Unknown character encodings in .emacs

Unknown character encodings in .emacs

[Alan E. Davis]

[-- Attachment #1: Type: text/plain, Size: 1917 bytes --]

I have stumbled on a problem in my ~/.emacs.el .  I am at a loss to even
explain it.  It seems to me like it might be related to character encoding.


For a number of months, or even years, I have been encountering messages
asking what encoding I wish to use to save files.  I have gotten into the
habit of saying "utf-8" because it gets me out of there, but I haven't a
clue what this might do or mean.

Now, my .emacs has gotten buggered.  I pasted some snippets from a web
site.  Later one, those snippets, it seems, have gotten expressed as garbage
characters.  I'be been poking around, but don't have a clue where to start.
I never did understand encoding---I mean, what it does, what parts of the
sysem are involved.  I cannot even get to first base debugging this.

As an example, I have an abbreviation table that has been in my init file
for 15 years, and now reads as gibberish.

(define-abbrev-table 'global-abbrev-table '(
    ("%`" "‚\°" nil 0)
    ("a`" "á" nil 1)
    ("A`" "Á" nil 1)
    ("o`" "ó" nil 2)
    ("O`" "ÃÂ"" nil 1)
    ("u`" "ú" nil 2)
    ("U`" "Ú" nil 1)
    ("n`" "ñ" nil 0)
    ("e`" "ÃÂ(c)" nil 4)
    ("E`" "É" nil 1)
    ("m`" "µ" nil 0)
    ("p`" "¶" nil 0)
    ("s`" "§" nil 0)
    ("y`" "\245" nil 0)   ;; ¥
    ("?`" "¿" nil 0)
    ("!`" "¡" nil 0)
    (":`" "÷" nil 0)
    ("<`" "«" nil 0)
    (">`" "\273" nil 0)
    ("/`" "\370" nil 0)   ;;  ø
    ("!`" "\241" nil 0)   ;;  ¡
    ("b`" "\337" nil 0)   ;;  ß
    ("c`" "\242" nil 0)   ;;  ¢
    ))

Can someone point me to an explanation of why this happened, and how to fix
it?

Thank you very much,

Alan


-- 
Alan Davis  :    lngndvs@gmail.com

"It's never a matter of liking or disliking ..."
---Santa Ynez Chumash Medicine Man

"We have no art. We do everything as well as we can." ---Balinese saying

[-- Attachment #2: Type: text/html, Size: 3150 bytes --]

Re: Unknown character encodings in .emacs

[Peter Dyballa]

Am 07.06.2008 um 15:32 schrieb Alan E. Davis:

> As an example, I have an abbreviation table that has been in my  
> init file
> for 15 years, and now reads as gibberish.


Check which encoding is used in a backup of your init file. Then open  
it with a prefix command: C-x RET c <the encoding> RET and then C-x C- 
f or e in dired-mode to open the init file. Now put into its first line:

	;; -*- mode: Emacs-Lisp; coding: utf-8; -*-

and with a prefix command: C-x RET c utf-8 RET C-x C-s. This will  
make sure that the file is read into GNU Emacs in its original  
encoding and is saved as UTF-8 *and* will ever be re-opened in UTF-8  
encoding.


The kind your abbreviation table looks like, makes it obvious that  
its UTF-8 file contents was opened in some 8-bit mode and again saved  
as UTF-8. 8-bit characters ("extended" US-ASCII) are in UTF-8 encoded  
as two 8-bit "characters." When you re-read them in 8-bit mode (1  
char = 1 char, while in UTF-8 it would be 2 char = 1 char) and save  
them again as UTF-8 they become four 8-bit "characters."

--
Greetings

   Pete

We have to expect it, otherwise we would be surprised.

