在 python 3.8 中，如何测试数据类中注释为 Literal 的字段在运行时有效

Question

Given the following example:

from typing import Literal
from dataclasses import dataclass
@dataclass
Class Example:
    answer: Literal['Y', 'N']

x = Example('N')
field = fields(x)[0]

How can I check that the variable field is of type Literal? issubclass(field.type, Literal) doesn't seem to work.

Secondly how can I then get the list ['Y', 'N'] from field.type , so that I could check the value at run time and raise an error when fail = Example('invalid')

pydantic does this but you would have to use their drop in dataclass...

Answer 1

Literal is not a normal type that a python object would have at runtime, it doesn't make sense to check that an object is a Literal

You can access the annotations for a class using __annotations__ , following on from your example:

>>> Example.__annotations__['answer'].__args__
('Y', 'N')

Answer 2

from dataclasses import dataclass
from typing import Literal

from validated_dc import ValidatedDC


@dataclass
class Example(ValidatedDC):
    answer: Literal['Y', 'N']


instance = Example('N')
assert instance.is_valid()


instance = Example('X')
assert not instance.is_valid()

Answer 3

I created a tiny Python library for this purpose: https://github.com/tamuhey/dataclass_utils

This library can be applied for such dataclass that holds another dataclass (nested dataclass), and nested container type (like Tuple[List[Dict... ). Of course, it can test Literal at runtime.