Adding decorator to a python class

Question

I have a few methods that require a certain condition to be applied to function properly. To avoid code repetition I've created a decorator to check if this condition is applied and only then execute the method. Here is the code:

class ClientHandler:
    def __init__(self):
        self.__is_dead = False
        self.c_ident = c_ident

    def __s_operation(func):
        @functools.wraps(func)
        def check(*args):
            self = args[0]
            if not self.__is_dead:
                func(*args)
            else:
                logger.error(
                    "couldn't proceed - client {} is dead!".format(
                        self.c_ident
                    )
                )
        return check

    @__s_operation
    def __receive_m(self, size) -> bytes:
        ...

    @__s_operation
    def close_c(self, message: str) -> None:
        ...

I was following this answer to be able to use class attributes and call class methods from my decorator, but I ran into 2 errors:

'ClientHandler' object is not callable on the func(*args) line, when I'm trying to call a class method from a decorator.
function '__s_operation' lacks a positional argument, when I'm trying to add my decorator @__s_operation over a class method.

I don't understand what is the problem and how to solve it, because in the other answer on stack overflow everything seems working. Also, I'm not sure if I will be able to reach my private attributes and private methods from this decorator. I would be grateful for your help. Thank you.

Gwang-Jin Kim · Accepted Answer · 2025-06-24 19:38:21Z

@Yzkodot's solution is good.

But I miss a @staticmethod in his example - to make explicitly clear that no self in the first argument position is needed.

import functools
import logging

logging.basicConfig(level=logging.INFO)
logger = logging.getLogger(__name__)

class ClientHandler:
    # Make this a staticmethod so it doesn't expect self
    @staticmethod
    def _s_operation(func):
        @functools.wraps(func)
        def wrapper(self, *args, **kwargs):
            if not self._is_dead:
                return func(self, *args, **kwargs)
            else:
                logger.error(f"Couldn't proceed - client {self.c_ident} is dead!")
                return None
        return wrapper

    def __init__(self, c_ident: str):
        self._is_dead = False
        self.c_ident = c_ident

    @_s_operation
    def receive_message(self, size: int) -> bytes:
        logger.info(f"{self.c_ident} received {size} bytes.")
        return b"x" * size

    @_s_operation
    def close_connection(self, message: str) -> None:
        logger.info(f"{self.c_ident} closed with message: {message}")
        self._is_dead = True

Use it as:

client = ClientHandler("alpha")
client.receive_message(10)      ## => Will work
client.close_connection("done") ## => Will close
client.receive_message(5)       ## => Blocked by the decorator

Output:

INFO:__main__:alpha received 10 bytes.
INFO:__main__:alpha closed with message: done
ERROR:__main__:Couldn't proceed - client alpha is dead!

Jaime · Accepted Answer · 2025-06-24 08:30:19Z

2

I think your problem lies on calling @functools.wraps(func) over the check() function. This one is already supposed to be the wrapper.

I slightly modified your code as following (only for testing) and it seems to work:

class ClientHandler:
    def __init__(self):
        self.__is_dead = False

    def __s_operation(func):
        def check(*args):
            print("The decorator code has worked!")

            self = args[0]
            if not self.__is_dead:
                func(*args)
            else:
                logging.error(
                    "couldn't proceed - client {} is dead!".format(
                        self.c_ident
                    )
                )
        return check

    @__s_operation
    def receive_m(self) -> bytes:
        print("message received")



if __name__ == "__main__":
    ch = ClientHandler()
    ch.receive_m()

Over the terminal I get correctly both print statements.

answered Jun 24 at 8:30

Jaime

1436 bronze badges

1 Comment

good user Jun 24 at 20:07

I thought I could use functools.wraps to not lose information about original function. As explained here: stackoverflow.com/a/309000/19329071

Collectives™ on Stack Overflow

Adding decorator to a python class

2 Answers 2

1 Comment

1 Comment

Your Answer

Linked

Hot Network Questions

Collectives™ on Stack Overflow

2 Answers 2

1 Comment

1 Comment

Your Answer

Sign up or log in

Post as a guest

Linked

Related