Function argument validation for humans for Python 3!

Highlights

• Straight port from the amazing library ow from Sindre Sorhus
• Expressive chainable API
• Pythonic approach via decorator
• Lots of built-in validations
• Supports custom validations
• Written in Python3 with Type hinting

Notes

Since this is a straight up port from the JavaScript library not all features are available. Partly since this is a port and I haven't caught up and also since Python doesn't support all usecases as JavaScript does.

Install

$ pip3 install pyow

Usage

from pyow import pyow

def unicorn(input):
	pyow(input, pyow.string.min_length(5))

	# ...

unicorn(3)
>>> ArgumentError: 'Expected argument to be of type `str` but received type `int`

unicorn('yo')
>>> ArgumentError: Expected string to have a minimum length of `3`, got `yo`

or via decorator

from pyow.decorator import validate

@validate(
    pyow.string.min_length(5)
)
def unicorn(input):
    return 1

API

pyow(value, predicate)

Test if value matches the provided predicate. Throws an ArgumentError if the test fails.

pyow.is_valid(value, predicate)

Returns True if the value matches the predicate, otherwise returns False.

pyow.create(predicate)

Create a reusable validator.

check_password = pyow.create(pyow.string.min_length(6))

check_password('foo')
>>> ArgumentError: ('Expected string to have a minimum length of `6`, got `foo`')

pyow.any(predicate: List[Predicate])

Returns a predicate that verifies if the value matches at least one of the given predicates.

pyow('foo', pyow.any(pyow.string.max_length(3), pyow.number))

pyow.{type}

All the below types return a predicate. Every predicate has some extra operators that you can use to test the value even more fine-grained.

Primitives

• string
• number
• boolean

Built-in types

• list
• set
• dict
• error

Predicates

The following predicates are available on every type.

nix/isnot

Inverts the following predicates.

pyow(1, pyow.number.nix.infinite)
pyow(1, pyow.number.isnot.infinite)

pyow('', pyow.string.isnot.empty);
>>> ArgumentError: [NOT] Expected string to be empty, got ``

is_(fn)

Use a custom validation function. Return True if the value matches the validation, return False if it doesn't.

pyow(1, pyow.number.is_(lambda x: x < 10))

pyow(1, pyow.number.is_(lambda x: x > 10))
>>> ArgumentError: Expected `1` to pass custom validation function

Instead of returning False, you can also return a custom error message which results in a failure.

def greater_than(max_number: int, x: int):
	return x > max_number or f'Expected `{x}` to be greater than `{max_number}`'
};

pyow(5, pyow.number.is_(lambda x: greater_than(10, x)))
>>> ArgumentError: Expected `5` to be greater than `10`

Maintainers

• Henrik Andersson

Related

• @sindresorhus/ow - Function argument validation for humans

License

MIT