beartype —[ the bare-metal type checker ]—

beartype continuous integration (CI) status beartype Read The Docs (RTD) status

Look for the bare necessities,
  the simple bare necessities.
Forget about your worries and your strife.

                        — The Jungle Book.

Beartype is an open-source pure-Python PEP-compliant constant-time runtime type checker emphasizing efficiency, portability, and thrilling puns.

Beartype brings Rust– and C++-inspired zero-cost abstractions into the lawless world of dynamically-typed Python by enforcing type safety at the granular level of functions and methods against type hints standardized by the Python community in O(1) non-amortized worst-case time with negligible constant factors. If the prior sentence was unreadable jargon, see our friendly and approachable FAQ for a human-readable synopsis.

Beartype is portably implemented in Python 3, continuously stress-tested via GitHub Actions + tox + pytest, and permissively distributed under the MIT license. Beartype has no runtime dependencies, only one test-time dependency, and only one documentation-time dependency. Beartype supports all actively developed Python versions, all Python package managers, and multiple platform-specific package managers.

tl;dr:

  1. Install beartype:

    pip3 install beartype
    
  2. Decorate functions and methods annotated by PEP-compliant type hints with the @beartype.beartype decorator:

    from beartype import beartype
    from collections.abc import Iterable
    from typing import Optional
    
    @beartype
    def print_messages(messages: Optional[Iterable[str]] = ('Hello, world.',)):
        print('\n'.join(messages))

Quality assurance has now been assured.

News

Beartype has a roadmap forward to our first major milestone: beartype 1.0.0, delivering perfect constant-time compliance with all annotation standards by late 2021. …in theory

Join the strangely enticing conversation and be a part of the spicy runtime type-checker that goes up to eleven.

Install

Let’s install beartype with pip, because community standards are good:

pip3 install beartype

Let’s install beartype with Anaconda, because corporate standards are (occasionally) good too:

conda config --add channels conda-forge
conda install beartype

macOS

Let’s install beartype with Homebrew on macOS courtesy our third-party tap:

brew install beartype/beartype/beartype

Let’s install beartype with MacPorts on macOS:

sudo port install py-beartype

A big bear hug to our official macOS package maintainer @harens for packaging beartype for our Apple-appreciating audience.

Linux

Let’s install beartype with emerge on Gentoo courtesy a third-party overlay, because source-based Linux distributions are the CPU-bound nuclear option:

emerge --ask app-eselect/eselect-repository
mkdir -p /etc/portage/repos.conf
eselect repository enable raiagent
emerge --sync raiagent
emerge beartype

Overview

Beartype is a novel first line of defense. In Python’s vast arsenal of software quality assurance (SQA), beartype holds the shield wall against breaches in type safety by improper parameter and return values violating developer expectations.

Beartype is unopinionated. Beartype inflicts no developer constraints beyond importation and usage of a single configuration-free decorator. Beartype is trivially integrated into new and existing applications, stacks, modules, and scripts already annotating callables with PEP-compliant industry-standard type hints.

Beartype is zero-cost. Beartype inflicts no harmful developer tradeoffs, instead stressing expense-free strategies at both:

Versus Static Type Checkers

Like competing static type checkers operating at the coarse-grained application level via ad-hoc heuristic type inference (e.g., Pyre, mypy, pyright, pytype), beartype effectively imposes no runtime overhead. Unlike static type checkers:

  • Beartype operates exclusively at the fine-grained callable level of pure-Python functions and methods via the standard decorator design pattern. This renders beartype natively compatible with all interpreters and compilers targeting the Python language – including PyPy, Numba, Nuitka, and (wait for it) CPython itself.
  • Beartype enjoys deterministic Turing-complete access to the actual callables, objects, and types being type-checked. This enables beartype to solve dynamic problems decidable only at runtime – including type-checking of arbitrary objects whose:

Versus Runtime Type Checkers

Unlike comparable runtime type checkers (e.g., enforce, pytypes, typeguard), beartype decorates callables with dynamically generated wrappers efficiently type-checking each parameter passed to and value returned from those callables in constant time. Since “performance by default” is our first-class concern, generated wrappers are guaranteed to:

Frequently Asked Questions (FAQ)

What is beartype?

Why, it’s the world’s first O(1) runtime type checker in any dynamically-typed lang… oh, forget it.

You know typeguard? Then you know beartype – more or less. beartype is typeguard‘s younger, faster, and slightly sketchier brother who routinely ingests performance-enhancing anabolic nootropics.

What is typeguard?

Okay. Work with us here, people.

You know how in low-level statically-typed memory-unsafe languages that no one should use like C and C++, the compiler validates at compilation time the types of all values passed to and returned from all functions and methods across the entire codebase?

$ gcc -Werror=int-conversion -xc - <<EOL
#include <stdio.h>
int main() {
    printf("Hello, world!");
    return "Goodbye, world.";
}
EOL
<stdin>: In function ‘main’:
<stdin>:4:11: error: returning ‘char *’ from a function with return type
‘int’ makes integer from pointer without a cast [-Werror=int-conversion]
cc1: some warnings being treated as errors

You know how in high-level duck-typed languages that everyone should use instead like Python and Ruby, the interpreter performs no such validation at any interpretation phase but instead permits any arbitrary values to be passed to or returned from any function or method?

$ python3 - <<EOL
def main() -> int:
    print("Hello, world!");
    return "Goodbye, world.";
main()
EOL

Hello, world!

Runtime type checkers like beartype and typeguard selectively shift the dial on type safety in Python from duck to static typing while still preserving all of the permissive benefits of the former as a default behaviour.

$ python3 - <<EOL
from beartype import beartype
@beartype
def main() -> int:
    print("Hello, world!");
    return "Goodbye, world.";
main()
EOL

Hello, world!
Traceback (most recent call last):
  File "<stdin>", line 6, in <module>
  File "<string>", line 17, in __beartyped_main
  File "/home/leycec/py/beartype/beartype/_decor/_code/_pep/_error/peperror.py", line 218, in raise_pep_call_exception
    raise exception_cls(
beartype.roar.BeartypeCallHintPepReturnException: @beartyped main() return
'Goodbye, world.' violates type hint <class 'int'>, as value 'Goodbye,
world.' not int.

When should I use beartype?

Use beartype to assure the quality of Python code beyond what tests alone can assure. If you have yet to test, do that first with a pytest-based test suite, tox configuration, and continuous integration (CI). If you have any time, money, or motivation left, annotate callables with PEP-compliant type hints and decorate those callables with the @beartype.beartype decorator.

Prefer beartype over other runtime type checkers whenever you lack control over the objects passed to or returned from your callables – especially whenever you cannot limit the size of those objects. This includes common developer scenarios like:

  • You are the author of an open-source library intended to be reused by a general audience.
  • You are the author of a public app accepting as input or generating as output sufficiently large data internally passed to or returned from app callables.

Prefer beartype over static type checkers whenever:

Even where none of the above apply, still use beartype. It’s free as in beer and speech and cost-free at installation- and runtime. Leverage beartype until you find something that suites you better, because beartype is always better than nothing.

Why should I use beartype?

The idea of beartype is that it never costs you anything. It might not do quite as much as you’d like, but it will always do something – which is more than Python’s default behaviour, which is to do nothing and ignore type hints altogether. This means you can always safely add beartype to any Python package, module, app, or script regardless of size, scope, funding, or audience and never worry about your back-end Django server taking a nosedive on St. Patty’s Day just because your front-end React client helpfully sent a 5MB JSON file serializing a doubly-nested list of integers.

The idea of typeguard is that it does everything. If you annotate a function decorated by typeguard as accepting a triply-nested list of integers and then pass that function a list containing 1,000 nested lists each containing 1,000 nested lists each containing 1,000 integers, every call to that function will check every integer transitively nested in that list – even if that list never changes. Did we mention that list transitively contains 1,000,000,000 integers in total?

$ python3 -m timeit -n 1 -r 1 -s '
from typeguard import typechecked
@typechecked
def behold(the_great_destroyer_of_apps: list[list[list[int]]]) -> int:
    return len(the_great_destroyer_of_apps)
' 'behold([[[0]*1000]*1000]*1000)'

1 loop, best of 1: 6.42e+03 sec per loop

Yes, 6.42e+03 sec per loop == 6420 seconds == 107 minutes == 1 hour, 47 minutes to check a single list once. Yes, it’s an uncommonly large list, but it’s still just a list. This is the worst-case cost of a single call to a function decorated by a naïve runtime type checker.

What does beartype do?

Generally, as little as it can while still satisfying the accepted definition of “runtime type checker.” Specifically, beartype performs a one-way random walk over the expected data structure of objects passed to and returned from @beartype-decorated functions and methods.

Consider the prior example of a function annotated as accepting a triply-nested list of integers passed a list containing 1,000 nested lists each containing 1,000 nested lists each containing 1,000 integers.

When decorated by typeguard, every call to that function checks every integer nested in that list.

When decorated by beartype, every call to the same function checks only a single random integer contained in a single random nested list contained in a single random nested list contained in that parent list. This is what we mean by the quaint phrase “one-way random walk over the expected data structure.”

$ python3 -m timeit -n 1024 -r 4 -s '
from beartype import beartype
@beartype
def behold(the_great_destroyer_of_apps: list[list[list[int]]]) -> int:
   return len(the_great_destroyer_of_apps)
' 'behold([[[0]*1000]*1000]*1000)'

1024 loops, best of 4: 13.8 usec per loop

13.8 usec per loop == 13.8 microseconds = 0.0000138 seconds to transitively check only a random integer nested in a single triply-nested list passed to each call of that function. This is the worst-case cost of a single call to a function decorated by an O(1) runtime type checker.

Usage

Beartype makes type-checking painless, portable, and possibly fun. Just:

Decorate functions and methods annotated by standard type hints with the @beartype.beartype decorator, which wraps those functions and methods in performant type-checking dynamically generated on-the-fly.

Toy Example

Let’s see what that looks like for a “Hello, Jungle!” toy example. Just:

  1. Import the @beartype.beartype decorator:

    from beartype import beartype
  2. Decorate any annotated function with that decorator:

    from sys import stderr, stdout
    from typing import TextIO
    
    @beartype
    def hello_jungle(
        sep: str = ' ',
        end: str = '\n',
        file: TextIO = stdout,
        flush: bool = False,
    ):
        '''
        Print "Hello, Jungle!" to a stream, or to sys.stdout by default.
    
        Optional keyword arguments:
        file:  a file-like object (stream); defaults to the current sys.stdout.
        sep:   string inserted between values, default a space.
        end:   string appended after the last value, default a newline.
        flush: whether to forcibly flush the stream.
        '''
    
        print('Hello, Jungle!', sep, end, file, flush)
  3. Call that function with valid parameters and caper as things work:

    >>> hello_jungle(sep='...ROOOAR!!!!', end='uhoh.', file=stderr, flush=True)
    Hello, Jungle! ...ROOOAR!!!! uhoh.
  4. Call that function with invalid parameters and cringe as things blow up with human-readable exceptions exhibiting the single cause of failure:

    >>> hello_jungle(sep=(
    ...     b"What? Haven't you ever seen a byte-string separator before?"))
    BeartypeCallHintPepParamException: @beartyped hello_jungle() parameter
    sep=b"What? Haven't you ever seen a byte-string separator before?"
    violates type hint <class 'str'>, as value b"What? Haven't you ever seen
    a byte-string separator before?" not str.

Industrial Example

Let’s wrap the third-party numpy.empty_like() function with automated runtime type checking to demonstrate beartype’s support for non-trivial combinations of nested type hints compliant with different PEPs:

from beartype import beartype
from collections.abc import Sequence
from numpy import dtype, empty_like, ndarray
from typing import Optional, Union

@beartype
def empty_like_bear(
    prototype: object,
    dtype: Optional[dtype] = None,
    order: str = 'K',
    subok: bool = True,
    shape: Optional[Union[int, Sequence[int]]] = None,
) -> ndarray:
    return empty_like(prototype, dtype, order, subok, shape)

Note the non-trivial hint for the optional shape parameter, synthesized from a PEP 484-compliant optional of a PEP 484-compliant union of a builtin type and a PEP 585-compliant subscripted abstract base class (ABC), accepting as valid either:

  • The None singleton.
  • An integer.
  • A sequence of integers.

Let’s call that wrapper with both valid and invalid parameters:

>>> empty_like_bear(([1,2,3], [4,5,6]), shape=(2, 2))
array([[94447336794963,              0],
       [             7,             -1]])
>>> empty_like_bear(([1,2,3], [4,5,6]), shape=([2], [2]))
BeartypeCallHintPepParamException: @beartyped empty_like_bear() parameter
shape=([2], [2]) violates type hint typing.Union[int,
collections.abc.Sequence, NoneType], as ([2], [2]):
* Not <class "builtins.NoneType"> or int.
* Tuple item 0 value [2] not int.

Note the human-readable message of the raised exception, containing a bulleted list enumerating the various ways this invalid parameter fails to satisfy its type hint, including the types and indices of the first container item failing to satisfy the nested Sequence[int] hint.

See the “Decoration” section for actual code dynamically generated by beartype for real-world use cases resembling those above. Fun!

Would You Like to Know More?

If you know type hints, you know beartype. Since beartype is driven entirely by tool-agnostic community standards, beartype’s public API is simply the summation of those standards. As the end user, all you need to know is that decorated callables magically begin raising human-readable exceptions when you pass parameter or return values that violate the PEP-compliant type hints annotating those parameter or return values.

If you don’t know type hints, this is your moment to go deep on the hardest hammer in Python’s SQA toolbox. Here are a few friendly primers to guide you on your maiden voyage through the misty archipelagos of type hinting:

Onward to the cheats!

Cheatsheet

Let’s type-check like greased lightning:

# Import the core @beartype decorator.
from beartype import beartype

# Import PEP 593-compliant type hints. Note this requires Python ≥ 3.9.
from typing import Annotated

# Import PEP 585-compliant type hints. Note this requires Python ≥ 3.9.
from collections.abc import (
    Callable, Generator, Iterable, MutableSequence, Sequence)

# Import PEP 544-compliant type hints. Note this requires Python ≥ 3.8.
from typing import Protocol, runtime_checkable

# Import PEP 484-compliant type hints, too. Note that many of these types
# have been deprecated by PEP 585-compliant type hints under Python ≥ 3.9,
# where @beartype emits non-fatal deprecation warnings at decoration time.
# See also: https://docs.python.org/3/library/typing.html
from typing import Any, List, Optional, Tuple, TypeVar, Union

# Import beartype-specific types to annotate callables with, too.
from beartype.cave import (
    NoneType, NoneTypeOr, RegexTypes, ScalarTypes, VersionTypes)

# Import standard abstract base classes (ABCs) for use with @beartype, too.
from numbers import Integral, Real

# Import user-defined classes for use with @beartype, too.
from my_package.my_module import MyClass

# User-defined PEP 544-compliant protocol referenced below in type hints.
# Note this requires Python ≥ 3.8 and that protocols *MUST* be explicitly
# decorated by the @runtime_checkable decorator to be usable with @beartype.
@runtime_checkable
class MyProtocol(Protocol):
    def my_method(self) -> str:
        return (
            'Objects satisfy this protocol only if their '
            'classes define a method with the same signature as this method.'
        )

# User-defined PEP 484-compliant type variable. Note that @beartype currently
# ignores type variables, but that @beartype 0.9.0 is expected to fully
# support type variables. See also: https://github.com/beartype/beartype/issues/7
T = TypeVar('T')

# Decorate functions with @beartype and...
@beartype
def my_function(
    # Annotate builtin types as is.
    param_must_satisfy_builtin_type: str,

    # Annotate user-defined classes as is, too. Note this covariantly
    # matches all instances of both this class and subclasses of this class.
    param_must_satisfy_user_type: MyClass,

    # Annotate PEP 593-compliant types, indexed by a type checked by
    # @beartype followed by arbitrary objects ignored by @beartype.
    param_must_satisfy_pep593: Annotated[dict[int, bool], range(5), True],

    # Annotate PEP 585-compliant builtin container types, indexed by the
    # types of items these containers are required to contain.
    param_must_satisfy_pep585_builtin: list[str],

    # Annotate PEP 585-compliant standard collection types, indexed too.
    param_must_satisfy_pep585_collection: MutableSequence[str],

    # Annotate PEP 544-compliant protocols, either unindexed or indexed by
    # one or more type variables.
    param_must_satisfy_pep544: MyProtocol[T],

    # Annotate PEP 484-compliant non-standard container types defined by the
    # "typing" module, optionally indexed and only usable as type hints.
    # Note that these types have all been deprecated by PEP 585 under Python
    # ≥ 3.9. See also: https://docs.python.org/3/library/typing.html
    param_must_satisfy_pep484_typing: List[int],

    # Annotate PEP 484-compliant unions of arbitrary types, including
    # builtin types, type variables, and PEP 585-compliant type hints.
    param_must_satisfy_pep484_union: Union[dict, T, tuple[MyClass, ...]],

    # Annotate PEP 484-compliant relative forward references dynamically
    # resolved at call time as unqualified classnames relative to the
    # current user-defined submodule. Note this class is defined below and
    # that beartype-specific absolute forward references are also supported.
    param_must_satisfy_pep484_relative_forward_ref: 'MyOtherClass',

    # Annotate PEP-compliant types indexed by similar references. Note that
    # forward references are supported everywhere standard types are.
    param_must_satisfy_pep484_hint_relative_forward_ref: (
        Union['MyPep484Generic', set['MyPep585Generic']]),

    # Annotate beartype-specific types predefined by the beartype cave.
    param_must_satisfy_beartype_type_from_cave: NoneType,

    # Annotate beartype-specific unions of types as tuples.
    param_must_satisfy_beartype_union: (dict, MyClass, int),

    # Annotate beartype-specific unions predefined by the beartype cave.
    param_must_satisfy_beartype_union_from_cave: ScalarTypes,

    # Annotate beartype-specific unions concatenated together.
    param_must_satisfy_beartype_union_concatenated: (Iterator,) + ScalarTypes,

    # Annotate beartype-specific absolute forward references dynamically
    # resolved at call time as fully-qualified "."-delimited classnames.
    param_must_satisfy_beartype_absolute_forward_ref: (
        'my_package.my_module.MyClass'),

    # Annotate beartype-specific forward references in unions of types, too.
    param_must_satisfy_beartype_union_with_forward_ref: (
        Iterable, 'my_package.my_module.MyOtherClass', NoneType),

    # Annotate PEP 484-compliant optional types. Note that parameters
    # annotated by this type typically default to the "None" singleton.
    param_must_satisfy_pep484_optional: Optional[float] = None,

    # Annotate PEP 484-compliant optional unions of types.
    param_must_satisfy_pep484_optional_union: (
        Optional[Union[float, int]]) = None,

    # Annotate beartype-specific optional types.
    param_must_satisfy_beartype_type_optional: NoneTypeOr[float] = None,

    # Annotate beartype-specific optional unions of types.
    param_must_satisfy_beartype_tuple_optional: NoneTypeOr[float, int] = None,

    # Annotate variadic positional arguments as above, too.
    *args: VersionTypes + (Real, 'my_package.my_module.MyVersionType'),

    # Annotate keyword-only arguments as above, too.
    param_must_be_passed_by_keyword_only: Sequence[Union[bool, list[str]]],

# Annotate return types as above, too.
) -> Union[Integral, 'MyPep585Generic', bool]:
    return 0xDEADBEEF

# Decorate generators as above but returning a generator type.
@beartype
def my_generator() -> Generator[int, None, None]:
    yield from range(0xBEEFBABE, 0xCAFEBABE)

