ispell change

ispell change

[Kai Großjohann]

I use the following code which allows me to use ispell in a Latin-9
environment:

(add-to-list 'ispell-dictionary-alist
	     '("ndeutsch8-15"
	       "[a-zA-Z\304\326\334\344\366\337\374]"
	       "[^a-zA-Z\304\326\334\344\366\337\374]"
	       "[']" t
	       ("-C" "-d" "ndeutsch")
	       "~latin1" iso-8859-15))

Thoughts?

kai
-- 
Silence is foo!

Re: ispell change

[Eli Zaretskii]

> From: Kai.Grossjohann@CS.Uni-Dortmund.DE
> Date: Fri, 10 May 2002 17:51:54 +0200
> 
> I use the following code which allows me to use ispell in a Latin-9
> environment:
> 
> (add-to-list 'ispell-dictionary-alist
> 	     '("ndeutsch8-15"
> 	       "[a-zA-Z\304\326\334\344\366\337\374]"
> 	       "[^a-zA-Z\304\326\334\344\366\337\374]"
> 	       "[']" t
> 	       ("-C" "-d" "ndeutsch")
> 	       "~latin1" iso-8859-15))
> 
> Thoughts?

Can we assume with high probability that `-d ndeutsch' will not pick
up the wrong dictionary?  I think someone (Gerd?) said when a similar
issue popped up in the past that `ndeutsch' is sometimes used for the
German new-orthography dictionaries.

Also, shouldn't \274 and \275 be part of the CASECHARS member of this
list?  Or are those characters unused in German?  (My German is so
rusty that I cannot be sure.)

Re: ispell change

[Andreas Schwab]

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

|> Also, shouldn't \274 and \275 be part of the CASECHARS member of this
|> list?  Or are those characters unused in German?  (My German is so
|> rusty that I cannot be sure.)

German does not use the œ character, it only has ö, which can be written
as the two character sequence oe where ö is not available.

Andreas.

-- 
Andreas Schwab, SuSE Labs, schwab@suse.de
SuSE GmbH, Deutschherrnstr. 15-19, D-90429 Nürnberg
Key fingerprint = 58CA 54C7 6D53 942B 1756  01D3 44D5 214B 8276 4ED5
"And now for something completely different."

Re: ispell change

[Kai Großjohann]

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

> Can we assume with high probability that `-d ndeutsch' will not pick
> up the wrong dictionary?

I'm sorry.  I totally forgot that I also have a site-local
customization to add the ngerman dictionary to
ispell-dictionary-alist.

Anyhow.  There is a german dictionary which uses Latin-1 and I think
we need a variant for Latin-9.  (Doing it for ngerman or ndeutsch or
whatever it is called is another step.)

kai
-- 
Silence is foo!

Re: ispell change

[Karl Eichwalder]

Andreas Schwab <schwab@suse.de> writes:

> German does not use the œ character, it only has ö, which can be written
> as the two character sequence oe where ö is not available.

Yes, but it occurs in composites like "der Œuvrekatalog des Maler
Hieronymus Bosch umfasst …" (cf. Duden 1996, p. 534).

On my system the dictionary with new spellings is installed as
german/german8 and the old one is available as deutsch/deutsch8.

-- 
ke@suse.de (work) / keichwa@gmx.net (home):              |
http://www.suse.de/~ke/                                  |      ,__o
Free Translation Project:                                |    _-\_<,
http://www.iro.umontreal.ca/contrib/po/HTML/             |   (*)/'(*)

Re: ispell change

[Andreas Schwab]

Karl Eichwalder <keichwa@gmx.net> writes:

|> Andreas Schwab <schwab@suse.de> writes:
|> 
|> > German does not use the œ character, it only has ö, which can be written
|> > as the two character sequence oe where ö is not available.
|> 
|> Yes, but it occurs in composites like "der Œuvrekatalog des Maler
|> Hieronymus Bosch umfasst …" (cf. Duden 1996, p. 534).

But that's not a German word.  With this rational you would have to add
every special character.

Andreas.

-- 
Andreas Schwab, SuSE Labs, schwab@suse.de
SuSE GmbH, Deutschherrnstr. 15-19, D-90429 Nürnberg
Key fingerprint = 58CA 54C7 6D53 942B 1756  01D3 44D5 214B 8276 4ED5
"And now for something completely different."

Re: ispell change

[Karl Eichwalder]

Andreas Schwab <schwab@suse.de> writes:

> But that's not a German word.

Arguable :)  It's listed in the "Duden"; thus it is a valid word of the
current German language.  It would be nice if Emacs would cover all
words listed in the Duden.

> With this rational you would have to add every special character.

I don't think so -- but it should be possible to check for Café (and
Café crème, Switzerland) or Cañon.  Must one use a exception list to
catch those words?

-- 
ke@suse.de (work) / keichwa@gmx.net (home):              |
http://www.suse.de/~ke/                                  |      ,__o
Free Translation Project:                                |    _-\_<,
http://www.iro.umontreal.ca/contrib/po/HTML/             |   (*)/'(*)

Re: ispell change

[Miles Bader]

Karl Eichwalder <keichwa@gmx.net> writes:
> it should be possible to check for Café

I agree; is there _any_ modern country that doesn't use the word café,
at least some of the time?

-Miles
-- 
Suburbia: where they tear out the trees and then name streets after them.

Re: ispell change

[Eli Zaretskii]

> From: Kai.Grossjohann@CS.Uni-Dortmund.DE
> Date: Fri, 10 May 2002 22:15:01 +0200
> 
> Anyhow.  There is a german dictionary which uses Latin-1 and I think
> we need a variant for Latin-9.

Yes, I agree.

Re: ispell change

[Eli Zaretskii]

> From: Karl Eichwalder <keichwa@gmx.net>
> Date: Fri, 10 May 2002 22:19:25 +0200
> 
> On my system the dictionary with new spellings is installed as
> german/german8 and the old one is available as deutsch/deutsch8.

I think someone told us that the new spellings dictionary is available
under different names on some GNU/Linux systems.

Re: ispell change

[Eli Zaretskii]

> From: Andreas Schwab <schwab@suse.de>
> Date: Sat, 11 May 2002 00:50:59 +0200
> 
> But that's not a German word.  With this rational you would have to add
> every special character.

I think if you never need to spell-check text that mixes different
languages, you indeed can add all word-constituent characters, even
those which are unused in German, to the regexp.  Those which are not
used in German will simply never happen.

Re: utf-7 encoding (was: ispell change)

[Eli Zaretskii]

> From: Andreas Schwab <schwab@suse.de>
> Date: Sat, 11 May 2002 00:50:59 +0200
> 
> Karl Eichwalder <keichwa@gmx.net> writes:
> 
> |> Andreas Schwab <schwab@suse.de> writes:
> |> 
> |> > German does not use the +AVM- character, it only has +APY-, which can be written
> |> > as the two character sequence oe where +APY- is not available.
> |> 
> |> Yes, but it occurs in composites like "der +AVI-uvrekatalog des Maler
> |> Hieronymus Bosch umfasst +ICY-" (cf. Duden 1996, p. 534).
> 
> But that's not a German word.

Andreas, your message came with this header:

   Content-Type: text/plain; charset=utf-7

AFAICS, only Gnus supports this encoding.  I think we should do
something to support it in RMAIL as well.

(I'm also curious how come utf-7 was selected even though Karl's
message was encoded in Latin-9--perhaps some local priority scheme on
your box?)

Re: utf-7 encoding (was: ispell change)

[Karl Eichwalder]

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

> Andreas, your message came with this header:
>
>    Content-Type: text/plain; charset=utf-7
>
> AFAICS, only Gnus supports this encoding.

My Oort Gnus (updated approx. 1 month ago) isn't able to decode uft-7
out of the box.

> (I'm also curious how come utf-7 was selected even though Karl's
> message was encoded in Latin-9--perhaps some local priority scheme on
> your box?)

FYI, my message was utf-8 encoded, probably because of the "..." ("…")
character – here it comes again ;)

-- 
ke@suse.de (work) / keichwa@gmx.net (home):              |
http://www.suse.de/~ke/                                  |      ,__o
Free Translation Project:                                |    _-\_<,
http://www.iro.umontreal.ca/contrib/po/HTML/             |   (*)/'(*)

Re: utf-7 encoding (was: ispell change)

[Eli Zaretskii]

> From: Karl Eichwalder <keichwa@gmx.net>
> Date: Sat, 11 May 2002 11:13:30 +0200
> 
> > (I'm also curious how come utf-7 was selected even though Karl's
> > message was encoded in Latin-9--perhaps some local priority scheme on
> > your box?)
> 
> FYI, my message was utf-8 encoded

Sorry, you are right.  I remembered that one of the previous messages
was in Latin-9, but it turns out it was from Andreas, not you.

Re: utf-7 encoding (was: ispell change)

[Andreas Schwab]

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

|> > From: Andreas Schwab <schwab@suse.de>
|> > Date: Sat, 11 May 2002 00:50:59 +0200
|> > 
|> > Karl Eichwalder <keichwa@gmx.net> writes:
|> > 
|> > |> Andreas Schwab <schwab@suse.de> writes:
|> > |> 
|> > |> > German does not use the +AVM- character, it only has +APY-, which can be written
|> > |> > as the two character sequence oe where +APY- is not available.
|> > |> 
|> > |> Yes, but it occurs in composites like "der +AVI-uvrekatalog des Maler
|> > |> Hieronymus Bosch umfasst +ICY-" (cf. Duden 1996, p. 534).
|> > 
|> > But that's not a German word.
|> 
|> Andreas, your message came with this header:
|> 
|>    Content-Type: text/plain; charset=utf-7

This is because I have Mule-UCS installed.  But there is an anomaly in
find-coding-systems-region: it is reversing the list of coding systems
each time the intersection with a new list of coding systems is computed.
So the order of coding systems depends on the number of non-ascii
characters in the string to be processed.  I have now changed the function
intersection in coding.c to keep the order of elements consistent.

Andreas.

-- 
Andreas Schwab, SuSE Labs, schwab@suse.de
SuSE GmbH, Deutschherrnstr. 15-19, D-90429 Nürnberg
Key fingerprint = 58CA 54C7 6D53 942B 1756  01D3 44D5 214B 8276 4ED5
"And now for something completely different."

Re: utf-7 encoding (was: ispell change)

[Stefan Monnier]

> This is because I have Mule-UCS installed.  But there is an anomaly in
> find-coding-systems-region: it is reversing the list of coding systems
> each time the intersection with a new list of coding systems is computed.

Huh? Aren't you talking about find-coding-systems-region-internal, not
find-coding-systems-region.  find-coding-systems-region should not suffer from
such problems because it calls `sort-coding-systems'.


	Stefan

Re: utf-7 encoding (was: ispell change)

[Andreas Schwab]

"Stefan Monnier" <monnier+gnu/emacs@RUM.cs.yale.edu> writes:

|> > This is because I have Mule-UCS installed.  But there is an anomaly in
|> > find-coding-systems-region: it is reversing the list of coding systems
|> > each time the intersection with a new list of coding systems is computed.
|> 
|> Huh? Aren't you talking about find-coding-systems-region-internal, not
|> find-coding-systems-region.  find-coding-systems-region should not suffer from
|> such problems because it calls `sort-coding-systems'.

sort-coding-systems does not change the order if two coding systems have
equal weight.  In my environment utf-8 and utf-7 (and others) have equal
weight.

Andreas.

-- 
Andreas Schwab, SuSE Labs, schwab@suse.de
SuSE GmbH, Deutschherrnstr. 15-19, D-90429 Nürnberg
Key fingerprint = 58CA 54C7 6D53 942B 1756  01D3 44D5 214B 8276 4ED5
"And now for something completely different."

Re: ispell change

[Richard Stallman]

    I use the following code which allows me to use ispell in a Latin-9
    environment:

Could you talk with the ispell.el maintainer about this?

Re: ispell change

[Kai Großjohann]

Richard Stallman <rms@gnu.org> writes:

>     I use the following code which allows me to use ispell in a Latin-9
>     environment:
>
> Could you talk with the ispell.el maintainer about this?

Will do.

kai
-- 
Silence is foo!