Custom Content Management System (CMS) built for New America

Contents

• First time setup
• Dependencies

Setup

Dependencies

This software should be installed on your computer:

1. Docker

First-time setup script

From inside the repo, run the setup script, which automates several steps needed to configure your environment so that the Docker containers can build successfully.

./setup.sh

Starting the environment

This command will start the web, database, elastic search, and redis servers needed for the site to operate:

docker compose up

The containers can also be run in detached mode with docker compose up -d. In the default mode, you see all the logs and output from the containers in the shell and will stop running with Control+C. Running with -d will cause docker compose to exit, but the containers will run in the background. They can be stopped with docker compose down.

The site will not immediately work without existing page data. You can download this from the staging or production environments using Fabric commands. Many of these commands require you be logged into Heroku, which you can do with this command:

docker compose exec web heroku login

Once the login is complete, you can use Fabric on the web container to obtain data from various sources. For example:

docker compose exec web fab pull-production-data

Other Fabric commands can be listed with

docker compose exec web fab -l

Visiting the site locally

After you have pull down data, you should be able to visit the Wagtail site at http://localhost:8000/

Updating dependencies

The "docker build" command below will create new requirements.txt (for production -- Heroku requires this file to be in the root of the project folder hierarchy) and requirements/local-dev.txt (for development) and requirements/ci.txt (for use with continuous integration services) from the information in pyproject.toml.

docker build --target=requirements-artifacts -f ./docker/Dockerfile --output type=local,dest=. .

To change a requirement, update the version boundaries in the project.dependencies or appropriate key in project.optional-dependencies in pyproject.toml and re-run the docker build command.