Reading the docs we have the following doubt:
It is not clear from the docs. For a 'native' Hive table it is clear that this is possible.
For Spark-only tables this is not possible. Confirmed with Ranger test.