使用动态键创建 Pydantic 模型模式

Question

我正在尝试为以下 JSON 实施 Pydantic 模式模型。

{
    "description": "Best Authors And Their Books",
    "authorInfo":
    {
        "KISHAN":
            {
            "numberOfBooks": 10,
            "bestBookIds": [0, 2, 3, 7]
            },
        "BALARAM":
            {
            "numberOfBooks": 15,
            "bestBookIds": [10, 12, 14]
            },
        "RAM":
            {
            "numberOfBooks": 6,
            "bestBookIds": [3,5]

        }
    }
}

这是 Pydantic 中的模式对象

from typing import List, Type, Dict
from pydantic import BaseModel

class AuthorBookDetails(BaseModel):
    numberOfBooks: int
    bestBookIds: List[int]

class AuthorInfoCreate(BaseModel):
    __root__: Dict[str, Type[AuthorBookDetails]]    
#pass

class ScreenCreate(BaseModel):
    description: str
    authorInfo: Type[AuthorInfoCreate]

我正在按如下方式解析 AuthorInfoCreate：

y = AuthorBookDetails( numberOfBooks = 10, bestBookIds = [3,5])
print(y)
print(type(y))

x = AuthorInfoCreate.parse_obj({"RAM" : y})
print(x)

我看到以下错误。

numberOfBooks=10 bestBookIds=[3, 5]

<class '__main__.AuthorBookDetails'>

Traceback (most recent call last):

  File "test.py", line 44, in <module>

    x = AuthorInfoCreate.parse_obj({"RAM": y})

  File "C:\sources\rep-funds\env\lib\site-packages\pydantic\main.py", line 402, in parse_obj

    return cls(**obj)

  File "C:\sources\rep-funds\env\lib\site-packages\pydantic\main.py", line 283, in __init__

    raise validation_error

pydantic.error_wrappers.ValidationError: 1 validation error for AuthorInfoCreate

__root__ -> RAM

  subclass of AuthorBookDetails expected (type=type_error.subclass; expected_class=AuthorBookDetails)

我想了解如何更改 AuthorInfoCreate 以便提及 json 架构。

Answer 1

实际上，您应该从类型注释中删除 Type。您需要 class 的实例，而不是实际的 class。尝试以下解决方案：

from typing import List,Dict
from pydantic import BaseModel

class AuthorBookDetails(BaseModel):
    numberOfBooks: int
    bestBookIds: List[int]

class AuthorInfoCreate(BaseModel):
    __root__: Dict[str, AuthorBookDetails]

class ScreenCreate(BaseModel):
    description: str
    authorInfo: AuthorInfoCreate

Answer 2

对于那些知道如何初始化 Pydantic 类的人，@SKhalymon 给出了答案，

from typing import List, Dict
from pydantic import BaseModel

class AuthorBookDetails(BaseModel):
    numberOfBooks: int
    bestBookIds: List[int]

class AuthorInfoCreate(BaseModel):
    __root__: Dict[str, AuthorBookDetails]

class ScreenCreate(BaseModel):
    description: str
    authorInfo: AuthorInfoCreate



kishan = AuthorBookDetails( numberOfBooks = 10, bestBookIds = [0, 2, 3, 7])

balram = AuthorBookDetails( numberOfBooks = 15, bestBookIds = [10, 12, 14])

ram = AuthorBookDetails( numberOfBooks = 6, bestBookIds = [3, 5])

aic = AuthorInfoCreate(__root__={"KISHAN": kishan, "BALRAM": balram, "RAM": ram})

sc = ScreenCreate( description = "Best Authors And Their Books", authorInfo = aic)

print(sc.json())

输出：

{"description": "Best Authors And Their Books", "authorInfo": {"__root__": {"KISHAN": {"numberOfBooks": 10, "bestBookIds": [0, 2, 3, 7]}, "BALRAM": {"numberOfBooks": 15, "bestBookIds": [10, 12, 14]}, "RAM": {"numberOfBooks": 6, "bestBookIds": [3, 5]}}}}

Answer 3

来自pydantic issue #2100

__root__ is only supported at parent level.

这就是为什么不能使用

class AuthorInfoCreate(BaseModel):
    __root__: Dict[str, AuthorBookDetails]

针对上述问题提出了以下解决方法

from typing import Any, Dict, List

from pydantic import BaseModel as PydanticBaseModel
from pydantic.utils import ROOT_KEY


class BaseModel(PydanticBaseModel):
    def __init__(__pydantic_self__, **data: Any) -> None:
        if __pydantic_self__.__custom_root_type__ and data.keys() != {ROOT_KEY}:
            data = {ROOT_KEY: data}
        super().__init__(**data)

...
...

现在可以正常使用 AuthorInfoCreate 模型

aic = AuthorInfoCreate(**{"KISHAN": kishan, "BALRAM": balram, "RAM": ram})

PR #2237 应该不需要上述解决方法。

使用动态键创建 Pydantic 模型模式

Creating Pydantic Model Schema with Dynamic Key

python

pydantic

fastapi