Handling Partitioned Target Tables

Data loading into a partitioned Hive target table is supported. The partitioned columns are indicated using a superscript P in the DI Mapping window.

You can set a static value to a partitioned column from the REV_TAB_PARTITIONS table. If it is set, you can view it from the DI Mapping window by pointing the mouse over the column name. You need not to map the target column to any source column. If you map a source column to a target partitioned column that already has a static value, the static value will get precedence.

If no static value is set to a partitioned column, you can pass a dynamic partitioned valued. You should map a source column to the target partitioned column. If there is no mapping and static value is not set, the empty or blank is passed as the partition value. Hive defaults the partition to _HIVE_DEFAULT_PARTITON_. There is no loss of data in the non-partitioned columns.

Note:

If you need to enable dynamic partition in non-strict mode, set the below property as a Prescript in the Data Mapping window:

set hive.exec.dynamic.partition.mode=nonstrict

Static partition value can also be set with placeholders. The placeholders supported in Data Mapping are $RUNID, $PHID, $EXEID, $RUNSK, $SYSDATE, $TASKID, and $MISDATE. Additionally, partition value can be provided as a parameter within square brackets. For example, [PARAM1]. Passing the parameter values at runtime from RRF/ Operations module is same as for the other Run-time parameters in Data Management Framework. Value for the placeholders/ additional parameters will be substituted as the static partition values during the Run-time. For more information, see Passing Runtime parameters in Data Mapping.