test_correctness.py

import pytest

from legacy_calculator import calculate_total as legacy_calc
from refactored_calculator import calculate_total as refactored_calc

TEST_CASES = [
    # (items, discount_code, customer_type, is_holiday)
    ([{"category": "electronics", "price": 1200}], None, "premium", True),
    ([{"category": "clothing", "price": 50}], "SUMMER", "regular", False),
    ([{"category": "books", "price": 30}], "VIP20", "premium", False),
    ([{"category": "electronics", "price": 800}], None, "regular", False),
    ([{"category": "books", "price": 100}], None, "premium", False),
    (
        [
            {"category": "electronics", "price": 1200},
            {"category": "clothing", "price": 50},
            {"category": "books", "price": 30},
        ],
        "SUMMER",
        "premium",
        True,
    ),
]


@pytest.mark.parametrize("items,discount,customer,holiday", TEST_CASES)
def test_refactored_matches_legacy(items, discount, customer, holiday):
    """Verify refactored code produces identical results."""
    legacy_result = legacy_calc(items, discount, customer, holiday)
    refactored_result = refactored_calc(items, discount, customer, holiday)
    assert abs(legacy_result - refactored_result) < 0.01, (
        f"Results differ: {legacy_result} vs {refactored_result}"
    )