You can use an Avro or Parquet format complex file data object as a source or target without using a Data Processor transformation. The Data Integration Service can directly read and write Avro and Parquet resources that contain flat structure to HDFS or local file system.

You can use a complex file data object with an intelligent structure model resource as a source in a mapping that runs over Spark. When you associate a complex file data object with an intelligent structure model, you can use any file input that the intelligent structure model applies to without using a Data Processor transformation.

When you use a binary complex file data object as a source, you can use a Data Processor transformation to parse the binary output of the complex file.

Configure the Data Processor transformation as follows:

Set an input port to buffer input and binary data type. Specify the port size. The port size that you specify in the complex file properties and the Data Processor transformation must be the same.
Set an output port to buffer output or set it for relational output. If you set the ports for relational output, specify the ports based on the number of relational groups of ports you want in the output. Specify the port size for the ports. You can use an XML schema reference that describes the XML hierarchy.
Set a Streamer object as a startup component.

If you configure a binary complex file data object with an intelligent structure model, you do not need to use a Data Processor transformation to parse the output of the complex file.

When you use a complex file data object as a target, you must use a Data Processor transformation to convert the source data into a binary format. Set the Data Processor transformation port to binary. You can then use the binary stream as an input to the complex file data object.