Before You Start

Locally

You need to have PostgreSQL (14) and Node.js (20) installed and working.

Start the postgress service:

For OS:

brew services start postgresql

For Linux:

sudo service postgresql start

Install dependencies:

npm install

Create the postgress user/password:

psql -U postgres -c "CREATE USER node_js_exercise_user WITH PASSWORD '123456789';"

Change the .env file with the following information:

POSTGRES_USER=node_js_exercise_user
POSTGRES_PASSWORD=123456789
POSTGRES_HOST=localhost

Push the schema into the db:

npx prisma db push

Docker

In case you do not want to install dependencies locally and you have already installed docker in your host machine, you can build the exercise image:

make build

Start the services:

make start

Set up db:

make db-setup

Run a bash container:

make terminal

Run your tests:

npm test

If you want to debug with VSCode, you need to run the Attach to Mocha in Docker debug task available at the top of the Run and Debug section. This task simply run npm test but allos you to use breakpoints successfully.

Note: You will have to restart the vscodemocha service every time after you run the task with:

docker-compose restart vscodemocha

Introduction

Razor has a collection of shows with ticket inventory numbers that need to be synchronized with a 3rd party API located at https://shows-remote-api.com. A job runs every hour to check which shows need to be synchronized.

When we synchronize shows, we don't want to synchronize them all at once because we could overwhelm the 3rd party API with web requests. We prefer to schedule the show updates at least 15 seconds apart, but if there are more shows than can fit into a 1 hour period spaced 15 seconds apart, the updates can be scheduled in shorter intervals.

Your job is to examine the shows in the database and determine which ones need to be updated, then output a hash with the IDs that need updating and when they should be updated. You will not update any shows as part of this project, only output the schedule of when each show should be updated. You only need to schedule updates for shows that exist in your local database, for which the quantity has changed, and which have not been updated in the last hour.

Schema With Examples

 shows
-------------------------------
| id | quantity | last_update |
|  1 |       35 |         154 |
|  2 |        4 |       10435 |
|  3 |       89 |        7343 |
|  4 |       15 |       12704 |
|  5 |        2 |        3865 |
-------------------------------

• last_update is represented in seconds since last update.

Example API Interaction

GET https://shows-remote-api.com/

Response:

[
  { "id": 1, "quantity": 34 },
  { "id": 2, "quantity": 4 },
  { "id": 3, "quantity": 91 },
  { "id": 4, "quantity": 12 },
  { "id": 5, "quantity": 1 }
]

Example Output

In our example, we need to update IDs 3, 4, and 5 because it has been more than an hour since they were last updated and their quantity has changed.

Hash format is { ID => seconds_until_next_update }

{ 3 => 0, 4 => 15, 5 => 30 }

Implementation Details

• If an ID has never been updated, it will not have a value for seconds_since_last_update
• You should test at least 2 things for each example file:
  1. Which IDs should be scheduled for update? The outputs for these are included in the spec.
  2. When should each ID be scheduled for update? The answer for this is determined by you, but should be in the specified format.

Other Information

• Your solution should use Test-Driven Development.
• Feel free to add any dependencies you may want to use.
• Any resource you want is available.
• Ask as many questions as you want.
• Note that the third-party API will only respond under the test execution lifecycle since it is not a real API.