On Thu, Jan 28, 2010 at 4:58 PM, brad anderson <solrinter...@gmail.com> wrote: > I think commiting code to solr would not fix the problem. I don't want to > have to go through the HTTP stack to create an index.
Why not? If you use something like SolrJ, it's an implementation detail if there is a transport, and what it is. Anyway, there is an embedded SolrJ client that doesn't go through HTTP. You can also get update handlers like CSV to directly read a local file, skipping HTTP: http://wiki.apache.org/solr/UpdateCSV#Methods_of_uploading_CSV_records -Yonik http://www.lucidimagination.com
