PART 2 · PYTHON FOUNDATIONS · CHAPTER 1

Names, objects and mutability

FreeLesson 24 of 78

A Python variable is a name tag on an object, not a box that holds a value. Two names can point at the same object, and that causes most beginner bugs.

TWO NAMES, ONE OBJECTPYTHON 3.12

a = [1, 2, 3]
b = a            # a second name, same list
b.append(4)
print(a, a is b)

c = a.copy()     # a new list
c.append(5)
print(a, c, a is c)

x, y = [1, 2], [1, 2]
print(x == y, x is y)   # equal value, different objects

Output

[1, 2, 3, 4] True
[1, 2, 3, 4] [1, 2, 3, 4, 5] False
True False

== compares values. is compares identity: same object in memory. Use is only for None, True and False.

Immutable (cannot change in place)	Mutable (can change in place)
`int float bool str tuple frozenset bytes None`	`list dict set bytearray`, most objects you create
"Changing" one makes a new object: s = s + "x"	Changes are seen by every name pointing at it

The mutable default argument trap

THE BUG AND THE FIXPYTHON 3.12

def add_row(row, batch=[]):      # default made ONCE
    batch.append(row)
    return batch

print(add_row("r1"))
print(add_row("r2"))             # r1 is still there

def add_row_fixed(row, batch=None):
    batch = [] if batch is None else batch
    batch.append(row)
    return batch

print(add_row_fixed("r1"), add_row_fixed("r2"))

Output

['r1']
['r1', 'r2']
['r1'] ['r2']

Shallow vs deep copy

COPIES OF NESTED DATAPYTHON 3.12

import copy

rows = [{"id": 1, "tags": ["new"]}]
shallow = rows.copy()          # new outer list only
deep = copy.deepcopy(rows)     # new everything

shallow[0]["tags"].append("vip")
print(rows[0]["tags"])         # changed through shallow
print(deep[0]["tags"])         # untouched

Output

['new', 'vip']
['new']

Practice before you move on

Make a list of order dicts, copy it with list(), change a status in the copy, and check whether the original changed.
Do the same with copy.deepcopy and write one line on why the result differs.
Write a function with a list as its default argument, call it twice, and fix it with None.