mercury-py (Mercury for Python) is a Python based microservice that allow to manage scheduled notifications' sending.
The project consists of a Flask-based RESTful Web API, through which it is possible to manage the scheduling of notifications.
SQLite database is used internally.
MongoDB database is used to store scheduled notifications.
The dispatch of notifications is managed by Celery-based submodules:
- A Celery Beat that acts as an activator for the Workers.
- A Celery Worker for each category of notification, who takes care of managing the specific type of dispatch.
RabbitMQ message-broker is used for interprocess communications.
The categories of notifications that can currently be sent are:
- Email.
Categories will be implemented:
- SMS;
- Push notification - Firebase Cloud Messaging (FCM);
- Telegram message.
To download mercury, either fork this GitHub repo or simply download it from PyPI via pip:
pip install mercury-py
Mercury needs:
- MongoDB: get it from the official site (https://www.mongodb.com/) or from the official repository of your Linux distro, for example, on Debian-like systems:
sudo apt install mongodb
- RabbitMQ: get it from the official site (https://www.rabbitmq.com/) or from the official repository of your Linux distro, for example, on Debian-like systems:
sudo apt install rabbitmq-server
After that, to use Mercury, we need to create a RabbitMQ user, a virtual host and allow that user access to that virtual host:
From terminal, type (substitute in appropriate values for [MYUSER] and [MYPASSWORD] below):
sudo rabbitmqctl add_user [MYUSER] [MYPASSWORD]
sudo rabbitmqctl add_vhost mercury
sudo rabbitmqctl set_permissions -p mercury [MYUSER] ".*" ".*" ".*"
See the RabbitMQ Admin Guide for more information about access control (http://www.rabbitmq.com/admin-guide.html#access-control).
Copy and paste file config-*.py, from folder configs, into folder mercury-instance and modify it according to your needs.
SECRET_KEY should be changed to some random bytes in production.
You can use the following command to output a random secret key:
$ python -c 'import os; print(os.urandom(16))'
Then place returned value into your config.py.
You can also embed beat inside the worker by enabling the workers -B option, this is convenient if you’ll never run more than one worker node, but it’s not commonly used and for that reason isn’t recommended for production use:
celery -A mercury.app.celery worker -B --logfile=instance\logs\celery.log --pidfile=instance\celerybeat\celery.pid --schedule=instance\celerybeat\celery-schedule.db --loglevel=DEBUG
On Windows systems, an additional dependency must be installed in order for Celery-based submodules to function properly.
Download and install Gevent from PyPI via pip:
pip install gevent
And run (Celery Beat and Celery Workers) from terminal:
celery -A mercury.app.celery worker --logfile=instance/logs/celeryworker.log --loglevel=DEBUG --pool=gevent
celery -A mercury.app.celery beat --logfile=instance/logs/celerybeat.log --loglevel=DEBUG --pidfile=instance/celerybeat/celerybeat.pid --schedule=instance/celerybeat/celerybeat-schedule.db