UTF-8 and ispell

[Paul Boekholt]

2007/9/20, G. Milde <milde at users.sourceforge.net>:

> I suppose that dictionaries-common needs to be refined:
>
> Example: the iogerman package provides in odeutsch.aff the following
> encodings::
>
>   altstringtype "tex" "TeX" ".tex" ".bib"
>   altstringtype "plaintex" "TeX" ".tex"
>   altstringtype "latin1" "TeX" ".latin1" ".txt" ".tex" ".bib"
>   altstringtype "utf8" "TeX" ".txt"
>   altstringtype "ascii" "nroff" ".ascii" ".txt"
>   altstringtype "pc" "TeX" ".pc" ".txt" ".tex"
>   altstringtype "HTML" "TeX" ".html" ".htm" ".sgml" ".xml"
>
> i.e. it is prepared for UTF-8 input, provided the right argument is given to
> ispell.

That's interesting, I didn't realise that ispell had any utf-8 support.
I've only used aspell for utf-8 spellchecking. I just tried to use
ispell.sl with ispell with the ngerman package, but it didn't work
quite right. See also line 215 of ispell.sl:

   % Aspell 0.60 counts multibyte characters
   % as one character in utf-8 mode, we can use go_right() which does not
   % expand tabs, but is utf-8 aware.

Ispell seems to count bytes here. So it looks like ispell.sl doesn't
work with ispell in utf-8 mode ATM.

Does anybody still use ispell? Why, when aspell is better?