reckon: Dead-simple, dynamic memoization

Installation

In order to install the latest version, simply pip3 install -U reckon.

This library requires Python 3.6 or greater.

What is it?

reckon implements a dynamic LRU cache by automatically monitoring the memory usage of your machine and purging entries as it approaches a pre-defined ratio (defaults to 90%).

reckon is largely inspired by the global_lru_cache package, so credit should be given for the initial implementation. This package brings those ideas into python3 and adds a local cache implementation as well.

Usage

Usage is simple:

import reckon

@reckon.memoize
def some_expensive_func(foo: int, bar: int):
    return foo ** bar

reckon will automatically make use of the global cache.

While the global cache is automatically maintained, it may be necessary to managed the cache manually. To that purpose, reckon provides the following global methods:

• reckon.glob.clear: Clear the global cache.
• reckon.glob.shrink: Shrink the global cache.
• reckon.glob.usage: Check the current usage ratio.
• reckon.glob.set_usage: Set the max memory usage ratio for the global cache.
• reckon.glob.info: View high-level information about the cache - similar to functools.lru_cache.cache_info

If you wish to only maintain a cache local to a function you can simply pass a flag to the decorator:

import reckon

@reckon.memoize(locale="local")
def some_expensive_func(foo: int, bar: int):
    return foo ** bar

Additionally, if you wish to maintain a cache local to a module, you can initialize your own instance of the LocalCache object:

import reckon

cache = reckon.local()

@cache.memoize
def some_expensive_func(foo: int, bar: int):
    return foo ** bar

The local cache instance maintains the same high-level API for management as the global cache:

• LocalCache.clear: Clear the local cache.
• LocalCache.shrink: Shrink the local cache.
• LocalCache.usage: Check the current usage ratio.
• LocalCache.set_usage: Set the max memory usage ratio for the local cache.
• LocalCache.info: View high-level information about the cache - similar to functools.lru_cache.cache_info

All memoized functions have introspection into their cache via the cache attribute.

Documentation

Full documentation coming soon!

How to Contribute

1. Check for open issues or open a fresh issue to start a discussion around a feature idea or a bug.
2. Create a branch on Github for your issue or fork the repository on GitHub to start making your changes to the master branch.
3. Write a test which shows that the bug was fixed or that the feature works as expected.
4. Send a pull request and bug the maintainer until it gets merged and published. :)