Implementation pattern (inspired by Rust)
Useful when it is needed to extend a class (usually 3d party) with some methods or interfaces
Please, be careful. This library is literally a class patcher.
Install
pip install impler
or
poetry add impler
Usage
Methods implementation
Using implementation pattern you can extend any class (even 3rd party) with regular, class or static methods.
from impler import impl
from pydantic import BaseModel
@impl(BaseModel)
def fields_count(self: BaseModel):
return len(self.__fields__)
class Point(BaseModel):
x: int = 0
y: int = 1
point = Point()
print(point.fields_count())
Class methods
@impl_classmethod(BaseModel)
def fields_count(cls):
return len(cls.__fields__)
# or
@impl(BaseModel)
@classmethod
def fields_count(cls):
return len(cls.__fields__)
Static methods
@impl_staticmethod(BaseModel)
def zero(cls):
return 0
# or
@impl(BaseModel)
@staticmethod
def zero(cls):
return 0
Async methods
@impl(BaseModel)
async def zero(cls):
await asyncio.sleep(1)
return 0
Interfaces implementation
The same way you can extend any class with the whole interface
Here is example of the base interface
from pathlib import Path
class BaseFileInterface:
def dump(self, path: Path):
...
@classmethod
def parse(cls, path: Path):
...
This is how you can implement this interface for Pydantic BaseModel
class:
from impler import impl
from pydantic import BaseModel
from pathlib import Path
@impl(BaseModel, as_parent=True)
class ModelFileInterface(BaseFileInterface):
def dump(self, path: Path):
path.write_text(self.json())
@classmethod
def parse(cls, path: Path):
return cls.parse_file(path)
If as_parent
parameter is True
the implementation will be injected to the list of the target class parents.
Then you can check if the class or object implements the interface:
print(issubclass(BaseModel, BaseFileInterfase))
# True
print(issubclass(Point, BaseFileInterfase))
# True
print(isinstance(point, BaseFileInterface))
# True
The whole api documentation could be found by the link