# User-defined class referenced in forward references above.
class MyOtherClass:
    # Decorate instance methods as above without annotating "self".
    @beartype
    def __init__(self, scalar: ScalarTypes) -> NoneType:
        self._scalar = scalar

    # Decorate class methods as above without annotating "cls". When
    # chaining decorators, "@beartype" should typically be specified last.
    @classmethod
    @beartype
    def bare_classmethod(cls, regex: RegexTypes, wut: str) -> (
        Callable[(), str]):
        import re
        return lambda: re.sub(regex, 'unbearable', str(cls._scalar) + wut)

    # Decorate static methods as above.
    @staticmethod
    @beartype
    def bare_staticmethod(callable: Callable, *args: str) -> Any:
        return callable(*args)

    # Decorate property getter methods as above.
    @property
    @beartype
    def bare_gettermethod(self) -> Iterator[int]:
        return range(0x0B00B135 + int(self._scalar), 0xB16B00B5)

    # Decorate property setter methods as above.
    @bare_gettermethod.setter
    @beartype
    def bare_settermethod(self, bad: Integral = 0xBAAAAAAD) -> NoneType:
        self._scalar = bad if bad else 0xBADDCAFE

# User-defined PEP 585-compliant generic referenced above in type hints.
# Note this requires Python ≥ 3.9.
class MyPep585Generic(tuple[int, float]):
    # Decorate static class methods as above without annotating "cls".
    @beartype
    def __new__(cls, integer: int, real: float) -> tuple[int, float]:
        return tuple.__new__(cls, (integer, real))

# User-defined PEP 484-compliant generic referenced above in type hints.
class MyPep484Generic(Tuple[str, ...]):
    # Decorate static class methods as above without annotating "cls".
    @beartype
    def __new__(cls, *args: str) -> Tuple[str, ...]:
        return tuple.__new__(cls, args)

Features

Let’s chart current and future compliance with Python’s typing landscape:

category feature versions partially supporting versions fully supporting
decoratable classes none none
  coroutines none none
  functions 0.1.0current 0.1.0current
  generators 0.1.0current 0.1.0current
  methods 0.1.0current 0.1.0current
parameters optional 0.1.0current 0.1.0current
  keyword-only 0.1.0current 0.1.0current
  positional-only none none
  variadic keyword none none
  variadic positional 0.1.0current 0.1.0current
hints covariant 0.1.0current 0.1.0current
  contravariant none none
  absolute forward references 0.1.0current 0.1.0current
  relative forward references 0.4.0current 0.4.0current
  tuple unions 0.1.0current 0.1.0current
builtins None 0.5.2current 0.5.2current
  dict 0.5.0current none
  frozenset 0.5.0current none
  list 0.5.0current 0.5.0current
  set 0.5.0current none
  tuple 0.5.0current 0.5.0current
  type 0.5.0current none
collections collections.ChainMap 0.5.0current none
  collections.Counter 0.5.0current none
  collections.OrderedDict 0.5.0current none
  collections.defaultdict 0.5.0current none
  collections.deque 0.5.0current none
collections.abc collections.abc.AsyncGenerator 0.5.0current none
  collections.abc.AsyncIterable 0.5.0current none
  collections.abc.AsyncIterator 0.5.0current none
  collections.abc.Awaitable 0.5.0current none
  collections.abc.ByteString 0.5.0current 0.5.0current
  collections.abc.Callable 0.5.0current none
  collections.abc.Collection 0.5.0current none
  collections.abc.Container 0.5.0current none
  collections.abc.Coroutine 0.5.0current none
  collections.abc.Generator 0.5.0current none
  collections.abc.ItemsView 0.5.0current none
  collections.abc.Iterable 0.5.0current none
  collections.abc.Iterator 0.5.0current none
  collections.abc.KeysView 0.5.0current none
  collections.abc.Mapping 0.5.0current none
  collections.abc.MappingView 0.5.0current none
  collections.abc.MutableMapping 0.5.0current none
  collections.abc.MutableSequence 0.5.0current 0.5.0current
  collections.abc.MutableSet 0.5.0current none
  collections.abc.Reversible 0.5.0current none
  collections.abc.Sequence 0.5.0current 0.5.0current
  collections.abc.Set 0.5.0current none
  collections.abc.ValuesView 0.5.0current none
contextlib contextlib.AbstractAsyncContextManager 0.5.0current none
  contextlib.AbstractContextManager 0.5.0current none
re re.Match 0.5.0current none
  re.Pattern 0.5.0current none
typing typing.AbstractSet 0.2.0current none
  typing.Annotated 0.4.0current 0.4.0current
  typing.Any 0.2.0current 0.2.0current
  typing.AnyStr 0.4.0current none
  typing.AsyncContextManager 0.4.0current none
  typing.AsyncGenerator 0.2.0current none
  typing.AsyncIterable 0.2.0current none
  typing.AsyncIterator 0.2.0current none
  typing.Awaitable 0.2.0current none
  typing.BinaryIO 0.4.0current none
  typing.ByteString 0.2.0current 0.2.0current
  typing.Callable 0.2.0current none
  typing.ChainMap 0.2.0current none
  typing.ClassVar none none
  typing.Collection 0.2.0current none
  typing.Container 0.2.0current none
  typing.ContextManager 0.4.0current none
  typing.Coroutine 0.2.0current none
  typing.Counter 0.2.0current none
  typing.DefaultDict 0.2.0current none
  typing.Deque 0.2.0current none
  typing.Dict 0.2.0current none
  typing.Final none none
  typing.ForwardRef 0.4.0current 0.4.0current
  typing.FrozenSet 0.2.0current none
  typing.Generator 0.2.0current none
  typing.Generic 0.4.0current 0.4.0current
  typing.Hashable 0.2.0current none
  typing.IO 0.4.0current none
  typing.ItemsView 0.2.0current none
  typing.Iterable 0.2.0current none
  typing.Iterator 0.2.0current none
  typing.KeysView 0.2.0current none
  typing.List 0.2.0current 0.3.0current
  typing.Literal none none
  typing.Mapping 0.2.0current none
  typing.MappingView 0.2.0current none
  typing.Match 0.4.0current none
  typing.MutableMapping 0.2.0current none
  typing.MutableSequence 0.2.0current 0.3.0current
  typing.MutableSet 0.2.0current none
  typing.NamedTuple 0.1.0current none
  typing.NewType 0.4.0current 0.4.0current
  typing.NoReturn 0.4.0current 0.4.0current
  typing.Optional 0.2.0current 0.2.0current
  typing.OrderedDict 0.2.0current none
  typing.Pattern 0.4.0current none
  typing.Protocol 0.4.0current 0.4.0current
  typing.Reversible 0.2.0current none
  typing.Sequence 0.2.0current 0.3.0current
  typing.Set 0.2.0current none
  typing.Sized 0.2.0current 0.2.0current
  typing.SupportsAbs 0.4.0current 0.4.0current
  typing.SupportsBytes 0.4.0current 0.4.0current
  typing.SupportsComplex 0.4.0current 0.4.0current
  typing.SupportsFloat 0.4.0current 0.4.0current
  typing.SupportsIndex 0.4.0current 0.4.0current
  typing.SupportsInt 0.4.0current 0.4.0current
  typing.SupportsRound 0.4.0current 0.4.0current
  typing.Text 0.1.0current 0.1.0current
  typing.TextIO 0.4.0current none
  typing.Tuple 0.2.0current 0.4.0current
  typing.Type 0.2.0current none
  typing.TypedDict 0.1.0current none
  typing.TypeVar 0.4.0current none
  typing.Union 0.2.0current 0.2.0current
  typing.ValuesView 0.2.0current none
  typing.TYPE_CHECKING 0.5.0current 0.5.0current
  @typing.final none none
  @typing.no_type_check 0.5.0current 0.5.0current
PEP 484 0.2.0current none
  544 0.4.0current 0.4.0current
  560 0.4.0current 0.4.0current
  563 0.1.1current 0.1.1current
  572 0.3.0current 0.4.0current
  585 0.5.0current 0.5.0current
  586 none none
  589 none none
  591 none none
  593 0.4.0current 0.4.0current
packages PyPI 0.1.0current
  Anaconda 0.1.0current
  Gentoo Linux 0.2.0current
  macOS Homebrew 0.5.1current
  macOS MacPorts 0.5.1current
Python 3.5 0.1.00.3.0
  3.6 0.1.0current
  3.7 0.1.0current
  3.8 0.1.0current
  3.9 0.3.2current

Timings

Let’s profile beartype against other runtime type-checkers with a battery of surely fair, impartial, and unbiased use cases:

$ bin/profile.bash

beartype profiler [version]: 0.0.2

python    [basename]: python3.9
python    [version]: Python 3.9.0
beartype  [version]: 0.5.2
typeguard [version]: 2.9.1

===================================== str =====================================
profiling regime:
   number of meta-loops:      3
   number of loops:           100
   number of calls each loop: 100
decoration         [none     ]: 100 loops, best of 3: 359 nsec per loop
decoration         [beartype ]: 100 loops, best of 3: 389 usec per loop
decoration         [typeguard]: 100 loops, best of 3: 13.5 usec per loop
decoration + calls [none     ]: 100 loops, best of 3: 14.8 usec per loop
decoration + calls [beartype ]: 100 loops, best of 3: 514 usec per loop
decoration + calls [typeguard]: 100 loops, best of 3: 6.34 msec per loop

=============================== Union[int, str] ===============================
profiling regime:
   number of meta-loops:      3
   number of loops:           100
   number of calls each loop: 100
decoration         [none     ]: 100 loops, best of 3: 1.83 usec per loop
decoration         [beartype ]: 100 loops, best of 3: 433 usec per loop
decoration         [typeguard]: 100 loops, best of 3: 15.6 usec per loop
decoration + calls [none     ]: 100 loops, best of 3: 17.7 usec per loop
decoration + calls [beartype ]: 100 loops, best of 3: 572 usec per loop
decoration + calls [typeguard]: 100 loops, best of 3: 10 msec per loop

