Building Solr index with Lucene

[Ben Bangert]

For performance reasons, we're attempting to build the index used with  
Solr, directly in Lucene. It works for the most part fine, but I'm  
having issue when it comes to stemming. I'm guessing this is due to a  
mismatch in how Lucene is stemming, with how Solr stems during its  
queries or something.

Has anyone built their Solr index using Lucene, and how did you handle  
stemmed fields in Lucene so that Solr worked properly with them?

Cheers,
Ben