I have a jpa entity class called QueryIndividual which contains 1 or more QueryIdentifier classes in a set.

If I do a query by accessing the database via a org.hibernate.search.jpa.FullTextQuery query class (using JPA) the QueryIndividual properties aswell as all the associated QueryIdentifiers are returned correctly.

However if I set a Projection on the query as follows "firstName","identifiers.idValue" and set a ResultTransformer on the query, I do not get all QueryIdentifiers back. Only one is returned.
('firstName' is the property in the QueryIndividual class, 'identifiers' is a list of QueryIdentifier classes in the QueryIndividual class. 'idValue' is a property in the QueryIdentifier class.)

I understand that the data is denormalized when writing to the lucene engine. It would be fine if I got duplicates in the result set as I can programmatically deal with this.

Example:
if the FirstName property in the QueryIndividual class is set to 'Jimmy' and there are 2 identifiers
'12345' and 'abcde' in the 'identifier' list.

If I search for FirstName='Jimmy'
I would like the resulting hitlist to be (2 records)
'Jimmy','12345'
'Jimmy','abcde'

However I only get
'Jimmy','12345'

Hi Garth,
what you're asking is not possible with the current released versions, but we could discuss details for such a feature if you're interested in helping out?

I am trying to implement the .setSort() feature on my query. I read that you have to set the field to not be tokenized and that this was changed from Index.UN_TOKENIZED to analyze = Analyze.NO.