Basedmypy is a Python type checker that is built on top of the work done by the mypy project. It adds based functionality and breaks compatibility with the cringe parts of pep 484.
Based features include:
- Typesafe by default (optional and dynamic typing still supported)
- Baseline functionality
- Support for
Intersectiontypes - Default return type of
Noneinstead ofAny - Generic
TypeVarbounds - Based type-guards
- Infer parameter type from default value
- Infer overload types
- Bare literals
- Tuple literal types
See the features for more information, or the docs for a comprehensive list.
Basedmypy can be installed using pip from PyPI or from this GitHub repo:
python -m pip install -U basedmypy Basedmypy is installed as an alternative to, and in place of, the mypy installation:
mypy test.py python -m mypy test.py Have you ever tried to use Python's type system and thought to yourself "This doesn't seem based"?
Well fret no longer as basedmypy has got you covered!
You can find a comprehensive list in the docs.
Basedmypy has baseline, baseline is based! It allows you to adopt new strictness or features without the burden of refactoring and fixing every new error, just save all current errors to the baseline file and resolve them at what ever pace you want. Only new code will report new errors.
Read more and see examples in the docs
Using the & operator or basedtyping.Intersection you can denote intersection types:
class Growable(ABC, Generic[T]): @abstractmethod def add(self, item: T): ... class Resettable(ABC): @abstractmethod def reset(self): ... def f(x: Resettable & Growable[str]): x.reset() x.add("first")Mypy joins types to their common base type:
a: int b: str reveal_type(a if bool() else b) # Revealed type is "builtins.object"Basedmypy joins types into unions instead:
a: int b: str reveal_type(a if bool() else b) # Revealed type is "int | str"Literal is so cumbersome! just use a bare literal instead.
class Color(Enum): RED = auto() a: 1 | 2 b: True | Color.REDThe default return type of functions is None instead of Any: (configurable with the default_return option.)
def f(name: str): print(f"Hello, {name}!") reveal_type(f) # (str) -> NoneBasedmpy allows the bounds of TypeVars to be generic.
So you are able to have functions with polymorphic generic parameters:
E = TypeVar("E") I = TypeVar("I", bound=Iterable[E]) def foo(i: I, e: E) -> I: assert e not in i return i reveal_type(foo(["based"], "mypy")) # N: Revealed type is "list[str]" reveal_type(foo({1, 2}, 3)) # N: Revealed type is "set[int]"Type-guards have been re-designed from the ground up:
# The target parameter of the typeguard can be specified def guard(name: str, x: object) -> x is int: ... # impossible type-guards show an error def bad(x: str) -> x is int: ... # error: A type-guard's type must be assignable to its parameter's type. (guard has type "int", parameter has type "str") class A: ... class B: ... def is_b(x: object) -> x is B: ... x = A() assert is_b(x) # type-guards narrow instead of resetting the type reveal_type(x) # A & B # type-guards work on instance parameters class Foo: def guard(self) -> self is int: ... f = Foo() assert f.guard() reveal_type(f) # Foo & intThe types in overload implementations (including properties) can be inferred:
@overload def f(a: int) -> str: ... @overload def f(a: str) -> int: ... def f(a): reveal_type(a) # int | str return None # error: expected str | int class A: @property def foo(self) -> int: ... @foo.setter def foo(self, value): ... # no need for annotationsInfer the type of a function parameter from its default value:
def f(a=1, b=True): reveal_type((a, b)) # (int, bool)Basedmypy allows denotation of tuple types with tuple literals:
a: (int, str) = (1, "a")Basedmypy makes significant changes to error and info messages, consider:
T = TypeVar("T", bound=int) def f(a: T, b: list[str | 1 | 2]) -> Never: reveal_type((a, b)) reveal_type(f)Mypy shows:
Revealed type is "Tuple[T`-1, Union[builtins.str, Literal[1], Literal[2]]]" Revealed type is "def [T <: builtins.int] (a: T`-1, b: Union[builtins.str, Literal[1], Literal[2]]) -> <nothing>" Basedmypy shows:
Revealed type is "(T@f, str | 1 | 2)" Revealed type is "def [T: int] (a: T, b: str | 1 | 2) -> Never" Feel free to start a discussion or raise an issue, we're happy to respond:
- basedmypy tracker for basedmypy issues
- basedtypeshed tracker for issues with specific modules
- basedtyping tracker for issues with the 'basedtyping' package (runtime functionality).

