Configuration Reference

Introduction

The configuration is a metadata object that describes the schema of your Databricks workspace, including tables, columns, and the connection details. This configuration is static — changes in your Databricks schema will require updating and reapplying the configuration file. See updating with introspection for more details.

Structure

The configuration object is a JSON object with the following fields:

{
  "connection_uri": {
    "variable": "JDBC_URL"
  },
  "tables": [],
  "primary_keys": [],
  "foreign_keys": {}
}

Property: connection_uri

This field defines the environment variable that contains the JDBC connection URI to connect to your Databricks instance.

Property: tables

This is an array of table definitions that describe the structure of the available data in your Databricks workspace.

Each table includes:

• Fully-qualified name (including schema and catalog)
• List of columns and their types
• Primary key and foreign key metadata (if available)

Updating with introspection

If your Databricks schema changes, you should re-introspect to update the configuration accordingly.

Running update in a configuration directory will:

• Connect to the Databricks workspace using the JDBC_URL
• Scan available tables and columns
• Generate an updated configuration file reflecting the current schema