Question

我有一张这样的表：

Title   | Version   | Condition | Count
-----------------------------------------------
Title1  | 1.0       | 1         | 10
Title1  | 1.1       | 2         | 5
Title1  | 1.1       | 2         | 10
Title1  | 1.1       | 1         | 10
Title2  | 1.0       | 2         | 10
Title2  | 1.5       | 1         | 5
Title2  | 1.5       | 2         | 5
Title3  | 1.5       | 2         | 10
Title3  | 1.5       | 1         | 10

我想总结＆＃34; Count＆＃34;的价值。对于每一行都有MAX（）＆＃34;版本＆＃34;和＆＃34;条件＆＃34; = 2.我希望这是结果数据集：

Title   | Version   | Condition | Count
-----------------------------------------------
Title1  | 1.1       | 2         | 15
Title2  | 1.5       | 2         | 5
Title3  | 1.5       | 2         | 10

我可以获得＆＃34;标题＆＃34;的列表和MAX（＆＃34;版本＆＃34;）＆＃34;条件＆＃34; = 2：

SELECT DISTINCT Title, MAX(Version) AS MaxVer FROM TABLE
WHERE Condition = 2
GROUP BY Title

我不知道如何添加所有＆＃34; Count＆＃34; s。

Answer 1

试试这个：

SELECT t1.Title, t1.Version, t1.Condition, SUM([Count]) AS Count
FROM mytable AS t1
JOIN (
  SELECT Title, MAX(Version) AS max_version
  FROM mytable
  WHERE Condition = 2
  GROUP BY Title
) AS t2 ON t1.Title = t2.Title AND t1.Version = t2.max_version
WHERE t1.Condition = 2
GROUP BY t1.Title, t1.Version, t1.Condition

Demo here

Answer 2

这只需要一次访问该表：

SELECT Title, Version, Condition, Count
FROM
 ( -- calculate all SUMs
   SELECT Title, Version, Condition, SUM(Count) AS Count,
      ROW_NUMBER() OVER (PARTITION BY Title ORDER BY Version DESC) AS rn
   FROM TABLE
   WHERE Condition = 2
   GROUP BY Title, Version, Condition
 ) AS dt
   -- and return only the row with the highest Version
WHERE rn = 1

SUM（），GROUP BY和WHERE

2 个答案: