Python The None Value

None means the value is absent, which is different from every other "empty-looking" value:

Keeping these apart lets a program give the right message. If an unknown estimate were stored as 0, the customer would be told the repair is free.

Use None as the starting value of a variable that will be filled in later, as the result of a search that found nothing (dict.get() returns None for a missing key), and for optional fields in records. Avoid using it to mean an error occurred; raise an exception instead, so the problem can't be ignored.

There is exactly one None object in a Python program, so the correct test is identity: value is None, or value is not None.

estimate = 0
print(estimate is None)   # False: 0 is a real estimate
print(not estimate)       # True: 0 is falsy, so this test is wrong here

not estimate and if estimate: treat None, 0, "" and empty lists the same way, which is exactly the confusion to avoid when 0 is meaningful. PEP 8 recommends is None over == None, because a class can redefine == to return something unexpected, while is can't be overridden.

Doing arithmetic or calling methods on None raises errors such as TypeError: unsupported operand type(s) for +: 'NoneType' and 'int' or AttributeError: 'NoneType' object has no attribute 'upper'. When you see NoneType in a traceback, find where the variable was expected to get a real value and didn't.

Every function returns something. If execution reaches the end of the body without a return, or hits a bare return, the result is None:

def reminder(name):
    print(f"Collect your bike, {name}")

result = reminder("Ana")   # prints the message
print(result)              # None

Printing is not returning. reminder() shows text on the screen, but the caller receives nothing it can store or reuse. Many built-in methods that change an object in place also return None on purpose, to signal that they didn't create a new object: list.sort(), list.append() and dict.update() all return None. Writing notes = notes.append("done") therefore replaces the list with None.

None is the standard default for an optional argument, especially when the real default is a mutable object such as a list:

def add_note(text, notes=None):
    if notes is None:
        notes = []          # a fresh list on every call
    notes.append(text)
    return notes

print(add_note("brakes"))   # ['brakes']
print(add_note("chain"))    # ['chain']

Writing notes=[] instead would create one list when the function is defined and share it between every call, so the second call would return ['brakes', 'chain']. Using None also lets a function tell "argument not given" apart from any real value, including 0 and empty strings. In type hints, an optional value is written str | None (or Optional[str] before Python 3.10).

Task: display separate messages for unknown, free and priced repairs. Write a function that takes an estimate and returns one of three messages, then test it with None, 0 and 30.

def repair_message(estimate):
    if estimate is None:
        return "We'll call you with a quote."
    if estimate == 0:
        return "Good news: no charge."
    return f"Your repair will cost £{estimate}."

for value in (None, 0, 30):
    print(repair_message(value))

The order of the checks matters: testing if not estimate: first would give None and 0 the same message. For the worked check, the test that specifically checks for None is value is None.

The documentation describes the None constant and truth value testing. Continue with function arguments for more on defaults, and booleans for truthiness. Practise with the Python exercises and Python quiz.