Question

我有一个自行车骑行数据库，我用它来学习使用Libreoffice Base和HSQL的SQL。该数据库由一个包含以下列的表组成：Index，Date，Bike，Miles。

我想做的是为我的公路自行车创建一个列有自行车骑行总和的列，另一个用于我的mtn自行车。我从子查询开始，但这给了我整个数据库的总数。这是我的公路自行车代码，另一个子查询将用于mtn自行车列：

SELECT YEAR( "RideDate" ) AS "Year", SUM( "Miles" ) AS "Miles", 
( SELECT SUM( "Miles" ) FROM "BikeDate" 
   WHERE( "Bike" = 'Fuji' OR "Bike" = 'Yfoil' )) AS "% Miles" 
FROM "BikeDate" AS "BikeDate" 
GROUP BY YEAR( "RideDate" ) 
ORDER BY YEAR( "RideDate" ) DESC

由于我按年分组，如何获得每个分组元素的总和 - >年？我为不知道正确的术语而道歉。我也不想做一个简单的总和，我按年和自行车分组，我有。我想用where子句打破年份。戴夫，感谢任何帮助。 LOBase 4.2.3.3。

Answer 1

让我根据您的真实需要提出两种可能性，以便像原始帖子一样返回值：

SELECT 
   YEAR( "RideDate" ) AS "Year", 
   SUM( "Miles" ) AS "Miles", 
   SUM(CASE WHEN "Bike" = 'Fuji' OR "Bike" = 'Yfoil' THEN "Miles" ELSE 0 END) as "% Miles"
FROM "BikeDate" AS "BikeDate" 
GROUP BY YEAR( "RideDate" ) 
ORDER BY YEAR( "RideDate" ) DESC

为了创建一个额外的列，它真正计算了百分比。。。类似的东西：

SELECT 
   YEAR( "RideDate" ) AS "Year", 
   SUM( "Miles" ) AS "Miles", 
   SUM(CASE WHEN "Bike" = 'Fuji' OR "Bike" = 'Yfoil' THEN "Miles" ELSE 0 END) as "Miles_Fuji_Yfoil",
   SUM(CASE WHEN "Bike" = 'Fuji' OR "Bike" = 'Yfoil' THEN "Miles" ELSE 0 END) * 100.0000000 / SUM("Miles") as "Pct_Miles_Fuji_Yfoil"
FROM "BikeDate" AS "BikeDate" 
GROUP BY YEAR( "RideDate" ) 
ORDER BY YEAR( "RideDate" ) DESC

解释：这使用HSQL函数 CASE WHEN。。。要么。。。值ELSE 0 END 所以它只会 SUM （聚合函数）这些值。

子组数据的LOBase HSQL子查询

1 个答案: