Factory for Generic types

Question

I guess most Python lovers know it but, in order to provide some context, the typing module provides a mechanism for defining type hints as follows:

from typing import List

x = List[int]

In the code above, x represents a List whose items are integers.

In the scope of microservices, FastAPI library allows defining the endpoint argument types through the use of Pydantic library.

from pydantic import BaseModel

class User(BaseModel):
    id: int
    name: str
    friends: List[int] = []

I am trying to generate an endpoint for a ML model without knowing its input format. My approach is to extract is metadata and see the input schema, and generate a Pydantic class from it.

My question is...

Is there a way of programmatically obtaining get a type like those referenced by x, so I can implement a factory for generating them based on some input?

Answer 1

You can create your own generic class:

from pydantic import GenericModel
from typing import Generic, TypeVar


F = TypeVar("F")


class User(GenericModel, Generic[F]):
    id: int
    name: str
    friends: List[F]