Mutable default argument bug
Quick summary
You used def f(x=[]) or def f(x={}) — the default is created once at definition time and shared across all calls that use the default.
Why this happens
A list, dict, or set is used as a default argument value, which is evaluated once when the function is defined
The function modifies the default argument in-place, and the mutation persists to the next call
The developer assumed the default is re-created fresh for each call, but Python evaluates defaults at definition time
Minimal example
def add_item(item, items=[]): # mutable default!
items.append(item)
return items
print(add_item(1)) # [1]
print(add_item(2)) # [1, 2] — not [2]!def add_item(item, items=None):
if items is None:
items = [] # fresh list each call
items.append(item)
return items
print(add_item(1)) # [1]
print(add_item(2)) # [2]Using None as the default and creating a new list inside the function ensures each call gets a fresh list, preventing state leakage between calls.
How to diagnose
Call the function multiple times without the argument and check if the result accumulates unexpectedly
Inspect the default argument with print(func.__defaults__) to see if it has been mutated
Review any function that modifies its argument in-place and has a mutable default value
How to fix
Use None as the default and create the mutable object inside the function body if the argument is None
Use an immutable default like a tuple or frozenset if the function does not need to modify it
Create a fresh copy of the default inside the function with list(default) or dict(default)
How to prevent
Never use mutable objects (list, dict, set) as default argument values — use None and initialize inside
Use a linter like flake8-bugbear (B006) that detects mutable default arguments