Day-28 : Model 多对多

今天我们要来讲多对多
什麽是多对多
举个例子：

商店has_many商品，商品也has_many商店，然後有一个第三方的资料表来存放两边Model的资讯。

接下来，奉上图解！

那我们来开始做多对多拉
1.第三方Model仅需存放两边（商店和商品）Model的id，若想要额外增加一些栏位也是可行的

这边会用到references的写法是因为会帮我们多做几件事情

1.自动加上索引，加快查询的速度
2.自动帮Model加上belongs_to

2.执行rails db:migrate

3.让我们来编辑planet这个档案

上面有说到，因为我们使用了references的写法，所以Ruby on Rails 会自动帮我们补上这两行

 belongs_to :store
 belongs_to :product

回到Store Model，加上下面这两行

5.Product Model加上这两行

Planet Model同时belongs_to Store 以及 Product这两个Model

Store 与 Product这两个Model也都has_many Planet

关系图解如下：

Store跟Product都是透过through Planet的纪录，也就是第三方，才会知道商店与商品之间的关系

接下来，进入 rails c看看
1.先取任意两间商店＝> store1和 store2

> store1 = Store.find(1)
  Store Load (0.3ms)  SELECT "stores".* FROM "stores" WHERE "stores"."id" = ? LIMIT ?  [["id", 1], ["LIMIT", 1]]
 => #<Store id: 1, title: "良心商店", tel: nil, address: nil, user_id: 1, created_at: "2021-10-10 09:...
> store2 = Store.find(2)
  Store Load (0.3ms)  SELECT "stores".* FROM "stores" WHERE "stores"."id" = ? LIMIT ?  [["id", 2], ["LIMIT", 1]]
 => #<Store id: 2, title: "三眼怪行星", tel: nil, address: nil, user_id: 2, created_at: "2021-10-10 0...

2.随意取出2件商品＝> product1和 product2

> product1 = Product.find(1)
  Product Load (0.1ms)  SELECT "products".* FROM "products" WHERE "products"."id" = ? LIMIT ?  [["id", 1], ["LIMIT", 1]]
 => #<Product id: 1, name: "吃吃喝喝商店", description: nil, price: 0.2e3, is_available: nil, store_i...
> product2 = Product.find(2)
  Product Load (0.3ms)  SELECT "products".* FROM "products" WHERE "products"."id" = ? LIMIT ?  [["id", 2], ["LIMIT", 1]]
 => #<Product id: 2, name: "我想要飞", description: nil, price: 0.1e4, is_available: nil, store_id: 1...

3.把product1, product2丢给store2,把product1, product2丢给store1

> store2.products = [product1, product2]
  Product Load (0.1ms)  SELECT "products".* FROM "products" INNER JOIN "planets" ON "products"."id" = "planets"."product_id" WHERE "planets"."store_id" = ?  [["store_id", 2]]
  TRANSACTION (0.0ms)  begin transaction
  Planet Create (0.5ms)  INSERT INTO "planets" ("store_id", "product_id", "created_at", "updated_at") VALUES (?, ?, ?, ?)  [["store_id", 2], ["product_id", 1], ["created_at", "2021-10-12 15:24:01.126751"], ["updated_at", "2021-10-12 15:24:01.126751"]]
  Planet Create (0.1ms)  INSERT INTO "planets" ("store_id", "product_id", "created_at", "updated_at") VALUES (?, ?, ?, ?)  [["store_id", 2], ["product_id", 2], ["created_at", "2021-10-12 15:24:01.128415"], ["updated_at", "2021-10-12 15:24:01.128415"]]
  TRANSACTION (0.5ms)  commit transaction
 => [#<Product id: 1, name: "吃吃喝喝商店", description: nil, price: 0.2e3, is_available: nil, store_...
 > store1.products = [product1, product2]
  Product Load (0.4ms)  SELECT "products".* FROM "products" INNER JOIN "planets" ON "products"."id" = "planets"."product_id" WHERE "planets"."store_id" = ?  [["store_id", 1]]
  TRANSACTION (0.1ms)  begin transaction
  Planet Create (0.5ms)  INSERT INTO "planets" ("store_id", "product_id", "created_at", "updated_at") VALUES (?, ?, ?, ?)  [["store_id", 1], ["product_id", 1], ["created_at", "2021-10-12 15:24:12.087711"], ["updated_at", "2021-10-12 15:24:12.087711"]]
  Planet Create (0.1ms)  INSERT INTO "planets" ("store_id", "product_id", "created_at", "updated_at") VALUES (?, ?, ?, ?)  [["store_id", 1], ["product_id", 2], ["created_at", "2021-10-12 15:24:12.089616"], ["updated_at", "2021-10-12 15:24:12.089616"]]
  TRANSACTION (1.0ms)  commit transaction
 => [#<Product id: 1, name: "吃吃喝喝商店", description: nil, price: 0.2e3, is_available: nil, store_...

查询看看，有几间商店在卖product1:

> product1.stores.count
   (0.5ms)  SELECT COUNT(*) FROM "stores" INNER JOIN "planets" ON "stores"."id" = "planets"."store_id" WHERE "planets"."product_id" = ?  [["product_id", 1]]
 => 2
> product2.stores.count
   (0.5ms)  SELECT COUNT(*) FROM "stores" INNER JOIN "planets" ON "stores"."id" = "planets"."store_id" WHERE "planets"."product_id" = ?  [["product_id", 2]]
 => 2

因为在store1和store2都有贩售product1与product2，因此为两家，我们看一下上面的SQL语法，我们就可以得知，他不再对Store与Product要资料，而是与第三方planet进行查询的动作！

与前面的差别在於，