I have a Solr text field and when I use Solr's field analysis tool, it shows that wildcard queries are being stemmed. But query results indicate that it is not. It looks like there is a bug in the tool.
Bill On Mon, Aug 3, 2009 at 7:21 AM, gateway0 <reiterwo...@yahoo.de> wrote: > > Ok thanks you´re right. > > But the thing is my users will often search for expressions like: "Harr" or > "har" etc.. > > So I thought I automatically add the wildcard "*" to every request. > > If that too gets me into trouble "Harr*"=>no result "harry*"=>no result > > What should I do? > > > > Otis Gospodnetic wrote: > > > > I believe it's because wildcard queries are not stemmed. During indexing > > "harry" probably got stemmed to "harr", so now "harry*" doesn't match, > > because there is no "harry" token in that string, only "harr". Why > > wildcard queries are not analyzed is described in the Lucene FAQ on the > > Lucene Wiki. > > > > You could also try searching for kunde:Harr* for example (not the > > upper-case Harr). I bet it won't result in a hit for the same reason - > at > > index time you probably lower-case tokens with LowerCaseFilter(Factory), > > and if you search for Harr*, the lower-casing won't happen because the > > query string with the wildcard character isn't analyzed. > > > > Otis > > -- > > Sematext is hiring -- http://sematext.com/about/jobs.html?mls > > Lucene, Solr, Nutch, Katta, Hadoop, HBase, UIMA, NLP, NER, IR > > > > > > > > ----- Original Message ---- > >> From: gateway0 <reiterwo...@yahoo.de> > >> To: solr-user@lucene.apache.org > >> Sent: Sunday, August 2, 2009 7:30:19 PM > >> Subject: Solr failing on "y" charakter in string? > >> > >> > >> Hi, > >> > >> I have the following setting: > >> schema.xml: > >> "" > >> the "text" field-type was updated with the "preserveOriginal=1" option > in > >> the schema > >> > >> I have the following string indexd in the field "kunde" > >> "Harry Heim KG" > >> > >> Now when I search for "kunde:harry*" it gives me an empty result. > >> > >> When I search for "kunde:harry" I get the right result. Also > >> "kunde:harr*" > >> works just fine. > >> > >> The strange thing is that with every other string (for example > >> "kunde:heim*") I will get the right result. > >> > >> So why not on "harry*" with an "y*" at the end? > >> > >> kind regards, S. > >> -- > >> View this message in context: > >> > http://www.nabble.com/Solr-failing-on-%22y%22-charakter-in-string--tp24783211p24783211.html > >> Sent from the Solr - User mailing list archive at Nabble.com. > > > > > > > > -- > View this message in context: > http://www.nabble.com/Solr-failing-on-%22y%22-charakter-in-string--tp24783211p24789070.html > Sent from the Solr - User mailing list archive at Nabble.com. > >
