2.4. Function Calls

The Python interpreter can compute new values with function calls. You are familiar with the idea of functions from high school algebra. There you might define a function f by specifying how it transforms an input into an output, f(x) = 3x + 2. Then, you might write f(5) and expect to get the value 17.

Python adopts a similar syntax for invoking functions. If there is a named function foo that takes a single input, we can invoke foo on the value 5 by writing foo(5).

There are many built-in functions available in Python. You’ll be seeing some in this chapter and the next couple of chapters.

Functions are like factories that take in some material, do some operation, and then send out the resulting object.

Icon that represents a function. Appears simliar to a factory with three places on top for inputs to come in and a place on the bottom for the output/return value to come out.

In this case, we refer to the materials as arguments or inputs and the resulting object is refered to as output or return value. This process of taking input, doing something, and then sending back the output is demonstrated in the gif below.

Animated gif that deomnstrates using the visual representation of a factory as used above. Shows three arrows coming into the function to represent that input or arguments that a function can require. It then shows the function object shaking, representing an action being completed by the function. Then it shows annother arrow leaving the function image, which represents a return value or output coming from the factory.

Note

Don’t confuse the “output value” of a function with the output window. The output of a function is a Python value and we can never really see the internal representation of a value. But we can draw pictures to help us imagine what values are, or we can print them to see an external representation in the output window.

To confuse things even more, print is actually a function. All functions produce output values. Only the print function causes things to appear in the output window.

It is also possible for programmers to define new functions in their programs. You will learn how to do that later in the course. For now, you just need to learn how to invoke, or call, a function, and understand that the execution of the function returns a computed value.

We’ve defined two functions above. The code is hidden so as not to bother you (yet) with how functions are defined. square takes a single input parameter, and returns that input multiplied by itself. sub takes two input parameters and returns the result of subtracting the second from the first. Obviously, these functions are not particularly useful, since we have the operators * and - available. But they illustrate how functions work. The visual below illustrates how the square function works.

a visual of the square function. Four is provided as the input, the function object shakes, and then sixteen comes out from the bottom of the function object.

Notice that when a function takes more than one input parameter, the inputs are separated by a comma. Also notice that the order of the inputs matters. The value before the comma is treated as the first input, the value after it as the second input.

Again, remember that when Python performs computations, the results are only shown in the output window if there’s a print statement that says to do that. In the activcode window above, square(5) produces the value 25 but we never get to see that in the output window, because it is not printed.

2.4.1. Function calls as part of complex expressions

Anywhere in an expression that you can write a literal like a number, you can also write a function invocation that produces a number.

For example:

Let’s take a look at how that last execution unfolds.

Notice that we always have to resolve the expression inside the innermost parentheses first, in order to determine what input to provide when calling the functions.

2.4.2. Functions are objects; parentheses invoke functions

Remember the earlier mention that some kinds of Python objects don’t have a nice printed representation? Functions are themselves just objects. If you tell Python to print the function object, rather than printing the results of invoking the function object, you’ll get one of those not-so-nice printed representations.

Just typing the name of the function refers to the function as an object. Typing the name of the function followed by parentheses () invokes the function.

Next Section - 2.5. Data Types