Shantanu, It sounds like all you have to do is switch to a field type that doesn't tokenize your mfg field. Try field type "string". You'll need to reindex once you make this change.
Otis -- Sematext -- http://sematext.com/ -- Lucene - Solr - Nutch ----- Original Message ---- > From: "Deo, Shantanu" <sd1...@att.com> > To: solr-user@lucene.apache.org > Sent: Tuesday, January 13, 2009 10:15:09 PM > Subject: faceted search returning multiple values for same field > > Hi, > I am using solr for indexing some product data, and wanted to use the > faceted search. My indexed field (mfg) sometimes contains two words > "sony erricson" for example. When I get the facets on the mfg, SOLR > return "sony" and "erricson" as separate hits. There are also some > facets that show up rather mysteriously. > > My Unique list of mfg's that is indexed is as follows: > AT&T > BlackBerry? > HTC > LG > Motorola > Nokia > Option > Palm > Pantech > Samsung > Sierra Wireless > Sony Ericsson > > > The resulting facets being returned is below: > "facet_fields":{ > "mfg":[ > "ericsson",195, > "soni",156, > "samsung",155, > "nokia",90, > "Ericsson",78, > "Sony",78, > "Samsung",62, > "motorola",55, > "lg",50, > "sony",39, > "Nokia",36, > "pantech",25, > "Motorola",22, > "LG",20, > "berri",16, > "black",16, > "blackberri",16, > "Pantech",10, > "BlackBerry",8, > "blackberry",4, > "AT",0, > "HTC",0, > "Option",0, > "Palm",0, > "Sierra",0, > "T",0, > "Wireless",0, > "at",0, > "att",0, > "htc",0, > "option",0, > "palm",0, > "sierra",0, > "t",0, > "wireless",0] > > > I have tried playing around with defining the fieldtype using the > following analyzers: > > positionIncrementGap="100" > > > > > > > words="manufacturer.txt"/> > > > > > Any ideas if its possible to get the same facets as are in the data > that's being indexed or would I have to write my own Filter for this > purpose ? > > Thanks > Shantanu Deo > AT&T eCommerce Web Hosting - Release Management > Office: (425)288-6081 > email: sd1...@att.com
