Nested Datatype
Lucene has no concept of inner objects and hence data is flattened when being indexed with not quite as expected behavior.
Data posted
{ "group" : "fans", "user" : [{ "first" : "John", "last" : "Smith" },{ "first" : "Alice", "last" : "White" } ] }
Indexed by Lucene as:
{ "group" : "fans", "user.first" : [ "alice", "john" ], "user.last" : [ "smith", "white" ] }
What you think would be a consequence of such indexing?
Should you be able to find person
user.first:john AND user.last:white
?Nested is a specialized version of the 'object' datatype that allows querying arrays of objects independently from each other addressing less than perfect search results
Last updated
Was this helpful?