多个内部连接 ​​- MySQL的

Question

我有2个表，Task和Transaction，它看起来像这样：

任务

交易

我想创建一个从两个表中返回信息的查询。

色谱柱：name, priority, waiting, error, done, total, status

其中：

• 名称= Task.name
• 优先= Task.priority
• 等待=计数（transaction.id）WHERE task.id =事务。任务和transaction.status = 1
• error = count（transaction.id）WHERE task.id = transaction.task and transaction.status = 2
• 做=计数（transaction.id）WHERE task.id = transaction.task和transaction.status = 3
• 总=计数（transaction.id）WHERE task.id = transaction.task
• 状态=任务。状态

我INNER试图加入，但我得到一个错误的结果：

SELECT tk.name, tk.priority, waiting.waiting, error.error, done.done, total.total 
    FROM task AS tk, transaction AS tran 
    INNER JOIN (
    SELECT count(id) AS waiting 
    FROM transaction 
    WHERE status = 1 
) AS waiting 
    INNER JOIN (
    SELECT count(id) AS error 
    FROM transaction 
    WHERE status = 3 
) AS error 
    INNER JOIN (
    SELECT count(id) AS done 
    FROM transaction 
    WHERE status = 4 
) AS done 
    INNER JOIN (
    SELECT count(id) AS total 
    FROM transaction 
) AS total; 

能否请你帮我创造这个查询？我得到了等待的列，错误，完成，总数与所有交易的计数。相反，它应该获得交易数量WHERE task.id = transaction.task和transaction.status = 1,2,3。

Answer 1

您现有查询存在一些问题。首先，你没有在任何栏目中将你的两个表格连在一起。看起来你可以加入task.id和transaction.task。第二。你应该能够通过使用聚合函数一些有条件的逻辑让每一个总的，就像一个CASE表达式：

SELECT tk.name, tk.priority, 
    sum(case when tran.status = 1 then 1 else 0 end) waiting, 
    sum(case when tran.status = 3 then 1 else 0 end) error, 
    sum(case when tran.status = 4 then 1 else 0 end) done, 
    count(*) total 
FROM task AS tk 
INNER JOIN transaction AS tran 
    ON tk.id = tran.task 
GROUP BY tk.name, tk.priority; 

这种类型的查询被称为一个支点，它就是你从取值您行并将它们转换成行。在CASE表达式中，您只会得到每个status的总数。

Answer 2

 

    select task.name, 
       task.priority , 
       count(tw.id) as waiting , 
       count(te.id) as error, 
       count(td.id) as done, 
       count(task.id) as total, 
       task.status 
     from transaction as tr 
        JOIN task 
         on task.id = tr.task 
     left JOIN task as tw 
         on (tw.id = tr.task and tr.status = 1) 
     left JOIN task as te 
         on (te.id = tr.task and tr.status = 2) 
     left JOIN task as td 
         on (td.id = tr.task and tr.status = 3) 
     group by task.id 

BY SHAHEN Babayan