You can use models as types in Pydantic.
nested_types_1.py
from pydantic import BaseModel, ValidationError
class Address(BaseModel):
street: str
city: str
country: str
class Employee(BaseModel):
id: int
name: str
age: int
address: Address
emp1 = Employee(id = 1, name = 'Ptr', age = 23, address = Address(street = 'Chowdeswari', city = "Bangalore", country = 'India'))
emp1Json = emp1.json(exclude = {'address' : {'street', 'country'}})
emp2Json = emp1.json(exclude = {'address' : {'street', 'country'}, 'age': ...})
print('emp1Json -> ' + emp1Json)
print('emp2Json -> ' + emp2Json)
Output
emp1Json -> {"id": 1, "name": "Ptr", "age": 23, "address": {"city": "Bangalore"}} emp2Json -> {"id": 1, "name": "Ptr", "address": {"city": "Bangalore"}}
Previous Next Home
No comments:
Post a Comment