Python中lambda、key与排序：sorted()与list.sort()相关疑问

Great questions—let’s break these down one by one to clear up the confusion!

1. Does sorted() have a return value? #

Absolutely! Unlike list.sort(), which sorts the list in-place and returns None, sorted() creates and returns a brand new sorted list. The original iterable you pass to it stays completely unchanged.

For example:

original = [3, 1, 2]
sorted_version = sorted(original)
print(original)    # Output: [3, 1, 2] (unchanged)
print(sorted_version)  # Output: [1, 2, 3] (new sorted list)

# Compare with list.sort()
original.sort()
print(original)    # Output: [1, 2, 3] (modified in-place)
print(original.sort())  # Output: None (it returns nothing)

2. Why use sorted() instead of list.sort()? #

There are three key scenarios where sorted() is the better choice:

• You need to keep the original data: If you still need the unsorted version of your iterable later, sorted() is perfect because it doesn’t alter the original. list.sort() destroys the original order permanently.
• You’re working with non-list iterables: list.sort() is a method that only exists on lists. If you want to sort a tuple, string, generator, or any other iterable, sorted() handles them all and returns a sorted list as output.
• Chaining operations: Since sorted() returns a list, you can pass it directly to another function in one line. For example:

  # Works perfectly
  print(sorted([5, 2, 7]))  # Output: [2,5,7]
  
  # Doesn't work as intended (prints None)
  print([5,2,7].sort())  # Output: None
  

3. What do x and y mean in sorting code? #

This depends on where they’re used—there are two common cases:

Case 1: In key functions #

When you use the key parameter (the most common approach in Python 3), x represents a single element from your iterable. The key function runs once per element, and returns a value that Python uses to determine the sort order.

For example:

people = [("Alice", 25), ("Bob", 20), ("Charlie", 30)]
# x is each tuple (e.g., ("Alice",25)), we sort by the second element (age)
sorted_people = sorted(people, key=lambda x: x[1])
print(sorted_people)  # Output: [("Bob",20), ("Alice",25), ("Charlie",30)]

Case 2: In comparison functions (via functools.cmp_to_key) #

If you’re using a custom comparison function (often wrapped with cmp_to_key in Python 3), x and y are two elements from your iterable that Python is comparing. The function returns:

• A negative number if x should come before y
• 0 if x and y are equal in sort order
• A positive number if y should come before x

Example:

from functools import cmp_to_key
numbers = [3, 1, 4, 2]
# x and y are pairs of numbers; we return y-x to sort descending
sorted_desc = sorted(numbers, key=cmp_to_key(lambda x, y: y - x))
print(sorted_desc)  # Output: [4,3,2,1]

内容的提问来源于stack exchange，提问作者Pal Kol