type_spec - Pre/Post conditions for Python functions

This is a python module that type checks for Python function arguments and return values.

Installation

pip install type_spec

Example Usage

A function that takes built-in types:

from type_spec import typeSpec

@typeSpec(int, int, int)
def add(a, b):
  return a + b

A function that takes any of a number of built in types with certain constraints:

from type_spec import typeSpec, AllOf, AnyOf, type_check

def Number(v):
    return type_check(v, AnyOf(int, float))

def GreaterThanZero(v):
    if v <= 0:
        return 'must be greater than zero'

@typeSpec(Number, AllOf(Number, GreaterThanZero), Number)
def div(a, b):
    return a / b

Usage Details

To specify a pre/post condition for your function you use the typeSpec decorator. The signature is:

typeSpec(arg_1_type, ..., arg_n_type, return_type)

The arguments/types provided to typeSpec can be any of the following:

• A built-in type such as int, float, str, bool, list, dict, set etc.
• A class, i.e. MyClass
• A function/callable that takes a value as argument and returns None if the value has a valid type and an error message (string) otherwise.
• A dict instance that specifies that a valid value must be a dictionary with a certain structure, i.e. {'name': str, 'enabled': bool}
• A list instance that specifies that a valid value must be a list with items of a certain type, i.e. [str].
• A predicate function that returns True or False, i.e. lambda n: n > 0. Note that you can use the callable built-in predicate function to type check functions.
• An example value/instance, i.e. None, "2017-01-01". This means the value needs to be an instance of the same type as the example value.

Resources

• typeannotations - uses the elegant approach of overriding isinstance for its type checks (through metaclasses) along with the built-in Python 3 type annotation syntax.
• PEP 362 - Function Signature Object - annotation checker