With code extension in Data 360, you can run custom Python batch transformations on a data lake object (DLO) backed by a zero-copy connector, or on a data model object (DMO) mapped from that DLO. The connector accesses the data directly on the external platform without copying it. Your code extension script runs the same custom Python transformations on zero-copy data as it does on any other DLO or DMO in Data 360.
Where: This change applies to Data 360 in Enterprise, Performance, Unlimited, and Developer editions. This feature isn't available in Government Cloud.
When: This feature is available on a rolling basis starting in August 2026.
Why: Transforming a zero-copy DLO or a DMO mapped from that DLO streamlines data pipelines and reduces storage overhead.
- Skip the ingestion step and the storage cost associated with copying large, external platform datasets into Data 360.
- Your Python code needs no special configuration or formatting to work with a zero-copy DLO or DMO. Whether the data is federated from the zero-copy source or ingested natively, your script runs the same way.
How: To set up and run Python transformations on zero-copy data, configure a supported zero-copy connector, and create a federated data stream that generates the DLO. If your script reads DMOs, map the DLO to a DMO in that data space. Next, write your custom script using the Code Extension plugin for the Salesforce CLI, and validate the script locally. Finally, deploy the script to Data 360, and then run it by creating a batch data transform. For setup details and the list of supported zero-copy sources, see Read Zero-Copy Data in Code Extension.