Well, that smells like a WordDelimiterFilterFactory [1]. It splits, as your debug output shows, value into three separate tokens. This means that (at least) the strings 'abc', '12' and 'def' are in your index and can be found. The abc12 value is not present. If you want to query for substrings, you can try NGramFilterFactory [2]. It's not really documented on the wiki but searching will help [3].
[1]: http://wiki.apache.org/solr/AnalyzersTokenizersTokenFilters#solr.WordDelimiterFilterFactory [2]: http://search.lucidimagination.com/search/document/CDRG_ch05_5.5.6 [3]: http://www.lucidimagination.com/blog/2009/09/08/auto-suggest-from-popular-queries-using-edgengrams/ -----Original message----- From: j <jta...@gmail.com> Sent: Sat 07-08-2010 19:18 To: solr-user@lucene.apache.org; Subject: dismax debugging hyphens dashes How does one debug index vs. dismax query parser? I have a solr instance with 1 document whose title is "ABC12-def". I am using dismax. While "abc", "12", and "def" do match, "abc12" and "def" do not. Here is a the parsedquery_toString, I'm having trouble understanding it: +(id:abc12^3.0 | title:"(abc12 abc) 12"^1.5) (id:abc12^3.0 | title:"(abc12 abc) 12"^1.5) Does anyone have advice for getting this to work?
