Sorting Lists of Dictionaries by Key Values in Python

Introduction

In many programming tasks, especially those involving data processing and manipulation, you may encounter situations where you need to sort collections of complex objects. In Python, one common scenario is sorting a list of dictionaries based on specific dictionary values. This tutorial will guide you through different techniques for achieving this using Python’s built-in functions and modules.

Understanding the Problem

Suppose you have a list of dictionaries, each representing an entity with various attributes. For example:

people = [
    {'name': 'Homer', 'age': 39},
    {'name': 'Bart', 'age': 10}
]

Your goal is to sort this list by one or more dictionary values, such as the name key or the age key.

Sorting with the sorted() Function

The sorted() function in Python returns a new sorted list from an iterable. It has several parameters that can help customize sorting behavior:

• key: A function to serve as a sort key.
• reverse: A boolean value indicating whether the list should be sorted in descending order.

Sorting by a Single Key

To sort the people list by the name key, you can use the sorted() function with a lambda function or operator.itemgetter.

Using a Lambda Function:

from pprint import pprint

# Original list
people = [
    {'name': 'Homer', 'age': 39},
    {'name': 'Bart', 'age': 10}
]

# Sorting by 'name' using lambda
sorted_by_name = sorted(people, key=lambda d: d['name'])
pprint(sorted_by_name)

Using operator.itemgetter:

from operator import itemgetter

# Sorting by 'name' using itemgetter
sorted_by_name = sorted(people, key=itemgetter('name'))
pprint(sorted_by_name)

Both methods will output:

[{'age': 10, 'name': 'Bart'}, {'age': 39, 'name': 'Homer'}]

Sorting in Descending Order

To sort the list in descending order by name, simply set reverse=True:

sorted_by_name_desc = sorted(people, key=itemgetter('name'), reverse=True)
pprint(sorted_by_name_desc)

Output:

[{'age': 39, 'name': 'Homer'}, {'age': 10, 'name': 'Bart'}]

In-Place Sorting with list.sort()

If you prefer to sort the list in place without creating a new list, use the sort() method:

# Sorts the original list by 'age' in place
people.sort(key=itemgetter('age'))
pprint(people)

Output:

[{'age': 10, 'name': 'Bart'}, {'age': 39, 'name': 'Homer'}]

Sorting by Multiple Keys

Python allows you to sort by multiple keys. For example, if two people have the same age, you might want to sort them by their name.

people = [
    {'name': 'Homer', 'age': 39},
    {'name': 'Milhouse', 'age': 10},
    {'name': 'Bart', 'age': 10}
]

# Sorting first by age, then by name
sorted_people = sorted(people, key=lambda elem: (elem['age'], elem['name']))
pprint(sorted_people)

Output:

[{'age': 10, 'name': 'Bart'}, {'age': 10, 'name': 'Milhouse'}, {'age': 39, 'name': 'Homer'}]

Best Practices and Tips

• Use itemgetter for Performance: While lambda functions are versatile and readable, operator.itemgetter can be more efficient for simple key extractions.
• Immutable Sorting: Use sorted() if you need to keep the original list unchanged. Use .sort() when you want an in-place modification.
• Key Function Flexibility: You can define custom key functions that return tuples or complex objects, providing a high degree of sorting customization.

Conclusion

Sorting lists of dictionaries by specific keys is a common task in Python programming. Using sorted(), list.sort(), and tools like lambda functions and operator.itemgetter, you can efficiently sort your data structures to meet various needs. Understanding these methods allows for flexible, readable, and efficient code when dealing with complex collections.