Python sorted(): Sorting Iterables with Custom Keys & Stability

1. Syntax & Formal Parameter Specifications 🔝

The sorted() function accepts one positional iterable and optional keyword arguments:

# Formal syntax of built-in sorted()
sorted_list = sorted(iterable, /, *, key=None, reverse=False)

2. Basic Usage: Numbers, Strings & Iterables 🔝

Pass any iterable collection to sorted() to generate a new ordered list without modifying the input:

# 1. Sorting a numeric list
numbers = [12, 3, 8, 7, 14]
sorted_nums = sorted(numbers)

print("Sorted copy:  ", sorted_nums)
print("Original list:", numbers)

# 2. Sorting strings and characters
fruits = ["Orange", "Banana", "Apple", "Mango"]
print("Sorted fruits: ", sorted(fruits))

text = "plus2net"
print("Sorted letters:", sorted(text))

Output:

Sorted copy:   [3, 7, 8, 12, 14]
Original list: [12, 3, 8, 7, 14]
Sorted fruits:  ['Apple', 'Banana', 'Mango', 'Orange']
Sorted letters: ['2', 'e', 'l', 'n', 'p', 's', 't', 'u']

3. Descending Sort with reverse=True 🔝

Pass reverse=True to arrange items in descending order directly:

# Descending sort using reverse=True
numbers = [12, 3, 8, 7, 14]
descending_nums = sorted(numbers, reverse=True)
print("Descending numbers:", descending_nums)

fruits = ["Orange", "Banana", "Apple", "Mango"]
print("Descending fruits: ", sorted(fruits, reverse=True))

Output:

Descending numbers: [14, 12, 8, 7, 3]
Descending fruits:  ['Orange', 'Mango', 'Banana', 'Apple']

4. Custom Sorting with key Functions 🔝

The key parameter accepts a callable function computing comparison proxy values:

# 1. Sorting by length using built-in len as key
words = ["Three", "One", "Seven", "thirtyfive", "Two"]
by_length = sorted(words, key=len)
print("Sorted by length:", by_length)

# 2. Case-insensitive sorting with str.lower
mixed_cases = ["banana", "Apple", "orange", "Cherry"]
case_insensitive = sorted(mixed_cases, key=str.lower)
print("Case-insensitive:", case_insensitive)

# 3. Custom key function for tuples
records = [(3, 5), (2, 1), (8, 4), (3, 7)]
def get_second(item):
    return item[1]

by_second = sorted(records, key=get_second)
print("Sorted by 2nd item:", by_second)

Output:

Sorted by length: ['One', 'Two', 'Three', 'Seven', 'thirtyfive']
Case-insensitive: ['Apple', 'banana', 'Cherry', 'orange']
Sorted by 2nd item: [(2, 1), (8, 4), (3, 5), (3, 7)]

5. sorted() vs. list.sort() Comparison 🔝

Compare non-destructive copying with in-place list mutation:

# sorted() creates a new list; list.sort() mutates in-place
original_data = [5, 2, 9, 1]

# sorted() leaves original untouched and returns a new list
result_copy = sorted(original_data)
print("result_copy:", result_copy)
print("original_data:", original_data)

# list.sort() modifies original and returns None
ret_val = original_data.sort()
print("original_data after sort():", original_data)
print("return value of sort():    ", ret_val)

Output:

result_copy: [1, 2, 5, 9]
original_data: [5, 2, 9, 1]
original_data after sort(): [1, 2, 5, 9]
return value of sort():     None

6. Production Recipes: Multi-Criteria & Object Sorting 🔝

Multi-level comparisons across dictionary fields or custom objects:

# Recipe A: Multi-attribute sorting using operator.itemgetter
from operator import itemgetter

employees = [
    {"name": "Alice", "dept": "HR", "salary": 75000},
    {"name": "Bob", "dept": "IT", "salary": 92000},
    {"name": "Charlie", "dept": "IT", "salary": 85000},
    {"name": "Diana", "dept": "HR", "salary": 80000},
]

# Sort by department ascending, then by salary descending
by_dept_salary = sorted(employees, key=lambda e: (e["dept"], -e["salary"]))
for emp in by_dept_salary:
    print(f"{emp['dept']} | {emp['salary']} | {emp['name']}")

# Recipe B: Stable sorting with custom object keys
class Student:
    def __init__(self, name: str, grade: int, gpa: float):
        self.name = name
        self.grade = grade
        self.gpa = gpa

students = [Student("Amy", 10, 3.8), Student("Ben", 11, 3.9), Student("Cara", 10, 3.95)]
top_students = sorted(students, key=lambda s: s.gpa, reverse=True)
print("Top student:", top_students[0].name, top_students[0].gpa)

Output:

HR | 80000 | Diana
HR | 75000 | Alice
IT | 92000 | Bob
IT | 85000 | Charlie
Top student: Cara 3.95

7. Common Pitfalls & Developer Traps 🔝

# Pitfall 1: Sorting unorderable mixed types in Python 3
try:
    sorted([10, "apple", 5])
except TypeError as err:
    print("Mixed types error:", err)

# Pitfall 2: Sorting sets returns a list, not a set
unique_numbers = {9, 1, 4, 3}
sorted_result = sorted(unique_numbers)
print("Type of sorted result:", type(sorted_result), sorted_result)

Output:

Mixed types error: '<' not supported between instances of 'str' and 'int'
Type of sorted result:  [1, 3, 4, 9]

8. Interactive Notebook & Practice Exercises 🔝

Practice stable sorting, multi-criteria lambdas, and benchmark sorting performance in our companion Jupyter Notebook:

Self-Verifying Assertion Suite

Run these unit checks directly in Python or your local environment to verify sorted() mechanics:

def test_sorted_mechanics():
    # 1. Basic sorting
    assert sorted([5, 1, 3]) == [1, 3, 5]
    assert sorted([5, 1, 3], reverse=True) == [5, 3, 1]
    
    # 2. Key function with length
    words = ["bb", "a", "ccc"]
    assert sorted(words, key=len) == ["a", "bb", "ccc"]
    
    # 3. Tuple key multi-criteria
    data = [(1, "b"), (2, "a"), (1, "a")]
    assert sorted(data) == [(1, "a"), (1, "b"), (2, "a")]
    
    # 4. Guaranteeing original iterable immutability
    original = [4, 2, 8]
    res = sorted(original)
    assert original == [4, 2, 8]
    assert res == [2, 4, 8]
    
    print("All sorted() unit assertions passed! 100% verified.")

test_sorted_mechanics()

Output:

All sorted() unit assertions passed! 100% verified.

9. Related Built-in Functions & Sequence Tools 🔝

Subhendu Mohapatra

Author & Instructor at plus2net

I write and maintain practical tutorials on Python, PHP, SQL, JavaScript, HTML, jQuery, and web development at plus2net. The tutorials focus on clear explanations, working examples, and code that readers can test and adapt while learning.

About the author   Watch on YouTube

Connect: