On 3/2/06, Davide Viti <[EMAIL PROTECTED]> wrote:
> > yesterday I managed to extend the spellchecker as to give infos about
> > the codepoints used for each language on each level
> > [...]
> > More about this in the next few days.
>
> I just committed the above modifications and forced a re-run of the
> spellchecker [1] for all levels; tested mainly on level1 translations
> where it seems to behave well.
>
> It's possible to browse / download the following:
>
> * specific list of codepoints for a particular languages ("Codepoints" column)something's wrong, there's only one code for arabic: http://d-i.alioth.debian.org/spellcheck/level1/latest/nozip/ar_codes.txt > * list of codepoints for the entire level (see link at the beginning of the > page) cool! this way we can check if we have uncovered codes (we can detect missing glyphs) > * a tarball containing all the infos related to codepoints for the entire > level > > using the lst2rng.sh script available in [2] it's possible to convert a file > of > codepoints in a format compatible with the stripping scripts i.e: > > $ cat it_codes.txt | sed "s|.*<U\(....\)>|\1|" | ./lst2rng.sh > u0020:u0022 u0025 u0027:u0029 u002B:u005D u005F u0061:u007A u00A0 u00AB u00B0 > u00BB u00C0 u00C8 u00E0 u00E8:u00E9 u00EC u00F2 u00F9 > > things still need to be polished, but we're getting closer. Hopefully this evening I will make my NSLU2 act as a router instead of my laptop running OS X and I will be able to jump in and help. -- Regards, EddyP ============================================= "Imagination is more important than knowledge" A.Einstein
