Health Dashboard

This repository contains all the source code of my personal health dashboard (see https://health.maximeheckel.com) I built this in order to provide a web interface for my healthkit data.

Tech stack

This project relies on:

• React for the frontend
• GraphQL as query language to contact the API
• Golang for the server side (both API and server)
• MongoDB for data storage

Installation

Running the project

The easiest way to get the project working in production mode is through Docker and docker-compose. This project contains the Dockerfile necessary to build the project

Note: Docker 17.05 at least is required to build this project as it uses multi-stage builds (see https://docs.docker.com/engine/userguide/eng-image/multistage-build/)

To run the project in production mode:

docker-compose up

This will download the required docker images and build the health dashboard docker image, spin them up, and link them under the same network.

The UI should then be accessible under http://yourhostname:8000

Pushing and accessing data

The backend accepts new data against http://yourhostname:8000/api/v1/healthhook Here's an example payload:

{
	"date":"20170224",
	"heartRate":
		[{
			"value":72,
			"startDate":"2017-02-24T18:58:26.947-0800",
			"endDate":"2017-02-24T18:58:26.947-0800"
		},
		...
		],
	"stepCountSamples":
		[{
			"value":34,
			"startDate":"2017-02-24T18:58:26.947-0800",
			"endDate":"2017-02-24T18:58:26.947-0800"
		},
		...
		],
	"sleepAnalysis":
		[{
			"value":"INBED",
			"startDate":"2017-02-24T18:58:26.947-0800",
			"endDate":"2017-02-24T18:58:26.947-0800"
		},
		...
		],
	"stepCount": {
		"value": 1000,
		"startDate":"2017-02-24T18:58:26.947-0800",
		"endDate":"2017-02-24T18:58:26.947-0800"
	},
	"cyclineDistance": {
		"value": 1000,
		"startDate":"2017-02-24T18:58:26.947-0800",
		"endDate":"2017-02-24T18:58:26.947-0800"
	},
	"walkinRunningDistance": {
		"value": 1000,
		"startDate":"2017-02-24T18:58:26.947-0800",
		"endDate":"2017-02-24T18:58:26.947-0800"
	},
	"weight": {
		"value": 7000,
		"startDate":"2017-02-24T18:58:26.947-0800",
		"endDate":"2017-02-24T18:58:26.947-0800"
	},
	"bodyfatpercentage": {
		"value": 12,
		"startDate":"2017-02-24T18:58:26.947-0800",
		"endDate":"2017-02-24T18:58:26.947-0800"
	},
	"bmi": {
		"value": 22,
		"startDate":"2017-02-24T18:58:26.947-0800",
		"endDate":"2017-02-24T18:58:26.947-0800"
	}
}

The data can be then accessed with a GraphQL client against: http://yourhostname:8000/api/v1/health/graphql

Note: If you want to require authentication on the http://yourhostname:8000/api/v1/healthhook endpoint you can add to the following environment variables: USER and PASS. You'll then need to include an Authorization header to any POST request against this endpoint:

Authorization: Basic USER:PASS where USER:PASS is in base64.

Contributing

If you want to contribute to the development of this project here's how you can run it in development mode:

Without Docker and Compose

1. You'll need at least node 7 and Golang 1.7 installed on your machine.
2. Pull the project in your GOPATH
3. Export the following envvars to target the host and port of your MongoDB instance: MONGODB_PORT_27017_TCP_ADDR and MONGODB_PORT_28017_TCP_PORT
4. Run yarn
5. Run yarn start. This will start the Golang server under http://localhost:8000 and the Webpack dev server for the frontend under http://localhost:3000.

With Docker and Compose

1. You'll need at least Docker 17.05
2. Run docker-compose -f docker-compose.dev.yml up --build --force-recreate. This will start the Golang server under http://localhost:8000 and the Webpack dev server for the frontend under http://localhost:3000 in a container as well as the mongo container linked to the previous one.

Every change in the frontend code is hot reloaded. For backend changes you will need to manually restart the Go server.

Note: The Golang part of the project uses govendor in order to manage dependencies