Pydantic

pydantic is a popular library for data validation. It could be used to define business entities of your domain if you want validation to be applied to data your entities encapsulate. It's possible to use a pydantic dataclasses to make defined entities compatible with generics library decorators. As with all previous examples we advice you to make your instances immutable. pydantic library has a first class support for immutability. Enable frozen setting on class and use replace function to copy instancies of it.

>>> from pydantic.dataclasses import dataclass
>>> from dataclasses import replace
>>> from generics import private

>>> @private
... @dataclass(frozen=True)
... class User:
...     name: str
...     def greet(self):
...         return f"Hello, {self.name}"
...     def rename(self, name):
...         return replace(self, name=name)
...

>>> User
Private::User

>>> user = User("Jeff")
>>> user
Private::User(name='Jeff')

>>> user.greet()
'Hello, Jeff'

>>> user.rename("Kate")
Private::User(name='Kate')

Arguments validation would be applied at the object construction time. This behavior would match regular dataclasses created by pydantic.

>>> from typing import Callable

>>> @private
... @dataclass(frozen=True)
... class User:
...     name: str
...     console: Callable
...
...     def greet(self):
...         self.console(f'Hello, {self.name}')

>>> User(name='Jeff', console=print)
Private::User(name='Jeff', console=<built-in function print>)

>>> User(name='Jeff', console=True)
Traceback (most recent call last):
  ...
pydantic.error_wrappers.ValidationError: 1 validation error for User
console
  True is not callable (type=type_error.callable; value=True)

— ⭐ —