Hi,

Currently, the connector mixes the trait InsertableRelation at DeaultSource which is the data source provider for Mongo relations. That is a kind of relation generator instead of a relation itself. Insert capabilities could be more easily leveraged if that interface was mixed at MongoRelation class (check InsertableRelation trait description: https://github.com/apache/spark/blob/v1.6.2/sql/core/src/main/scala/org/apache/spark/sql/sources/interfaces.scala#L294).

Once the implementation is moved to MongoRelation, INSERT INTO SELECT statements will be available.