Question

我有3张看起来像这样的表：

表1用户：

ID
姓名
得分

表2游戏：

ID
姓名
犯规

表3 Game2

ID
姓名
犯规

我想做的是在Table1中选择所有用户。然后选择Table1.Name = Table2.Name或Table1.Name = Table3.Name但具有更高的Table2优先级的所有数据，因此如果匹配Table1.Name = Table2.Name，则不会进行表3的连接。然后，当我选择正确的时候，我也想计算得分和犯规的总和。

这是如何在SQL中完成的，是否可能？我正在使用MSSQL

Answer 1

SELECT
  SUM(t1.Score) AS TotalScore,
  SUM(COALESCE(t2.Fouls, t3.Fouls)) AS TotalFouls
FROM Table1 t1
  LEFT JOIN Table2 t2 ON t2.Name = t1.Name
  LEFT JOIN Table3 t3 ON t3.Name = NULLIF(t1.Name, t2.Name)

Answer 2

以下内容为您提供了很多帮助。不过，我不确定你是想要每用户还是整体聚合。

此外 - 如果您要使用Game3，Game4等，那么您最好的选择是更改您的桌面设计，而不是进一步使这（和其他）查询复杂化。

SELECT      SUM(U.Score) AS Score,
        SUM(Fouls.Fouls) AS Fouls
FROM        dbo.Users U
INNER JOIN  (
    SELECT      COALESCE(G.Name, G2.Name) AS Name,
            COALESCE(G.Fouls, G2.Fouls) AS Fouls
    FROM        dbo.Game G
    FULL JOIN   dbo.Game2 G2
    ON      G.Name = G2.Name
) Fouls
ON      U.Name = Fouls.Name

Answer 3

您可以通过在连接到Table3 Table2.Name IS NULL的条件中指定连接到表2。如果你想要每个球员的总得分和犯规，你可以这样做：

SELECT derived_table.Name, SUM(derived_table.Score), SUM(derived_table.Fouls)
FROM (
    SELECT Name = Table1.Name, Score = Table1.Score,
        Fouls = COALESCE(Table2.Fouls, Table3.Fouls)
    FROM Table1
        LEFT OUTER JOIN Table2 ON Table2.Name = Table1.Name
        LEFT OUTER JOIN Table3 ON Table3.Name = Table1.Name
            AND Table2.Name IS NULL
    WHERE Table2.Name IS NOT NULL OR Table3.Name IS NOT NULL
    ) derived_table
GROUP BY derived_table.Name

总计：

SELECT SUM(derived_table.Score), SUM(derived_table.Fouls)
FROM (
    SELECT Score = Table1.Score,
        Fouls = COALESCE(Table2.Fouls, Table3.Fouls)
    FROM Table1
        LEFT OUTER JOIN Table2 ON Table2.Name = Table1.Name
        LEFT OUTER JOIN Table3 ON Table3.Name = Table1.Name
            AND Table2.Name IS NULL
    WHERE Table2.Name IS NOT NULL OR Table3.Name IS NOT NULL
    ) derived_table

多个group by，count和join

表1用户：

表2游戏：

3 个答案: