Re: Bug 130397 (Was: Emacs - Ispell problem with i[no]german dictionary)

[Agustin Martin]

On Wed, Dec 22, 2004 at 09:37:32PM +0900, Kenichi Handa wrote:

> Please try the same thing with the latest CVS code.  With
> that, when you type e-grave in address@hidden locale, e-grave of
> latin-iso8859-15 should be inserted in a buffer.  So, as far
> as you are using a dictionary that uses iso-8859-15 encoding
> (or in general, using a dictionary that uses the same
> encoding as your locale), you should not face the above
> problem.
> 

Thanks for the tip. I am not maintaining emacs, but a package for the common
dictionaries setup (dictionaries-common) that provides a recent and patched
ispell.el for all the diferent emacsen flavours ({x}emacs) to integrate the
different dicts and spellchecking engines in some way. I will be happy
to test this once is included in sid emacs.

> > I am playing with redefining ispell-get-coding-system function in ispell.el
> > so dict coding-system is changed to iso-8859-15 if was originally
> > iso-8859-1 and emacs has iso-8859-15 as buffer-file-coding-system, something
> > like
> 
> At least you should check if buffer-file-coding-system is
> nil or not before callding coding-system-get.  

Thanks for pointing put this, change added.

> But, anyway,
> I think the above function is too ad-hoc.  As iso-8859-1 and
> iso-8859-15 contains different set of characters (even if
> they are few), it's not good to treat them as the same
> thing.
> 
> For instance, if a dictionary uses iso-8859-1 encoding, it
> doesn't contain "\264" in CASECHARS entry.  But, if a
> dictionary uses iso-8859-15 encoding, it should contain
> "\264" (Z-WITH-CARON) in CASECHARS entry.
> 
> So, if you are going to check the spell of some word
> containing Z-WITH-CARON by iso-8859-1 dictionary, something
> goes wrong.
> 

I was aware of this, but anyway thanks for reminding. Code is probably too
ad-hoc, but latin{0,1} thing is also a somewhat ad-hoc scenario, where
latin0 should have really be named as something like iso-8859-1v2, that is,
a revision. I cannot imagine somebody using a iso-8859-2 dict and trying to
write in a iso8859-1 buffer, but with iso-8859-1 and iso-8859-15 that is
happening too frequently. 

So we have a lot of people that blindly select the locale @euro variant
without realizing its implications, and that iso-8859-1 and iso-8859-15
are different, but very close encodings (from a practical point of view,
they are fully equivalent for most languages but IIRC french (oe,"Y) and
finnish {sSzZ}^, ^ stands for caron; the euro symbol seems not significant
to spellchecking). 

Furthermore (this is probably fixed by the CVS code you mentioned above),
in current sid emacs utf-8 files can be checked with a latin1 dict (of
course if they do not use chars outside latin1) using the ispell.el
internal reencodings, but fails for iso-8859-15 declared dict.

The current state of ispell dicts in Debian is that ifrench is iso-8859-15
as default (although has a real latin1 entry), while finnish do not set at
all the {s,z}-caron chars, so it is a fully latin1 entry. aspell-fr and
aspell-fi are set to plain latin1.

So the only language that might currently require extra work is french, and
for it I find reasonable to use for emacs as default the iso-8859-15 entry
(tagged as iso-8859-1 for the above sustem to work). For this I would like
to hear Lionel's point of view, since he has put a lot of effort to make
iso-8859-15 available for spellchecking (Hi, Lionel). 

I personally do not like having separate iso-8859-15 entries unless they are
really required. For the above dicts, that would be for french, and I am not
at all sure that it is really required.

Thanks a lot for your feedback, Handa.

Cheers,

-- 
Agustin