Understanding the Caching Architecture¶

consumo supports caching for URLs, as that path takes longer than the file one. For a better library experience, consumo uses Dependency Injection to provide a database-agnostic caching experience.

The Core Logic is Database-Agnostic¶

The URL version of the get_duration function, get_url_duration relies on two function parameters for caching needs:

get_cached_resolver(program_name: str, key: str) -> int

Gets the value assigned to key from the cache.

cache_resolver(program_name: str, value: int, time_to_live: int) -> int

Saves the value related to key in the cache.

These parameters are dummy by default so one has to implement them with the caching backend they have in mind before considering the use of caching.

How the CLI Uses It¶

consumo has a built-in SQLite implementation. SQLite was picked because it is a serverless database. That means you don't need to have a massive database server running in the background just to use a CLI.

Why Build It This Way?¶

This allows developers to scale up consumo's implementation of consumption time calculation, in case they want to integrate it into a larger program, that may need a larger database.