Setup

• Database: Postgresql (could've used sqlite3 but to deploy it in Heroku it required pg gem)
• Test suite: Rspec
• Search: Ransack
• Pagination: Kaminari
• Authentication: JWT and bcrypt
• Serialization: Activemodel Serialziers (:json format)

To setup the database locally just add your configurations to a new .env file. Use the included .env.example in the project

Create your own keybase using:

EDITOR=vim rails credentials:edit

And save the file, this will create a master.key file. Otherwise the JWT encryption process will fail.

bundle install
rails db:setup

# or 

rails db:create
rails db:migrate
rails db:seed

Postman

Collection of Postman Requests and examples in: https://www.getpostman.com/collections/d33434f8851577711add

It contains examples of successfull and failures for each request.

Heroku temporal server

I uploaded a test server in Heroku, it can be found in: https://library-api-blue.herokuapp.com/

Check seeds for password and email. It can be tested with Postman, just change the {{service_url}} variable or string in the path to https://library-api-blue.herokuapp.com/

API

Authenticate through the /authenticate endpoint sending email and password:

# POST /authenticate

{
  "email": "[email protected]",
  "password": "password-in-seeds-rb-file"
}

Authentication when successfull will set the token automatically in the environment variables of postman, if not, just change it manually.

• enjoy :)

Endpoints

• POST /authenticate authentication

Books

• GET /books list of books
• GET /books/:id a book information, limited to book fields
• PATCH /books/:id update a book name
• DELETE /books/:id delete a book from database
• POST /books/:id create a book

Pages

• POST /books/:book_id/pages create a page for a book (page number is automatic incremental)
• GET /books/:book_id/pages/:page_number Fetches a page by the page number, not the ID of the page record

Content

• GET /pages/:page_id/contents?format=html Returns the content html for a page, NOTE: it uses the page.id not the page_number. This is something I decided because the path would be very very long otherwise. It would also require the book_id and at this point you already have the page id from the page show endpoint. It can be changed though to use page number. Since I don't have any specifics on the Front end approach I did what made more sense to me.
• POST /pages/:page_id/contents Creates a content for a page, it requires the format and the body. Currently it will only allow HTML.

Steps to load a page

• GET /books find the book that you want
• GET /books/:book_id/pages/:page_number load a page from the selected book and save the page id
• GET /pages/:page_id/contents?format=html Load the content for that page using the id

Resources

User

Used for authentication

Book

Base resource it has many pages

Page

Belongs to a book and it has many contents, the intention was to allow different sources. One source or Content for each format in a page, so you could upload a PDF, HTML, XML, etc.

NOTE: files formats are not implemented yet as those were not part of the challenge. But I left a TODO comment where it should be added.

Content

Source of truth for the content, it has body and file fields, you should use one or the other, not both. Each content should always have a ContentFormat so that it can be identified

ContentFormat

As a kind of list, one and only one should be created for each format, in the seeds.rb file you can find an example of the "HTML" ContentFormat.

Test suite

Used Rspec for test suite, just run: rspec spec/