Python associates types with objects, not variables. A name can refer to objects of different types, but an unexpected type may cause a failure only when the code runs. Type hints optionally state the intended types of variables, function parameters, and return values. Python does not enforce them at runtime; type checkers and IDEs can use them to flag mismatches earlier. section 1.3.2 section 4.4
def add_ints(x: int, y: int) -> int: #1
return x + y
p: int = 2.3 #2
z = add_ints(1, 2)
w = add_ints(1.5, 2.5) #3
A colon annotates a variable or parameter; an arrow annotates a return value. Here, the hints say that both arguments and the result of `add_ints` should be integers. Python runs the code, but mypy flags three mismatches: assigning a float to `p` and passing two floats to `add_ints`. section 4.4
Hints are most useful where an expected type is otherwise unclear or a mismatch would be hard to trace through a large codebase. They can be added progressively, starting where type information helps most. In a small informal script, the extra notation may not be worth the effort; annotating everything can also make definitions harder to read. Type hints offer a way to check expectations without requiring every piece of Python code to be typed. section 4.4.1 section 4.4.2 section 4.4.3
