Chapter 6 - Python Comprehensions¶
The Python language has a couple of methods for creating lists and dictionaries that are known as comprehensions. There is also a third type of comprehension for creating a Python set. In this chapter we will learn how to use each type of comprehension. You will find that the comprehension constructs build on the knowledge you have acquired from the previous chapters as they contain loops and conditionals themselves.
List Comprehensions¶
List comprehensions in Python are very handy. They can also be a little hard to understand when and why you would use them. List comprehensions tend to be harder to read than just using a simple for loop as well. You may want to review the looping chapter before you continue.
If you are ready, then we’ll spend some time looking at how to construct list comprehensions and learn how they can be used. A list comprehension is basically a one line for loop that produces a Python list data structure. Here’s a simple example:
>>> x = [i for i in range(5)]
Let’s break this down a bit. Python comes with a range function that can return a list of numbers. By default, it returns integers starting at 0 and going up to but not including the number you pass it. So in this case, it returns a list containing the integers 0-4. This can be useful if you need to create a list very quickly. For example, say you’re parsing a file and looking for something in particular. You could use a list comprehension as a kind of filter:
if [i for i in line if "SOME TERM" in i]:
# do something
I have used code similar to this to look through a file quickly to parse out specific lines or sections of the file. When you throw functions into the mix, you can start doing some really cool stuff. Say you want to apply a function to every element in a list, such as when you need to cast a bunch of strings into integers:
>>> x = ['1', '2', '3', '4', '5']
>>> y = [int(i) for i in x]
>>> y
[1, 2, 3, 4, 5]
This sort of thing comes up more often than you’d think. I have also had to loop over a list of strings and call a string method, such as strip on them because they had all kinds of leading or ending white space:
>>> myStrings = [s.strip() for s in myStringList]
There are also occasions where one needs to create a nested list comprehension. One reason to do that is to flatten multiple lists into one. This example comes from the Python documentation:
>>> vec = [[1,2,3], [4,5,6], [7,8,9]]
>>> [num for elem in vec for num in elem]
[1, 2, 3, 4, 5, 6, 7, 8, 9]
The documentation shows several other interesting examples for nested list comprehensions as well. I highly recommend taking a look at it! At this point, you should now be capable of using list comprehensions in your own code and use them well. Just use your imagination and you’ll start seeing lots of good places where you too can use them.
Now we’re ready to move on to Python’s dictionary comprehensions!
Dictionary Comprehensions¶
Dictionary comprehensions started life in Python 3.0, but were backported to Python 2.7. They were originally proposed in the Python Enhancement Proposal 274 (PEP 274) back in 2001. They are pretty similar to a list comprehension in the way that they are organized.
The best way to understand is to just do one!
>>> print( {i: str(i) for i in range(5)} )
{0: '0', 1: '1', 2: '2', 3: '3', 4: '4'}
This is a pretty straightforward comprehension. Basically it is creating an integer key and string value for each item in the range. Now you may be wondering how you could use a dictionary comprehension in real life. Mark Pilgrim mentioned that you could use a dictionary comprehension for swapping the dictionary’s keys and values. Here’s how you would do that:
>>> my_dict = {1:"dog", 2:"cat", 3:"hamster"}
>>> print( {value:key for key, value in my_dict.items()} )
{'hamster': 3, 'dog': 1, 'cat': 2}
This will only work if the dictionary values are of a non-mutable type, such as a string. Otherwise you will end up causing an exception to be raised.
I could also see a dictionary comprehension being useful for creating a table out of class variables and their values. However, we haven’t covered classes at this point, so I won’t confuse you with that here.
Set Comprehensions¶
Set comprehensions are created in much the same way as dictionary comprehensions. Now a Python set is much like a mathematical set in that it doesn’t have any repeated elements. You can create a normal set like this:
>>> my_list = [1, 2, 2, 3, 4, 5, 5, 7, 8]
>>> my_set = set(my_list)
>>> my_set
set([1, 2, 3, 4, 5, 7, 8])
As you can see from the example above, the call to set has removed the duplicates from the list. Now let’s rewrite this code to use a set comprehension:
>>> my_list = [1, 2, 2, 3, 4, 5, 5, 7, 8]
>>> my_set = {x for x in my_list}
>>> my_set
set([1, 2, 3, 4, 5, 7, 8])
You will notice that to create a set comprehension, we basically changed the square brackets that a list comprehension uses to the curly braces that the dictionary comprehension has.
Wrapping Up¶
Now you know how to use the various Python comprehensions. You will probably find the list comprehension the most useful at first and also the most popular. If you start using your imagination, I am sure you will be able to find uses for all three types of comprehensions. Now we’re ready to move on and learn about exception handling!