Hi. Did you get this working? I think this is a problem on many German languages. Using "brute force dictionary splitting" of compound words will generate to many words.
There is some words about using myspell dictionaries in the tsearch project here: http://www.sai.msu.su/~megera/wiki/Tsearch_V2_compound_words In this approach a word must be marked with a special flag to be able to participate in a compound word. Did anyone tried to implement a this in solr/lucene? Is there a good reason to not do this? To get this functionality I think there must be implemented a new CompoundWordTokenFilter which “eat” ispell dictionaries. Has anyone looked at this? Best regards Kjell.
