Aggregating data in Django using database views — Mikuláš Poul

Database views in Django provide a way to store and reuse complex queries as virtual tables that behave like regular models

The Django-PgViews-Redux library offers PostgreSQL-specific support for implementing database views with features like:

• Automatic field detection and type serialization
• Support for indexes and materialized views
• Management commands for refreshing views
• Integration with Django REST framework

Materialized views store query results on disk rather than computing them each time:

• Useful for expensive calculations that don’t need real-time updates
• Good for batch processing and reporting use cases
• Requires periodic refreshing to update stale data
• Takes additional disk space but can be indexed for faster queries

Key use cases for database views:

• Aggregating statistics and reports
• Backwards compatibility when changing schemas
• Denormalizing data for read performance
• Storing expensive calculated fields
• Gradual migration of legacy code

Database views have limitations:

• Read-only by default
• Regular views still evaluate query each time
• Materialized views need full recalculation on refresh
• Not suitable when real-time data is required
• Can become performance bottleneck at scale

Views can be defined using raw SQL or through Django ORM-style syntax

• Must include an ID field for Django compatibility
• Support foreign key relationships
• Can use custom managers and querysets
• Integrate with Django admin and REST framework