Thank you both for responsing. Hoss,
what you've pointed out was exactly what I am looking for. However, I would *always* prefer the second implementation, because of the fact that you have to compute the number of terms for all records only for *one* time. :-) At the moment I would feel like writing a TokenCountingTokenFilter and implement the QParser this way: extending my favorite QParser and in the constructor I would do something like: - creating a StringReader from the query-string - let a Tokenizer tokenize my query-string (without a factory, just instantiate something like Tokenizer t = new WhitespaceTokenizer(reader);) - maybe filtering the tokenized query with other filters - give my query to the TokenCountingTokenFilter and set the number of tokens of the query with its help. - getting MAX_LEN with the help of a getParam-Method. However, I got some doubts on this: What about queries that should be filtered with the WordDelimiterFilter. This could make a large difference to a none-delimiter-filtered MAX_LEN *and* it has got a protwords param. I can't instantiate a new WordDelimiterFilter everytime I do a query, so how can I put my already instantiated Filters into a cache for such usecases? I think solving this problem perhaps would also lead to a possibility to make multiword synonyms at query-time possible. Do you know which class stores the produced filters from the FilterFactories and how I can access them? Kind regards - Mitch -- View this message in context: http://n3.nabble.com/Minimum-Should-Match-the-other-way-round-tp694867p698683.html Sent from the Solr - User mailing list archive at Nabble.com.
