Hey all,
I may not be using the best description/verbiage for the problem as I haven't been able to find an example on my own (and I am new to hibernate search):

Given a large resultset that has, say 4 relevant fields. Each of the fields may have a much smaller set of distinct values, but aren't necessarily FK's as it could contain freeform information.

So given an example table for the problem:
person |age | favorite color | searchengine | ide | programminglanguage

I may want to return all results for a given age range (30-40 for example), and then want to view all the distinct available colors and counts, distinct available searchengine and counts, distinct available ide and counts, and distinct programming languages and counts.

Looking for an efficient method to handle this problem, and rather than writing a bunch of SQL scripts and working the database was curious if Hibernate Search with a local lucene index (and stored values) would be better able to solve this problem in a more maintainable fashion?

thanks for any feedback,
-D

Hi,
yes Lucene if very efficient to count term frequencies.
You could implement something like this using the Faceting API

Excellent, that is what I'm looking for, thank you!! (it's all about the right keyword/term)


Question, can you reuse a jpa query or jpa2 criteria query with Facets, or is it limited to only with the Hibernate Search QueryBuilder?

i.e., I already have queries built with named queries and/or jpa2 criteria, but want to add facets. I am attempting to convert them to equivalent search querybuilders, but there are some mis-matches here and there where the Facet may not match the original query result.

No, the two are different: you either run a fulltext query or a criteria query, they are send to two different query engines.

You can set a criteria for a fulltext query, but only to optimize the loading (join fetch), not to add restrictions.