It should be easy to configure SOLR Schema, & use SOLRJ client; does not matter jetty/tomcat etc.; stick with simple SOLRJ java client: access database, generate SOLR document, update SOLR, execute query, parse (SOLRJ->SolrDocument), generate content, etc... Much easier, scalable, and more effective than XSLT (initial version of my website was Cocoon&SAXON powered; it was ugly...). www.tokenizer.org
> > Hey hi... > > Ya the content is generated dynamically from a database...but > all data in > the xml docs(parameters in it) will have same structure as > specified in the > schema..e.g all will have uniquekey parameter set as "csid". > > I am adding particular case studies in my database with all > information like > name, address...also tags. Then I want to search those case > studies on the > basis of tags added. I have set up solr for that but I am > starting it using > jetty server. I think I will have to start it thru tomcat itself. > > Any quotes? > > Funtick wrote: > > > > Hi, > > > > As I understood, you need a search for your web application. > > > > - How many pages it has? > > - Is content generated dynamically from a database (for instance)? > > > > Another problem: after updating solrschema.xml you need > > - restart SOLR > > - reindex SOLR > > > > With changing unique ID in schema... I don't remember, but > it should be of > > specific type. You probably need to reindex SOLR. > > > > If your application generates dynamic content, it would be > better to index > > database directly adding 'URL' field to SOLR schema. > > > > > > > > -- > View this message in context: > http://www.nabble.com/solr-to-work-for-my-web-application-tp15 450968p15474490.html Sent from the Solr - User mailing list archive at Nabble.com.
