IBM FISE Lounge

IBM FISE Lounge is an application that acts as a smart, collaborative workout coach for the elderly to use in the current pandemic, where loneliness is becoming a major issue. It provides a simple interface with a standalone dashboard for more tech-savvy relatives to set up the Lounge app and preferences on the elderly relative's behalf.

Contributors:

• Adam Peace (App, Tensorflow Exercise Assistant and Server)
• Emil Almazov (Server API & Backend)
• Rikaz Rameez (IBM Watson Integration).

This repository contains both the web app for FISE (in /app) as well as the dashboard and API (in /server)

Server Installation

Ensure you have yarn installed

• cd server
• yarn install
• cp .env.local.example .env.local (This file should never be tracked, only the example)
• Fill in all the missing details in server/.env.local
• To begin development: yarn dev
• To build for production: yarn build
• To serve production build: yarn start

App Installation

Ensure you have yarn installed

• cd app
• yarn install
• cp .env.local.example .env.local (This file should never be tracked, only the example)
• Fill in all the missing details in app/.env.local
• To begin development: yarn dev
• To build for production: yarn build
• To serve production build: yarn start

Server Documentation

Overview

• Running on NextJS 9 (different to Express, look it up and be familiar with ES5/ES6 syntax)
• Auth handled by Passport
• Storage in MongoDB (Can easily set up using Atlas)
• All API routes at localhost:3000/api/*
• All dashboard pages at localhost:3000/*
• All API routes are "pages" in NextJS 9
• All pages are in /server/pages, with API routes in /server/pages/api and the file path from pages corresponding to the actual path

Data Types

User

• The actual account administrator
• They manage the Consumer's FISE account and set it up on the Consumer's device
• They can add mutiple Consumers and add multiple Contacts per Consumer
• This is the only person who ever accesses the Dashboard

Consumer

• The elderly person with difficulty contacting their relatives
• This person's data will show up on the FISE app
• Can be related to multiple Contacts

Contact

• Someone the Consumer can call
• Will receive an email when called through the app

Deploying the Server

Easiest way: Vercel

• Connect repo to Vercel
• Add all env variables that are in /server/.env.local.example
• (NB: Adding VERCEL_URL will autopopulate based on whether the build is production or preview)

Otherwise: Server

• Follow the above instructions for Server Installation and expose port 3000

Server API Routes

/api/login

POST

Parameters:

• email
• password

/api/logout

POST

Parameters: None

/api/signup

POST

Parameters:

• email
• password
• name

/api/user

GET

Gets the current (based on session) user's data

/api/user/delete

DELETE

Revokes the current user's session

/api/consumer

POST

Create a new consumer

Success: returns {message: ..., data: consumer}

/api/consumer/:consumer_id

GET

Get the corresponding consumer data

PUT

Update the corresponding consumer's data

DELETE

Delete the corresponding consumer

POST

Refreshes one-time-code of the corresponding consumer

/api/contact

POST

Create a new contact

Parameters:

• consumer_id
• name
• email
• relation

/api/contact/:contact_id

GET

Get the corresponding contact data

PUT

Update the corresponding contact's data

Parameters:

• [name]
• [email]
• [relation]

DELETE

Delete the corresponding contact

/api/otc/:otc

GET

Get the corresponding consumer's data

POST

Sends a call notification email to the desired contact

Parameters:

• contact_id

/api/otc/watson/:otc

POST

Parse audio through IBM Watson API

Parameters:

• req.body should be Base64 encoded audio/mp3