Skip to content

When you need just a little more than standard typing. (Probably the #1 library with the highest downloads/star ratio!✨)

License

Notifications You must be signed in to change notification settings

ramonhagenaars/typish

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

image Downloads Pypi version codecov Scrutinizer Code Quality

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

Function Description
subclass_of(cls: type, *args: type) -> bool Returns whether cls is a sub type of all types in args
instance_of(obj: object, *args: type) -> bool Returns whether cls is an instance of all types in args
get_origin(t: type) -> type Return the "origin" of a generic type. E.g. get_origin(List[str]) gives list.
get_args(t: type) -> typing.Tuple[type, ...] Return the arguments of a generic type. E.g. get_args(List[str]) gives (str, ).
get_alias(cls: T) -> typing.Optional[T] Return the typing alias for a type. E.g get_alias(list) gives List.
get_type(inst: T, use_union: bool = False) -> typing.Type[T] Return the (generic) type of an instance. E.g. a list of ints will give List[int].
common_ancestor(*args: object) -> type Return the closest common ancestor of the given instances.
common_ancestor_of_types(*args: type) -> type Return the closest common ancestor of the given classes.
get_args_and_return_type(hint: typing.Type[typing.Callable]) -> typing.Tuple[typing.Optional[typing.Tuple[type]], typing.Optional[type]] Get the argument types and the return type of a callable type hint (e.g. Callable[[int], str]).
get_type_hints_of_callable(func: typing.Callable) -> typing.Dict[str, type] Return the type hints of the parameters of the given callable.
is_type_annotation(item: typing.Any) -> bool Returns whether item is a type or a typing type.
is_from_typing(cls: type) -> bool Returns whether cls is from the typing module.
is_optional_type(cls: type) -> bool Returns whether cls is considered to be an optional type.
get_mro(obj: typing.Any) -> typing.Tuple[type, ...] Wrapper around getmro from inspect to also support typing types.

Types

Type Description
T A generic Type var.
KT A Type var for keys in a dict.
VT A type var for values in a dict.
Empty The type of emptiness (= Parameter.empty).
Unknown The type of something unknown.
Module The type of a module.
NoneType The type of None.
EllipsisType The type of ....

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.

About

When you need just a little more than standard typing. (Probably the #1 library with the highest downloads/star ratio!✨)

Resources

License

Stars

Watchers

Forks

Packages

No packages published

Languages