Mutable Default Arguments

The Pitfall

This is perhaps the most famous "gotcha" in Python.

If you use a mutable object (like a list, dict, or set) as a default argument in a function signature, that object is created exactly once when the function is defined, NOT every time the function is called.

The Broken Code

def append_to_list(val, my_list=[]):
    my_list.append(val)
    return my_list

# Interviewer: "What does this print?"
print(append_to_list(1))
print(append_to_list(2))

What You Think Happens:

[1]
[2]

What Actually Happens:

[1]
[1, 2]

Because my_list is created once during function definition, the second call to append_to_list(2) appends to the exact same list that was used in the first call.

---

When Does This Happen in Interviews?

This bug frequently occurs when candidates write helper functions for Trees or Graphs, and they try to initialize a visited set or a path list in the signature.

# DANGEROUS
def dfs(node, visited=set()):
    if node in visited: return
    visited.add(node)
    # ...

If the interviewer runs your dfs() function twice on two different test cases, the second test case will fail immediately because visited still contains all the nodes from the first test case!

---

The Fix

Always use None as the default value, and initialize the mutable object inside the function body.

# CORRECT
def dfs(node, visited=None):
    if visited is None:
        visited = set()

    if node in visited: return
    visited.add(node)
    # ...

This guarantees that a fresh, empty set is created every single time the function is called without a visited argument.