Code Wars 2022

It's Tuesday and you're walking into the best office on the planet: JG HQ.

Huh - the elevator - how silly. I think I'll take the stairs!

huff huff puff wheeeew

Made it up five whole flights, ahh look at that nice sign reminding me to log my stairs activity - haha yep I'll be forgetting to do that almost immediately once I sit down and become consumed with client work.

If only there was a way to log my stair climbing efforts where I had to do no logging on my own and they just got tallied up somewhere in the cloud ... automagically. Seems like a pipe dream... or is it?

JG EffingLegs takes your traditional stair logging workflow to the next level of awesome. Logging your stairs no longer needs to be done in a web browser with your smart device or at your desk once you're done climbing.

Forgetting how many flights you climbed no longer is an issue.

JG EffingLegs uses that wonderful face of yours, and it's kinda hard to forget that no matter how late you were staying up bashing the cron the night before.

The Velominati crew has taken it upon themselves to import high quality headshots of all JG employees (~96 or so as of 3/4/2022) into Amazon for usage within their Rekognition API. Anyone can begin logging stair efforts using our prototype and their wonderful face.

• Challenge other JGers to a stair-off for any time period over a day
• Oauth2 integration
• Deploy protoype to RaspPi or other low-cost hardware with USB camera and physical button/gpio
• Add more metrics and reports so JGers can dive into past performance
• Allow JGers to add additional headshots to increase accuracy of AI or support that killer new beard
• Sweep the Legs
• Scheduled jobs to compile aggregate challenge/climb data for management
• Add a dev environment

This is a lightweight Web 2 stack using React and AWS SAM.

The stack can be easily developed and tested locally.

For persistence we're using an RDS/PostgreSQL cluster to store users and their stair climbing achievements ingested from the IoT edge.

Functional / smoke testing is provided by Nightwatch.js to ensure basic functionality exists before shipping the app out automatically. Upon deploy to prod we have functional tests wired in to ensure the site can be navigated to and is responsive.

You'll need AWS SAM, Node.js 16.x and Python 3.9 installed locally before getting started with this.

To start the Angular webclient:

# optional
sudo npm install -g yarn
# required
cd client
npm install
yarn install
yarn start
# navigate to http://localhost:3000 and enjoy!

To start the Python API locally:

sam build (--use-container, maybe?)
sam local start-api

This application utilizes GitHub Actions and a mixture of custom and pre-made workflows from vendors like Amazon.

CI/CD pipelines are located within the https://github.com/kevholmes/effinglegsdocs/tree/main/.github/workflows folder.

The database Schema is relatively simple, we have four tables that currently track users, their efforts, and challenges between users. We also have a table to link users and their challenges.

SQL seed entries can be found in the https://github.com/kevholmes/effinglegsdocs/tree/main/sql folder of this project.