Faceting on analyzed text can eat a lot of RAM. This strategy might not scale.
On Tue, Oct 5, 2010 at 4:00 PM, Savvas-Andreas Moysidis <savvas.andreas.moysi...@googlemail.com> wrote: > Good point, > > so you could have an unanalyzed counterpart field set with a <copyfield /> > and facet on that.. > > On 5 October 2010 23:49, Markus Jelsma <markus.jel...@buyways.nl> wrote: > >> It is a good practice (for many cases as seen on the list) to search >> (usually with fq) on analzyed fields but return the facet list based on the >> unanalyzed counterparts. >> >> -----Original message----- >> From: Savvas-Andreas Moysidis <savvas.andreas.moysi...@googlemail.com> >> Sent: Wed 06-10-2010 00:46 >> To: solr-user@lucene.apache.org; >> Subject: Re: Umlaut in facet name attribute >> >> Hello, >> >> It seems that your analysis process removes punctuation and therefore >> indexes terms without it. What you see in the faceted result is the text >> that has been indexed. >> >> If you select a Tokenizer/Token Filter which preserves punctuation you >> should be able to see what you want. >> >> Cheers, >> -- Savvas >> >> On 5 October 2010 20:25, alexander sulz <a.s...@digiconcept.net> wrote: >> >> > Good Evening and Morning. >> > >> > I noticed that if I do a facet search on a field which value contains >> > umlaute (öäü), >> > the facet list returned converted the value of the field into a normal >> > character (oau).. >> > >> > How do I precent this from happening? >> > >> > I cant seem to find the configuration for faceting in theschema or config >> > xml files. >> > >> > thx >> > alex >> > >> > -- Lance Norskog goks...@gmail.com