=========================== List[int] of 1000 items ===========================
profiling regime:
   number of meta-loops:      1
   number of loops:           1
   number of calls each loop: 7485
decoration         [none     ]: 1 loop, best of 1: 10.1 usec per loop
decoration         [beartype ]: 1 loop, best of 1: 1.3 msec per loop
decoration         [typeguard]: 1 loop, best of 1: 41.1 usec per loop
decoration + calls [none     ]: 1 loop, best of 1: 1.24 msec per loop
decoration + calls [beartype ]: 1 loop, best of 1: 18.3 msec per loop
decoration + calls [typeguard]: 1 loop, best of 1: 104 sec per loop

============ List[Sequence[MutableSequence[int]]] of 10 items each ============
profiling regime:
   number of meta-loops:      1
   number of loops:           1
   number of calls each loop: 7485
decoration         [none     ]: 1 loop, best of 1: 11.8 usec per loop
decoration         [beartype ]: 1 loop, best of 1: 1.77 msec per loop
decoration         [typeguard]: 1 loop, best of 1: 48.9 usec per loop
decoration + calls [none     ]: 1 loop, best of 1: 1.19 msec per loop
decoration + calls [beartype ]: 1 loop, best of 1: 81.2 msec per loop
decoration + calls [typeguard]: 1 loop, best of 1: 17.3 sec per loop

Note

  • sec = seconds.
  • msec = milliseconds = 10-3 seconds.
  • usec = microseconds = 10-6 seconds.
  • nsec = nanoseconds = 10-9 seconds.

ELI5

beartype is:

  • At least twenty times faster (i.e., 20,000%) and consumes three orders of magnitude less time in the worst case than typeguard – the only comparable runtime type-checker also compatible with most modern Python versions.
  • Asymptotically faster in the best case than typeguard, which scales linearly (rather than not at all) with the size of checked containers.
  • Constant across type hints, taking roughly the same time to check parameters and return values hinted by the builtin type str as it does to check those hinted by the unified type Union[int, str] as it does to check those hinted by the container type List[object]. typeguard is variable across type hints, taking significantly longer to check List[object] as as it does to check Union[int, str], which takes roughly twice the time as it does to check str.

beartype performs most of its work at decoration time. The @beartype decorator consumes most of the time needed to first decorate and then repeatedly call a decorated function. beartype is thus front-loaded. After paying the initial cost of decoration, each type-checked call thereafter incurs comparatively little overhead.

Conventional runtime type checkers perform most of their work at call time. The @typeguard.typechecked and similar decorators consume almost none of the time needed to first decorate and then repeatedly call a decorated function. They are thus back-loaded. Although the initial cost of decoration is essentially free, each type-checked call thereafter incurs significant overhead.

How Much Does All This Cost?

Beartype dynamically generates functions wrapping decorated callables with constant-time runtime type-checking. This separation of concerns means that beartype exhibits different cost profiles at decoration and call time. Whereas standard runtime type-checking decorators are fast at decoration time and slow at call time, beartype is the exact opposite.

At call time, wrapper functions generated by the @beartype decorator are guaranteed to unconditionally run in O(1) non-amortized worst-case time with negligible constant factors regardless of type hint complexity or nesting. This is not an amortized average-case analysis. Wrapper functions really are O(1) time in the best, average, and worst cases.

At decoration time, performance is slightly worse. Internally, beartype non-recursively iterates over type hints at decoration time with a micro-optimized breadth-first search (BFS). Since this BFS is memoized, its cost is paid exactly once per type hint per process; subsequent references to the same hint over different parameters and returns of different callables in the same process reuse the results of the previously memoized BFS for that hint. The @beartype decorator itself thus runs in:

  • O(1) amortized average-case time.
  • O(k) non-amortized worst-case time for k the number of child type hints nested in a parent type hint and including that parent.

Since we generally expect a callable to be decorated only once but called multiple times per process, we might expect the cost of decoration to be ignorable in the aggregate. Interestingly, this is not the case. Although only paid once and obviated through memoization, decoration time is sufficiently expensive and call time sufficiently inexpensive that beartype spends most of its wall-clock merely decorating callables. The actual function wrappers dynamically generated by @beartype consume comparatively little wall-clock, even when repeatedly called many times.

That’s Some Catch, That Catch-22

Beartype’s greatest strength is that it checks types in constant time.

Beartype’s greatest weakness is that it checks types in constant time.

Only so many type-checks can be stuffed into a constant slice of time with negligible constant factors. Let’s detail exactly what (and why) beartype stuffs into its well-bounded slice of the CPU pie.

Standard runtime type checkers naïvely brute-force the problem by type-checking all child objects transitively reachable from parent objects passed to and returned from callables in O(n) linear time for n such objects. This approach avoids false positives (i.e., raising exceptions for valid objects) and false negatives (i.e., failing to raise exceptions for invalid objects), which is good. But this approach also duplicates work when those objects remain unchanged over multiple calls to those callables, which is bad.

Beartype circumvents that badness by generating code at decoration time performing a one-way random tree walk over the expected nested structure of those objects at call time. For each expected nesting level of each container passed to or returned from each callable decorated by @beartype starting at that container and ending either when a check fails or all checks succeed, that callable performs these checks (in order):

  1. A shallow type-check that the current possibly nested container is an instance of the type given by the current possibly nested type hint.
  2. A deep type-check that an item randomly selected from that container itself satisfies the first check.

For example, given a parameter’s type hint list[tuple[Sequence[str]]], beartype generates code at decoration time performing these checks at call time (in order):

  1. A check that the object passed as this parameter is a list.
  2. A check that an item randomly selected from this list is a tuple.
  3. A check that an item randomly selected from this tuple is a sequence.
  4. A check that an item randomly selected from this sequence is a string.

Beartype thus performs one check for each possibly nested type hint for each annotated parameter or return object for each call to each decorated callable. This deep randomness gives us soft statistical expectations as to the number of calls needed to check everything. Specifically, it can be shown that beartype type-checks on average all child objects transitively reachable from parent objects passed to and returned from callables in O(n log n) calls to those callables for n such objects. Praise RNGesus!

Beartype avoids false positives and rarely duplicates work when those objects remain unchanged over multiple calls to those callables, which is good. Sadly, beartype also invites false negatives, because this approach only checks a vertical slice of the full container structure each call, which is bad.

We claim without evidence that false negatives are unlikely under the optimistic assumption that most real-world containers are homogenous (i.e., contain only items of the same type) rather than heterogenous (i.e., contain items of differing types). Examples of homogenous containers include (byte-)strings, ranges, streams, memory views, method resolution orders (MROs), generic alias parameters, lists returned by the dir builtin, iterables generated by the os.walk function, standard NumPy arrays, Pandas DataFrame columns, PyTorch tensors, NetworkX graphs, and really all scientific containers ever.

Nobody Expects the Linearithmic Time

Math time, people. it’s happening

Most runtime type-checkers exhibit O(n) time complexity (where n is the total number of items recursively contained in a container to be checked) by recursively and repeatedly checking all items of all containers passed to or returned from all calls of decorated callables.

beartype guarantees O(1) time complexity by non-recursively but repeatedly checking one random item at all nesting levels of all containers passed to or returned from all calls of decorated callables, thus amortizing the cost of deeply checking containers across calls. (See the subsection on @beartype-generated code deeply type-checking arbitrarily nested containers in constant time for what this means in practice.)

beartype exploits the well-known coupon collector’s problem applied to abstract trees of nested type hints, enabling us to statistically predict the number of calls required to fully type-check all items of an arbitrary container on average. Formally, let:

  • E(T) be the expected number of calls needed to check all items of a container containing only non-container items (i.e., containing no nested subcontainers) either passed to or returned from a @beartype-decorated callable.
  • γ ≈ 0.5772156649 be the Euler–Mascheroni constant.

Then:

https://render.githubusercontent.com/render/math?math=%5Cdisplaystyle+E%28T%29+%3D+n+%5Clog+n+%2B+%5Cgamma+n+%2B+%5Cfrac%7B1%7D%7B2%7D+%2B+O%5Cleft%28%5Cfrac%7B1%7D%7Bn%7D%5Cright%29

The summation ½ + O(1/n) is strictly less than 1 and thus negligible. While non-negligible, the term γn grows significantly slower than the term nlogn. So this reduces to:

https://render.githubusercontent.com/render/math?math=%5Cdisplaystyle+E%28T%29+%3D+O%28n+%5Clog+n%29

We now generalize this bound to the general case. When checking a container containing no subcontainers, beartype only randomly samples one item from that container on each call. When checking a container containing arbitrarily many nested subcontainers, however, beartype randomly samples one random item from each nesting level of that container on each call.

In general, beartype thus samples h random items from a container on each call, where h is that container’s height (i.e., maximum number of edges on the longest path from that container to a non-container leaf item reachable from items directly contained in that container). Since h ≥ 1, beartype samples at least as many items each call as assumed in the usual coupon collector’s problem and thus paradoxically takes a fewer number of calls on average to check all items of a container containing arbitrarily many subcontainers as it does to check all items of a container containing no subcontainers.

Ergo, the expected number of calls E(S) needed to check all items of an arbitrary container exhibits the same or better growth rate and remains bound above by at least the same upper bounds – but probably tighter: e.g.,

https://render.githubusercontent.com/render/math?math=%5Cdisplaystyle+E%28S%29+%3D+O%28E%28T%29%29+%3D+O%28n+%5Clog+n%29%0A

Fully checking a container takes no more calls than that container’s size times the logarithm of that size on average. For example, fully checking a list of 50 integers is expected to take 225 calls on average.

Compliance

beartype is fully compliant with these Python Enhancement Proposals (PEPs):

beartype is currently not compliant whatsoever with these PEPs:

See also the PEP and typing categories of our features matrix for further details.

Full Compliance

beartype deeply type-checks (i.e., directly checks the types of and recursively checks the types of items contained in) parameters and return values annotated with these typing types:

beartype also fully supports callables decorated by these typing decorators:

Lastly, beartype fully supports these typing constants:

Partial Compliance

beartype currently only shallowly type-checks (i.e., only directly checks the types of) parameters and return values annotated with these typing types:

Subsequent beartype versions will deeply type-check these typing types while preserving our O(1) time complexity (with negligible constant factors) guarantee.

No Compliance

beartype currently silently ignores these typing types at decoration time:

beartype currently raises exceptions at decoration time when passed these typing types:

Subsequent beartype versions will first shallowly and then deeply type-check these typing types while preserving our O(1) time complexity (with negligible constant factors) guarantee.

Tutorial

Let’s begin with the simplest type of type-checking supported by @beartype.

Builtin Types

Builtin types like dict, int, list, set, and str are trivially type-checked by annotating parameters and return values with those types as is.

Let’s declare a simple beartyped function accepting a string and a dictionary and returning a tuple:

from beartype import beartype

@beartype
def law_of_the_jungle(wolf: str, pack: dict) -> tuple:
    return (wolf, pack[wolf]) if wolf in pack else None

Let’s call that function with good types:

>>> law_of_the_jungle(wolf='Akela', pack={'Akela': 'alone', 'Raksha': 'protection'})
('Akela', 'alone')

Good function. Let’s call it again with bad types:

>>> law_of_the_jungle(wolf='Akela', pack=['Akela', 'Raksha'])
Traceback (most recent call last):
  File "<ipython-input-10-7763b15e5591>", line 1, in <module>
    law_of_the_jungle(wolf='Akela', pack=['Akela', 'Raksha'])
  File "<string>", line 22, in __law_of_the_jungle_beartyped__
beartype.roar.BeartypeCallTypeParamException: @beartyped law_of_the_jungle() parameter pack=['Akela', 'Raksha'] not a <class 'dict'>.

The beartype.roar submodule publishes exceptions raised at both decoration time by @beartype and at runtime by wrappers generated by @beartype. In this case, a runtime type exception describing the improperly typed pack parameter is raised.

Good function! Let’s call it again with good types exposing a critical issue in this function’s implementation and/or return type annotation:

>>> law_of_the_jungle(wolf='Leela', pack={'Akela': 'alone', 'Raksha': 'protection'})
Traceback (most recent call last):
  File "<ipython-input-10-7763b15e5591>", line 1, in <module>
    law_of_the_jungle(wolf='Leela', pack={'Akela': 'alone', 'Raksha': 'protection'})
  File "<string>", line 28, in __law_of_the_jungle_beartyped__
beartype.roar.BeartypeCallTypeReturnException: @beartyped law_of_the_jungle() return value None not a <class 'tuple'>.

Bad function. Let’s conveniently resolve this by permitting this function to return either a tuple or None as detailed below:

>>> from beartype.cave import NoneType
>>> @beartype
... def law_of_the_jungle(wolf: str, pack: dict) -> (tuple, NoneType):
...     return (wolf, pack[wolf]) if wolf in pack else None
>>> law_of_the_jungle(wolf='Leela', pack={'Akela': 'alone', 'Raksha': 'protection'})
None

The beartype.cave submodule publishes generic types suitable for use with the @beartype decorator and anywhere else you might need them. In this case, the type of the None singleton is imported from this submodule and listed in addition to tuple as an allowed return type from this function.

Note that usage of the beartype.cave submodule is entirely optional (but more efficient and convenient than most alternatives). In this case, the type of the None singleton can also be accessed directly as type(None) and listed in place of NoneType above: e.g.,

>>> @beartype
... def law_of_the_jungle(wolf: str, pack: dict) -> (tuple, type(None)):
...     return (wolf, pack[wolf]) if wolf in pack else None
>>> law_of_the_jungle(wolf='Leela', pack={'Akela': 'alone', 'Raksha': 'protection'})
None

Of course, the beartype.cave submodule also publishes types not accessible directly like RegexCompiledType (i.e., the type of all compiled regular expressions). All else being equal, beartype.cave is preferable.

Good function! The type hints applied to this function now accurately document this function’s API. All’s well that ends typed well. Suck it, Shere Khan.

Arbitrary Types

Everything above also extends to:

  • Arbitrary types like user-defined classes and stock classes in the Python stdlib (e.g., argparse.ArgumentParser) – all of which are also trivially type-checked by annotating parameters and return values with those types.
  • Arbitrary callables like instance methods, class methods, static methods, and generator functions and methods – all of which are also trivially type-checked with the @beartype decorator.

Let’s declare a motley crew of beartyped callables doing various silly things in a strictly typed manner, just ’cause:

from beartype import beartype
from beartype.cave import GeneratorType, IterableType, NoneType

class MaximsOfBaloo(object):
    @beartype
    def __init__(self, sayings: IterableType):
        self.sayings = sayings

@beartype
def inform_baloo(maxims: MaximsOfBaloo) -> GeneratorType:
    for saying in maxims.sayings:
        yield saying

For genericity, the MaximsOfBaloo class initializer accepts any generic iterable (via the beartype.cave.IterableType tuple listing all valid iterable types) rather than an overly specific list or tuple type. Your users may thank you later.

For specificity, the inform_baloo generator function has been explicitly annotated to return a beartype.cave.GeneratorType (i.e., the type returned by functions and methods containing at least one yield statement). Type safety brings good fortune for the New Year.

Let’s iterate over that generator with good types:

>>> maxims = MaximsOfBaloo(sayings={
...     '''If ye find that the Bullock can toss you,
...           or the heavy-browed Sambhur can gore;
...      Ye need not stop work to inform us:
...           we knew it ten seasons before.''',
...     '''“There is none like to me!” says the Cub
...           in the pride of his earliest kill;
...      But the jungle is large and the Cub he is small.
...           Let him think and be still.''',
... })
>>> for maxim in inform_baloo(maxims): print(maxim.splitlines()[-1])
       Let him think and be still.
       we knew it ten seasons before.

Good generator. Let’s call it again with bad types:

>>> for maxim in inform_baloo([
...     'Oppress not the cubs of the stranger,',
...     '     but hail them as Sister and Brother,',
... ]): print(maxim.splitlines()[-1])
Traceback (most recent call last):
  File "<ipython-input-10-7763b15e5591>", line 30, in <module>
    '     but hail them as Sister and Brother,',
  File "<string>", line 12, in __inform_baloo_beartyped__
beartype.roar.BeartypeCallTypeParamException: @beartyped inform_baloo() parameter maxims=['Oppress not the cubs of the stranger,', '     but hail them as Sister and ...'] not a <class '__main__.MaximsOfBaloo'>.

Good generator! The type hints applied to these callables now accurately document their respective APIs. Thanks to the pernicious magic of beartype, all ends typed well… yet again.

Unions of Types

That’s all typed well, but everything above only applies to parameters and return values constrained to singular types. In practice, parameters and return values are often relaxed to any of multiple types referred to as unions of types. You can thank set theory for the jargon… unless you hate set theory. Then it’s just our fault.

Unions of types are trivially type-checked by annotating parameters and return values with the typing.Union type hint containing those types. Let’s declare another beartyped function accepting either a mapping or a string and returning either another function or an integer:

from beartype import beartype
from collections.abc import Callable, Mapping
from numbers import Integral
from typing import Any, Union

@beartype
def toomai_of_the_elephants(memory: Union[Integral, Mapping[Any, Any]]) -> (
    Union[Integral, Callable[(Any,), Any]]):
    return memory if isinstance(memory, Integral) else lambda key: memory[key]

For genericity, the toomai_of_the_elephants function both accepts and returns any generic integer (via the standard numbers.Integral abstract base class (ABC) matching both builtin integers and third-party integers from frameworks like NumPy and SymPy) rather than an overly specific int type. The API you relax may very well be your own.

Let’s call that function with good types:

>>> memory_of_kala_nag = {
...     'remember': 'I will remember what I was, I am sick of rope and chain—',
...     'strength': 'I will remember my old strength and all my forest affairs.',
...     'not sell': 'I will not sell my back to man for a bundle of sugar-cane:',
...     'own kind': 'I will go out to my own kind, and the wood-folk in their lairs.',
...     'morning':  'I will go out until the day, until the morning break—',
...     'caress':   'Out to the wind’s untainted kiss, the water’s clean caress;',
...     'forget':   'I will forget my ankle-ring and snap my picket stake.',
...     'revisit':  'I will revisit my lost loves, and playmates masterless!',
... }
>>> toomai_of_the_elephants(len(memory_of_kala_nag['remember']))
56
>>> toomai_of_the_elephants(memory_of_kala_nag)('remember')
'I will remember what I was, I am sick of rope and chain—'

Good function. Let’s call it again with a tastelessly bad type:

>>> toomai_of_the_elephants(
...     'Shiv, who poured the harvest and made the winds to blow,')
BeartypeCallHintPepParamException: @beartyped toomai_of_the_elephants()
parameter memory='Shiv, who poured the harvest and made the winds to blow,'
violates type hint typing.Union[numbers.Integral, collections.abc.Mapping],
as 'Shiv, who poured the harvest and made the winds to blow,' not <protocol
ABC "collections.abc.Mapping"> or <protocol "numbers.Integral">.

Good function! The type hints applied to this callable now accurately documents its API. All ends typed well… still again and again.

Optional Types

That’s also all typed well, but everything above only applies to mandatory parameters and return values whose types are never NoneType. In practice, parameters and return values are often relaxed to optionally accept any of multiple types including NoneType referred to as optional types.

Optional types are trivially type-checked by annotating optional parameters (parameters whose values default to None) and optional return values (callables returning None rather than raising exceptions in edge cases) with the typing.Optional type hint indexed by those types.

