UFO 🛸

uniform functional orchestrators

Delicious monads and functional programming patterns in python!

Check out the api docs for reference

Project Goals

This library is meant to implement a simple tools for functional programming in python in a way that:

• Can be intergrated quickly/simply with existing code
• Maintains type safety
• Depends only on the standard library

Quick Start

Install libary with pip:

pip install monadcontainers

You can now import monad classes for handling functional chains over data.

Say we want to pass a value through a list of functions like so:

def add_six(x: int) -> int:
  return x + 6

def minus_two(x: int) -> int:
  return x - 2

x = minus_two(add_six(4))  # x == 8

This can be a little messy with long calls, more importantly monads can handle repeated work for us with each call (but we'll cover that in a sec).

A basic identity Monad implementing the above looks like this:

from monadcontainers.monads import Monad

x = (
  Monad(4)
  .bind(add_six)
  .bind(minus_two)
)  # x == Monad(8)

y = x.value  # y == 8

This above pattern chains the Monad contained value through the relevant functions. For alternative syntactical sugar to the bind method, you can use >>:

x = (
  Monad(4)
  >> add_six
  >> minus_two
)  # x == Monad(8)

We can also use Monads to handle additional functionality for us, like error handling with the Result monad:

def divide_by_zero(x: int) -> int:
    return x / 0

x = (
  Result(4)
  >> add_six
  >> divide_by_zero
  >> minus_two
)  # x is a result monad

x.unwrap()  # <- raises divide by zero exception
x.unwrap_or(42)  # <- evaluates to 42
x.value  # <- is None (since exception has been raised)
x.exception  # <- is a ZeroDivisionError
x.recover(add_six)  # <- will apply the given function to the last non-error state

Monad's can be a helpful pattern for functional programming, see the docs for more info on available Monad classes.

Enjoy! 👍