Sort. To perform sorting in Python, we use built-in methods. We use sort and sorted(). The sorted() built-in returns a view (not a list) that is ordered.
Sorting things. We can sort from low to high (ascending) or the opposite (descending). We can even sort based on a property or function call result.
Sort example. Let us first reorder a list of integers. To order a list's elements from lowest to highest (starting with negative numbers, moving to positive ones), we call sort().
Result The list is sorted in-place. Nothing needs to be assigned to the sort() call.
Sorted, reversed. These are views that transform lists without modifying them. Here, we apply the sorted() view and then the reversed() view and print the results.
Info We iterate over the reversed view with a for-loop. Reversed returns an iterator, not a list.
Tip You can apply a reversed sort directly with the sorted keyword. Simply set the "reverse" argument to True.
elements = [22, 333, 0, -22, 1000]
# Use a reversed, sorted view: a descending view.
view = reversed(sorted(elements))
# Display our results.
for element in view:
print(element)1000
333
22
0
-22
Sorted, arguments. This built-in can receive more than one argument. The first is the collection we want to sort. But we also can specify a key function.
Note The key function receives an element, and returns a value that is used to sort.
Note 2 A lambda expression may be used for the key argument. Here we receive an argument with name "color," and return its length.
Tip We can specify that the collection is to be sorted in descending order (from high to low) by setting reverse to true.
# An array of color names.
colors = ["blue", "lavender", "red", "yellow"]
# Sort colors by length, in reverse (descending) order.
for color in sorted(colors, key=lambda color: len(color), reverse=True):
print(color)lavender
yellow
blue
red
Sort dictionary. How can you sort the keys in a dictionary? First you must acquire a list of the keys using the keys() method. Then, you can use the sorted built-in to order those keys.
Note This approach requires a lookup to get the values from the sorted key view. Another approach may be faster.
# A dictionary with three pairs.
furniture = {"table": 1, "chair": 2, "desk": 4}
# Get sorted view of the keys.
s = sorted(furniture.keys())
# Display the sorted keys.
for key in s:
print(key, furniture[key])chair 2
desk 4
table 1
Error. An error will occur if you try to sort the result of the keys() method. This is because keys does not return a list. It returns a dict_keys object.
So We have to use sorted() here to avoid further conversions. We are not actually dealing with lists.
However If the program has to repeatedly get the keys and sort them, using a cached list of the sorted keys would likely be faster.
AttributeError: 'dict_keys' object has no attribute 'sort'
Objects. A list of class instances (objects) can be sorted. Here we introduce the Bird class. Each Bird has a weight. And the repr method returns a string representation of each object.
Start We append three new Bird objects to our list. The Birds are not ordered by their weights when we add them.
Next We use the sort() method with a lambda "key" argument. The lambda selects the key that each Bird is sorted upon.
Finally We display the sorted list of birds. This approach can be used with any object type.
class Bird:
def __init__(self, weight):
self.__weight = weight
def weight(self):
return self.__weight
def __repr__(self):
return "Bird, weight = " + str(self.__weight)
# Create a list of Bird objects.
birds = []
birds.append(Bird(10))
birds.append(Bird(5))
birds.append(Bird(200))
# Sort the birds by their weights.
birds.sort(lambda b: b.weight())
# Display sorted birds.
for b in birds:
print(b)Bird, weight = 5
Bird, weight = 10
Bird, weight = 200
String chars. We can sort the characters in a string. We first convert the string to a list of characters with a list comprehension. Then, after sort(), we join the chars together.
value = "boat"# Get list comprehension of characters and sort the list.
list = [c for c in value]
list.sort()
# Join the characters together.
result = "".join(list)
print(result)abot
Benchmark, sorted. I benchmarked the sort() and reverse() methods, and compare them to the sorted() and reversed() views. I found that the sort and reverse methods were more efficient here.
Version 1 This version of the code uses the sort() and reverse() methods. It loops over the returned elements.
Version 2 Here we use the sorted() and reversed() methods, and loop over the returned collections.
Result We discover that using sorted() and reversed() views is not always a performance advantage.
import time
data = ["carrot", "apple", "peach", "nectarine"]
print(time.time())
# Version 1: use sort and reverse.
i = 0
while i < 1000000:
v = 0
data.sort()
for element in data:
v += 1
data.reverse()
for element in data:
v += 1
i += 1
print(time.time())
# Version 2: use sorted and reversed.
i = 0
while i < 1000000:
v = 0
for element in sorted(data):
v += 1
for element in reversed(data):
v += 1
i += 1
print(time.time())1389905688.967582
1389905691.0347 Call sort(), reverse(): 2.07 s
1389905693.901864 Call sorted(), reversed(): 2.87 s
Custom. The data structures we construct before sorting are important. For example, building a list of tuples, and then sorting on an item in each tuple, is effective.
Info We sort files by their sizes. We store each file and its size to a tuple, and then append that tuple to a list.
Summary. In Python, we can directly use sort, or use the sorted() view. Where possible, using sort() is faster because it copies less. But often the simplest solution is preferred.
Dot Net Perls is a collection of tested code examples. Pages are continually updated to stay current, with code correctness a top priority.
Sam Allen is passionate about computer languages. In the past, his work has been recommended by Apple and Microsoft and he has studied computers at a selective university in the United States.
This page was last updated on Mar 22, 2024 (edit link).
