Work at SourceForge, help us to make it a better place! We have an immediate need for a Support Technician in our San Francisco or Denver office.
I'm in the process of making Northern and Lule Sami hunspell dictionaries. During this work I've experienced this problem:
The word ašeahtažan is accepted, but not ašeahtaža, even though ža and žan have identical flags.
SFX 28 Y 1
SFX 28 0 0/63003 . NIE
One similar peculiarity:
A short wordlist contains these four words. The third word is a constructed word, the other ones are real words.
ađđamaž is not accepted, but ađđamažs is, same issue with the flags as above.
SFX 1 Y 1
SFX 1 0 seaskka/63003 . NIE
SFX 2 Y 1
SFX 2 0 0/63003 . NIE
SFX 3 Y 1
SFX 3 0 eamet/63003 . NIE
This is tested with hunspell 1.1.9 (kubuntu package) and 1.2.2 (self built) on Kubuntu 8.04.
Is there an error in the dic or aff files? Could it be an issue with non-ascii chars?
I was wondering what was wrong, and it suddenly struck me that I had to use COMPOUNDMIN in the aff file. So adding COMPOUNDMIN 2 in the aff file solves the problem