What is a deep copy in the context of working with data?

A deep copy creates a new composite object (e.g., list, dictionary), and then recursively inserts copies of the contents of the original object into it. This means that if the original object contains nested objects (lists, dictionaries, etc.), the deep copy will also copy them, creating a completely independent structure.

import copy

original_list = [[1, 2], [3, 4]]
# Deep copy will create new nested lists
deep_copied_list = copy.deepcopy(original_list)

# Changing a nested list in the original will not affect the deep copy
original_list[0][0] = 99
# deep_copied_list will remain [[1, 2], [3, 4]]

Comparison table with shallow copy:

Characteristic	Shallow Copy	Deep Copy
Object creation	New composite object	New composite object
Content copying	Copies references to nested objects	Recursively copies nested objects
Independence	Nested objects are shared	Nested objects are fully independent
Changing nested	May affect the original	Does not affect the original

Deep copy is used when full independence between the original object and its copy is required, including all nested structures.