[Python] How can I tell Python that the result type of a pipeline is a certain type?

I have a pipeline class based on the reduce function. I was factoring the code today to be able to trace type information and noticed that the reduce function removes the type information so I cannot trace the type. See below code:

from functools import reduce
from typing import Callable, TypeVar, TypeAlias

def int_to_str(i: int) -> str:
return str(i)

def str_to_float(s: str) -> float:
return float(s)

pipes = [int_to_str, str_to_float]

_S = TypeVar("S")
_T = TypeVar("T")
Pipe : TypeAlias = Callable[[_S], _T]

def reducer(prev: _S, pipe: Pipe) -> _T:
return pipe(prev)

initial_arg = 100
# Pylance shows that the type of what_type is int,
# because reduce return type annotation(_T@reduce) is same as the type of initial_arg.
what_type = reduce(reducer, pipes, initial_arg)
# what_type=100.0 (float value)
print(f"{what_type=}")
# type(what_type)=<class 'float'> (float type)
print(f"{type(what_type)=}")


The elements in the list as pipes are literal. So I thinked it might be staticly traceable. But the code result and the Pylance report are different. What can I do to avoid this problem?

Evironment:

• Python 3.10

Continue reading...