Borg pattern shared state in Python

Implement the Borg pattern to share state across class instances by assigning a class-level dictionary to each instance's __dict__.

Medium Python 3.9+ Aug 9, 2026 OOP & classes 14 views 0 copies

borg monostate shared-state design-pattern oop

Python code

30 lines

Python 3.9+

class Borg:
    _shared_state = {}

    def __init__(self):
        self.__dict__ = Borg._shared_state


class ConfigManager(Borg):
    def __init__(self):
        super().__init__()
        if not hasattr(self, "settings"):
            self.settings = {}

    def set(self, key, value):
        self.settings[key] = value

    def get(self, key):
        return self.settings.get(key)


if __name__ == "__main__":
    config1 = ConfigManager()
    config2 = ConfigManager()

    config1.set("theme", "dark")

    print(config1.get("theme"))
    print(config2.get("theme"))
    print(config1 is config2)
    print(config1.__dict__ is config2.__dict__)

Output

stdout

dark
dark
False
True

How it works

The Borg pattern (also called Monostate) ensures all instances share the same internal state by pointing each instance's __dict__ to the same class-level dictionary _shared_state. Because __dict__ holds instance attributes, assigning it once in __init__ makes every instance read and write to the same storage. The super().__init__() call in ConfigManager ensures the shared dictionary is set before accessing settings, and the hasattr guard prevents resetting existing shared data. Unlike the Singleton pattern, instances remain distinct objects (is returns False), but their attribute dictionaries are identical, so state is synchronized automatically.

Common mistakes

Forgetting to call `super().__init__()` in subclasses, leaving instance dicts unset
Reassigning `self.settings = {}` unconditionally, wiping shared state on each new instance
Assuming `config1 is config2` will be True—Borg keeps separate objects
Mutating `_shared_state` directly instead of through instance methods

Variations

Use `__new__` to return a shared instance for a true Singleton pattern
Set `_shared_state` as a class attribute updated via `__getattr__` for lightweight access

Real-world use cases

Sharing application configuration across modules without passing objects explicitly.
Maintaining a global connection pool or cache that all service instances reference.
Tracking user session data across multiple handler objects in a web application.

Borg pattern shared state in Python

Python code

Output

How it works

Common mistakes

Variations

Real-world use cases

More from OOP & classes

Tutorials

Quizzes

Python code

Output

How it works

Common mistakes

Variations

Real-world use cases

More from OOP & classes

Keep learning

Tutorials

Quizzes