MySQL查询返回错误的计数值

这里我分享了我的疑问。所有的总值都给出了错误的结果，即非常大的数值，而不是期望的计数值。我的问题出了什么问题

           SELECT bd.business_id,
                  bd.business_name,
                  bd.business_address,
                  bd.created_at,
                  bd.updated_at AS extent_expiry_date,
                  bd.manager_id as user_id,
                  COUNT(m.manager_id) AS total_managers,
                  COUNT(a.agent_id) AS total_agents,
                  COUNT(c.customer_id) AS total_customers,
                  COUNT(t.task_id) AS total_tasks,
                  COUNT(t.order_id) AS total_orders
            FROM business_details bd
            LEFT JOIN managers m ON m.business_id = bd.business_id
            LEFT JOIN agents a ON a.business_id = bd.business_id
            LEFT JOIN customers c ON c.business_id = bd.business_id
            LEFT JOIN tasks t ON t.business_id = bd.business_id
            GROUP BY bd.business_id

---

由于您的问题中没有样本数据集，我猜您得到的是相同的经理id、代理id。。。多次由于左连接而获得正确的计数使用distinct

SELECT bd.business_id,
      bd.business_name,
      bd.business_address,
      bd.created_at,
      bd.updated_at AS extent_expiry_date,
      bd.manager_id as user_id,
      COUNT(distinct m.manager_id) AS total_managers,
      COUNT(distinct a.agent_id) AS total_agents,
      COUNT(distinct c.customer_id) AS total_customers,
      COUNT(distinct t.task_id) AS total_tasks,
      COUNT(distinct t.order_id) AS total_orders
FROM business_details bd
LEFT JOIN managers m ON m.business_id = bd.business_id
LEFT JOIN agents a ON a.business_id = bd.business_id
LEFT JOIN customers c ON c.business_id = bd.business_id
LEFT JOIN tasks t ON t.business_id = bd.business_id
GROUP BY bd.business_id

根据新版本的规定，还应将所有非聚合列包含在group by中，并且大多数RDBMS拒绝此类查询

GROUP BY 
bd.business_id,
bd.business_name,
bd.business_address,
bd.created_at,
bd.updated_at,
bd.manager_id

---

由于您的问题中没有样本数据集，我猜您得到的是相同的经理id、代理id。。。多次由于左连接而获得正确的计数使用distinct

SELECT bd.business_id,
      bd.business_name,
      bd.business_address,
      bd.created_at,
      bd.updated_at AS extent_expiry_date,
      bd.manager_id as user_id,
      COUNT(distinct m.manager_id) AS total_managers,
      COUNT(distinct a.agent_id) AS total_agents,
      COUNT(distinct c.customer_id) AS total_customers,
      COUNT(distinct t.task_id) AS total_tasks,
      COUNT(distinct t.order_id) AS total_orders
FROM business_details bd
LEFT JOIN managers m ON m.business_id = bd.business_id
LEFT JOIN agents a ON a.business_id = bd.business_id
LEFT JOIN customers c ON c.business_id = bd.business_id
LEFT JOIN tasks t ON t.business_id = bd.business_id
GROUP BY bd.business_id

根据新版本的规定，还应将所有非聚合列包含在group by中，并且大多数RDBMS拒绝此类查询

GROUP BY 
bd.business_id,
bd.business_name,
bd.business_address,
bd.created_at,
bd.updated_at,
bd.manager_id

---

我假设您的所有表在

business\u id

上都不是唯一的？如果没有，那么如果一个企业有3个经理和3个代理人，你会得到9行。。。您是否查看了单个业务的原始数据以确定是否选择了所需的数据？我假设您的所有表在

业务id

上都不是唯一的？如果没有，那么如果一个企业有3个经理和3个代理人，你会得到9行。。。您是否查看了单个业务的原始数据，以确定是否选择了所需的数据？