PHE Coronavirus Resource Centre

---

• About this application
  • Purpose of this app
  • Tech stack
• Getting Started
  • Prerequisites
    • Without Docker
    • With Docker
  • Running the database
  • Environment variables
  • Running the site
    • Running without Docker
    • Running with Docker
      • Useful commands
      • Adding new dependencies
• Contributing
• Data Reporting
• License

About this application

Purpose of this app

The purpose of this site is to provide PHE with a performant platform to distribute their advice and guidance related to the Coronavirus outbreak.

Tech stack

This application is based on the Wagtail CMS. It utilises:

• Wagtail CMS
• Python/Django framework
• jQuery
• SASS

Getting Started

Prerequisites

Without Docker

You need to have Python (3.6.8), PIP and PostgreSQL installed. There are multiple ways to install Python, either download from the official Python site or use the package manager Homebrew brew install python3. PIP comes installed with Python 3.4(or greater) by default.

To install Postgres you can also use Homebrew

brew update
brew doctor
brew install postgresql

With Docker

You need to have Docker and PostgreSQL installed. Docker installation guidance can be found in the Docker Docs.

To install Postgres you can also use Homebrew

brew update
brew doctor
brew install postgresql

Running the database

To start Postgres:

pg_ctl -D /usr/local/var/postgres start && brew services start postgresql

Create a local database

CREATE DATABASE sampledb;
CREATE USER manager WITH PASSWORD 'supersecretpassword';
GRANT ALL PRIVILEGES ON DATABASE sample TO manager;

Environment variables

Variable	Default	Description
DBHOST	host.docker.internal	DB host url/string
DBPORT	5432	DB connection port
DBNAME		DB name to use
DBUSER		DB user
DBPASSWORD		DB password
AWS_ACCESS_KEY_ID	None	AWS account ID for S3 static file storage
AWS_SECRET_ACCESS_KEY	None	AWS account key for S3 static file storage
AWS_STORAGE_BUCKET_NAME	None	AWS S3 bucket for static file storage
LOCAL	False	Tells the site to use external API or mocks
AWS_ACCESS_KEY_ID_DEPLOYMENT	None	AWS account ID for site deployment to S3
AWS_SECRET_ACCESS_KEY_DEPLOYMENT	None	AWS account key for site deployment to S3
AWS_STORAGE_BUCKET_NAME_DEPLOYMENT	None	AWS S3 bucket to deploy site to
AWS_REGION_DEPLOYMENT	None	AWS region for site deployment
SCHEDULE_CRON_JOBS	False	(bool) Schedule management tasks with cron

Running the site

Running without Docker

Create a Python 3.6.8 virtual environment and run the following commands from the root directory of the project:

pip install -r requirements.txt
python manage.py migrate
python manage.py runserver

Running with Docker

docker-compose build
docker-compose up

The first command builds the docker image. The second command starts the docker container, running on port 8000.

Initial access

The migrations in the core module will create a superuser to allow initial access to the CMS.

Username - superuser
Password - adminpassword

This can be used for accessing the system during development, but should be removed when the code is deployed to Production.

Useful commands

There are commands setup in the bin directory of the project, that allow easy use of common commands inside the docker container.

These commands use the container name to access the container, this is based on the root directory name of the repository. If the directory name is changed these commands will need to be updated.

• container - takes you on to the containers command line.
• coverage - runs the unit tests with coverage reporting.
• lint - runs the code format linter against the code.
• manage - can be passed arguments to run standard django manage.py commands.
• shell - takes you on to the python shell command line for the project.
• test - runs the unit tests.

Scheduling Django/Wagtail management commands with Cron

Management tasks - anything that can be run as an argument for manage.py (e.g. 'publish_scheduled_pages') - can also be scheduled to run at regular intervals using cron. Jobs can be configured in the cron_jobs array specified in `management_cron_jobs.yml', found in the project root. The syntax is as follows (this info is also in the .yml file itself):

# define cron jobs below. Fields are:
#   - name (for logging)
#   - mgmt_command (the command or commands to run). NB, these are assumed to be
#     django or wagtail management tasks ONLY, rather than loose shell statements.
#     see CMS.settings.management_cron_jobs.py for more details.
#   - period_units (either second, minute, hour or day. NB will accept either
#     singular (minute) or plural (minutes)).
#   - period (how often, in period_units, to run)
#   - logging: True or False to write logs to the logfile set above.
# e.g run publish_scheduled_pages every 30 minutes and log the output:

cron_jobs:
  - name: "Publish Scheduled Pages"
    mgmt_command: "publish_scheduled_pages"
    period_units: "minutes"
    period: 30
    logging: True

By default, the user for cron jobs will be the whatever user runs the server (root, presumeably). This can be overridden by changing the cron_user value from current_user to a specific user.

If logging: True is set in a cron_job spec, then the job will log timestamped stdout and stderr to whatever path is set in logfile, by default /var/log/cron.log.

NB: IN ORDER FOR CRON JOBS TO BE SCHEDULED, THE ENVIRONMENTAL VARIABLE 'SCHEDULE_CRON_JOBS' MUST BE SET TO 'True'.

Adding new dependencies

Adding a new dependency requires rebuilding docker image for Django app.

Stop docker container

docker-compose down

Rebuild Docker image

docker-compose build

Start server again

docker-compose up

Contributing

See CONTRIBUTING for details.

Producing Usage Reports

See Data Reporting for details.

License

See LICENSE for details.