如果 :a has_many :b 和 :b has_many :c，我们可以通过 :a 使用 JSON / Rails 渲染 :c

Question

有3个嵌套模型（见底部的模型文件）

• :用户
• ：测验，
• :问题

是否可以通过user 渲染questions？

可以通过user 渲染quizzes。但是我不知道为什么每个测验的问题也不包括在内...

1. {{user}} 呈现该用户的测验

{"id":1,"email":"example@example.com","username":"jim","quizzes":[{"id":1,"name":"quiz1", "user_id":1},{"id":2,"name":"quiz2","user_id":1},]}

1. 但{{user.quizzes}} 只呈现测验，而不是那些测验的问题。

[{"id":1,"name":"quiz1","user_id":1},{"id":2,"name":"quiz2","user_id":1}]

1. 另外，{{user.questions}} 将呈现一个空白数组 []。

所以users 和questions 之间缺少连接，或者它只能渲染低一级的对象？

user.rb

class User < ActiveRecord::Base
  has_many :questions
  has_many :quizzes 

  def as_json(options = {})
    super(options.merge(include: [:quizzes, :questions]))
  end
end

quiz.rb

class Quiz < ActiveRecord::Base
  belongs_to :user
  has_many :questions

    def as_json(options = {})
        super(options.merge(include: [:user, :questions]))
    end    
end

question.rb

class Question < ActiveRecord::Base
    belongs_to :user    
    belongs_to :quiz

    def as_json(options = {})
        super(options.merge(include: [:user, :quiz]))
    end
end

请记住，我对数据库的经验并不丰富。任何建议将不胜感激。

==============

更新

==============

下面添加了架构。

ActiveRecord::Schema.define do

  create_table "questions", force: :cascade do |t|
    t.string   "ask"
    t.string   "answer"
    t.datetime "created_at", null: false
    t.datetime "updated_at", null: false
    t.integer  "user_id"
    t.integer  "quiz_id"
    t.integer  "easy_count"
    t.integer  "diff_count"
  end

  add_index "questions", ["quiz_id"], name: "index_questions_on_quiz_id"
  add_index "questions", ["user_id"], name: "index_questions_on_user_id"

  create_table "quizzes", force: :cascade do |t|
    t.string   "name"
    t.integer  "user_id"
    t.datetime "created_at", null: false
    t.datetime "updated_at", null: false
  end

  add_index "quizzes", ["user_id"], name: "index_quizzes_on_user_id"

  create_table "users", force: :cascade do |t|
    t.string   "email",                  default: "", null: false
    t.string   "encrypted_password",     default: "", null: false
    t.string   "reset_password_token"
    t.datetime "reset_password_sent_at"
    t.datetime "remember_created_at"
    t.integer  "sign_in_count",          default: 0,  null: false
    t.datetime "current_sign_in_at"
    t.datetime "last_sign_in_at"
    t.string   "current_sign_in_ip"
    t.string   "last_sign_in_ip"
    t.datetime "created_at"
    t.datetime "updated_at"
    t.string   "username"
  end

  add_index "users", ["email"], name: "index_users_on_email", unique: true
  add_index "users", ["reset_password_token"], name: "index_users_on_reset_password_token", unique: true
  add_index "users", ["username"], name: "index_users_on_username", unique: true

end

Answer 1

虽然我不是 100% 确定，但我认为您没有正确创建关系。我猜user.questions 给了你一个空白数组，因为在你的数据库中，questions 只与quiz 相关。如果可以的话，你能提供你的数据库模式吗？

==================

更新

我完全按照你的模式创建了一个数据库，下面是我运行User.first.as_json后得到的示例结果

{"id"=>1, "email"=>"", "encrypted_password"=>"", "reset_password_token"=>nil, "reset_password_sent_at"=>nil, "remember_created_at"=>nil, "sign_in_count"=>0, "current_sign_in_at"=>nil, "last_sign_in_at"=>nil, "current_sign_in_ip"=>nil, "last_sign_in_ip"=>nil, "created_at"=>Sat, 02 Apr 2016 05:42:33 UTC +00:00, "updated_at"=>Sat, 02 Apr 2016 05:42:33 UTC +00:00, "username"=>nil, "quizzes"=>[{"id"=>1, "name"=>nil, "user_id"=>1, "created_at"=>Sat, 02 Apr 2016 05:42:42 UTC +00:00, "updated_at"=>Sat, 02 Apr 2016 05:42:42 UTC +00:00}], "questions"=>[{"id"=>1, "ask"=>nil, "answer"=>nil, "created_at"=>Sat, 02 Apr 2016 05:43:11 UTC +00:00, "updated_at"=>Sat, 02 Apr 2016 05:43:11 UTC +00:00, "user_id"=>1, "quiz_id"=>1, "easy_count"=>nil, "diff_count"=>nil}]}

更新 2

一个快速修复将是

class User < ActiveRecord::Base
  has_many :questions
  has_many :quizzes

  def as_json(options = {})
    super(options.merge(include: [quizzes: {include: :questions}]))
  end
end

但再次如您所见，使用 include 并覆盖 as_json 非常难看。我建议你最终切换到 Jbuilder 或active_model_serializers

Answer 2

在您的情况下，您需要使用 has_many through，因为您的问题与测验相关联，而不是与用户相关联。 实际上，您所有的麻烦都来自您将 user_id 用于提问和测验的事实。 假设您的问题属于属于用户的测验，您应该只将 user_id 放在测验中。

没有数据重复，您可以确保问题不属于与问题不属于同一用户的测验（！）

并使用委托方法访问用户ID。

这里有一个改进：

class User < ActiveRecord::Base
  has_many :quizzes 
  has_many :questions, through: quizzes

  def as_json(options = {})
    super(options.merge(include: [:quizzes, :questions]))
  end
end

class Quizz < ActiveRecord::Base
  has_many :questions
  belongs_to :user
end

class Question < ActiveRecord::Base
  belongs_to :quizz
  delegate :user, to: :quizz
end

Answer 3

我建议使用更强大的解决方案，例如 ActiveModelSerializers。

class UserSerializer < ActiveModel::Serializer
  attributes :id, :email
  has_many :quizzes, serializer: QuizSerializer
end

class QuizSerializer < ActiveModel::Serializer
  attributes :name
  has_many :questions, serializer: QuestionSerializer
end

class QuestionSerializer < ActiveModel::Serializer
  attributes :ask, :answer
end

class UserController < ApplicationController
  def show
    render json: user, serializer: UserSerializer
  end
end

Answer 4

正如 Yacine 所指出的，has_many :through 可能是一个更好的解决方案，因此您在 quizzes 和 questions 表中都没有重复的 user_id。

如果您确实进行了更改，您还可以从questions 表中删除user_id 列。