Scripts

Enhance your data using automated Python scripts

Use Python scripts that run after ingest to create new signals and add dataset metadata.

Go to Process → Scripts to define and test your custom scripts. Add scripts to the ingestion pipeline of a datastream under Datastream → Configure → Pipeline.

Scripts run at the end of the ingestion pipeline for every new dataset in the datastream.

Script library

The script library gives you an overview of all scripts in your workspace. Each script has several settings available:

• General — Set the script's name and description, or delete the script.

• Editor — Edit the script definition.

• History — Browse the script's version history.

• Runs — View an overview of all runs of this script across all datasets.

Script specifics

Scripts run in a Python sandbox on Python 3.13. In addition to the Python standard library, the following packages are available for use in your scripts:

• marpledata (Read our SDK docs for more info)

• numpy

• pandas

• pyarrow

• scipy

• requests

• polars

Each script can run for a maximum of 15 minutes.

Test a script

Via Marple DB

To test a script, select a datastream and a dataset, then press Run. Be careful: this executes the script for real; it is not a dry run.

Locally

To iterate faster, develop your script locally in an IDE of your choice. See the example in our SDK docs for a faster local development workflow: Processing scripts tutorial.

Example scripts

The script library includes several example scripts to help you get started. Use one as a starting point and tailor it to your specific needs.

License limits

Your license includes a set amount of compute minutes for running scripts each month. You can check your usage against your license limits under Settings → License. If you're approaching your limit, contact us at [e-mail].