Skip to content

Getting Started with Computed Join

A Computed Join combines data from two or more source containers into a single result you can query and check. Define it pairwise (two containers on a single key) or in SQL mode (a Spark SQL query over multiple aliased sources). Use it to merge data across datastores, write reusable joins, and run quality checks on the joined output. A Computed Join lives inside its parent datastore and appears alongside that datastore's other containers.

Add Computed Join form with left and right reference selectors, select expression, and filter clause

Permissions

To create, edit, or delete a Computed Join you need:

  • Editor on the parent datastore, or Author on the parent datastore plus ownership of the Computed Join.
  • At least Viewer on every other datastore an input reads from (cross-datastore joins).

The parent datastore is the left container's datastore in pairwise mode, or the first source's datastore in SQL mode.

Deep Dive

  • Introduction

    What a Computed Join is, why it matters, and where it fits in the platform.

    Introduction

  • How It Works

    Form fields, execution flow, cross-datastore behavior, and output behavior.

    How It Works

  • Supported Inputs

    Which container types you can use on either side and how to work around restrictions.

    Supported Inputs

  • Query Diff

    How configuration changes are tracked and how to inspect the diff.

    Query Diff

  • Best Practices

    Recommended patterns for prefixes, filters, performance, and cross-datastore joins.

    Best Practices

  • Permissions

    Who can view, create, edit, and delete a Computed Join, including ownership exceptions and cross-datastore rules.

    Permissions

How-tos

  • Create a Computed Join

    Step-by-step walkthrough of the Add Computed Join form.

    Create a Computed Join

  • Edit a Computed Join

    What can and cannot be edited, and how the "Fields will be marked as missing" warning works.

    Edit a Computed Join

  • View Query Diff

    Open the diff dialog from the History section on the Overview tab and review the From / To changes.

    View Query Diff

  • Delete a Computed Join

    What gets removed when deleting, and how to handle blocked deletes.

    Delete a Computed Join

Reference

  • Troubleshooting

    Common validation and execution errors with cause and resolution.

    Troubleshooting

  • API

    REST endpoints for creating, editing, and deleting a Computed Join.

    API

  • FAQ

    Frequently asked questions about Computed Join.

    FAQ