Maybe you can avoid reindexing the whole index then. If you add a synonym for term "foo" then the only documents you really need to reindex are the ones that contain that term "foo".
Otis -- Sematext -- http://sematext.com/ -- Lucene - Solr - Nutch ----- Original Message ---- From: anuvenk <[EMAIL PROTECTED]> To: solr-user@lucene.apache.org Sent: Sunday, January 6, 2008 12:38:11 PM Subject: Re: parsedquery_ToString I log the search phrases the users on my site are using to search, and review them regularly. Based on that i add synonyms for certain phrases to help increase the relevant results. The reason i don't have the synonym filter at index time is because i can't re-index whole/portion of data everytime i add synonyms because my index is huge. Any suggestions on how to overcome this? hossman wrote: > > > : Is the parsedquery_ToString, the one passed to solr after all the > tokenizing > : and analyzing of the query? > > yes. > > : For the search term 'chapter 7' i have this parsedquery_ToString > ... > : I have these synonyms > : chap 7 => bankruptcy > ... > : But seem to have a little bit of trouble understanding how its building > this > : parsedquery_Tostring > > 1) if you really want to understand the lucene query structure built by > the dismax handler, you should definitely start with some simpler queries > (without > synonyms) .. the query structure is definitely non trivial, and requies > some "expert level" reading. > > 2) you should read the disclaimer in the wiki about using > synonyms that involve multiple words .. applying them at query time is not > a good idea for several reasons, one of which is causing your problems. > do them at indexing time instead... > > http://wiki.apache.org/solr/AnalyzersTokenizersTokenFilters#SynonymFilter > > > > > -Hoss > > > -- View this message in context: http://www.nabble.com/parsedquery_ToString-tp14627131p14651154.html Sent from the Solr - User mailing list archive at Nabble.com.
