Alabama Election Map

This repository contains the files necessary to deploy a Bokeh figure on Heroku using Flask. The code visualizes the Alabama Senate Special Election 2017 results on a county level. The results of the US Presidential Election 2016 are also visualized as a comparison.

Description

There are two modules, process and app.

process: generates the requisite CSV files needed to visualize the election results.
app: serves the Bokeh visualization using Flask.

While we could use one set of dependencies for the entire application, we use pip-tools so that the deployment environment only contains the necessary packages to run.

Running the application

The first step to run the application is to clone the repository: git clone https://github.com/xofbd/alabama_election_map.

If you have make installed you can run the application several ways.

make deploy-standalone to deploy with Bokeh directly.
make deploy-dev to deploy with Flask.
make deploy-prod to deploy with Flask and Gunicorn as the WSGI.

Note, running make all deploys with Flask in development mode. If you don't have make installed, you can following these steps instead.

python3 -m venv venv
source venv/bin/activate && pip install -r requirements.txt
bin/deploy dev (with the virtual environment activated)

Deploying to Heroku

Heroku is a service to easily host web applications. Sign up and download the Heroku CLI tool. Make sure the application is ready to be deployed to Heroku by running the application locally. Follow the steps below to deploy to Heroku.

clone the repository.
heroku login and enter your credentials.
heroku create or heroku create <app-name> where <app-name> is a custom application name.
git push heroku master.
heroku open or open the application online through your Heroku profile.

License

This project is distributed under the MIT license. Please see LICENSE for more information.