Welcome to GRAF

GitHub Repository Analytics with Filtering (GRAF) is a Rails web application that has the ability to collect and visually represent useful statistics on various GitHub repositories.

How it works?

GRAF collects and stores GitHub information in its local data warehouse using a GitHub API wrapper called Octokit. Through Octokit, GRAF gathers information on every Pull Request and Commit for the selected repositories. Once this data has been successfully collected, our application allows the user to generate figures that are customized by timeframe, repository, company, etc.

Company affiliation is determined by querying each GitHub user's profile. If a user does not have an affiliated company listed on their profile, the application then checks to see whether the user is a member of any corporate Github organizations. If they are not, the user email is used to determine company affiliation.

What statistics are available?

These are the available metrics and options found on the Analytics page:

Metrics

• Commits
• Pull requests
• Average number of days a pull request is open
• Percent of pull requests merged

Views

• Pie Chart
• Bar Chart
• Line Chart
• Table

Group By / Filters

• Organization
• Month
• Quarter
• Year
• State
• Repo
• Company
• User Name
• Login Name

Additionally start and end date filters are available.

Deploying a GRAF instance on IBM's BlueMix

From the app's root directory do a push (see following command) and create an mysql db service.

cf push -c 'bundle exec rake db:create db:migrate' graf

This command will fail but will setup your database. Now delete your application using but do NOT delete your orphan database

cf delete graf

Now do another push and bind to the mysql database you created in the first push:

cf push graf

Your app should now be up and running.

Initializing the data

1. Modify the "config/graf/orgs.json" JSON file to select the organizations to load
2. Open GRAF's /login page in a browser and enter a username / password
3. Begin an initial load by going to /load and click the "Start Load" button
4. Once the load is complete, the user can view the data on the /analytics and /report page
5. Subsequent loads can be executed using "Start Load" button /load (Ensure there are enough api requests at /info)

Updating a GRAF instance on BlueMix

cf delete graf
cf create-service mysql 300 graf-db
cf push graf
http://graf.stage1.ng.bluemix.net/info
cf delete graf
cf push -c 'bundle exec rake db:create db:migrate' graf
cf delete graf
cf push graf
cf push graf

Deploying GRAF on local machine

1. Install Rails and MySQL
2. Install bundled gems
3. Create database in mysql console

CREATE DATABASE graf_db CHARACTER SET utf8 COLLATE utf8_general_ci;
GRANT ALL PRIVILEGES ON graf_db.* TO 'graf_user'@'localhost' IDENTIFIED BY 'graf_password';

4. Setup tables

bundle exec rake db:migrate

5. Start server

rails s

Setting up GitHub API access

Set up a Github Oauth token:

1. Login to your github account at https://github.com/login
2. Go to Github's Settings page at https://github.com/settings/applications
3. Register an application (https://github.com/settings/applications/new) to get a Client ID and a Client Secret
4. Generate a token (https://github.com/settings/tokens/new)
5. Place the Client ID, Client Secret, and Access token into the settings.json file, which is located in the root directory (graf/config/graf/settings.json)