Catalog-Based UDTF Registration
Introduction
Catalog-based UDTF registration registers User-Defined Table Functions (UDTFs) and Views in Unity Catalog, providing permanent, discoverable, and governable access to CDF data. This approach is ideal for:
- Production Deployments: Permanent registration with Unity Catalog governance
- Data Discovery: Views are indexed and searchable in the Databricks UI
- Access Control: Unity Catalog permissions (GRANT/REVOKE) for fine-grained access control
- Enterprise Security: Credentials stored securely in Databricks Secret Manager
- Team Collaboration: Shared, discoverable data assets across teams
Unity Catalog provides a three-level hierarchy: catalog.schema.object where UDTFs and Views are registered as objects within schemas, which are organized within catalogs.
Overview
This documentation covers the complete workflow for using catalog-based UDTFs and Views:
- Quickstart: Get from zero to registered UDTFs and Views in one notebook (catalog-based)
- Prerequisites: System requirements and access permissions
- Secret Manager: Set up secure credential storage
- Registration: Register UDTFs and Views in Unity Catalog
- Views: Understand Views and their benefits
- Querying: Query Views and UDTFs directly
- Filtering: Filter data using WHERE clauses with predicate pushdown
- Joining: Join data from different Views based on
external_idandspace - Time Series: Work with template-generated time series UDTFs in Unity Catalog (same template-based generation as Data Model UDTFs)
- SQL-Native Time Series (Alpha): SQL-native time series UDTF with pushdown hints (experimental)
- Governance: Set up Unity Catalog permissions
- Troubleshooting: Common issues and solutions
Quick Links
Quickstart
- Catalog-Based Quickstart: End-to-end flow based on the init quickstart notebook (install → generate → Secret Manager → register UDTFs and Views)
Examples
- Registration and Views: Secret Manager setup, UDTF/View registration, Unity Catalog verification
- Querying Views: Query Views (no credentials) and UDTFs directly
- Filtering Views: Filter Views with WHERE clauses
- Joining Views: Join Views and join Views with time series UDTFs
Related Documentation
- Session-Scoped UDTF Registration: For development and testing
- Technical Plan: CDF Databricks Integration (UDTF-Based)