You could develop your own tokenizer to extract the different forms of your ids.
It is possible to extend the pattern tokenizer. Ludovic. Le 25 mars 2011 21:13, "David Yang [via Lucene]" < ml-node+2732007-1439913827-383...@n3.nabble.com> a écrit : > > > Hi, > > > > I am having some really strange issues matching "N61JQ-B2". If I had a > field "N61JQ-B2", and I wanted to match "N61JQ", "N61JQB2", "N61JQ-B2" > and "N61JQ B2" in dismax, what fieldtype should it have? My final > fallback is to use ngrams but that would impose a pretty large overhead, > since the field could be a long normal string with one model number in > it. > > > > I noticed when I used WordDelimiterFilterFactory the dismax would > convert the parsed query to some pre-analyzed query. > > > > Cheers, > > David > > > > _______________________________________________ > If you reply to this email, your message will be added to the discussion below: > http://lucene.472066.n3.nabble.com/Dismax-and-worddelimiterfilter-tp2732007p2732007.html > To start a new topic under Solr - User, email ml-node+472068-1765922688-383...@n3.nabble.com > To unsubscribe from Solr - User, visit http://lucene.472066.n3.nabble.com/template/NamlServlet.jtp?macro=unsubscribe_by_code&node=472068&code=Ym91dHJvc2xAZ21haWwuY29tfDQ3MjA2OHw0Mzk2MDUxNjE= ----- Jouve France. -- View this message in context: http://lucene.472066.n3.nabble.com/Dismax-and-worddelimiterfilter-tp2732007p2732245.html Sent from the Solr - User mailing list archive at Nabble.com.
