Sorry to keep beating this to death. I could be looking for perfection which isn't possible.
I'm trying to spell check the user term "C Wishes" to collate to "B Wishes". If I use a tokenizer on the spell check field which splits "c wishes" into 2 tokens then there are no suggestions for "c". Obviously it's probably difficult to spellcheck a single letter. It doesn't seem possible to collate either. I do get suggestions for "wishes" just not for "c". I'm testing with: <float name="accuracy">0.3</float> <float name="thresholdTokenFrequency">1</float> <int name="minPrefix">0</int> <str name="spellcheck.collateMaxCollectDocs">5000</str> <str name="spellcheck.maxCollationEvaluations">10000</str> <str name="spellcheck.maxCollationTries">3000</str> So I'm guessing this can't be done. I have tried a spell check field that uses the KeywordTokenizerFactory so that "c wishes" is only one token but haven't had any luck with this either. I suspect that I'm asking for too much. Corey
