answer-map REST web service

• Run docker compose up.
• Run psql -d postgres -U user1 -a -f .\sql\create_schema.sql. May need to install psql.
• Enter password: pass1.
• The service will be available on port 8080.
• answer-map.postman_collection.json can be imported to connect to your local instance on 8080

1. How would you support multiple users?
   • there is an answer_user table with a primary key to identify each answer_user (user_id)
   • this user_id can be referenced in the answer_map table and with a unique constraint to make each answer_map unique to the user_id and answer_key
2. How would you support answers with types other than string?
   • Suppose that the answer value type is not known, then we can avoid decoding it from json by defining it as AnswerValue *json.RawMessage in the AnswerMap struct. That raw message can be marshalled into a byte array before being stored as a byte array in the database.
3. What are the main bottlenecks of your solution?
   • one replica of the docker image that runs the service
   • uses a sql docker image with limited resources (no volumes, small storage, not load-balanced)
4. How would you scale the service to cope with thousands of requests?
   • deploy more replicas of the docker image that runs the service and use a load-balancer
   • use a datasource that can autoscale resources