#Solr geodist full

All the deal details were stored in promotions while the location details were stored in merchants.

We have lots of promotions and lots of merchants. Where Is The Dataīefore going further let me show you how things were implemented in the database. Copied the schema.xml file from the sunspot project to our solr conf folder (SOLR_ROOT/solr/conf), restarted the solr server and bingo everything worked like a breeze. The culprit here is schema.xml file, which contains all details about fields your documents can contain, and how those fields should be dealt with when adding documents to the index, or when querying those fields. So we headed to solr’s official website, downloaded the latest version of solr 4, started the solr server and ran rake sunspot:solr:reindex to make the index. The first thing we have done was to see whether solr 4 integrates with sunspot or not. But we need to show our users the distance of the outlet from his/her present location, so the effort started to make it work our way. But you can’t get distance as search output because it was only supported by solr 4.0 and above while sunspot uses solr 3.

Using sunspot one can search by distance (lat, long and radius are required) or sort by distance (lat and long are required). We started configuring sunspot with rails 3.2 and initial results were very encouraging. At that time, the most popular plugin for solr was sunspot which comes with decent features and an active community. We were also looking at a solution which comes with geospatial search working out of box. We found out that the previous plugin we were using for solr (acts_as_solr) will not be able to serve the purpose now. We started the redesign of khojguru around April 2012 and the new version was to be build grounds up using rails 3. We were able to run the same setup for around 3.5 years and then the next challenge arrived.

We faced a lot of challenges configuring it for a reasonably write heavy load (a lot of data entry is going on those days), ultimately opting for lazy indexing based upon number of documents and time elapsed from last update.

#Solr geodist full

Our first encounter with solr was to use it for full text search and faceting in the old version of khojguru (when we were still a local search player). Apache SOLR is a really great search platform based on apache lucene with powerful full-text search, hit highlighting, faceted search, near real-time indexing, dynamic clustering, database integration and geospatial search.