Default coding system ?

Default coding system ?

[François Fleuret]

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain; charset=utf-8, Size: 944 bytes --]

Hi,

I read my mail with gnus, and from time to time, especially with spam
mails, gnus asks for a coding system with the following warning

,----
| These default coding systems were tried to encode text
| in the buffer ` *temp*':
|   iso-latin-1
| However, each of them encountered these problematic characters:
|   iso-latin-1: ¡ ¡ ¡ Ë Î Ì å Ë Î Ì ...
| The first problematic character is at point in the displayed buffer,
| and C-u C-x = will give information about it.
| 
| Select one of the following safe coding systems, or edit the buffer:
|   utf-8
| Or specify any other coding system
| on your risk of losing the problematic characters.
`----

I looked for a list of default coding systems with apropos-value
"iso-latin-1" but could not figure what list to change ... And this
coding system issue is such a mess that I can not find the courage to
go deeper in the documentations.

Any idea ?

Thanks in advance,

-- 
François Fleuret

Re: Default coding system ?

[Joakim Hove]

François Fleuret <francois.fleuret@inria.fr> writes:

> ... And this coding system issue is such a mess that I can not find
> the courage to go deeper in the documentations.

I really agree wholeheartedly - I love emacs and use it more than 50%
of my time all working days - but getting the Norwegian characters æøå
correct on all occasians (which I believe is a concept of coding
systems) - is just too hard.

Sorry I am only contributing to your 'rant' - and not providing any
real help :-(

Joakim


-- 
  /--------------------------------------------------------------------\
 / Joakim Hove  / hove@bccs.no  /  (55 5) 84076       |                 \
 | Unifob AS, Avdeling for Beregningsvitenskap (BCCS) | Stabburveien 18 |
 | CMU                                                | 5231 Paradis    |
 \ Thormøhlensgt.55, 5020 Bergen.                     | 55 91 28 18     /
  \--------------------------------------------------------------------/

Re: Default coding system ?

[Eli Zaretskii]

> From: =?iso-8859-1?q?Fran=E7ois_Fleuret?= <francois.fleuret@inria.fr>
> Newsgroups: gnu.emacs.help,gnus.emacs.gnus
> Date: 03 Dec 2003 13:22:56 +0100
> 
> I read my mail with gnus, and from time to time, especially with spam
> mails, gnus asks for a coding system with the following warning

Please tell when Emacs asks these questions.  They are generally asked
only when Emacs needs to write text to a file or send it to another
process.

> | Select one of the following safe coding systems, or edit the buffer:
> |   utf-8
> | Or specify any other coding system
> | on your risk of losing the problematic characters.
> `----
> 
> I looked for a list of default coding systems with apropos-value
> "iso-latin-1" but could not figure what list to change ...

Why didn't you select the only encoding suggested by Emacs, which is
UTF-8?  That'd do the job, I think, since the characters you showed
in your message are not part of ISO 8859-1.  Can you try that next
time and see if it solves the problem?

> And this
> coding system issue is such a mess that I can not find the courage to
> go deeper in the documentations.

Why do you need courage to read the documentation?  The manual doesn't
bite, you know.  So please do try reading it, and if something there
isn't clear, please complain on gnu.emacs.bug about documentation
quality.  That will allow us to make the docs better for the next
release.

TIA

Re: Default coding system ?

[Jesper Harder]

"Eli Zaretskii" <eliz@elta.co.il> writes:

> François Fleuret writes:
>
>> I read my mail with gnus, and from time to time, especially with spam
>> mails, gnus asks for a coding system with the following warning

You could try Gnus 5.10.  There has been quite a lot of changes in
this area since 5.9.

>> And this coding system issue is such a mess that I can not find the
>> courage to go deeper in the documentations.
>
> Why do you need courage to read the documentation?  The manual doesn't
> bite, you know.  So please do try reading it, and if something there
> isn't clear, please complain on gnu.emacs.bug about documentation
> quality.

I agree with François that it's a mess.  It don't really blame the
documentation -- it's more the design that's hard to understand and
use, i.e. the multibyte vs. unibyte distinction and character-set
vs. coding-system vs. mime-charset.

