Rails has_many,habtm SQL join没有返回任何结果

时间:2013-09-17 16:54:31

标签: sql ruby-on-rails ruby-on-rails-3 scope associations

我正在尝试购买至少购买两件商品的顾客。关联非常简单,但我在测试中没有得到任何结果。也许另一组眼睛可以看到我所缺少的东西。

# customer.rb
class Customer < Person
  has_many :orders, foreign_key: 'person_id'
  has_many :items, through: :orders


  # This is what I'm needing help with:
  def self.min_2_orders
    joins(:items).group('items_orders.item_id').having('COUNT(items_orders.item_id) >= ?', 2)
  end
end

-

# order.rb
class Order < ActiveRecord::Base
  belongs_to :customer, foreign_key: 'person_id'
  has_and_belongs_to_many :items
end

-

# item.rb
class Item < ActiveRecord::Base
    has_and_belongs_to_many :orders
end

-

# customer_spec.rb
it "finds customers who have ordered a min of 2 items" do
  @order1, @order2  = FactoryGirl.create_list(:order, 2) #returns 2 created orders

  @order2.items << FactoryGirl.create_list(:item, 2) #returns 2 created items

  @customer1  = @order2.customer
  @customer2  = @order1.customer

  ## Debug tests
    Order.count.should            == 2 #pass
    Customer.count.should         == 2 #pass
    Item.count.should             == 2 #pass
    @order_min.items.count.should == 2 #pass
    @customer1.items.count.should == 2 #pass

    puts Order.all.to_yaml    # debug. see below
    puts Customer.all.to_yaml # debug. see below
    puts Item.all.to_yaml     # debug. see below
  # End debugging

  # The final test
  Customer.min_2_orders.should == [@customer1]  # FAIL. returns []
end

以下是调试日志(sql查询和YAML db数据):

"SELECT 'people'.* FROM 'people'
  INNER JOIN 'orders' ON 'orders'.'person_id' = 'people'.'id'
  INNER JOIN 'items_orders' ON 'items_orders'.'order_id' = 'orders'.'id'
  INNER JOIN 'items' ON 'items'.'id' = 'items_orders'.'item_id'
WHERE 'people'.'type' IN ('Customer')
  GROUP BY items_orders.item_id
  HAVING COUNT(items_orders.item_id) >= 2"

---
- !ruby/object:Order
  attributes:
    id: 1
    person_id: 1
    created_at: 2013-06-18 16:42:17.558683000 Z
    updated_at: 2013-09-17 16:42:17.597082000 Z
- !ruby/object:Order
  attributes:
    id: 2
    person_id: 2
    created_at: 2013-09-17 16:42:17.600267000 Z
    updated_at: 2013-09-17 16:42:17.600267000 Z
---
- !ruby/object:Customer
  attributes:
    id: 1
    first_name: Wes
    type: Customer
    created_at: 2013-09-17 16:42:17.565677000 Z
    updated_at: 2013-09-17 16:42:17.565677000 Z
- !ruby/object:Customer
  attributes:
    id: 2
    first_name: Wes
    type: Customer
    created_at: 2013-09-17 16:42:17.599013000 Z
    updated_at: 2013-09-17 16:42:17.599013000 Z
---
- !ruby/object:Item
  attributes:
    id: 1
    name: Product 1
    created_at: 2013-09-17 16:42:17.632347000 Z
    updated_at: 2013-09-17 16:42:17.632347000 Z
- !ruby/object:Item
  attributes:
    id: 2
    name: Product 2
    created_at: 2013-09-17 16:42:17.633920000 Z
    updated_at: 2013-09-17 16:42:17.633920000 Z

1 个答案:

答案 0 :(得分:1)

由于您按item_id进行分组,因此每item_id只能获得一条记录,因此您永远不会有count(item_id)大于{1}的记录。您应该按people.id进行分组。

在相关点上,MySQL的独特之处在于:

  

MySQL扩展了GROUP BY的使用,以便选择列表可以引用   未在GROUP BY子句中命名的非聚合列

http://dev.mysql.com/doc/refman/5.1/en/group-by-extensions.html

中所述