Return all extra passed to pydantic model

Question

I'm trying to get a list of all extra fields not defined in the schema. I saw solution posted here but it ignores any nested models. Optimal solution would create a variable in the Pydantic model with extras that I could access after new object with passed data is created but not sure if this is even possible.

Here is the code I'm working with.

Edit: I want .extras to be something like a property that returns not just the extra data directly on that instance, but also the extra data on any nested model instances it holds

from typing import Any, Dict, Optional
from pydantic import BaseModel
from pydantic import BaseModel, Field, root_validator

unnecessary_data = {
    "name": "Lévy",
    "age": 3,
    "key_parent": "value",  # unnecessary
    "key2_parent": "value2",  # unnecessary x2
    "address": {
        "city": "Wonderland",
        "zip_code": "ABCDE",
        "number": 123,
        "key_child": 1232 # unnecessary x
    }
}


class NewBase(BaseModel):
    versio: Optional[str] = Field(alias='version')  # just to show that it supports alias too
    extra: Dict[str, Any]

    @root_validator(pre=True)
    def build_extra(cls, values: Dict[str, Any]) -> Dict[str, Any]:
        all_required_field_names = {field.alias for field in cls.__fields__.values() if field.alias != 'extra'}  # to support alias

        extra: Dict[str, Any] = {}
    for field_name in list(values):
        if field_name not in all_required_field_names:
            extra[field_name] = values.pop(field_name)
    values['extra'] = extra
    return values

class Address(NewBase):

    """
    Cat API Address definition
    """
    city: str
    zip_code: str
    number: int


class CatRequest(NewBase):
    """
    Cat API Request definition
    """
    name: str
    age: int
    address: Address


validated = CatRequest(**unnecessary_data)
print(validated.extras)
>> ["key_parent", "key2_parent", "address.key_child"]

Answer 1

Here is a version updated to Pydantic v2:

from typing import Any, Dict, Optional
from pydantic import BaseModel, Field, model_validator

unnecessary_data = {
    "name": "Lévy",
    "age": 3,
    "key_parent": "value",  # unnecessary
    "key2_parent": "value2",  # unnecessary x2
    "address": {
        "city": "Wonderland",
        "zip_code": "ABCDE",
        "number": 123,
        "key_child": 1232 # unnecessary x
    }
}


class NewBase(BaseModel):
    versio: Optional[str] = Field(alias="version", default=None)
    extra: Dict[str, Any] = Field(default={}, exclude=True)

    @model_validator(mode="before")
    @classmethod
    def validator(cls, values):
        extra, valid = {}, {}

        for key, value in values.items():
            if key in cls.model_fields:
                valid[key] = value
            else:
                extra[key] = value

        valid["extra"] = extra
        return valid

    @property
    def extra_flat(self):
        extra_flat = {**self.extra}

        for name, value in self:
            if isinstance(value, NewBase) and value.extra:
                data = {f"{name}.{k}": v for k, v in value.extra_flat.items()}
                extra_flat.update(data)
            
        return extra_flat



class Address(NewBase):

    """
    Cat API Address definition
    """
    city: str
    zip_code: str
    number: int


class CatRequest(NewBase):
    """
    Cat API Request definition
    """
    name: str
    age: int
    address: Address


validated = CatRequest(**unnecessary_data)
print(validated.extra_flat)

Which prints:

{'key_parent': 'value', 'key2_parent': 'value2', 'address.key_child': 1232}

I hope this helps!