Also, stuff like CCL and MULE isn't particularly well documented.

Re: Default coding system ?

[Eli Zaretskii]

> Newsgroups: gnu.emacs.help
> From: Jesper Harder <harder@myrealbox.com>
> Date: Thu, 04 Dec 2003 20:10:43 +0100
> >
> > Why do you need courage to read the documentation?  The manual doesn't
> > bite, you know.  So please do try reading it, and if something there
> > isn't clear, please complain on gnu.emacs.bug about documentation
> > quality.
> 
> I agree with Francois that it's a mess.

Care to explain why?

> It don't really blame the documentation -- it's more the design
> that's hard to understand and use, i.e. the multibyte vs. unibyte
> distinction and character-set vs. coding-system vs. mime-charset.

It's a complex issue, so it's no surprise that it's hard to
understand.  However, understanding of the design is not (and should
not, IMHO) be an obstacle to efficient use.

> Also, stuff like CCL and MULE isn't particularly well documented.

So you do blame the documentation...

Re: Default coding system ?

[Jesper Harder]

"Eli Zaretskii" <eliz@elta.co.il> writes:

> Jesper Harder <harder@myrealbox.com> writes:
>
>> I agree with Francois that it's a mess.
>
> Care to explain why?

1. Non-unification of characters.  Completely baffling for most
   people.

2. It's iso-2022 based.  Iso-2022 concepts are strange and unfamiliar
   to non-CJK users.

3. Terminology.  No ordinary user knows what a »coding system« is,
   they might know what a »character set« is (in the MIME sense).  And
   it's confusing that the term »character set« in Emacs does not
   correspond to the ordinary meaning.

   Also, why invent new names for coding systems, when what most
   people will be familiar with are MIME names, e.g. korean-iso-8bit
   vs. euc-kr.

4. Conflation of coding system and EOL convention.  I don't know any
   other application than Emacs that mix these two concepts.

5. unibyte-mode.  A frequent source of errors.  XEmacs is even worse
   since you can compile it without MULE.

>> Also, stuff like CCL and MULE isn't particularly well documented.
>
> So you do blame the documentation...

Well, CCL isn't documented *at all* -- all you get is the BNF syntax
stuffed away in a docstring somewhere.

Re: Default coding system ?

[Kai Grossjohann]

Jesper Harder <harder@myrealbox.com> writes:

>    Also, why invent new names for coding systems, when what most
>    people will be familiar with are MIME names, e.g. korean-iso-8bit
>    vs. euc-kr.

Why did the MIME folks invent new names for coding systems, given that
Mule was available already?

;-)

Kai

Re: Default coding system ?

[François Fleuret]

Hi people,

Eli Zaretskii wrote on 03 Dec 2003 17:38:43 MET:

> Please tell when Emacs asks these questions.  They are generally
> asked only when Emacs needs to write text to a file or send it to
> another process.

It seems to happen when shell-command-to-string is called (I have a
function somewhere which invokes lynx to decode html embedded in mails
... I guess I should have done something cleaner in the first place)

> Why didn't you select the only encoding suggested by Emacs, which is
> UTF-8?

My original post is not very clear. The problem is that I'd like this
encoding choice to be done by default. And I do not have the slightest
idea what is the standard variable to set to specify that.

> The manual doesn't bite, you know.

YES IT DOES!

But please note that I do not blame emacs or the documentation in any
way. Terminal and key encoding in general are (in my opinion) the
expression of the dark side in the beautiful and harmonious world of
unix.

Regards,

-- 
François Fleuret

Re: Default coding system ?

[Eli Zaretskii]

> From: =?iso-8859-1?q?Fran=E7ois_Fleuret?= <francois.fleuret@inria.fr>
> Newsgroups: gnu.emacs.help
> Date: 05 Dec 2003 17:35:04 +0100
> 
> > Why didn't you select the only encoding suggested by Emacs, which is
> > UTF-8?
> 
> My original post is not very clear. The problem is that I'd like this
> encoding choice to be done by default.

It's not clear to me what you would like to be done by default in this
case.  Did you want Emacs to _always_ select UTF-8 or just in this
specific situation (and if the latter, then why)?

So please elaborate.