Re: Unknown character encodings in .emacs

[Nikolaj Schumacher]

"Alan E. Davis" <lngndvs@gmail.com> wrote:

> I'be been poking around, but don't have a clue where to start.
> I never did understand encoding

Well, that's a clue where to start, I suppose. :)

>     ("a`" "á" nil 1)

Looks like the file has been written using UTF-8 and read using
Latin-1.  Try adding:

(set-language-environment "UTF-8")

However, I believe this shouldn't be necessary if the system is
configured correctly.  What OS do you use?


regards,
Nikolaj Schumacher

Re: Unknown character encodings in .emacs

[Alan E. Davis]

[-- Attachment #1: Type: text/plain, Size: 2491 bytes --]

Thank you very much, Nikolaj and Peter:

As of now, I see:
GDM_LANG=en_US.UTF-8
LANG=en_US.UTF-8

Emacs can read .emacs.el now, I have incrementally deleted or commented out
parts that emacs tripped over when reading .emacs . I also placed the header
into the file that peter suggested.

       ;; -*- mode: Emacs-Lisp; coding: utf-8; -*-

Now the question comes up: what is the most appropriate coding system to be
using?  I guess that is a matter for another post.  Or is it?  Should I be
using a two-byte language encoding at all?  All characters I ordinarily use
are availalbe in a latin-1 encoding.  The original reason I turned to a text
editor is the need for raw text input and output for language data (I was
working on a lexicon of Chuukese at the time).

I also was put off by the Windows encodings that emacs recommended when a
file was being saved.  I avoid all proprietary file formats, so felt  a bit
taken aback by that suggestion.

I have started to read more about this problem/issue.  I have gotten away
with a sloppy .emacs.el file for a while, and sloppy conventions on
encoding.  I need to look into it now.  However, it is not something I would
choose to spend a great deal of time on: it's only of secondary importance
to my work.

It is suspicious that the problem only recently happened.  The first
instance I discovered was a cut and paste from firefox of snippets for a
.emacs.  I've been using this same .emacs.el (with many changes and
additions) for 15 years, all on GNU/Linux systems.  I have had to adapt, but
nothing this radical.



Thank you again,

Alan



On Sun, Jun 8, 2008 at 4:38 AM, Nikolaj Schumacher <n_schumacher@web.de>
wrote:

> "Alan E. Davis" <lngndvs@gmail.com> wrote:
>
> > I'be been poking around, but don't have a clue where to start.
> > I never did understand encoding
>
> Well, that's a clue where to start, I suppose. :)
>
> >     ("a`" "á" nil 1)
>
> Looks like the file has been written using UTF-8 and read using
> Latin-1.  Try adding:
>
> (set-language-environment "UTF-8")
>
> However, I believe this shouldn't be necessary if the system is
> configured correctly.  What OS do you use?
>
>
> regards,
> Nikolaj Schumacher
>



-- 
Alan Davis, Kagman High School, Saipan lngndvs@gmail.com

"It's never a matter of liking or disliking ..."
---Santa Ynez Chumash Medicine Man

"We have no art. We do everything as well as we can." ---Balinese saying

[-- Attachment #2: Type: text/html, Size: 3234 bytes --]

Re: Unknown character encodings in .emacs

[Alan E. Davis]

[-- Attachment #1: Type: text/plain, Size: 3137 bytes --]

PARTLY SOLVED.

Thank you to people who responded.  I have gotten part way through this.
Temporarily, I am able to boot.  I also have backups of my .emacs.el, so can
check them out too.

Thank you again,

Alan

On Sun, Jun 8, 2008 at 8:36 AM, Alan E. Davis <lngndvs@gmail.com> wrote:

