Configured Similarity should be used in lucene index queries #375

Closed
tinwelint opened this Issue Dec 16, 2012 · 1 comment

Projects

None yet

2 participants

@tinwelint
Neo4j member

Configuring an index with "similarity": "my.own.SimilarityClass" doesn't affect the IndexSearcher, such that the default similarity is always used either way.

@jakewins

Closing since I imagine we don't want to invest more in lucene-specific APIs (although maybe this would be a useful feature to have on the Cypher level, eg CREATE INDEX with similarity mechanisms blah)? In any case, I don't think this is actionable right now. Please re-open if you feel differently MP.

@jakewins jakewins closed this Nov 6, 2013
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment