This article describes the steps to push down analytical functions like RANK() when using CDATA Adapters.

This article describes the steps to resolve No Push Reason when using Analytical functions like RANK() for any CDATA adapter in TIBCO Data Virtualization.

All support environments

The below steps can be used to push down the RANK() function in SparkSQL. The same steps can be used for any of the CDATA adapters.

- Navigate to <TDV_Install_Dir>/packages/SparkSQL_1/conf/ and open the file SparkSQL_properties.yaml.
- Append the following lines to the end of the file:
----------------------------------------------
rank:
      display_name: RANK
      native_expression: RANK()
      arguments:
        combination1:
          arguments_in_order: []
    rank.analytic:
      display_name: RANK
      native_expression: RANK()
      arguments:
        combination1:
          arguments_in_order: []
----------------------------------------------
- Restart the TDV Server.

Once the restart has been done, the execution plan of the query with RANK() function can be observed in TDV to verify that the function is being pushed down.

Note 1: Indentations are important in any YAML file. Ensure that the function names are being indented correctly.
Note 2: The same format as mentioned above can be used to push down any other Analytical function.