Symptom Tracker

This a mono repo implementing a web application made for collecting symptoms related to COVID-19. This would help:

• tracking evolution of symptoms amongst confined populations
• anticipate trafic in healthcare facilities

Collected data is meant to be open and non-identifying.

This application consists of:

• a lightweight (~80KB) react, next.js server-side-rendered frontend (see ./pages)
• a single lambda endpoint to store submissions sent by users (see ./api and ./lib/db)
• a Postgres instance where data will be stored (see config in ./.env and ./docker-compose.yml)

Local deployment

Install dependencies

yarn install

Run client and lambda

now dev

In case you only want to run the client:

next dev

Start local postgres instance

docker-compose up

When instanciating this db for the first time, you should create all tables etc using

yarn run typeorm schema:sync

Create or generate migrations

Requirements:

• You should have ts-node installed globally (yarn global add ts-node).

Commands:

typeorm migration:create -n migrationName
typeorm migration:generate -n migrationName

Run migrations

Requirements:

• Typeorm only handles .js migrations, so one needs to transpile them first:

tsc lib/db/**/*.ts --outDir build

Commands:

yarn run typeorm migration:run

Deploy to prod

Deploy client and lambda

now --prod

Run migrations to prod database

Requirements:

• you should have dotenv-cli installed globally (yarn global add dotenv-cli)
• you should have a ./.env.prod file with necessary variables setup (all necessary variables are listed in ./.env)

Commands:

dotenv -e .env.prod yarn run typeorm migration:run