Let’s declare another beartyped function accepting either an enumeration type or None and returning either an enumeration member or None:

from beartype import beartype
from beartype.cave import EnumType, EnumMemberType
from typing import Optional

@beartype
def tell_the_deep_sea_viceroys(story: Optional[EnumType] = None) -> (
    Optional[EnumMemberType]):
    return story if story is None else list(story.__members__.values())[-1]

For efficiency, the typing.Optional type hint creates, caches, and returns new tuples of types appending NoneType to the original types it’s indexed with. Since efficiency is good, typing.Optional is also good.

Let’s call that function with good types:

>>> from enum import Enum
>>> class Lukannon(Enum):
...     WINTER_WHEAT = 'The Beaches of Lukannon—the winter wheat so tall—'
...     SEA_FOG      = 'The dripping, crinkled lichens, and the sea-fog drenching all!'
...     PLAYGROUND   = 'The platforms of our playground, all shining smooth and worn!'
...     HOME         = 'The Beaches of Lukannon—the home where we were born!'
...     MATES        = 'I met my mates in the morning, a broken, scattered band.'
...     CLUB         = 'Men shoot us in the water and club us on the land;'
...     DRIVE        = 'Men drive us to the Salt House like silly sheep and tame,'
...     SEALERS      = 'And still we sing Lukannon—before the sealers came.'
>>> tell_the_deep_sea_viceroys(Lukannon)
<Lukannon.SEALERS: 'And still we sing Lukannon—before the sealers came.'>
>>> tell_the_deep_sea_viceroys()
None

You may now be pondering to yourself grimly in the dark: “…but could we not already do this just by manually annotating optional types with typing.Union type hints explicitly indexed by NoneType?”

You would, of course, be correct. Let’s grimly redeclare the same function accepting and returning the same types – only annotated with NoneType rather than typing.Optional:

from beartype import beartype
from beartype.cave import EnumType, EnumMemberType, NoneType
from typing import Union

@beartype
def tell_the_deep_sea_viceroys(story: Union[EnumType, NoneType] = None) -> (
    Union[EnumMemberType, NoneType]):
    return list(story.__members__.values())[-1] if story is not None else None

Since typing.Optional internally reduces to typing.Union, these two approaches are semantically equivalent. The former is simply syntactic sugar simplifying the latter.

Whereas typing.Union accepts an arbitrary number of child type hints, however, typing.Optional accepts only a single child type hint. This can be circumvented by either indexing typing.Optional by typing.Union or indexing typing.Union by NoneType. Let’s exhibit the former approach by declaring another beartyped function accepting either an enumeration type, enumeration type member, or None and returning either an enumeration type, enumeration type member, or None:

from beartype import beartype
from beartype.cave import EnumType, EnumMemberType, NoneType
from typing import Optional, Union

@beartype
def sang_them_up_the_beach(
    woe: Optional[Union[EnumType, EnumMemberType]] = None) -> (
    Optional[Union[EnumType, EnumMemberType]]):
    return woe if isinstance(woe, (EnumMemberType, NoneType)) else (
        list(woe.__members__.values())[-1])

Let’s call that function with good types:

>>> sang_them_up_the_beach(Lukannon)
<Lukannon.SEALERS: 'And still we sing Lukannon—before the sealers came.'>
>>> sang_them_up_the_beach()
None

Behold! The terrifying power of the typing.Optional type hint, resplendent in its highly over-optimized cache utilization.

Decoration

Let’s take a deep dive into the deep end of runtime type checking – the beartype way. In this subsection, we show code generated by the @beartype decorator in real-world use cases and tell why that code is the fastest possible code type-checking those cases.

Identity Decoration

We begin by wading into the torpid waters of the many ways beartype avoids doing any work whatsoever, because laziness is the virtue we live by. The reader may recall that the fastest decorator at decoration- and call-time is the identity decorator returning its decorated callable unmodified: e.g.,

from collections.abc import Callable

def identity_decorator(func: Callable): -> Callable:
    return func

beartype silently reduces to the identity decorator whenever it can, which is surprisingly often. Our three weapons are laziness, surprise, ruthless efficiency, and an almost fanatical devotion to constant-time type checking.

Unconditional Identity Decoration

Let’s define a trivial function annotated by no type hints:

def law_of_the_jungle(strike_first_and_then_give_tongue):
    return strike_first_and_then_give_tongue

Let’s decorate that function by @beartype and verify that @beartype reduced to the identity decorator by returning that function unmodified:

>>> from beartype import beartype
>>> beartype(law_of_the_jungle) is law_of_the_jungle
True

We’ve verified that @beartype reduces to the identity decorator when decorating unannotated callables. That’s but the tip of the iceberg, though. @beartype unconditionally reduces to a noop when:

Shallow Identity Decoration

Let’s define a trivial function annotated by the PEP 484-compliant typing.Any type hint:

from typing import Any

def law_of_the_jungle_2(never_order_anything_without_a_reason: Any) -> Any:
    return never_order_anything_without_a_reason

Again, let’s decorate that function by @beartype and verify that @beartype reduced to the identity decorator by returning that function unmodified:

>>> from beartype import beartype
>>> beartype(law_of_the_jungle_2) is law_of_the_jungle_2
True

We’ve verified that @beartype reduces to the identity decorator when decorating callables annotated by typing.Any – a novel category of type hint we refer to as shallowly ignorable type hints (known to be ignorable by constant-time lookup in a predefined frozen set). That’s but the snout of the crocodile, though. @beartype conditionally reduces to a noop when all type hints annotating the decorated callable are shallowly ignorable. These include:

  • object, the root superclass of Python’s class hierarchy. Since all objects are instances of object, object conveys no meaningful constraints as a type hint and is thus shallowly ignorable.
  • typing.Any, equivalent to object.
  • typing.Generic, equivalent to typing.Generic[typing.Any], which conveys no meaningful constraints as a type hint and is thus shallowly ignorable.
  • typing.Protocol, equivalent to typing.Protocol[typing.Any] and shallowly ignorable for similar reasons.
  • typing.Union, equivalent to typing.Union[typing.Any], equivalent to Any.
  • typing.Optional, equivalent to typing.Optional[typing.Any], equivalent to Union[Any, type(None)]. Since any union subscripted by ignorable type hints is itself ignorable, [1] typing.Optional is shallowly ignorable as well.
[1] Unions are only as narrow as their widest subscripted argument. However, ignorable type hints are ignorable because they are maximally wide. Unions subscripted by ignorable arguments are thus the widest possible unions, conveying no meaningful constraints and thus themselves ignorable.

Deep Identity Decoration

Let’s define a trivial function annotated by a non-trivial PEP 484-, 585– and 593-compliant type hint that superficially appears to convey meaningful constraints:

from typing import Annotated, NewType, Union

hint = Union[str, list[int], NewType('MetaType', Annotated[object, 53])]
def law_of_the_jungle_3(bring_them_to_the_pack_council: hint) -> hint:
    return bring_them_to_the_pack_council

Despite appearances, it can be shown by exhaustive (and frankly exhausting) reduction that that hint is actually ignorable. Let’s decorate that function by @beartype and verify that @beartype reduced to the identity decorator by returning that function unmodified:

>>> from beartype import beartype
>>> beartype(law_of_the_jungle_3) is law_of_the_jungle_3
True

We’ve verified that @beartype reduces to the identity decorator when decorating callables annotated by the above object – a novel category of type hint we refer to as deeply ignorable type hints (known to be ignorable only by recursive linear-time inspection of subscripted arguments). That’s but the trunk of the elephant, though. @beartype conditionally reduces to a noop when all type hints annotating the decorated callable are deeply ignorable. These include:

Constant Decoration

We continue by trundling into the turbid waters out at sea, where beartype reluctantly performs its minimal amount of work with a heavy sigh.

Constant Builtin Type Decoration

Let’s define a trivial function annotated by type hints that are builtin types:

from beartype import beartype

@beartype
def law_of_the_jungle_4(he_must_be_spoken_for_by_at_least_two: int):
    return he_must_be_spoken_for_by_at_least_two

Let’s see the wrapper function @beartype dynamically generated from that:

def __beartyped_law_of_the_jungle_4(
    *args,
    __beartype_func=__beartype_func,
    __beartypistry=__beartypistry,
    **kwargs
):
    # Localize the number of passed positional arguments for efficiency.
    __beartype_args_len = len(args)
    # Localize this positional or keyword parameter if passed *OR* to the
    # sentinel value "__beartypistry" guaranteed to never be passed otherwise.
    __beartype_pith_0 = (
        args[0] if __beartype_args_len > 0 else
        kwargs.get('he_must_be_spoken_for_by_at_least_two', __beartypistry)
    )

    # If this parameter was passed...
    if __beartype_pith_0 is not __beartypistry:
        # Type-check this passed parameter or return value against this
        # PEP-compliant type hint.
        if not isinstance(__beartype_pith_0, int):
            __beartype_raise_pep_call_exception(
                func=__beartype_func,
                pith_name='he_must_be_spoken_for_by_at_least_two',
                pith_value=__beartype_pith_0,
            )

    # Call this function with all passed parameters and return the value
    # returned from this call.
    return __beartype_func(*args, **kwargs)

Let’s dismantle this bit by bit:

  • The code comments above are verbatim as they appear in the generated code.
  • __beartyped_law_of_the_jungle_4() is the ad-hoc function name @beartype assigned this wrapper function.
  • __beartype_func is the original law_of_the_jungle_4() function.
  • __beartypistry is a thread-safe global registry of all types, tuples of types, and forward references to currently undeclared types visitable from type hints annotating callables decorated by @beartype. We’ll see more about the __beartypistry in a moment. For know, just know that __beartypistry is a private singleton of the beartype package. This object is frequently accessed and thus localized to the body of this wrapper rather than accessed as a global variable, which would be mildly slower.
  • __beartype_pith_0 is the value of the first passed parameter, regardless of whether that parameter is passed as a positional or keyword argument. If unpassed, the value defaults to the __beartypistry. Since no caller should access (let alone pass) that object, that object serves as an efficient sentinel value enabling us to discern passed from unpassed parameters. beartype internally favours the term “pith” (which we absolutely just made up) to transparently refer to the arbitrary object currently being type-checked against its associated type hint.
  • isinstance(__beartype_pith_0, int) tests whether the value passed for this parameter satisfies the type hint annotating this parameter.
  • __beartype_raise_pep_call_exception() raises a human-readable exception if this value fails this type-check.

So good so far. But that’s easy. Let’s delve deeper.

Constant Non-Builtin Type Decoration

Let’s define a trivial function annotated by type hints that are pure-Python classes rather than builtin types:

from argparse import ArgumentParser
from beartype import beartype

@beartype
def law_of_the_jungle_5(a_cub_may_be_bought_at_a_price: ArgumentParser):
    return a_cub_may_be_bought_at_a_price

Let’s see the wrapper function @beartype dynamically generated from that:

def __beartyped_law_of_the_jungle_5(
    *args,
    __beartype_func=__beartype_func,
    __beartypistry=__beartypistry,
    **kwargs
):
    # Localize the number of passed positional arguments for efficiency.
    __beartype_args_len = len(args)
    # Localize this positional or keyword parameter if passed *OR* to the
    # sentinel value "__beartypistry" guaranteed to never be passed otherwise.
    __beartype_pith_0 = (
        args[0] if __beartype_args_len > 0 else
        kwargs.get('a_cub_may_be_bought_at_a_price', __beartypistry)
    )

    # If this parameter was passed...
    if __beartype_pith_0 is not __beartypistry:
        # Type-check this passed parameter or return value against this
        # PEP-compliant type hint.
        if not isinstance(__beartype_pith_0, __beartypistry['argparse.ArgumentParser']):
            __beartype_raise_pep_call_exception(
                func=__beartype_func,
                pith_name='a_cub_may_be_bought_at_a_price',
                pith_value=__beartype_pith_0,
            )

    # Call this function with all passed parameters and return the value
    # returned from this call.
    return __beartype_func(*args, **kwargs)

The result is largely the same. The only meaningful difference is the type-check on line 20:

if not isinstance(__beartype_pith_0, __beartypistry['argparse.ArgumentParser']):

Since we annotated that function with a pure-Python class rather than builtin type, @beartype registered that class with the __beartypistry at decoration time and then subsequently looked that class up with its fully-qualified classname at call time to perform this type-check.

So good so far… so what! Let’s spelunk harder.

Constant Shallow Sequence Decoration

Let’s define a trivial function annotated by type hints that are PEP 585-compliant builtin types subscripted by ignorable arguments:

from beartype import beartype

@beartype
def law_of_the_jungle_6(all_the_jungle_is_thine: list[object]):
    return all_the_jungle_is_thine

Let’s see the wrapper function @beartype dynamically generated from that:

def __beartyped_law_of_the_jungle_6(
    *args,
    __beartype_func=__beartype_func,
    __beartypistry=__beartypistry,
    **kwargs
):
    # Localize the number of passed positional arguments for efficiency.
    __beartype_args_len = len(args)
    # Localize this positional or keyword parameter if passed *OR* to the
    # sentinel value "__beartypistry" guaranteed to never be passed otherwise.
    __beartype_pith_0 = (
        args[0] if __beartype_args_len > 0 else
        kwargs.get('all_the_jungle_is_thine', __beartypistry)
    )

    # If this parameter was passed...
    if __beartype_pith_0 is not __beartypistry:
        # Type-check this passed parameter or return value against this
        # PEP-compliant type hint.
        if not isinstance(__beartype_pith_0, list):
            __beartype_raise_pep_call_exception(
                func=__beartype_func,
                pith_name='all_the_jungle_is_thine',
                pith_value=__beartype_pith_0,
            )

    # Call this function with all passed parameters and return the value
    # returned from this call.
    return __beartype_func(*args, **kwargs)

We are still within the realm of normalcy. Correctly detecting this type hint to be subscripted by an ignorable argument, @beartype only bothered type-checking this parameter to be an instance of this builtin type:

if not isinstance(__beartype_pith_0, list):

It’s time to iteratively up the ante.

Constant Deep Sequence Decoration

Let’s define a trivial function annotated by type hints that are PEP 585-compliant builtin types subscripted by builtin types:

from beartype import beartype

@beartype
def law_of_the_jungle_7(kill_everything_that_thou_canst: list[str]):
    return kill_everything_that_thou_canst

Let’s see the wrapper function @beartype dynamically generated from that:

def __beartyped_law_of_the_jungle_7(
    *args,
    __beartype_func=__beartype_func,
    __beartypistry=__beartypistry,
    **kwargs
):
    # Generate and localize a sufficiently large pseudo-random integer for
    # subsequent indexation in type-checking randomly selected container items.
    __beartype_random_int = __beartype_getrandbits(64)
    # Localize the number of passed positional arguments for efficiency.
    __beartype_args_len = len(args)
    # Localize this positional or keyword parameter if passed *OR* to the
    # sentinel value "__beartypistry" guaranteed to never be passed otherwise.
    __beartype_pith_0 = (
        args[0] if __beartype_args_len > 0 else
        kwargs.get('kill_everything_that_thou_canst', __beartypistry)
    )

    # If this parameter was passed...
    if __beartype_pith_0 is not __beartypistry:
        # Type-check this passed parameter or return value against this
        # PEP-compliant type hint.
        if not (
            # True only if this pith shallowly satisfies this hint.
            isinstance(__beartype_pith_0, list) and
            # True only if either this pith is empty *OR* this pith is
            # both non-empty and deeply satisfies this hint.
            (not __beartype_pith_0 or isinstance(__beartype_pith_0[__beartype_random_int % len(__beartype_pith_0)], str))
        ):
            __beartype_raise_pep_call_exception(
                func=__beartype_func,
                pith_name='kill_everything_that_thou_canst',
                pith_value=__beartype_pith_0,
            )

    # Call this function with all passed parameters and return the value
    # returned from this call.
    return __beartype_func(*args, **kwargs)

We have now diverged from normalcy. Let’s dismantle this iota by iota:

  • __beartype_random_int is a pseudo-random unsigned 32-bit integer whose bit length intentionally corresponds to the number of bits generated by each call to Python’s C-based Mersenne Twister internally performed by the random.getrandbits function generating this integer. Exceeding this length would cause that function to internally perform that call multiple times for no gain. Since the cost of generating integers to this length is the same as generating integers of smaller lengths, this length is preferred. Since most sequences are likely to contain fewer items than this integer, pseudo-random sequence items are indexable by taking the modulo of this integer with the sizes of those sequences. For big sequences containing more than this number of items, beartype deeply type-checks leading items with indices in this range while ignoring trailing items. Given the practical infeasibility of storing big sequences in memory, this seems an acceptable real-world tradeoff. Suck it, big sequences!
  • As before, @beartype first type-checks this parameter to be a list.
  • @beartype then type-checks this parameter to either be:
    • not __beartype_pith_0, an empty list.
    • isinstance(__beartype_pith_0[__beartype_random_int % len(__beartype_pith_0)], str), a non-empty list whose pseudo-randomly indexed list item satisfies this nested builtin type.

Well, that escalated quickly.

Constant Nested Deep Sequence Decoration

Let’s define a trivial function annotated by type hints that are PEP 585-compliant builtin types recursively subscripted by instances of themselves, because we are typing masochists:

from beartype import beartype

@beartype
def law_of_the_jungle_8(pull_thorns_from_all_wolves_paws: (
    list[list[list[str]]])):
    return pull_thorns_from_all_wolves_paws

Let’s see the wrapper function @beartype dynamically generated from that:

def __beartyped_law_of_the_jungle_8(
    *args,
    __beartype_func=__beartype_func,
    __beartypistry=__beartypistry,
    **kwargs
):
    # Generate and localize a sufficiently large pseudo-random integer for
    # subsequent indexation in type-checking randomly selected container items.
    __beartype_random_int = __beartype_getrandbits(32)
    # Localize the number of passed positional arguments for efficiency.
    __beartype_args_len = len(args)
    # Localize this positional or keyword parameter if passed *OR* to the
    # sentinel value "__beartypistry" guaranteed to never be passed otherwise.
    __beartype_pith_0 = (
        args[0] if __beartype_args_len > 0 else
        kwargs.get('pull_thorns_from_all_wolves_paws', __beartypistry)
    )

    # If this parameter was passed...
    if __beartype_pith_0 is not __beartypistry:
        # Type-check this passed parameter or return value against this
        # PEP-compliant type hint.
        if not (
            # True only if this pith shallowly satisfies this hint.
            isinstance(__beartype_pith_0, list) and
            # True only if either this pith is empty *OR* this pith is
            # both non-empty and deeply satisfies this hint.
            (not __beartype_pith_0 or (
                # True only if this pith shallowly satisfies this hint.
                isinstance(__beartype_pith_1 := __beartype_pith_0[__beartype_random_int % len(__beartype_pith_0)], list) and
                # True only if either this pith is empty *OR* this pith is
                # both non-empty and deeply satisfies this hint.
                (not __beartype_pith_1 or (
                    # True only if this pith shallowly satisfies this hint.
                    isinstance(__beartype_pith_2 := __beartype_pith_1[__beartype_random_int % len(__beartype_pith_1)], list) and
                    # True only if either this pith is empty *OR* this pith is
                    # both non-empty and deeply satisfies this hint.
                    (not __beartype_pith_2 or isinstance(__beartype_pith_2[__beartype_random_int % len(__beartype_pith_2)], str))
                ))
            ))
        ):
            __beartype_raise_pep_call_exception(
                func=__beartype_func,
                pith_name='pull_thorns_from_all_wolves_paws',
                pith_value=__beartype_pith_0,
            )

    # Call this function with all passed parameters and return the value
    # returned from this call.
    return __beartype_func(*args, **kwargs)

We are now well beyond the deep end, where the benthic zone and the cruel denizens of the fathomless void begins. Let’s dismantle this pascal by pascal:

  • __beartype_pith_1 := __beartype_pith_0[__beartype_random_int % len(__beartype_pith_0)], a PEP 572-style assignment expression localizing repeatedly accessed random items of the first nested list for efficiency.
  • __beartype_pith_2 := __beartype_pith_1[__beartype_random_int % len(__beartype_pith_1)], a similar expression localizing repeatedly accessed random items of the second nested list.
  • The same __beartype_random_int pseudo-randomly indexes all three lists.
  • Under older Python interpreters lacking PEP 572 support, @beartype generates equally valid (albeit less efficient) code repeating each nested list item access.

In the kingdom of the linear-time runtime type checkers, the constant-time runtime type checker really stands out like a sore giant squid, doesn’t it?

See the Developers section for further commentary on runtime optimization from the higher-level perspective of architecture and internal API design.

Developers

Let’s contribute pull requests to beartype for the good of typing. The primary maintainer of this repository is a friendly beardless Canadian guy who guarantees that he will always be nice and congenial and promptly merge all requests that pass continuous integration (CI) tests.

And thanks for merely reading this! Like all open-source software, beartype thrives on community contributions, activity, and interest. This means you, stalwart Python hero.

beartype has two problem spots (listed below in order of decreasing importance and increasing complexity) that could always benefit from a volunteer army of good GitHub Samaritans.

Workflow

Let’s take this from the top.

  1. Create a GitHub user account.

  2. Login to GitHub with that account.

  3. Click the “Fork” button in the upper right-hand corner of the “beartype/beartype” repository page.

  4. Click the “Code” button in the upper right-hand corner of your fork page that appears.

  5. Copy the URL that appears.

  6. Open a terminal.

  7. Change to the desired parent directory of your local fork.

  8. Clone your fork, replacing {URL} with the previously copied URL.

    git clone {URL}
    
  9. Add a new remote referring to this upstream repository.

    git remote add upstream https://github.com/beartype/beartype.git
    
  10. Uninstall all previously installed versions of beartype. For example, if you previously installed beartype with pip, manually uninstall beartype with pip.

    pip uninstall beartype
    
  11. Install beartype with pip in editable mode. This synchronizes changes made to your fork against the beartype package imported in Python. Note the [dev] extra installs developer-specific mandatory dependencies required at test or documentation time.

    pip3 install -e .[dev]
    
  12. Create a new branch to isolate changes to, replacing {branch_name} with the desired name.

    git checkout -b {branch_name}
    
  13. Make changes to this branch in your favourite Integrated Development Environment (IDE). Of course, this means Vim.

  14. Test these changes. Note this command assumes you have installed all major versions of both CPython and PyPy supported by the next stable release of beartype you are hacking on. If this is not the case, install these versions with pyenv. This is vital, as type hinting support varies significantly between major versions of different Python interpreters.

    tox
    

    The resulting output should ideally be suffixed by a synopsis resembling:

    ________________________________ summary _______________________________
    py36: commands succeeded
    py37: commands succeeded
    py38: commands succeeded
    py39: commands succeeded
    pypy36: commands succeeded
    pypy37: commands succeeded
    congratulations :)
    
  15. Stage these changes.

    git add -a
    
  16. Commit these changes.

    git commit
    
  17. Push these changes to your remote fork.

    git push
    
  18. Click the “Create pull request” button in the upper right-hand corner of your fork page.

  19. Afterward, routinely pull upstream changes to avoid desynchronization with the “beartype/beartype” repository.

    git checkout main && git pull upstream main
    

Moar Depth

So, you want to help beartype deeply type-check even more type hints than she already does? Let us help you help us, because you are awesome.

First, an egregious lore dump. It’s commonly assumed that beartype only internally implements a single type-checker. After all, every other static and runtime type-checker only internally implements a single type-checker. Why would a type-checker internally implement several divergent overlapping type-checkers and… what would that even mean? Who would be so vile, cruel, and sadistic as to do something like that?

We would. beartype often violates assumptions. This is no exception. Externally, of course, beartype presents itself as a single type-checker. Internally, beartype is implemented as a two-phase series of orthogonal type-checkers. Why? Because efficiency, which is the reason we are all here. These type-checkers are (in the order that callables decorated by beartype perform them at runtime):

  1. Testing phase. In this fast first pass, each callable decorated by @beartype only tests whether all parameters passed to and values returned from the current call to that callable satisfy all type hints annotating that callable. This phase does not raise human-readable exceptions (in the event that one or more parameters or return values fails to satisfy these hints). @beartype highly optimizes this phase by dynamically generating one wrapper function wrapping each decorated callable with unique pure-Python performing these tests in O(1) constant-time. This phase is always unconditionally performed by code dynamically generated and returned by:
    • The fast-as-lightning pep_code_check_hint() function declared in the “beartype._decor._code._pep._pephint” submodule, which generates memoized O(1) code type-checking an arbitrary object against an arbitrary PEP-compliant type hint by iterating over all child hints nested in that hint with a highly optimized breadth-first search (BFS) leveraging extreme caching, fragile cleverness, and other salacious micro-optimizations.
  2. Error phase. In this slow second pass, each call to a callable decorated by @beartype that fails the fast first pass (due to one or more parameters or return values failing to satisfy these hints) recursively discovers the exact underlying cause of that failure and raises a human-readable exception precisely detailing that cause. @beartype does not optimize this phase whatsoever. Whereas the implementation of the first phase is uniquely specific to each decorated callable and constrained to O(1) constant-time non-recursive operation, the implementation of the second phase is generically shared between all decorated callables and generalized to O(n) linear-time recursive operation. Efficiency no longer matters when you’re raising exceptions. Exception handling is slow in any language and doubly slow in dynamically-typed (and mostly interpreted) languages like Python, which means that performance is mostly a non-concern in “cold” code paths guaranteed to raise exceptions. This phase is only conditionally performed when the first phase fails by:
    • The slow-as-molasses raise_pep_call_exception() function declared in the “beartype._decor._code._pep._error.peperror” submodule, which generates human-readable exceptions after performing unmemoized O(n) type-checking of an arbitrary object against a PEP-compliant type hint by recursing over all child hints nested in that hint with an unoptimized recursive algorithm prioritizing debuggability, readability, and maintainability.

This separation of concerns between performant O(1) testing on the one hand and perfect O(n) error handling on the other preserves both runtime performance and readable errors at a cost of developer pain. This is good! …what?

Secondly, the same separation of concerns also complicates the development of @beartype. This is bad. Since @beartype internally implements two divergent type-checkers, deeply type-checking a new category of type hint requires adding that support to (wait for it) two divergent type-checkers – which, being fundamentally distinct codebases sharing little code in common, requires violating the Don’t Repeat Yourself (DRY) principle by reinventing the wheel in the second type-checker. Such is the high price of high-octane performance. You probably thought this would be easier and funner. So did we.

Thirdly, this needs to be tested. After surmounting the above roadblocks by deeply type-checking that new category of type hint in both type-checkers, you’ll now add one or more unit tests exhaustively exercising that checking. Thankfully, we already did all of the swole lifting for you. All you need to do is add at least one PEP-compliant type hint, one object satisfying that hint, and one object not satisfying that hint to:

You’re done! Praise Guido.

Moar Compliance

So, you want to help beartype comply with even more Python Enhancement Proposals (PEPs) than she already complies with? Let us help you help us, because you are young and idealistic and you mean well.

You will need a spare life to squander. A clone would be most handy. In short, you will want to at least:

You’re probably not done by a long shot! But the above should at least get you fitfully started, though long will you curse our names. Praise Cleese.

License

beartype is open-source software released under the permissive MIT license.

Funding

beartype is currently financed as a purely volunteer open-source project – which is to say, it’s unfinanced. Prior funding sources (yes, they once existed) include:

  1. A Paul Allen Discovery Center award from the Paul G. Allen Frontiers Group under the administrative purview of the Paul Allen Discovery Center at Tufts University over the period 2015—2018 preceding the untimely death of Microsoft co-founder Paul Allen, during which beartype was maintained as the private @type_check decorator in the Bioelectric Tissue Simulation Engine (BETSE). Phew!

Authors

beartype is developed with the grateful assistance of a volunteer community of enthusiasts, including (in chronological order of issue or pull request):

  1. Cecil Curry (@leycec). Hi! It’s me. From beartype’s early gestation as a nondescript @type_check decorator in the Bioelectric Tissue Simulation Engine (BETSE) to its general-audience release as a public package supported across multiple Python and platform-specific package managers, I shepherd the fastest, hardest, and deepest runtime type-checking solution in any dynamically-typed language towards a well-typed future of PEP-compliance and boundless quality assurance. Cue epic taiko drumming.
  2. Felix Hildén (@felix-hilden), the Finnish computer vision expert world-renowned for his effulgent fun-loving disposition and:
  3. @harens, the boisterous London developer acclaimed for his defense of British animals that quack pridefully as they peck you in city parks as well as maintaining our first third-party package, a macOS-specific MacPorts Portfile estimated to solve all your problems.
  4. @Heliotrop3, the perennial flowering plant genus from Peru whose “primal drive for ruthless efficiency makes overcoming these opportunities for growth [and incoming world conquest] inevitable” as well as:

See Also

External beartype resources include:

Runtime type checkers (i.e., third-party mostly pure-Python packages dynamically validating Python callable types at Python runtime, typically via decorators, explicit function calls, and import hooks) include:

Static type checkers (i.e., third-party tooling not implemented in Python statically validating Python callable and/or variable types across a full application stack at tool rather than Python runtime) include:

  • mypy.
  • Pyre, published by FaceBook. …yah.
  • pyright, published by Microsoft.
  • pytype, published by Google.

GitHub

https://github.com/beartype/beartype