I have built an ispell hash file for Yiddish/UTF8/YIVO.

I will put it in http://www.cs.uky.edu/~raphael/yiddish.utf8.hash.gz tomorrow.
(I had transfer problems today.)

Here is how I decided on the various questions that Andrew Dunbar raises:

> Firstly, ispell and aspell (which we use for our spell-checking) can only
> handle 8-bit encodings, so you need to find an encoding which you can convert
> the Unicode to this format.

I used ispell's tools and let them pretend to be using an 8-bit encoding.  All
of UTF8 can be considered to be just bytes if you want.

> that there is more than one orthography or spelling standard, so we need to
> be able to tell people which one it uses.

I am using the YIVO (yidishe visnshaftlekhe institutsye = Yiddish Scientific
Institute) "takones far der eynhaytlekher oysleyg" = rules for uniform
spelling, published around 1950, and generally accepted since then.

> Some spelling standards require vowel points in certain places whereas in
> others they are always optional.

Vowel points are required in certain places.  There are no options.

> find if there is an Encoding which
> supports the vowel points - hopefully they do.

UTF8 supports the full range of Yiddish characters.

Raphael Finkel

Reply via email to