Richard Stallman schrieb:
Well the problem is, that this is hard to implement in aspell, as there
a various way's to build compound
words in german and - at least theoreticaly - an infinite number of
possible compund words. The real
problem is to find out about the word bounderies.
Those facts are true; but if they do not preclude implementing this in
Emacs, why would they preclude implementing this inside aspell?
extend flyspell. It is sure not very elegant and it would be really nice
to implement this in aspell or at least flyspell.
But there would be need of a totaly different aproche.
If this approach is good in Emacs, why isn't the same approach
good in aspell?
Well simply because gspell.el let the user mark the boundary of the
parts of a compound word. It so needs user interaction.
This is also the reason why it so far only work for TeX documents
where you can put hyphenation marks, that are not printed
unless the word needs hyphenation.
So it's quite is to check in
Hunde"-haar"-wasch"-mittel
each of the four word-parts. For aspell, however, it would be rather
difficult to guess where within Hundehaarwaschmittel to search for
possibly correct word parts.
Also the word-parts the gspell.el guesses to be true are mark with a
special face, because even so it cannot tell for sure, it really is correct.
Hund"-haar is wrong, but Hunde"-haar is correct. In both cases gspell.el
will mark Hund as probably correct. So it is a help, but still needs
the user to know about correct spelling.
_______________________________________________
gnu-emacs-sources mailing list
[email protected]
http://lists.gnu.org/mailman/listinfo/gnu-emacs-sources
