Update cache on materialization

[sakce]

Problem

When an endpoint's materialization changes (due to a new version or a materialization run), the existing cache might contain stale results, leading to users seeing outdated data. The cache needs to be invalidated if it predates these updates.

Changes

• Introduced _get_cache_age_for_version to dynamically calculate the effective cache age for inline endpoints, ensuring cached results older than the current version's creation are considered stale.
• Introduced _get_cache_age_for_materialization to set the effective cache age for materialized endpoints based on the last_run_at of the associated DataWarehouseSavedQuery, ensuring cached results older than the last materialization run are considered stale.
• Updated _execute_inline_endpoint and _execute_materialized_endpoint to use these dynamically calculated cache_age_seconds during query execution.

How did you test this code?

Added new comprehensive unit tests in products/endpoints/backend/tests/test_endpoint_materialization.py:

• test_cache_invalidated_on_version_change: Verifies cache invalidation when a new endpoint version is created.
• test_cache_invalidated_on_materialization_update: Verifies cache invalidation when a materialized table is updated.
• test_version_cache_age_respects_endpoint_cache_setting: Ensures version-based cache age does not exceed the endpoint's configured cache_age_seconds.
  All existing and new tests pass.

Changelog: (features only) Is this feature complete?

No

---

Slack Thread

 

[cursor]

Cursor Agent can help with this pull request. Just @cursor in comments and I'll start working on changes in this branch.
_{Learn more about Cursor Agents}