Solution 2: Deep Copying Nested State

Completed idea

Replace copy.copy() with copy.deepcopy().

This makes the clone’s nested objects — the widgets list, each Widget, and the filters dictionary — fully independent from the original.

---

Solution

def clone(self, **changes) -> "Dashboard":
    cloned = deepcopy(self)

    for field_name, value in changes.items():
        setattr(cloned, field_name, value)

    return cloned

---

Why shallow copy breaks here

With copy.copy(), the clone and the original share the same filters dictionary and the same Widget objects in memory.

Mutating one affects the other:

cloned.filters["region"] = "UK"
print(template.filters["region"])  # "UK" — unexpected

With deepcopy(), the entire object graph is copied recursively.

cloned.filters becomes a new dictionary, and cloned.widgets[0] becomes a new Widget object.

Changing either no longer affects the original.

---

Key distinction

copy.copy()    -> new outer object, shared nested objects
copy.deepcopy() -> new outer object, new nested objects (recursively)

For Prototype, the question to ask is:

Should the clone share nested objects with the original, or own independent copies?

If independence is required — as it almost always is for mutable nested state — use deepcopy().

---

Back to exercise · Solution script