Functions and returned values

We’ve used our intake decision a few times now. It would be nice to write that decision once and reuse it, especially when we need to fix a mistake. A function lets us give the decision a name and try it with different inputs.

Defining and calling a function¶

def tells Python that we’re defining a function. We give it a name, then list its inputs inside parentheses. These input names are called parameters. The colon and indented body work like the blocks you’ve already seen.

Here’s an easy detail to miss: defining a function doesn’t run it. It gives Python instructions to keep ready for when we call that function.

Calling choose_output(0.4, True) runs the function with those arguments: the parameter request receives 0.4 and enabled receives True. return sends a value back to the caller and leaves the function immediately. The caller can save that value, print it, or use it in another decision.

Printing a value is different from returning it. A function that only prints doesn’t give the caller that number; with no return, Python returns None. Names created inside a function are normally local to that call. That means the caller can’t use those names directly. Pass in what the function needs through its parameters, then return the result you want the caller to use. This keeps the connection between the two parts of your program easier to follow.

Read it before running it¶

Put this worked example in student-work/foundations/functions.py. Read it once and write down what you think it prints. Then save and run python student-work/foundations/functions.py using your environment’s Python.

def choose_output(request, enabled):
    if not enabled:
        return 0.0
    return request

result = choose_output(0.4, True)
print("Enabled:", result)
print("Disabled:", choose_output(0.4, False))

The first call returns 0.4; the second returns 0.0. Follow the disabled call: it reaches the first return, so it never executes the final return. Try changing the first return to print(0.0) temporarily. The function now continues and returns the request, which is the wrong disabled behavior. Restore it.

This is why we describe a function’s contract: what inputs it accepts and what result it promises. A clear contract makes the function easier to test and reuse in a mechanism later.

Try it yourself¶

def intake_output(request, enabled, has_piece):
    if not enabled:
        return 0.0
    if has_piece and request > 0:
        return 0.0
    return request

print(intake_output(0.5, True, False))
print(intake_output(0.5, True, True))
print(intake_output(-0.5, True, True))
print(intake_output(-0.5, False, True))

Before moving on¶

Explain parameters versus arguments. Show that calling the function with two different requests does not require editing its definition.