Home

Awesome

Chicago Justice Project backend app

Local installation instructions

Postgres installation

macOS

The easiest way to install PostgreSQL for Mac is with a prebuilt Postgres installation, like Postgres.app.

Alternatively, you may use Homebrew:

brew install postgres
brew services start postgresql

GNU/Linux

The version of PostgreSQL provided in most distros' repositories should be adequate and can be installed through your distro's package manager.

Ubuntu 16.04:

sudo apt-get update
sudo apt-get install postgresql

Arch Linux:

sudo pacman -S postgresql
sudo -u postgres initdb --locale $LANG -E UTF8 -D '/var/lib/postgres/data'
sudo systemctl start postgresql.service

Postgres setup

Once PostgreSQL is installed and running, you can create the database you'll use locally for this app.

As a user with Postgres database privileges:

createdb cjpdb

The name of the database (e.g., cjpdb) may be anything you choose, but keep track of what you name it along with the user and password we're about to create. You'll need these for setting up your virtual environment.

Create the Postgres user and give it a password:

createuser --interactive --pwprompt

Finally, grant privileges on the database you just created to the user you just created. For instance, if we created database cjpdb and the user cjpuser:

psql -d postgres -c "GRANT ALL ON DATABASE cjpdb TO cjpuser;"

Setup Environment Variables

Certain settings are read from environment variables. There are two ways you can set variables: 1) Use a .env file in the root directory; 2) setup a python virtual environment and use virtualenv's postactivate and predeactivate hooks. Both methods are detailed below.

Create a .env environment variable file

An example .env file is provided. You should copy it:

cp .env-example .env

Then, you can edit the file in your preferred editor.

Create a python virtual environment

Alternatively, you can create a virtual environment to house the environment variables and the app's dependencies.

If not already installed, install python's virtualenv and virtualenvwrapper:

pip install virtualenv virtualenvwrapper
mkdir ~/.virtualenvs

Add the following to your .bashrc file:

export WORKON_HOME=~/.virtualenvs
source /usr/local/bin/virtualenvwrapper.sh

Find out the path to your python installation:

which python

Create your working environment, naming it whatever you'd like (e.g., cjp_dev), where usr/local/bin/python is whatever path the previous command returned:

mkvirtualenv --python=/usr/local/bin/python cjp_dev

You may now use workon cjp_dev and deactivate to activate and deactivate the virtual environment. Setup hooks so that when the virtual environment is activated, the proper environment variables will be set. Be sure to substitute cjp_dev, cjpdb, cjpuser, and cjppassword with your setup. You can also generate a unique secret key with something like this Django Secret Key Generator

Add the following to ~/.virtualenvs/cjp_dev/bin/postactivate:

export DJANGO_SETTINGS_MODULE="cjp.settings.local"
export DATABASE_NAME="cjpdb"
export DATABASE_USER="cjpuser"
export DATABASE_PASSWORD="cjppassword"
export SECRET_KEY='#&ubnzmo6$-0nk7i&hmii=e$7y-)nv+bm#&ps)6eq@!k+n-nq5'

To make sure these variables are unset upon deactivating the virtual environment, add the following to ~/.virtualenvs/cjp_dev/bin/predeactivate:

unset DJANGO_SETTINGS_MODULE
unset DATABASE_NAME
unset DATABASE_USER
unset DATABASE_PASSWORD
unset SECRET_KEY

Install Dependencies

With the environment variables set, we're now ready to install the necessary dependencies:

pip install -r requirements.txt

Initialize Django models and start server

./manage.py migrate
./manage.py loaddata category news_source
./manage.py runserver

Running news scrapers

./manage.py runscrapers

To run a single scraper, enter the scraper name as an argument, e.g.:

./manage.py runscrapers crains

Deployment

CLI setup

The app runs on AWS Elastic Beanstalk. In order to manage the production app, a project maintainer must grant you an AWS login and access key.

The Elastic Beanstalk CLI is separate from the main AWS CLI. Install it as described in the docs.

The most reliable way to configure your credentials is to set the key ID and secret as environment variables. If you use a different AWS account normally, you can create a file that sets the envvars to the CJP account, and only source the file when working on the project.

Create a file cjp-aws.env with the following lines, or add them to your shell configuration. Make sure that these values don't get checked into version control!

export AWS_ACCESS_KEY_ID=XXXXXXXXXXXXXX
export AWS_SECRET_ACCESS_KEY=XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX

If you create a standalone file, you can enable the CJP credentials in your current terminal session with source cjp-aws.env.

Test that you have the CLI configured correctly by running the following from the chicago-justice project directory:

eb status

Deploying

To deploy to production, run eb deploy from the project directory. It will deploy whatever is on your local filesystem, even if it isn't checked into git. To maintain consistency between production and git, it's recommended to merge changes to master and then git checkout master && git pull before deploying.

Elastic Beanstalk will run any database migrations as part of the deployment. You can check on the status of the deployment with eb status, or eb logs for the most recent logs from various important logfiles.

Environment variables can also be configured with the CLI or from the AWS web interface.