Adding fields/methods to existing classes.
A Python approach to method/field extensions like in C# and Kotlin, probably the closest.
pip install extclass
Suppose you have a Person class in an already installed package:
from dataclasses import dataclass
@dataclass
class Person:
first_name: str
last_name: strThen, you could add an extra method outside the source package somewhere in your project.
from extclass import extension
@extension
class PersonExt(Person):
age: int = 10
def full_name(self):
return f'{self.first_name} {self.last_name}'With the help of the @extension decorator, new field age and full_name are added to the origin Person class. You can just call them as usual.
if __name__ == '__main__':
person = Person(first_name='rick', last_name='sanchez')
print(person.full_name())
print(person.age)which print as
rick sanchez
10
However, the IDE cannot know the dynamic informations of the added attributes, so there would be a warning message:
Unresolved attribute reference 'full_name' for class 'Person'
And no autocomplete for the attributes in the IDE.
To handle the warnings also enabling autocomplete, you can simply typing.cast the origin object to the new extended type.
from typing import cast
person = cast(PersonExt, person)In this case, person is still the old object but only marked as the extended type in order to cheat the IDE for better programming experiences.
Same method/field name in the extending class will not affect the origin class. So this package is very safe in any use-cases.
Classes of pydantic.BaseModel can only attach new methods. Dynamic fields are internally forbidden by the pydantic package. So please do NOT extend fields for pydantic models, or an AttributeError would be raised during runtime.
See test.py