HCA Coding Assessment

This repository represents a simple API that stores and retrieves "Transactions".

Your task will be to add a new API end point that accepts a csv of transactions from one of three different sources, transforms then inserts the data into our transaction repository. See the TODO section below for more details!

API Overview

To start the API, run:

./mvnw spring-boot:run 

Now you can create a transaction with a POST to localhost:8080/transactions

{
  "id": "46c0917b-e62f-41fd-b111-f09b575ef2c3",
  "basketSize": 29.39,
  "budtenderId": "6908f98f-da5b-4566-8f8f-49f83b3806ed",
  "locationId": "3e3d99b2-d2e9-4142-b6ba-da9592e92b51",
  "timestamp": "2023-08-09T22:44:00.000+00:00"
}

You can then retrieve transaction with a GET to that same resource:

[
  {
    "id": "46c0917b-e62f-41fd-b111-f09b575ef2c3",
    "basketSize": 29.39,
    "budtenderId": "6908f98f-da5b-4566-8f8f-49f83b3806ed",
    "locationId": "3e3d99b2-d2e9-4142-b6ba-da9592e92b51",
    "timestamp": "2023-08-09T22:44:00.000+00:00"
  }
]

Note that this project uses an in-memory embedded database for persistence, so if you restart the running service anything you create will be gone!

TODO

1. Add an API that accepts a CSV of transactions in one of the three formats defined in src/main/fixtures.
2. Take that csv and transform it into a Transaction, depending on the format. Insert the transaction into our repository. The CSVs we will use for testing can be found in src/main/fixtures.

Note there are three CSVs, each with data stored slightly differently.

good_pos.csv is the most similar to our transaction model.

wonky_pos.csv has a couple of curve balls in the data...

Additional Questions

1. Imagine one of these data sources started providing us individual transactions in JSON via a webhook. How would you refactor your code to handle this with minimal duplication?
2. How easy would it be for us to add a fourth or fifth data source? How would you refactor your code to make it easier to handle new data sources?