Typish

• Functions for thorough checks on types
• Instance checks considering generics
• Typesafe Duck-typing

Example

>>> from typing import Iterable
>>> from typish import instance_of
>>> instance_of([1, 2, 3], Iterable[int])
True

Installation

pip install typish

Content

Functions

Types

Decorators

hintable

This decorator allows one to capture the type hint of a variable that calls a function. If no hint is provided, None is passed as a value for hint.

Just remember: with great power comes great responsibility. Use this functionality wisely. You may want to make sure that if you hinted a variable with a certain type, your hintable function does indeed return a value of that type.

@hintable
def cast(arg: Any, hint: Type[T]) -> T:
    return hint(arg)

# The type hint on x is passed to cast:
x: int = cast('42')

# It works with MyPy hints as well:
y = cast('42')  # type: int

# Not something you would normally do, but the type hint takes precedence:
z: int = cast('42')  # type: str

Classes

SubscriptableType

This metaclass allows a type to become subscriptable.

Example:

class MyClass(metaclass=SubscriptableType):
    ...

Now you can do:

MyClass2 = MyClass['some args']
print(MyClass2.__args__)
print(MyClass2.__origin__)

Output:

some args
<class '__main__.MyClass'>

Something

Define an interface with typish.Something.

Example:

Duck = Something['walk': Callable[[], None], 
                 'quack': Callable[[], None]]

Anything that has the attributes defined in Something with the right type is considered an instance of that Something (classes, objects, even modules...).

The builtin isinstance is supported as well as typish.instance_of.

ClsDict

A dictionary that uses instance checking to determine which value to return. It only accepts types as keys.

This is particularly useful when a function accepts multiple types for an argument and you want to split the implementation into separate functions.

Example:

def _handle_str(item):
    ...

def _handle_int(item):
    ...

def func(item):
    # Suppose item can be a string or an int, you can use ClsDict to
    # pick a handler function.

    cd = ClsDict({
        str: _handle_str,
        int: _handle_int,
    })

    handler = cd[item]  # Pick the right handler.
    handler(item)       # Call that handler.

ClsFunction

A callable that uses ClsDict to call the right function. Below is the same example as above, but slightly modified in that it uses ClsFunction.

Example:

def _handle_str(item):
    ...


def _handle_int(item):
    ...


def func(item):
    # Suppose item can be a string or an int, you can use ClsFunction to
    # delegate to the right handler function.

    function = ClsFunction({
        str: _handle_str,
        int: _handle_int,
    })

    function(item)

Literal

A backwards compatible variant of typing.Literal (Python3.8). When importing Literal from typish, you will get the typing.Literal if it is available.