PreviousNext
AI-assisted engineering/Reviewing AI code

Fix the discount screen for an order endpoint

Review the AI's codeEasy10 minShopify and Stripe style

Your task

An AI wrote a DRF endpoint that applies a percent discount to an order. Its serializer checked only that the percent was a number, so finance found orders with 300 percent discounts and negative totals. Your team now screens every discount request with business rules before it touches an order.

An assistant wrote screen_discount_requests(percents, max_percent). It is in the editor. It has a bug. Find it and fix it.

Input and output:

  • percents: the requested discounts, as numbers that may have decimals.
  • max_percent: the most this user's role may grant, as a number. A bad config can make it 0 or negative.
  • Return one verdict per request, in the same order.

Rules the tests check:

  • A request's verdict is the first of these that applies:
    1. "rejected_negative" when the percent is below 0.
    2. "rejected_above_100" when the percent is above 100.
    3. "rejected_above_policy" when the percent is above max_percent.
    4. "accepted" otherwise.
  • Compare the numbers exactly, with no rounding.

Examples

screen_discount_requests(percents: list[float], max_percent: float) -> list[str]
Example 1
Inputscreen_discount_requests([10, 300, -50], 25)
Output["accepted", "rejected_above_100", "rejected_negative"]
Example 2
Inputscreen_discount_requests([25, 25.5], 25)
Output["accepted", "rejected_above_policy"]
Example 3
Inputscreen_discount_requests([], 50)
Output[]