How to parse ObjectId in a pydantic model?

Question

I am trying to parse MongoDB records to a pydantic model but failing to do so for ObjectId

From what I understood, I need to setup validator for ObjectId and did try to both extend ObjectId class and add the validator decorator to my class using ObjectId. which I did as follows.

from pydantic import BaseModel, validator
from bson.objectid import ObjectId


class ObjectId(ObjectId):
    pass
    @classmethod
    def __get_validators__(cls):
        yield cls.validate
    @classmethod
    def validate(cls, v):
        if not isinstance(v, ObjectId):
            raise TypeError('ObjectId required')
        return str(v)


class User(BaseModel):
    who: ObjectId


class User1(BaseModel):
    who: ObjectId
    @validator('who')
    def validate(cls, v):
        if not isinstance(v, ObjectId):
            raise TypeError('ObjectId required')
        return str(v)

data = {"who":ObjectId('123456781234567812345678')}

Unfortunately, both "solution" are failing as follows:

>>> test = User(**data)
Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
  File "pydantic/main.py", line 274, in pydantic.main.BaseModel.__init__
pydantic.error_wrappers.ValidationError: 1 validation error for User
id
  field required (type=value_error.missing)

>>> test = User1(**data)
Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
  File "pydantic/main.py", line 274, in pydantic.main.BaseModel.__init__
pydantic.error_wrappers.ValidationError: 1 validation error for User1
who
  ObjectId required (type=type_error)

There is definitely something that I am missing here.

Answer 1

You first test case works fine. The problem is with how you overwrite ObjectId.

from pydantic import BaseModel
from bson.objectid import ObjectId as BsonObjectId


class PydanticObjectId(BsonObjectId):
    @classmethod
    def __get_validators__(cls):
        yield cls.validate

    @classmethod
    def validate(cls, v):
        if not isinstance(v, BsonObjectId):
            raise TypeError('ObjectId required')
        return str(v)


class User(BaseModel):
    who: PydanticObjectId


print(User(who=BsonObjectId('123456781234567812345678')))

prints

who='123456781234567812345678'

Only pydantic should use pydantic type. Mongo will provide you with bsons ObjectId. So instantiate your data with real ObjectId. So data = {"who":ObjectId('123456781234567812345678')} is wrong, as it uses your child ObjectId class.

Answer 2

Just another way to do this is with pydantic that i found useful from another source is:

Define a file called PyObjectId.py in a models folder.

from pydantic import BaseModel, Field as PydanticField
from bson import ObjectId

class PyObjectId(ObjectId):
    @classmethod
    def __get_validators__(cls):
        yield cls.validate
    @classmethod
    def validate(cls, v):
        if not ObjectId.is_valid(v):
            raise ValueError("Invalid objectid")
        return ObjectId(v)
    @classmethod
    def __modify_schema__(cls, field_schema):
        field_schema.update(type="string")

Then you can use this in any of your object files like this users.py

from models.PyObjectId import PyObjectId
from pydantic import BaseModel, Field as PydanticField
from bson import ObjectId
class Users(BaseModel):
    id: PyObjectId = PydanticField(default_factory=PyObjectId, alias="_id")
    class Config:
        allow_population_by_field_name = True
        arbitrary_types_allowed = True #required for the _id 
        json_encoders = {ObjectId: str}

Answer 3

Getting Started with MongoDB and FastAPI

Mongo Developers

This code help you to use json encoder

from bson import ObjectId
from pydantic import BaseModel


class ObjId(ObjectId):
    @classmethod
    def __get_validators__(cls):
        yield cls.validate

@classmethod
def validate(cls, v: str):
    try:
        return cls(v)
    except InvalidId:
        raise ValueError("Not a valid ObjectId")


class Foo(BaseModel):
    object_id_field: ObjId = None

    class Config:
        json_encoders = {
            ObjId: lambda v: str(v),
        }



obj = Foo(object_id_field="60cd778664dc9f75f4aadec8")
print(obj.dict())
# {'object_id_field': ObjectId('60cd778664dc9f75f4aadec8')}
print(obj.json())
# {'object_id_field': '60cd778664dc9f75f4aadec8'}