have installed
ucs-tables and added the following lines in my .emacs:

(require 'ucs-tables)
(unify-8859-on-encoding-mode 1)

But nothing happens. When I view a UTF-8 encoded post in a newsgroup
it comes out at two byte chars for me, when I follow up to such
message, my follow up will contain the UTF-8 chars quoted and my own
in ISO-8859-1.

That's rather bad. I have gone through .emacs and .gnus and eliminated
all things that does anything with charset I can find, but to no
effect.

I would be very happy if someone could talk me through getting support
for UTF-8 to work. I would not mind to be able to post in it sometimes
myself either, such as when responding to someone who post in UTF-8.

-- 
AA #769 ICQ: 1645566 Yahoo: Ichimusai AOL: Ichimusai1972 MSN: Ichimusai 
IRC: Ichimusai#AmigaSWE@IRCnet URL: http://www.ichimusai.org/  
"The backbone of surprise is fusing speed with secrecy." 
    -- Von Clausewitz (1780-1831) 

Re: UTF-8 still a problem

[Ichimusai]

kai.grossjohann@gmx.net (Kai Großjohann) writes:

> Ichimusai <ichi@ichimusai.org> writes:
>
>> (require 'ucs-tables)
>> (unify-8859-on-encoding-mode 1)
>>
>> But nothing happens. When I view a UTF-8 encoded post in a newsgroup
>> it comes out at two byte chars for me, when I follow up to such
>> message, my follow up will contain the UTF-8 chars quoted and my own
>> in ISO-8859-1.
>
> What happens if you hit `1 g' on the article and enter utf-8 as the
> charset?

No difference at all.

-- 
AA #769 ICQ: 1645566 Yahoo: Ichimusai AOL: Ichimusai1972 MSN: Ichimusai 
IRC: Ichimusai#AmigaSWE@IRCnet URL: http://www.ichimusai.org/  
"I shall not waste my days in trying to prolong them." 
    -- Ian L. Fleming (1908-1964)

Re: UTF-8 still a problem

[Ichimusai]

kai.grossjohann@gmx.net (Kai Großjohann) writes:

> Ichimusai <ichi@ichimusai.org> writes:
>
>> kai.grossjohann@gmx.net (Kai Großjohann) writes:
>>
>>> What happens if you hit `1 g' on the article and enter utf-8 as the
>>> charset?
>>
>> No difference at all.
>
> Hm. Does Emacs itself grok UTF-8? Can you create a UTF-8 file and do
> C-x RET c utf-8 RET C-x C-f /path/to/the/file RET? Then you should
> be able to see if it works.

For some reason it doesn't. It makes no difference at all. There must
be something broken with the support in my Emacs, but I can't figure
out what it is...

-- 
AA #769 ICQ: 1645566 Yahoo: Ichimusai AOL: Ichimusai1972 MSN: Ichimusai 
IRC: Ichimusai#AmigaSWE@IRCnet URL: http://www.ichimusai.org/  
"If everything seems under control, you're just not going fast
enough." 
    -- Mario Andretti

Re: UTF-8 still a problem

[Ichimusai]

kai.grossjohann@gmx.net (Kai Großjohann) writes:

> Ichimusai <ichi@ichimusai.org> writes:
>
>> kai.grossjohann@gmx.net (Kai Großjohann) writes:
>>
>>> I don't understand what you mean by "difference". For a difference,
>>> two things are required, but I don't know what they are. (One of
>>> then is probably C-x RET c utf-8 C-x C-f /path/to/utf-8/file RET.)
>>
>> It looks as if I just opened it and viewed it in ISO 8859 without
>> trying to tell Emacs it is a utf-8 encoded file. The multibyte
>> characters are not changed into something intelligeble.
>
> Ah.  So you mean that
>
>     C-x RET c latin-1 RET C-x C-f /path/to/file RET
>
> and
>
>     C-x RET c utf-8 RET C-x C-f /path/to/file RET
>
> produce the same visible result?

Yes, both look like it is trying to do Latin-1.

> That's strange.  What kind of file are we talking about?

The test file you recommended earlier. And some posts I saved in UTF.

>>> Maybe post in gnu.emacs.help that UTF-8 isn't working and describe
>>> what you do, what you see, what you expect to see.
>>
>> I have trouble with even describeing the problem since I don't know
>> what is the problem. Am I doing something I should not? Or missing
>> to do something I should not. Is my version of Emacs requiring
>> something special in .emacs to turn on MULE or what it is that
>> handle different charsets.
>
> Emacs 21.2 has Mule turned on by default. So unless you do "emacs
> --unibyte" or set the variable default-enable-multibyte-characters,
> it will be on. (Even with those things, Mule will be on, but behave
> differently.)

Okay, I am not messing around with any of those.

> Hm.  Also (standard-display-european 1) will do bad things.  Don't
> use it.

I had that set before, but commented it out some fortnights ago. 

> What happens with the C-x RET c utf-8 RET C-x C-f /path/to/file RET
> command in an Emacs started with "emacs -q -no-site-file"?

Damn. Though I do not have all the glyphs, now the encoding looks
right.

I am running RedHat, is it possible they got a site-file somewhere
that is screwing this up? Where is such a file usually located?


-- 
AA #769 ICQ: 1645566 Yahoo: Ichimusai AOL: Ichimusai1972 MSN:
Ichimusai IRC: Ichimusai#AmigaSWE@IRCnet URL:
http://www.ichimusai.org/ We're almost equals I purr to show I love
you Want to smell my butt. -- Cat Haiku