python-empty-list-parameters.md

theme	default
background	https://source.unsplash.com/collection/94734566/1920x1080
class	text-center
highlighter	shiki
lineNumbers	true
drawings	persist

The Dangers of Empty Lists as Default Parameters in Python

---

The Problem

def add_item(item, my_list=[]):
    my_list.append(item)
    return my_list

print(add_item(1))  # [1]
print(add_item(2))  # [1, 2] 😱

Wait, where did 1 come from?

---

What's Happening?

• Default parameters are evaluated only once at function definition time
• The empty list is created when the function is defined
• The same list object is reused for every call where a list isn't provided
• Modifications persist between function calls

---

The Explanation

When Python loads the function definition:

def add_item(item, my_list=[]):
    ...

1. It creates an empty list object in memory
2. It binds the parameter my_list to reference this object
3. This happens just once when the module loads
4. All calls to add_item() without a second argument will use the same list object

---

The Solution: Use None as Default

def add_item(item, my_list=None):
    if my_list is None:
        my_list = []
    my_list.append(item)
    return my_list

print(add_item(1))  # [1]
print(add_item(2))  # [2] ✅

---

Why This Works

• None is immutable and safe to use as a default value
• A new empty list is created each time the function is called
• Each call gets its own unique list
• No more unexpected shared state between calls

---

The Rule

Never use mutable objects as default parameters:

• ❌ Lists: def func(param=[])
• ❌ Dictionaries: def func(param={})
• ❌ Sets: def func(param=set())
• ✅ Immutable types are safe: None, 0, "", (), etc.

---

Bonus: Python Visualization

First call

add_item(1)

Creates output: [1]

Second call

add_item(2)

Creates output: [1, 2]

Both calls reference the same list!

---

Resources

• Python documentation on default parameters
• Common Gotchas in Python
• Created with Slidev