Coupon Service

Tech Stack

• FastAPI
• SQLAlchemy

Running the service

• Clone the repository
• Install the dependencies

poetry install

• Run the service

uvicorn app.main:app

API Documentation

• Swagger UI: http://localhost:8000/docs

Running the tests

pytest

Testing API Endpoints

1. Add Repeat Counts to a Coupon Code

curl -X POST "http://127.0.0.1:8000/api/coupons" \
  -H "Content-Type: application/json" \
  -d '{
        "code": "SUMMER_SALE",
        "repeat_counts": {
          "user_total": 3,
          "user_daily": 1,
          "user_weekly": 1,
          "global_total": 10000
        }
      }'

2. Verify a Coupon Code Validity

curl -X GET "http://127.0.0.1:8000/api/coupons/verify?code=SUMMER_SALE&user_id=user123"

3. Redeem a Coupon Code

curl -X POST "http://127.0.0.1:8000/api/coupons/apply" \
  -H "Content-Type: application/json" \
  -d '{
        "code": "SUMMER_SALE",
        "user_id": "user123"
      }'

Future Improvements

Trade-offs

• The current implementation uses an in-memory database for simplicity. For production use, we should use a persistent database like PostgreSQL.
• The CouponUsage table can grow large over time. In this implementation, we query this table frequently, which might affect performance. Indexing and archiving old data can help mitigate this.

Scalability Challenges

• With many users applying coupons simultaneously, there may be race conditions leading to overuse of coupons beyond their limits.
• Caching coupon configurations can reduce database load but may lead to stale data.
• Using a asynchronous implementation can help in handling more requests concurrently.