Today, Mark Gray
Svend Tollak Munkejord
writes: I use GNU Emacs and SuSE Linux 8.0. In Emacs, when I do M-x ispell on a LaTeX file containing Norwegian characters, using the British dictionary, Ispell fails with the following error message:
ispell-process-line: Ispell misalignment: word `ørn' point 4613; probably incompatible versions
Does anyone have a better solution?
I think you might be able to fix the problem by editing:
/usr/lib/ispell/emacs/english.el
and adding the characters in question to the proper character classes as it is done for the Norwegian version of english.el --> this is just a wild guess as I do not have the Norwegian dictionary installed.
Thanks for your quick reply. I tried to change the contents of british.el from (append-ispell-dict-alist "british.hash" '("british" "[A-Za-z]" "[^A-Za-z]" "[']" nil ("-B" "-d" "british") nil ni\ l)) to (append-ispell-dict-alist "british.hash" '("british" "[A-Za-z\305\306\310\311\322\324\330\345\346\350\351\362\364\370]" "[^A-Za-z\305\306\310\311\322\324\330\345\346\350\351\362\364\370]" "[']" nil ("-C" "-d" "british") nil nil)) like the Norwegian one, and checked that the variable ispell-dictionary-alist was set correctly. Unfortunately, it did not work. It seems like the problem lies somewhere else. If I start Emacs with the --no-site-file flag (so that /usr/lib/ispell/ispell-emacs-menu.el is not loaded), then Ispell happily spell-checks my file with the British dictionary, using the default (?) value ... ("british" "[A-Za-z]" "[^A-Za-z]" "[']" nil ("-B" "-d" "british") ... for ispell-dictionary-alist. By the way, I have problems not only with Norwegian characters, but also with accented characters, as in Citroën. Regards, -- Svend Tollak Munkejord