How to define UserBooksResource for grouped JSON output by status?

[reckyy]

Hello,
I'm having trouble defining the UserBooksResource and would appreciate any advice you could offer.

Background

Only the parts relevant to this question are included here.
To show the relationship between models, I've prepared the following ER diagram:

erDiagram
	users  ||--o{ user_books        : ""
	books  ||--o{ user_books        : ""
  
	users {
		int id PK
	}
  
	books {
		int id PK
		string title
		string author
		string cover_image_url
	}
  
	user_books {
		int id PK
		int user_id FK
		int book_id FK
                integer status
	}

Loading

For managing the status in the UserBook model, I plan to use an enum.

# app/models/user_book.rb

class UserBook < ApplicationRecord
  enum :status, { unread: 0, reading: 1, finished: 2 }, prefix: true
end

Here’s the setup I plan to use for UserBookResource and BookResource:

class UserBookResource < BaseResource
  attributes :id, :status

  one :book, resource: BookResource do
    attributes :id, :author, :title
  end
end

class BookResource < BaseResource
  attributes :id, :title, :author

  attribute :coverImageUrl, &:cover_image_url
end

Question

I want to return JSON in the format below, with each status grouped into its own array. I’m having trouble achieving this structure, so any guidance or examples of how to implement this would be greatly appreciated.
If any part of my question is unclear, or if you need additional information, please let me know. Thanks!

{
  unread_books: [{
    id: 1,
    status: 'unread',
    book: {
      id: 1,
      title: 'hogehoge',
      author: 'hogehoge',
    }
  },
  { 
    id: 2,
    status: 'unread',
    book: {
      id: 2,
      title: 'hogehoge',
      author: 'hogehoge',
    }
  }],
  reading_books: [{
    same structure as above
  }],
  finished_books: [{
    same structure as above
  }]
}

[okuramasafumi]

https://github.com/okuramasafumi/alba?tab=readme-ov-file#inline-definition-with-albaserialize This might help.

[reckyy]

@okuramasafumi
Sorry for the delayed response, and thank you for the link and your suggestion. I looked into the example provided in the documentation, but I struggled to understand how to apply it to my use case. I ended up implementing the feature as follows:

# UnreadUserBooksResource and FinishedUserBooksResource are almost identical.
# Only the class name and the enum status are different.
# app/resources/reading_user_books_resource.rb
class ReadingUserBooksResource < BaseResource
  many :user_books,
       proc { |user_books|
         user_books.order(:position).select(&:status_reading?)
       },
       resource: UserBookResource
end

#app/resources/user_books_resource.rb
class UserBooksResource < BaseResource
  attribute :unread do
    UnreadUserBooksResource.new(object).serializable_hash['user_books']
  end

  attribute :reading do
    ReadingUserBooksResource.new(object).serializable_hash['user_books']
  end

  attribute :finished do
    FinishedUserBooksResource.new(object).serializable_hash['user_books']
  end
end

While this approach worked for my requirements, I’m curious if there’s a better or more idiomatic way to implement this using Alba. Could you provide some insights or an example based on the approach you suggested earlier? I’d like to learn more about how to use Alba effectively for scenarios like this. Thank you!

[okuramasafumi]

@reckyy OK, how about this?

require 'alba'

Book = Data.define(:id, :title, :author, :status)

class BookResource
  include Alba::Resource

  attributes :id, :title, :author, :status
end

book1 = Book.new(1, 'Ruby Book', 'Matz', 'finished')
book2 = Book.new(2, 'Rails Book', 'DHH', 'reading')
book3 = Book.new(3, 'Elixir Book', 'Jose Valim', 'unread')
book4 = Book.new(4, 'Phoenix Book', 'Chris McCord', 'reading')

BookCollection = Data.define(:unread_books, :reading_books, :finished_books)
book_collection = BookCollection.new([book3], [book2, book4], [book1])

class BookCollectionResource
  include Alba::Resource

  many :unread_books, resource: BookResource
  many :reading_books, resource: BookResource
  many :finished_books, resource: BookResource
end

pp BookCollectionResource.new(book_collection).to_json
# => "{\"unread_books\":[{\"id\":3,\"title\":\"Elixir Book\",\"author\":\"Jose Valim\",\"status\":\"unread\"}],\"reading_books\":[{\"id\":2,\"title\":\"Rails Book\",\"author\":\"DHH\",\"status\":\"reading\"},{\"id\":4,\"title\":\"Phoenix Book\",\"author\":\"Chris McCord\",\"status\":\"reading\"}],\"finished_books\":[{\"id\":1,\"title\":\"Ruby Book\",\"author\":\"Matz\",\"status\":\"finished\"}]}"

[reckyy]

Wow, I hadn’t thought of creating data for that purpose!
I’ll give it a try right away.
Thank you so much!