Apache Spark Component Guide
Also available as:
loading table of contents...

Enhancements for Connection URL Support.

The connection URL format for Hive is described in Apache Hive documentation. In user impersonation mode, the Spark Thrift server supports a default database and hivevar variables.

Specifying Default Database in the Connection URL

Specifying the connection URL as jdbc:hive2://$HOST:$PORT/my_db results in an implicit “use my_db” when a user connects.

For an example, see the preceding Beeline example where the !connect command specifies the connection URL for “foo_db”.

Support for hivevar Variables

Hive variables can be used to parameterize queries. To set a Hive variable, use the set hivevar command:

set hivevar:key=value

You can also set a Hive variable as part of the connection URL. In the following Beeline example, plan=miles is appended to the connection URL. The variable is referenced in the query as ${hivevar:plan}.