04 · Functions, parameters and testable logic

Understand the mechanism

A function body runs when called. Parameters are inputs and return provides output; paths without a return produce None. Pure formatting or calculation functions are easier to verify than duplicated output expressions.

Keyword arguments clarify calls. Default values are evaluated at definition time, so a default list can be shared across calls. Use a None sentinel and allocate a fresh container inside when needed.

Prefer returning results over hidden global mutation. Document intentional mutation of arguments. Type annotations assist readers and tools but do not automatically validate runtime values.

Test requirements such as zero duration and negative input. Assertions are useful for exercise checks, not as the only production input validation because optimized execution can omit them.

Runnable example

Use Python 3.11 or later. Save lesson.py in a separate practice directory and run python3 lesson.py, or py lesson.py on Windows. Examples use only the standard library.

def duration_label(minutes):
    if type(minutes) is not int or minutes < 0:
        raise ValueError("minutes must be a nonnegative integer")
    hours, rest = divmod(minutes, 60)
    return f"{hours}h {rest}m"

def collect(title, items=None):
    if items is None:
        items = []
    items.append(title)
    return items

assert duration_label(125) == "2h 5m"
assert duration_label(0) == "0h 0m"
print(duration_label(125))
print(collect("Python"), collect("SQL"))

Trace and verify

Expect 2h 5m and separate one-item lists. Temporarily use a default list to observe sharing, then restore. The strict integer check deliberately rejects booleans despite their integer subclass relationship.

Common mistakes and boundaries

Do not hide unclear contracts behind broad exception handling. This example deliberately mutates a list supplied by the caller; document ownership.

Practice challenge

Add average calculation with None for empty data, no source mutation and an explicit numeric policy.

def average_minutes(values):
    if any(type(value) is not int or value < 0 for value in values):
        raise ValueError("minutes must be nonnegative integers")
    if not values:
        return None
    return sum(values) / len(values)

print(average_minutes([]))
print(average_minutes([0, 30, 60]))
for bad in [[-1], [True], ["30"]]:
    try:
        average_minutes(bad)
    except ValueError:
        print("rejected")
assert average_minutes([0]) == 0

Further reading

• Python official tutorial