prompt: "Select one of these safe coding systems"

prompt: "Select one of these safe coding systems"

[D. Goel]

hi

Here, emacs (21.2) always prompts for a "safe" coding system (when
trying to save the bdbb file)--->

,----
| 
| These default coding systems were tried:
|   iso-2022-7bit-unix iso-latin-1
| However, none of them safely encodes the target text.
| 
| Select one of the following safe coding systems:
|   utf-8 raw-text emacs-mule no-conversion. 
`----

So, I tried to set-language-environment (used to be english) to
Latin-1, but emacs continued to prompt the first option being utf-8.

So, I tried to setq default-buffer-file-coding-system to 'utf-8, but
emacs still continued to show the same prompt.

I then tried the same coding-system utf-8 with english as well, but
still got the same prompt.

So

[1] It seems I cannot get emacs to follow a system I want (utf-8) by
default.  It still tries other systems and then shows me utf-8 among
the list of choices.  What am i doing wrong?

[2] If emacs cannot "safely" encode, how do i ask it to please not
prompt me, but simply select the first from the "safe" "alternatives"
shows me?


[3] Any other hints or workarounds, considering this situation?
I really want to get rid of this prompt :(, esp. since the prompt
shows up for not me but for an infobot, which is supposed to be run
noninteractively.  I don't care too much how it stores any
international characters that may have shown up, as long as it stores
the 99.9% english part just fine---i just want emacs to select
somethigng and just do it. The main problem is that the bot gets stuck
in the minibuffer which is really bad for other bot functions.  BTW,
FWIW, emacs is running in console for the bot.

Thanks for any hints. 
DG                                 http://deego.gnufans.org/~deego/
--

Re: prompt: "Select one of these safe coding systems"

[Hubert Chan]

[-- Attachment #1: Type: text/plain, Size: 1021 bytes --]

>>>>> "D" == D Goel <deego@gnufans.org> writes:

D> hi Here, emacs (21.2) always prompts for a "safe" coding system (when
D> trying to save the bdbb file)--->

D> ,----
D> | 
D> | These default coding systems were tried: | iso-2022-7bit-unix
D> iso-latin-1 | However, none of them safely encodes the target text.
D> | 
D> | Select one of the following safe coding systems: | utf-8 raw-text
D> emacs-mule no-conversion.  `----

D> So, I tried to set-language-environment (used to be english) to
D> Latin-1, but emacs continued to prompt the first option being utf-8.

D> So, I tried to setq default-buffer-file-coding-system to 'utf-8, but
D> emacs still continued to show the same prompt.

[...]

It seems BBDB uses its own variable:

(setq bbdb-file-coding-system 'utf-8)

-- 
Hubert Chan <hubert@uhoreg.ca> - http://www.uhoreg.ca/
PGP/GnuPG key: 1024D/124B61FA
Fingerprint: 96C5 012F 5F74 A5F7 1FF7  5291 AF29 C719 124B 61FA
Key available at wwwkeys.pgp.net.   Encrypted e-mail preferred.

[-- Attachment #2: Type: application/pgp-signature, Size: 188 bytes --]

Re: prompt: "Select one of these safe coding systems"

[D. Goel]

Hubert Chan <hubert@uhoreg.ca> writes:
> 
> It seems BBDB uses its own variable:
> 
> (setq bbdb-file-coding-system 'utf-8)


That did solve my problem.  Thanks a lot Hubart.



(Answers to the general problem of getting emacs to not prompt for
"safe" coding system are still welcome :-)




DG                                 http://deego.gnufans.org/~deego/
--

Re: prompt: "Select one of these safe coding systems"

[Eli Zaretskii]

On 24 Nov 2002, D. Goel wrote:

> [1] It seems I cannot get emacs to follow a system I want (utf-8) by
> default.  It still tries other systems and then shows me utf-8 among
> the list of choices.  What am i doing wrong?

How did you try to achieve that?  The right way is to use 
prefer-coding-system, I think.

> [2] If emacs cannot "safely" encode, how do i ask it to please not
> prompt me, but simply select the first from the "safe" "alternatives"
> shows me?

IIRC, you can't.  Wrong encoding is deemed a disaster (because it could 
lose information and corrupt your precious files), so Emacs always prompts.  
The only exception is when the buffer's coding-system is undecided and 
the buffer can be encoded using the language environment's default 
encoding.  For example, if you start with a pure-ASCII text and then add 
Latin-1 characters to it, and your language environment is Latin-1, then 
Emacs will silently use Latin-1 encoding when you save the text.

> [3] Any other hints or workarounds, considering this situation?
> I really want to get rid of this prompt :(

Just don't type characters that cannot be encoded with the buffer's 
file-coding-system ;-)

> esp. since the prompt
> shows up for not me but for an infobot, which is supposed to be run
> noninteractively.

Non-interactive Lisp programs can bind coding-system-for-write to 
something appropriate before they invoke functions that write to files.

> I don't care too much how it stores any
> international characters that may have shown up, as long as it stores
> the 99.9% english part just fine---i just want emacs to select
> somethigng and just do it.

If the encoding of non-ASCII characters _really_ doesn't matter (but 
please think carefully about this and make a few experiments before you 
decide, since the results might surprise you), then bind 
coding-system-for-write to raw-text.

Re: prompt: "Select one of these safe coding systems"

[D. Goel]

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

> On 24 Nov 2002, D. Goel wrote:
> 
> > [1] It seems I cannot get emacs to follow a system I want (utf-8) by
> > default.  It still tries other systems and then shows me utf-8 among
> > the list of choices.  What am i doing wrong?
> 
> How did you try to achieve that?  The right way is to use 
> prefer-coding-system, I think.

ahh, i was not using that.  I was trying to setq
default-buffer-coding-system. 




Thanks for all your other hints, Eli.  Will try those out.