You are welcome to go through our tutorials but please keep in mind that as we rely on community PRs for maintainance they may be out of date.

Fun with Functions

In this tutorial we are going to look at what functions are, how we use them and why.

Calling a function

A function is a collection of statements grouped together, that can be called later to run these statements. Python comes with plenty of functions already built in that you can use to perform common tasks. One example should be something you’re already familiar with:

>>> print("This is a function")
This is a function

It is also possible to store the output of a function in a variable for future use. Let’s try this with the built in function len(), which will provide us with the length of a string:

>>> length_of_my_string = len("This is my string")
>>> print(length_of_my_string)
17

The function len() takes one argument: the string it’s measuring, and it returns an integer: the length of the string. Here, we’re storing the output of that function into a variable and then printing it out.

We can also chain functions to use the output of other functions as parameters. In the earlier example, we stored the integer provided by the len() function in a variable, and then passed that variable as a parameter for the print() function, which then printed the list of numbers to the screen. That’s more steps than are really necessary. We can just as easily use the output of len() as an argument for print():

>>> print(len("This is my string"))
17

This is called nesting and it can be very useful in making code more concise, but using it too much could make code harder to read. Knowing when to use variables instead of chaining functions comes with experience; in reading and in writing code.

Try using len() to measure different strings with different lengths. Looking at the documentation is always a good idea when learning more about the language. You can also learn more by purposefully trying to break the function: What happens when you don’t provide any parameters, or try to measure an integer?

Defining your own functions

Now we are going to create our own function. To do this we use the def keyword, provide a unique (and descriptive) name for the function and specify any parameters that may be accepted by the function. Then we write the code we want to be run when the function is called. Let’s start with a simple function that takes no parameters:

>>> def print_name():
...     print("My name is Bart")

We can then call this function like we would any other:

 >>> print_name()
 My name is Bart

We can change the function to use a parameter. Let’s change it so that it allows the user to specify the name they want to print:

>>> def print_name(name):
...     print("My name is " + name)
...
>>> print_name("Lisa")
My name is Lisa

You may have noticed that the code inside the function is indented. This is to let the Python interpreter know what is part of the function we are defining, and what is not. That is why when we write print_name("Lisa") directly after, it knows to call the function and not include it in the function definition, it is not indented.

Now that we have defined a parameter, the function will raise an error if no parameter is provided. Try it yourself, it’s good to get used to understanding how Python errors work and what they mean. It’s an important skill when debugging more complex code.

We can enhance the function to run with a default name if one is not provided. Let’s do that by specifying a default value.

>>> def print_name(name='Bart'):
...     print("My name is " + name)
...
>>> print_name("Lisa")
My name is Lisa
>>> print_name()
My name is Bart

Different kinds of functions

Our function prints stuff out to the string, but returns None. Let’s make a function that returns a value:

>>> def get_statement(name='Bart'):
...    return "My name is " + name

Now the function returns a value that we can work with. We could store this in a variable to be used later. Let’s try storing it as a value, and using that value as a parameter in another print() function call.

>>> statement = get_statement("Lisa")
>>> print(statement)
My name is Lisa

Notice that we now need to use the print function to print to the screen. While the REPL will print what the function returns to the console, if run as a script, simply calling the function without a print statement would not print anything to the screen. This is because the function is returning a value, and no longer printing a string.

A subtle indicator of this are the quote marks around the sentence that denotes a string is being output to the screen, rather than something being printed with the print function:

>>> statement = get_statement("Lisa")
>>> statement
'My name is Lisa'

Why use functions

At first glance it may not seem obvious why it’s worth expending the extra effort of defining and calling a function, rather than just writing the code independent of such things. Defining tasks as functions reduces the need to copy and paste the same code multiple times to achieve the same effect. Simply calling the function multiple times makes it much easier to not only write code, but to read it as well.

Using functions also makes it a lot easier to fix and change code. If you are performing the same tasks in multiple places and discover a bug, without functions you would need to fix the same bug multiple times. However with functions, you simply fix the bug once, and all of the subsequent function calls will now behave accordingly.

Further reading

Python comes with a lot of built-in functions that are worth learning about, as they have been created specifically to perform the most common tasks that developers require. They are listed in the documentation. To begin with, try using a few to see what they do, and how they behave with varying parameters. Once you have grasped how a few of them work, try writing your own functions that call the built in functions within them, like we did when we used the print function in our own custom print_name() function.

Additionally, the site LearnXinYminutes has some good material on Python.