Question

我（再次）这个表：

User:
id | name
-------------
1  | 'John'
2  | 'Peter'
3  | 'Luke'

Accounts:
id | userid | amount | date
--------------------------
1  | 1      | 1000   | '2012-01-26'
2  | 1      | 2000   | '2011-12-25'
3  | 1      | 1000   | '2012-01-25'
4  | 2      | 1500   | '2012-01-15'
5  | 3      | 2500   | '2011-11-30'

我需要收集数据，因此我同时拥有用户的结果集，每个用户的帐户数量，每个用户的帐户金额总和，以及每个用户的帐户，但仅限于当前年份的日期......

这样的事情：

Some-query:
UserName | CountAccts | SumAccts | CountAcctsThisYear
-----------------------------------------------------
'John'   | 3          | 4000     | 2
'Peter'  | 1          | 1500     | 1
'Luke'   | 1          | 2500     | 0

我设法通过此查询获得前3列：

SELECT u.name as Username,
       COUNT(a.id) as CountAccts,
       SUM(a.amount) as SumAccts
FROM   User u
LEFT JOIN Accounts a ON u.id = a.userid
GROUP BY u.id;

另外，我可以用另一个列获得第一列和最后一列：

SELECT u.name as Username,
       COUNT(acctsthisyear.id) as CountAcctsThisYear
FROM   User u
LEFT JOIN (SELECT a.id
             FROM Accounts a
            WHERE DATE_FORMAT(a.date,'%Y') = DATE_FORMAT(CURDATE(),'%Y')
          ) AS acctsthisyear ON u.id = acctsthisyear.userid
GROUP BY u.id;

但是，这就是问题，我无法以这种方式使用这两个查询来获得我期待的最终结果......

我有...... - 尝试将Accounts表和acctsthisyear子查询与User表（由us.id字段分组）连接起来，但我得到的是这样的：

UserName | CountAccts | SumAccts | CountAcctsThisYear
-----------------------------------------------------
'John'   | 6          | 6000     | 6
'Peter'  | 1          | 1500     | 1
'Luke'   | 0          | 0        | 0

（不记得所有细节，但问题是，例如，在CountAcctsThisyear中'John'的2个结果，在结果中也得到重复，并且所有细节都成倍增加......）

尝试将每个查询作为子查询： SELECT * FROM （首先使用u.id进行查询，因为ufid也被选中）AS首先，（使用u.id作为usid的第二个查询也被选中）AS第二个 GROUP BY ufid，usid

得到这样的东西：

UserName | CountAccts | SumAccts | CountAcctsThisYear
-----------------------------------------------------
'John'   | 3          | 4000     | 2
'Peter'  | 1          | 1500     | 2
'Luke'   | 1          | 2500     | 2
'John'   | 3          | 4000     | 1
'Peter'  | 1          | 1500     | 1
'Luke'   | 1          | 2500     | 1
'John'   | 3          | 4000     | 0
'Peter'  | 1          | 1500     | 0
'Luke'   | 1          | 2500     | 0

（如果我按元素反转Group，我得到的只是相同但按不同的顺序分组......）

任何方式......就是这样......任何关于实现我正在寻找什么的正确方法的想法？我真的很沮丧，因为我真的不是SQL专家，特别是关于Joins： - /

Answer 1

试试这个：

SELECT u.`Name` as UserName,
       Count(*) as CountAccts,
       SUM(a.`amount`) as SumAccts,
       COALESCE(iTable.CountAcctsThisYear, 0) as CountAcctsThisYear
FROM    `User` u INNER JOIN `Accounts` a
                    ON u.`ID` = a`.`ID`
               LEFT JOIN
                   (SELECT `ID`, Count(*) as CountAcctsThisYear
                    FROM `Accounts`
                    WHERE DATE_FORMAT(`date`,'%Y') = DATE_FORMAT(CURDATE(),'%Y')
                    GROUP BY `ID`) as iTable
                ON `User`.`ID` = iTable.`ID`
GROUP BY `User`.`Name`

Answer 2

SELECT u.name as Username,
       COUNT(a.id) as CountAccts,
       SUM(a.amount) as SumAccts,
       SUM( CASE WHEN YEAR(a.date) = YEAR(CURDATE()) THEN 1 ELSE 0 END ) 
           as CountAcctsThisYear
FROM   User u
LEFT JOIN Accounts a ON u.id = a.userid
GROUP BY u.id;

Answer 3

SELECT u.name as Username,
       COUNT(a.userid) as CountAccts,
       SUM(a.amount) as SumAccts,
       SUM( CASE WHEN YEAR(a.date) = YEAR(CURDATE()) THEN 1 ELSE 0 END ) 
           as CountAcctsThisYear
FROM Accounts a 
INNER JOIN  User u ON u.id = a.userid
GROUP BY a.userid;

Answer 4

SELECT u.name as Username,
       COUNT(a.id) as CountAccts,
       SUM(a.amount) as SumAccts
FROM   User u
INNER JOIN Accounts a ON u.id = a.userid
GROUP BY u.id,YEAR(a.date)

尝试一次!!!

多个表（一个是子查询）连接

4 个答案: