A web app to check whether followers of a given Twitter account are bots using Botometer. This repository started as a fork of Twitter Clean-up.

If you're looking for the CLI version, it's tagged.

• Twitter API keys
• Botometer API key
• Docker Compose for the development mode instructions
• Dokku for the production mode (deploy) instructions

Copy .env.sample as .env and fill the environment values as appropriated. I tried to use meaningful variables names, but fell free to ask if anything is not clear.

To start the services use the default docker-compose up.

Having a Dokku-ready server, install the following Dokku plug-ins:

• PostgreSQL (or other database of your choice)
• RabbitMQ (or other Celery broker of your choice)
• Let's Encrypt

Create an app for Bot Followers in Dokku, add it as a remote repository in your local Git repository, and activate the plug-ins.

For each variable in .env.sample, create an equivalent environment variable for your Dokku app.

The commands in this section might be prefixed by:

• docker-compose run --rm django in development mode
• dokku run <app name> in production mode

In order to get the app ready, before visiting / at your server, you need to run migrations and create a user(s) to access the dashboard.

1. Run migrations with python manage.py migrate
2. Create a super user for yourself with python manage.py createsuperuser

It's recommended to create a proper user to access the dashboard without super powers: all you need to do is to:

1. Login in as superuser create a new user that is staff
2. Get back to the edit page of this user
3. In the Permissions menu, add only the permission to view report to this new user

There is a simple JSON API at /api/ to share the report data without the need of user or login.

$ python manage.py import /path/to/borsalino.sqlite3

$ python manage.py updatecelerytasks

$ python manage.py purgecelerytasks

Please, write and run tests, and format your code with Black:

$ black .
$ python manage.py test