Reportr
"Your life's personal dashboard."
Reportr is a complete application which works like a dashboard for tracking events in your life (using a very simple API). With a simple interface, it helps you track and display your online activity (with trackers for Facebook, Twitter, GitHub, ...) or your real-life activity (with hardware trackers or applications like Runkeeper).
The project is entirely open source and you can host your own Reportr instance on your own server or Heroku.
A simple instance of Reportr runs at www.reportr.io. This is a very small instance and you can't use it yet for tracking a lot of events, but I'll probably scale this in the future.
Why is Reportr great?
- Host your own Reportr instance and keep your data private
- Track events from anywhere (web server, client application, connected hardware, ...)
- It's very simple to track an event: HTTP API
- Simple, but powerful web dashboard
- Realtime display
- One place for all your personal analytics data
- Very simple to export data for analytics (machine learning, ...)
How to host your own Reportr instance?
Get the application source code
# Clone the source code
$ git clone https://github.com/SamyPesse/reportr.git && cd ./reportr
Configuration
Configure the Reportr instance using Heroku config vars or define these variables in a .env file for local use. Configure the differents trackers using Trackers list below.
# Host name
HOST
# Express Secret session
SESSION_SECRET
Create your application and deploy it
# Create your heroku application
$ heroku create
# Add MongoHQ or MongoLab addon for heroku
$ heroku addons:add mongohq:small
or
$ heroku addons:add mongolab:sandbox
# Deploy the application
$ git push heroku master
# Open the application in your browser
$ heroku open
Run it locally
For running it in local, use foreman from heroku:
foreman start
Run it on Ubuntu 12.04
# Install dependencies:
sudo apt-get install -y mongodb
# Clone reportr
git clone https://github.com/SamyPesse/reportr.git && cd ./reportr
npm install
# Build client
make build
# Write a run script with your configuration
echo -e "
PORT=8000 \
FACEBOOK_CLIENTID=1 \
FACEBOOK_CLIENTSECRET=1 \
TWITTER_CLIENTID=1 \
TWITTER_CLIENTSECRET=1 \
FOURSQUARE_CLIENTID=1 \
FOURSQUARE_CLIENTSECRET=1 \
GITHUB_CLIENTID=1 \
GITHUB_CLIENTSECRET=1 \
RUNKEEPER_CLIENTID=1 \
RUNKEEPER_CLIENTSECRET=1 \
FITBIT_CLIENTID=1 \
FITBIT_CLIENTSECRET=1 \
node bin/reportr.js run --mode=all" > run.sh
chmod +x run.sh
# Start Reportr
./run.sh
Run for free on heroku (only one web dynoo)
You can run Reportr for free on Heroku by using the Procfile named Procfile-free, it'll only run one web dyno (with the worker and background task in it).
# Change mode
make mode-free
# Run it
make run
# Deploy it
make deploy
Trackers
State | Name | Description | Configuration |
---|---|---|---|
X | Web Navigation | Track web navigation using a Chrome extension | |
X | Ping | Ping state of the Reportr instance | |
X | Foursquare | Track your checkins | Callback url is /auth/foursquare/callback |
FOURSQUARE_CLIENTID | |||
FOURSQUARE_CLIENTSECRET | |||
X | GitHub | Track coding activity on GitHub | Callback url is /auth/github/callback |
GITHUB_CLIENTID | |||
GITHUB_CLIENTSECRET | |||
X | Track relations on Facebook | Callback url is /auth/facebook/callback | |
FACEBOOK_CLIENTID | |||
FACEBOOK_CLIENTSECRET | |||
X | Track twitter activity (tweets, mentions, ...) | Callback url is /auth/twitter/callback | |
TWITTER_CLIENTID | |||
TWITTER_CLIENTSECRET | |||
X | Runkeeper | Track fitness activity from Runkeeper | Callback url is /auth/runkeeper/callback |
RUNKEEPER_CLIENTID | |||
RUNKEEPER_CLIENTSECRET | |||
X | Fitbit | Track fitness activity from Fitbit | Callback url is /auth/fitbit/callback |
FITBIT_CLIENTID | |||
FITBIT_CLIENTSECRET | |||
- | Temperature | Track the temperature in a room (using tessel.io) |
APIs
Reportr uses a HTTP REST API to track event and manage models.
Data is always JSON encoded and Base64 encoded and passed as a "data" argument. You can pass a "callback" arguments for using HTTP API in a client side application.
Python and Javascript libraries for Reportr are in the examples directory. To build the Javascript client library, you will need 'browserify' (npm install -g browserify).
Track events
Events are defined by 'event', 'namespace' and 'properties'. You can alse specify optional paramaters such as 'id' for updating a unique event if existant or define 'timestamp' to track date & time for the event.
<host>/api/<token>/events/track
{
"namespace": "string", // Event namespace
"event": "(string)", // Event name
"properties": {
// Properties for the event
}
}
List events
# List last events
<host>/api/<token>/events/last
# List specific events
<host>/api/<token>/events/<namespace>/<event>
Get data series
<host>/api/<token>/data/<namespace>/<event>?arguments
arguments could contains :
interval: Interval between data (in ms) (default: 1000)
period: Period for events (in ms) (default: -1)
property: Property to calculate (default: null)
transform: Transofrmation to do (default: 'sum')
fill: Fill empty time with 0 (default: true)
Define models
Event models define information about how to display an event in the dashboard.
<host>/api/<token>/model/set
{
"namespace": "string", // Event namespace
"event": "(string)", // Event name
"name": "(string)", // Display name for the event
"icon": "(string)", // Url for a 64x64 icon image
"description": "(string)", // Description text for the event
}
List models
<host>/api/<token>/models
Special properties for events
When you track events using the API, you can define some special properties that can be use by Reportr for advanced use. Special properties always begin with '@'.
@lat : latitude for location
@lng : longitude for location