You got my problem right. I had seen your solution (which works), and just wanted to know if there is a better one. This one allow me to get the nb of hits, which is also what I want.
The only drawback is that I have to filter the returnrd facet field on the first letters myself. No big deal, but it needs some resources. I will use this for a prototype, and try a better approach if we go further on the project. Thanks ! Paul Terray - Consultant Avant-Vente SOLLAN Tel : +33 (0)1 48 51 15 44 Fax : +33 (0)1 48 51 15 48 [EMAIL PROTECTED] -----Message d'origine----- De: "Yonik Seeley" <[EMAIL PROTECTED]> Objet: Re: Index term list - using facets ? Date: ven 15 sep 2006 16:29 Taille: 932 octets A:solr-user@lucene.apache.org On 9/15/06, Yonik Seeley <[EMAIL PROTECTED]> wrote: > On 9/15/06, Paul Terray <[EMAIL PROTECTED]> wrote: > > I think I am close to have a list of index terms, using facet searching. > > However, I still have a question: I would like to limit the terms to a > > query. My goal is to do a simple google suggest type of search (with just > > one term), so I have to limit the terms to the first letters typed. Of > > course, I can filter the facet results to get this, but it seems there > > should be a better solution. > > So instead of documents matching field:foo* you want the actual terms > matching field:foo* > I think exposing lower level info from an index like this can make > sense, but it's not currently supported. Oh, I think you can get what you want via q=myfield:foo*&facet=on&facet.field=myfield&facet.zeros=false It's a roundabout way of getting the info and is likely to be slow if your index is large. -Yonik