Both of those features are needed at indexing time, right? If it is, the best place to put it is on an UpdateRequestProcessor. See http://wiki.apache.org/solr/UpdateRequestProcessor
Tomás On Mon, Aug 29, 2011 at 11:06 AM, samuele.mattiuzzo <samum...@gmail.com>wrote: > I've posted a similar question few days ago, but our needs have gone a bit > further. > > I need to develop two plugins which need to be persistent throu the whole > indexing and updating process > > The first one need to open a connection to a mysql instance (and query that > connection during every document processing) > > The second one uses a java library (classifier4j) which is a Bayesian > categorization system (and doesn't talk to a db). This one learns while > matching, so the object needs to be created at the very beginning of the > indexing and should be available for all the documents (i cannot create the > object for every object, because i'd miss the learning feature) > > For the first one i could use a dataimporthandler, but i'm not sure about > it: i don't need to import the whole db, but just the occurencies matching > a > particular condition for each document. About the second one, i'm blind. > > Is there a place in solr where i can create the connection object and the > categorizer object before everything else, and have them available to all > documents? > > Thanks all in advance! > > -- > View this message in context: > http://lucene.472066.n3.nabble.com/Solr-custom-plugins-is-it-possible-to-have-them-persistent-tp3292781p3292781.html > Sent from the Solr - User mailing list archive at Nabble.com. >
