I've been using Pydantic to create models that I use for input validation / data cleaning. These work properly when a framework calls the constructor (like Flask) but if I call the constructor by hand I get type errors because I'm not passing the "final" types into it. For example here I have a model that has Decimal attributes:

class MarketOrder(BaseModel):
    """Order creation response."""
    symbol: str
    price: Decimal = Field(alias="price")


    @field_validator('price', mode='before')
    @classmethod
    def validate_order_decimals(cls, v):
        """Convert order values to Decimal."""
        try:
            return Decimal(str(v))
        except (ValueError, TypeError):
            raise ValueError(f"Invalid price value: {v}")

but when I'm trying to write a test for this the linter complains because I'm passing strs, and not Decimals (even though the actual code runs fine):

def test_create_order_command_valid():
    # Test valid BUY order
    cmd = MarketOrder(
        symbol="BTCUSDT",
        price="40000"
        #     ^^^--- linter error
    )

The linter error I get is

Argument of type "Literal['40000']" cannot be assigned to parameter "price" of type "Decimal" in function "__init__"
  "Literal['40000']" is not assignable to "Decimal"

Is there a way to modify the constructor that Pydantic generates to include this conversion function (validate_order_decimals)? Alternatively, is there some other libarary that has similar functionality and supports this? I'm coming from Typescript and I've used zod, schema, and other similar libraries that support this, so I'm pretty sure this exists in Python too.