Tuples and Unpacking 🔒

Round brackets instead of square

point = (3, 7)
colours = ("red", "green", "blue")

print(point, colours)
print(point[0], colours[-1])
print(len(colours))

(3, 7) ('red', 'green', 'blue')
3 blue
3

Everything you can read from a list works. Everything that would change it does not:

point = (3, 7)
point[0] = 10

TypeError: 'tuple' object does not support item assignment

Why would you want that?

• It documents intent. A tuple says "these belong together and this set will not change": a coordinate, an RGB colour, a database row.
• It cannot be broken by accident. Hand a list to a function and the function might append to it. Hand a tuple and it cannot.
• It can be a dictionary key. Lists cannot (Lesson 13 explains why).
• It is slightly smaller and faster. True, and almost never the reason to choose one.

not_a_tuple = (5)
actually_a_tuple = (5,)

print(type(not_a_tuple))
print(type(actually_a_tuple))

<class 'int'>
<class 'tuple'>

Unpacking: the feature you will use every day

point = (3, 7)
x, y = point
print(x, y)

# the brackets are optional, which is why this works
name, role, age = "Guybrush", "pirate", 24
print(f"{name} the {role}, aged {age}")

# and why swapping is one line
a, b = 1, 2
a, b = b, a
print(a, b)

3 7
Guybrush the pirate, aged 24
2 1

The number of names on the left must match the number of values on the right, or Python raises ValueError: too many values to unpack. Unless you use a star:

scores = [95, 88, 72, 61, 40]

best, *rest = scores
print(best, rest)

first, *middle, last = scores
print(first, middle, last)

95 [88, 72, 61, 40]
95 [88, 72, 61] 40

The starred name soaks up everything left over, and always becomes a list. There can be at most one of them, for the obvious reason.

Returning several things at once

def split_name(full_name):
    parts = full_name.split()
    return parts[0], parts[-1]


first, last = split_name("Guybrush Ulysses Threepwood")
print(first, last)

both = split_name("Elaine Marley")
print(both, type(both))

Guybrush Threepwood
('Elaine', 'Marley') <class 'tuple'>

A function can only return one object, but that object can be a tuple, so in practice Python functions return as many values as they like. Notice you did not have to write any brackets: a bare comma-separated list of values is a tuple. This is the single most common use of tuples in real code.

Tuples in loops

crew = [
    ("Guybrush", "pirate", 24),
    ("Elaine", "governor", 28),
    ("Otis", "prisoner", 41),
]

for name, role, age in crew:
    print(f"{name:10} {role:10} {age}")

Guybrush   pirate     24
Elaine     governor   28
Otis       prisoner   41

The loop unpacks each tuple automatically. This is exactly how enumerate and zip from Lesson 9 work: they hand you tuples, and you unpack them in the for line without thinking about it.

Named tuples, when positions get confusing

from collections import namedtuple

Pirate = namedtuple("Pirate", ["name", "role", "insults"])

guy = Pirate("Guybrush", "pirate", 7)

print(guy)
print(guy.name, guy.insults)
print(guy[0])                 # still a tuple underneath
name, role, insults = guy     # still unpacks
print(role)

Pirate(name='Guybrush', role='pirate', insults=7)
Guybrush 7
Guybrush
pirate

guy.name beats guy[0] the moment you have more than two fields. For anything more elaborate, Lesson 33's dataclass is the modern answer, but named tuples are perfect when you want something immutable and light.

def summarise(numbers):
    return min(numbers), max(numbers), sum(numbers) / len(numbers)


lowest, highest, average = summarise([88, 42, 95, 61, 73])

print(f"Lowest:  {lowest}")
print(f"Highest: {highest}")
print(f"Average: {average:.1f}")

Lowest:  42
Highest: 95
Average: 71.8

data = [
    ("Monkey Island", 9, 8, 10),
    ("Disco Elysium", 10, 10, 9),
    ("Grim Fandango", 9, 9, 9),
]

data = [
    ("Monkey Island", 9, 8, 10),
    ("Disco Elysium", 10, 10, 9),
    ("Grim Fandango", 9, 9, 9),
]

for title, *ratings in data:
    print(f"{title:15} best score {max(ratings)}")

Monkey Island   best score 10
Disco Elysium   best score 10
Grim Fandango   best score 9