Ruby Interview Task

Download this repository. The data directory contains 2 JSON files, articles.json and users.json.

The following Ruby should help you parse the files:

require 'json'
USERS = JSON.parse(File.read('./data/users.json'), symbolize_names: true)
ARTICLES = JSON.parse(File.read('./data/articles.json'), symbolize_names: true)

This will create 2 global constant arrays to use in your code.

ARTICLES is an array of hashes containing:

id - a unique integer
concepts - an array of short tag-like strings, which are the concepts that the article is about. The same concept can appear in many different articles.

USERS is an array of hashes containing:

id - a unique integer
read_article_ids - an array of the ids of articles that the user has read

(Feel free to write any other code or global constants that you think might help or optimise your implementation.)

Can you do the following?

write classes Article and User
- objects of both classes should have an id method
- User objects should have a read_article_ids method
- Article objects should have a concepts method
write class methods Article.all and User.all that will produce the expected arrays of the corresponding objects
write methods Article.find(id:) and User.find(id:) that will quickly instantiate the indicated element
add a User.read_articles method that will return Article objects that the user has read

Further work

Following this general line of development, we would like to see how close you can get to these goals:

What are a user’s favourite concepts?
Can you work out a way to recommend an article to the user, that
- the user hasn’t read
- most closely matches the user’s interests

Submission

When you are finished, add your modified files to a public repository, either on Github or Bitbucket

Notes:

Obviously, concepts which come up most often in a user’s read articles will be concepts the user is most interested in.

Each of an article's concepts could be of equal weight – and code which treats them this way is fine.

But if you want to get fancy, consider the possibility that the concepts most relevant to an article appear closest to the beginning of the article's concepts array

Additional Support