> Thank you very much, Nikolaj and Peter:
>
> As of now, I see:
> GDM_LANG=en_US.UTF-8
> LANG=en_US.UTF-8
>
> Emacs can read .emacs.el now, I have incrementally deleted or commented out
> parts that emacs tripped over when reading .emacs . I also placed the header
> into the file that peter suggested.
>
>        ;; -*- mode: Emacs-Lisp; coding: utf-8; -*-
>
> Now the question comes up: what is the most appropriate coding system to be
> using?  I guess that is a matter for another post.  Or is it?  Should I be
> using a two-byte language encoding at all?  All characters I ordinarily use
> are availalbe in a latin-1 encoding.  The original reason I turned to a text
> editor is the need for raw text input and output for language data (I was
> working on a lexicon of Chuukese at the time).
>
> I also was put off by the Windows encodings that emacs recommended when a
> file was being saved.  I avoid all proprietary file formats, so felt  a bit
> taken aback by that suggestion.
>
> I have started to read more about this problem/issue.  I have gotten away
> with a sloppy .emacs.el file for a while, and sloppy conventions on
> encoding.  I need to look into it now.  However, it is not something I would
> choose to spend a great deal of time on: it's only of secondary importance
> to my work.
>
> It is suspicious that the problem only recently happened.  The first
> instance I discovered was a cut and paste from firefox of snippets for a
> .emacs.  I've been using this same .emacs.el (with many changes and
> additions) for 15 years, all on GNU/Linux systems.  I have had to adapt, but
> nothing this radical.
>
>
>
> Thank you again,
>
> Alan
>
>
>
> On Sun, Jun 8, 2008 at 4:38 AM, Nikolaj Schumacher <n_schumacher@web.de>
> wrote:
>
>> "Alan E. Davis" <lngndvs@gmail.com> wrote:
>>
>> > I'be been poking around, but don't have a clue where to start.
>> > I never did understand encoding
>>
>> Well, that's a clue where to start, I suppose. :)
>>
>> >     ("a`" "á" nil 1)
>>
>> Looks like the file has been written using UTF-8 and read using
>> Latin-1.  Try adding:
>>
>> (set-language-environment "UTF-8")
>>
>> However, I believe this shouldn't be necessary if the system is
>> configured correctly.  What OS do you use?
>>
>>
>> regards,
>> Nikolaj Schumacher
>>
>
>
>
> --
> Alan Davis, Kagman High School, Saipan lngndvs@gmail.com
>
> "It's never a matter of liking or disliking ..."
> ---Santa Ynez Chumash Medicine Man
>
> "We have no art. We do everything as well as we can." ---Balinese saying
>
>


-- 
Alan Davis, Kagman High School, Saipan lngndvs@gmail.com

"It's never a matter of liking or disliking ..."
---Santa Ynez Chumash Medicine Man

"We have no art. We do everything as well as we can." ---Balinese saying

[-- Attachment #2: Type: text/html, Size: 4325 bytes --]

Re: Unknown character encodings in .emacs

[Giorgos Keramidas]

On Sat, 7 Jun 2008 23:32:32 +1000, "Alan E. Davis" <lngndvs@gmail.com> wrote:
> I have stumbled on a problem in my ~/.emacs.el .  I am at a loss to even
> explain it.  It seems to me like it might be related to character encoding.
>
>
> For a number of months, or even years, I have been encountering messages
> asking what encoding I wish to use to save files.  I have gotten into the
> habit of saying "utf-8" because it gets me out of there, but I haven't a
> clue what this might do or mean.
>
> Now, my .emacs has gotten buggered.  I pasted some snippets from a web
> site.  Later one, those snippets, it seems, have gotten expressed as garbage
> characters.  I'be been poking around, but don't have a clue where to start.
> I never did understand encoding---I mean, what it does, what parts of the
> sysem are involved.  I cannot even get to first base debugging this.
>
> As an example, I have an abbreviation table that has been in my init file
> for 15 years, and now reads as gibberish.
>
> (define-abbrev-table 'global-abbrev-table '(
>     ("%`" "‚\°" nil 0)
>     ("a`" "á" nil 1)
[...]
>     ("c`" "\242" nil 0)   ;;  ¢
>     ))
>
> Can someone point me to an explanation of why this happened, and how to fix
> it?

Which Emacs version are you using?

Re: Unknown character encodings in .emacs

[Alan]

On Jun 8, 12:04 am, Giorgos Keramidas <keram...@ceid.upatras.gr>
wrote:

> Which Emacs version are you using?

emacs-snaptshot on Ubuntu GNU/Linux 8.04.  I seldom use emacs 21 or
emacs 22.  Not at all over the past few months, but they are installed
because of dependencies.

Thank you.

Alan