Hive Dimension Table

Function

You can use Hive tables as temporal tables and associate them through temporal joins. For more information on temporal joins, refer to temporal join.

Flink supports processing-time temporal joins with Hive tables, which always join the latest version of the temporal table. Flink supports temporary joins with both partitioned and non-partitioned Hive tables. For partitioned tables, Flink automatically tracks the latest partition of the Hive table. For details, see Apache Flink Hive Read & Write.

Caveats

Currently, Flink does not support event-time temporal joins with Hive tables.
The "Temporal Join The Latest Partition" feature is only supported in Flink STREAMING mode.

When you create a Flink OpenSource SQL job, set Flink Version to 1.15 in the Running Parameters tab. Select Save Job Log, and specify the OBS bucket for saving job logs.
For details about how to use data types, see Format.
Flink 1.15 currently only supports creating OBS tables and DLI lakehouse tables using Hive syntax, which is supported by Hive dialect DDL statements.
- To create an OBS table using Hive syntax:
  - For the default dialect, set hive.is-external to true in the with properties.
  - For the Hive dialect, use the EXTERNAL keyword in the create table statement.
- To create a DLI lakehouse table using Hive syntax:
  - For the Hive dialect, add 'is_lakehouse'='true' to the table properties.
When creating a Flink OpenSource SQL job, enable checkpointing in the job editing interface.