FastAPI-Pydantic校验异常：字段齐全时返回无效输入错误排查

问题场景 #

使用@model_validator(mode='before')结合model_validate校验请求是否匹配子类时，缺少必填字段能正常返回"字段缺失"提示；但字段齐全时，尽管模型校验通过、后续逻辑可执行，却抛出Input should be a valid dictionary or object to extract fields from: ('body',)错误。

示例代码与请求 #

请求载荷 #

{
  "name": "BC",
  "connectorType": "SNOWFLAKE",
  "productName": "bc",
  "secretArn": "foobararn",
  "warehouse": "foobarwarehouse",
  "tableSchema": "foobarschema",
  "database": "bardatabse"
}

模型代码 #

class SnowflakeConnector(SpotApiBaseModel):
    """Snowflake Connector."""
    model_config = ConfigDict(from_attributes=True)

    secretArn: str = Field(description="AWS secret arn")
    warehouse: str = Field(description="Snowflake warehouse. Required for connector type: SNOWFLAKE")
    database: str = Field(description="Database name. Required for connector type: SNOWFLAKE")
    tableSchema: str = Field(description="Database table schema. Required for connector type: SNOWFLAKE")


class ConnectorInterface(BaseConnector):

    @model_validator(mode='before')
    def check_allowed_products(cls, values):
        connector = values.get('connectorType').lower()
        if connector == "s3":
            S3Connector.model_validate(values)
        elif connector == "snowflake":
            SnowflakeConnector.model_validate(values)

异常信息 #

{"time": "2024-06-18T10:46:45.641Z", "l": "ERROR", "rid": "2558770a-66ff-4da0-b8a2-977afc5043e6", "org": "", "act": "", "message": "An error occurred: [{'type': 'model_attributes_type', 'loc': ('body',), 'msg': 'Input should be a valid dictionary or object to extract fields from', 'input': None, 'url': 'https://errors.pydantic.dev/2.6/v/model_attributes_type'}]", "component": {"name": "/app/common/exceptions/base_exception_handler.py", "version": "0.1.3"}, "exception": "Traceback (most recent call last):\n  File \"/Library/Caches/pypoetry/virtualenvs/service-Qx3_L5c2-py3.12/lib/python3.12/site-packages/starlette/middleware/exceptions.py\", line 68, in __call__\n    await self.app(scope, receive, sender)\n  File \"/Library/Caches/pypoetry/virtualenvs/service-Qx3_L5c2-py3.12/lib/python3.12/site-packages/fastapi/middleware/asyncexitstack.py\", line 20, in __call__\n    raise e\n  File \"/Library/Caches/pypoetry/virtualenvs/service-Qx3_L5c2-py3.12/lib/python3.12/site-packages/fastapi/middleware/asyncexitstack.py\", line 17, in __call__\n    await self.app(scope, receive, send)\n  File \"/Library/Caches/pypoetry/virtualenvs/service-Qx3_L5c2-py3.12/lib/python3.12/site-packages/starlette/routing.py\", line 718, in __call__\n    await route.handle(scope, receive, send)\n  File \"/Library/Caches/pypoetry/virtualenvs/service-Qx3_L5c2-py3.12/lib/python3.12/site-packages/starlette/routing.py\", line 276, in handle\n    await self.app(scope, receive, send)\n  File \"Library/Caches/pypoetry/virtualenvs/service-Qx3_L5c2-py3.12/lib/python3.12/site-packages/starlette/routing.py\", line 66, in app\n    response = await func(request)\n               ^^^^^^^^^^^^^^^^^^^\n  File \"/Library/Caches/pypoetry/virtualenvs/service-Qx3_L5c2-py3.12/lib/python3.12/site-packages/fastapi/routing.py\", line 273, in app\n    raise RequestValidationError(_normalize_errors(errors), body=body)\nfastapi.exceptions.RequestValidationError: [{'type': 'model_attributes_type', 'loc': ('body',), 'msg': 'Input should be a valid dictionary or object to extract fields from', 'input': None, 'url': 'https://errors.pydantic.dev/2.6/v/model_attributes_type'}]"}

原因分析 #

问题根源在于@model_validator(mode='before')的返回规则：Pydantic要求该模式下的校验器必须返回处理后的输入值，若未显式返回，默认会返回None。当前校验器仅调用子类的model_validate完成校验，未返回原始values，导致Pydantic后续处理时拿到的输入为None，触发model_attributes_type错误。

解决方案 #

修改校验器，在完成子类校验后显式返回原始values：

class ConnectorInterface(BaseConnector):

    @model_validator(mode='before')
    def check_allowed_products(cls, values):
        connector = values.get('connectorType').lower()
        if connector == "s3":
            S3Connector.model_validate(values)
        elif connector == "snowflake":
            SnowflakeConnector.model_validate(values)
        # 必须返回输入值，否则Pydantic会将None作为后续处理的输入
        return values

补充说明 #

• model_validate仅做规则校验，不会修改原始values，直接返回即可。
• 若需对输入值做修改（如补充默认字段），可在返回前完成处理，再返回修改后的字典。

内容的提问来源于stack exchange，提问作者electromecca