Available on GitHub
Previously I covered a more sophisticated caching solution for Firestore. However, you don't always need something that complex.
Sometimes you just want an expensive function call to be cached for 5 or 10 minutes to reduce load. This is often the case for read-focused operations where it's ok if the results are a little stale. Especially things like search results, image caching, certain computationally expensive operations, etc.
For that purpose I built this transparent caching wrapper.
- The cache is periodically updated in the background without blocking the primary call. So it's always fast.
- Simplicity. Just wrap any function and it becomes cached on the next call.
- Includes both local LRU cache and Redis cache levels. This improves speed and as a bonus minor network interrupts don't effect serving from the local cache.
In the most basic case, you can just supply the redis configuration, and then wrap the function.
Beyond that, you can specify global defaults for cache sizes and TTL.
And you can override any defaults at the moment the function is wrapped.
That's it! A simple caching solution for read-heavy operations.