[Python] Creating a dummy Pydantic model generator

I want to create a method on a base Pydantic model to instantiate child models with dummy data.

from __future__ import annotations
from pydantic import BaseModel


class BaseModelWrapper(BaseModel):
@classmethod
def make_dummy(cls) -> BaseModelWrapper:
for name, field in cls.model_fields.items():
if not field.is_required():
continue

# How can I create values based on the field type?
print(field.annotation)
return cls()


class XXX(BaseModelWrapper):
a: int | None
b: str
c: int
d: int | None = None
e: list[str]


# These should be equivalent
XXX.make_dummy()
XXX(a=None, b="", c=0, e=[])


The part I'm struggling with is how to programmatically map type annotations to values.

Let's say field.annotation is int | None. I could just create a dictionary to map that to None, but there are tons of possible combinations of types, so this doesn't scale. There must be a cleaner way to create a value for each field.

Continue reading...