What’s New in Python 3.10? Explore Type Annotations, Union Operator, and More

New Features in Python 3.10 Alpha 2

Python 3.10’s second alpha, released early November, adds several language enhancements, especially around type annotations.

Type Annotation Extensions

Python 3.9 already overhauled type hints; 3.10 continues this trend by extending annotation capabilities. Annotations are now evaluated lazily, allowing forward references to be expressed as strings and reducing import overhead. Annotations are stored in __annotations__ and evaluated centrally.

Union Operator

Python 3.10 introduces the | operator as a concise way to express unions in type hints (e.g., int | float), replacing the older typing.Union[int, float] syntax.

def f(x: int | float) -> float:
    return x * 3.142

f(1)      # pass
f(1.5)    # pass
f('str')  # linter will show annotation error

Explicit TypeAlias

To avoid confusion when using string literals for forward references, the new TypeAlias from typing_extensions marks a name as a type alias.

from typing_extensions import TypeAlias
MyType: TypeAlias = "ClassName"

def foo() -> MyType:
    ...
# Or, if ClassName is already defined:
MyType: TypeAlias = ClassName

Why Type Annotations Matter

Type hints improve code readability and maintainability, allowing developers to understand expected input and output types without reading implementation details. They are optional and do not affect runtime behavior.

Other Core Updates

zip strict mode (PEP 618) : zip(..., strict=True) raises ValueError if iterables have different lengths.

int.bit_count() : Returns the population count (number of set bits) of an integer.

Dictionary view mapping : Views returned by dict.keys(), dict.values(), and dict.items() now expose a mapping attribute that returns the original dictionary.

# zip strict example
x = [1,2,3,4,5,6]
y = [1,2,3,4]
z = zip(x, y, strict=True)  # raises ValueError

# int.bit_count example
for x in [0,1,2,3,10,11,12,100,101,102]:
    print(f"{x} = {x.bit_count()}")

# dict view mapping example
x = {'hello': 0, 'world': 1}
keys = x.keys()
print(keys)          # dict_keys(['hello', 'world'])
print(keys.mapping)  # {'hello': 0, 'world': 1}