[pcauthorn]

This was born out of necessity as data comes from multiple systems and takes around 4 minutes per run. Now I can iterate in like 5 seconds

Not something I seen, a decorator used only for development that is deployed in production.

Thoughts?

[dwrodri]

Interesting project! I like the config file. The readme also has a good example.

Can you go a bit more into detail as to why I would pick this over functional.lru_cache?

[pcauthorn]

Thanks for checking it out!

Good question, this is slightly different than lru_cache in that it persists to disk so it retrieves from the cache even when you rerun the Python interpreter.

Also it gives you a bit of flexibility as when to refresh the cache and to use or ignore function parameters. I often make a change in one 'group', db data for example, and am able to just refresh those methods that gather data from the db.