Node Challenge

Functionalities

• Paginated results.
• Sorting by amount_in_cents or date_created.
• Filtering by status and currency and merchant_name.
• Query using GraphQL.
• Store new expense data.
• GET, POST request validation.
• Caching for resilence and performance.
• dotenv for easy env management.
• pre-start validation to cut out runtime/enviroment issues.

API

• 
• link to doc

Running App

• Create a .env in the base directory from the .env-example file included.
• Make sure your node version is >=16
• Note: the app pre-startup check fill fails and app would not start if you don't have the right env variables or node version is below the required version.

Run:

yarn

psql challenge < dump.sql

Start

yarn start

Test

yarn test

yarn test:acceptance

Summary

• I only allow sorting by amount_in_cents and date_created because they were the only data points that make sense to sort the records by, merchant_name is another candidate but not as strong a candidate as the other two.
• I liberally looked to what was in the user domain for inspiration in term of folder structures.
• I added pre-start validation to ensure all env values are available and node runtime has the required version.
• I added dontenv to easy env management.
• I added request validation for improved security.
• Thinking of API design, I wanted to make the expense route RESTFUL since I add feature to create new expense record but didn't do that at the end.
• Might have to revisit the data caching going on, since I am parsing and stringifying records the overhead introduced by JSON methods is one to watch out for.
• Adding test to the GraphQL endpoint is something that can be revisited.
• My decision making process is camptured in my commit messages.
• Finally, I spend about 5hr over 2 days putting these together.