Skip to content
Go to file


Failed to load latest commit information.
Latest commit message
Commit time

Public Whip Stories in Ready Build Status Code Climate Test Coverage


In our democracy the definitive exercise of the power we give our politicians when we vote them into office is how they vote in our parliaments on our behalf. Yet you probably don't know how your MP votes. This isn't your fault.

Parliamentary voting information is notoriously difficult to find and analyse. This project changes that by making it understandable and easy to use.

Over 10 years ago the pioneering Public Whip project was created in the UK. This is an evolution of that original PHP application into a modern Rails application.

Process overview


The project parses the Australian Federal Hansard into ParlParse format (this due to it's history of being a fork of the UK TheyWorkForYou project). The debates XML files the parser creates, also available on, contain voting data and we load this into a Rails application.


The UK Public Whip site still operates from the original PHP codebase however it's very possible for it to be upgraded to Rails in the future. During the development of They Vote For You, the OpenAustralia Foundation was careful to ensure there is an upgrade path.

To upgrade, checkout the php-compatibility branch and point the Rails application at a copy of the UK production database. Test the site out and fix any bugs - there are likely to be some UK-specific additions needed to the Rails application.

Once the site is working you can then checkout a more recent version of the codebase and run rake db:migrate to upgrade the database schema. This also is likely to need some UK-specific changes.

The final step is to customise the site language and interface. The best way to achieve this would be to develop some sort of theming system.


People data is collected by a scraper and fed into EveryPolitician. This produces Popolo formatted data that is then loaded into TVFY using a Rake task, e.g.:

bundle exec rake application:load:popolo[]

Once the people data has been loaded you can start loading votes. These are scraped by another scraper, that saves data in a flat format that can easily be converted to Popolo. The conversion is handled by a small proxy application and the results are imported using another Rake task, e.g.:

bundle exec rake application:load:popolo[]


If your machine is already set up to develop Rails applications with MySQL just carry out the following steps and you should be good to go. Developing with Vagrant is also possible (see below) but was mainly useful with the retired PHP application.

Before beginning, install MySQL, HTMLTidy and Ruby:

# OS X ...
brew install tidy-html5 mysql rbenv ruby-build
rbenv install $(cat .ruby-version)

# ... or Linux (Debian)
sudo apt-get install tidy mysql-server mysql-client libmysqlclient-dev
# then follow: to get rbenv and ruby-build

Steps required to configure, install and start the Rails application:

# Install bundle
bundle install

# Copy the default config files over.
# (Edit config/database.yml and fill in your username, password and database settings.)
bundle exec rake application:config:dev
cp config/database.yml.example config/database.yml

# Copy secrets config
cp config/secrets.yml.example config/secrets.yml

# Set up your database (including seed data)
bundle exec rake db:setup

# Run tests
bundle exec rake

# Start the server
bundle exec rails server

With Vagrant

Once you have vagrant and virtualbox installed and have cloned this repository run vagrant up. This will download the base virtualbox image and set up the development environment, be prepared for a bit of a wait.

Run the tests from inside the VM like this:

  • vagrant ssh
  • cd /vagrant
  • bundle exec rake

Assuming they pass, you can start the rails server:

  • bundle exec rails server

Once it is up you can browse to http://localhost:3000

When manually testing the site, the "sign up" confirmation emails will automatically go to a dummy smtp server called mailcatcher. To check the emails, browse to http://localhost:1080

If vagrant reports that it can't mount the /vagrant virtualbox shared folder, it's becuase the VM has had it's kernel updated. Run vagrant provision && vagrant reload and you should be back in business.

The original PHP app is also available at http://localhost:8080 but only if you're running an older branch (out of scope for this guide).

Loading data


These rake tasks are the ones you're most likely to need to run. You can run them as the deploy user in /srv/www/production/current, for instance:

deploy@ip-172-31-37-36:/srv/www/production/current$ RAILS_ENV=production bundle exec rake application:load:divisions[2018-10-18]
  • application:load:members loads members, offices and electorates. You always need this to run the site. Strictly speaking it only needs to run when details need updating but can be run as often as you like as it only updates data.
  • application:load:divisions[from_date,to_date] load division[s]. to_date is optional and if omitted, allows you to load a single date.
  • application:cache this namespace contains cache updating tasks that are necessary for the site to run. They should be self-explanatory.

Daily updates are carried out by the application:load:daily Rake task, which is run daily at 09:15 by cron.


Countries that use Popolo, e.g. Ukraine, only need to know about the application:load:popolo Rake task. It will load people or country data, depending on what it finds in the file.

Better Search

You can enable elasticsearch for a better search experience. Enable the setting in config/settings.yml then download the .deb for Linux or on Mac run brew install elasticsearch.

We're currently using a very old version of ElasticSearch in production, version 1.4.0, so we can't use the latest versions of the searchkick gem either.

Add data to your index the first time with bundle exec rake searchkick:reindex:all and Searchkick should take care of updates from there.


Extra Requirements

  • Memcached



The code is deployed using Capistrano. To deploy to production run:

bundle exec cap production deploy

You'll need a local copy of config/newrelic.yml that includes your licence key to be able to record deployments to New Relic.


Server provisioning

Ukraine's server has its configuration management in another repository. Once you've run the server provisioning tasks you can follow the instructions below to deploy the application.


After provisioning your development server, set up and deploy using Mina:

bundle exec mina ukraine-dev setup
bundle exec mina ukraine-dev deploy

# Now you can load people data
bundle exec mina ukraine-dev rake[application:load:popolo[]]

# And some vote data
bundle exec mina ukraine-dev rake[application:load:popolo[]]

# Setup caches
bundle exec mina ukraine-dev rake[application:cache:all_except_member_distances]

# Then build the index so search works
bundle exec mina ukraine-dev rake[searchkick:reindex:all]

To deploy to the production server, replace ukraine-dev with ukraine-production in the above commands.

Other Credits

This project uses some icons from the noun project under under creative commons licenses:


Making parliamentary voting information accessible, understandable, and easy to use so that you can hold your elected representatives to account.





No packages published