This is a very simple simulation of a Recommendation system - matching a User to Products.

ProductApplication.py models a product service and returns a list of Products.

UserApplication.py models a user service and provides a function to get a User.

Both Users and Products inherit the Entity.AttributedEntity. They merely contain a set of "tags".

Tags.py contains a list of tag-types (eg. "formal", "retro", "blue", etc.) and provides a function to return a random set of tags with random values.

Strategies.py contains functions to calculate the Euclidian distance, Cosine similarity, and Hamming distance between two objects that have tags.

RecommendationCalculator.py provides functions to get "recommendations" based on the function in Strategies.py.

Workers.py provides handler functions to handle events.

[simluate.py](simluate.py0 is basically a test program which dumps users to a simple memory based PubSub (https://pypi.org/project/PyPubSub/).

Listeners are implemented using the 3 strategies from Strategies.py and results are just dumped to standard out.