Inheritance and Reuse

If a query method or DataLoader can return data of type Car, then a subclass CarWithRepairHistories that inherits from Car can directly replace Car in any scenario.

After resolving, the data under CarWithRepairHistories will be loaded.

The prerequisite is that the fields extended by the subclass all have default values.

For example:

class Car(BaseModel):
    id: int
    name: str

def query_car():
    return dict(id=1, name='Ford')

class ReturnData(BaseModel):
    car: Optional[Car] = None
    def resolve_car(self):
        return query_car()

You can directly replace Car with CarWithRepairHistories without modifying query_car and other query methods.

class Car(BaseModel):
    id: int
    name: str

def query_car():
    return dict(id=1, name='Ford')

class CarWithRepairHistories(A):
    histories: List[RepairHistory] = []  # Default value
    def resolve_histories(self, loader=LoaderDepend(RepairHistoriesLoader)):
        return loader.load(self.id)

class ReturnData(BaseModel):
    # car: Optional[Car] = None
    car: Optional[CarWithRepairHistories] = None
    def resolve_car(self):
        return query_a()  # No adjustment needed

Therefore, the methods for obtaining data or DataLoader can be reused to provide data for various subclasses.

Subclass inheritance can reuse the fields of the parent class, reducing the overhead of repeatedly defining fields.

By making good use of inheritance, we can lay the groundwork for the efficient development model of "ERD-oriented development".