SQL Server 2008中SQL应用系列及BI笔记系列--目录索引
导读:本文介绍集合(Sets)的进阶内容,本文将包括以下内容:
■1、Filter函数
■2、NonEmpty函数
■3、Set的Union、InterSection和Exception
■4、Generate和Extract函数
本文所用数据库和所有源码,请到微软官网下载
1、使用Filter函数(http://msdn.microsoft.com/zh-cn/library/ms146037.aspx)
例6-14
SELECT
{
([Measures].[Reseller Sales Amount]),
([Measures].[Internet Sales Amount])
} ON COLUMNS,
Filter(
{[Product].[Product].[Product].Members},
([Measures].[Internet Sales Amount]) > ([Measures].[Reseller Sales Amount])
) ON ROWS
FROM [Step-by-Step]
;
例6-15
SELECT
{
([Measures].[Reseller Sales Amount]),
([Measures].[Internet Sales Amount])
} ON COLUMNS,
Filter(
{[Product].[Product].[Product].Members},
([Measures].[Internet Sales Amount]) > ([Measures].[Reseller Sales Amount]) AND
(
[Product].[Category].CurrentMember Is [Product].[Category].[Clothing] OR
[Product].[Category].CurrentMember Is [Product].[Category].[Accessories]
)
) ON ROWS
FROM [Step-by-Step]
;
例6-16
SELECT
{
([Measures].[Reseller Sales Amount]),
([Measures].[Internet Sales Amount])
} ON COLUMNS,
Filter(
{[Product].[Product].[Product].Members},
([Measures].[Internet Sales Amount]) >
([Measures].[Reseller Sales Amount]) AND
NOT (
[Product].[Category].CurrentMember Is [Product].[Category].[Clothing] OR
[Product].[Category].CurrentMember Is [Product].[Category].[Accessories]
)
) ON ROWS
FROM [Step-by-Step]
;
2、使用NonEmpty函数(http://msdn.microsoft.com/en-us/library/ms145988.aspx)
例6-17
SELECT
{([Measures].[Reseller Sales Amount])} ON COLUMNS,
NonEmpty(
{[Product].[Product].[Product].Members},
{([Measures].[Reseller Sales Amount])}
) ON ROWS
FROM [Step-by-Step]
;
例6-18
SELECT
{([Measures].[Reseller Sales Amount])} ON COLUMNS,
NON EMPTY {[Product].[Product].[Product].Members} ON ROWS
FROM [Step-by-Step]
;
这两个例子中,使用NonEmpty函数和Non Empty关键字其实效果是一样的。
3、组合结果集
结果集的组合有三种形式:Union,intersection和Exception
如下图所示:
看原始例子:
例6-19:
SELECT
{([Measures].[Reseller Sales Amount])} ON COLUMNS,
TopCount(
{[Product].[Product].[Product].Members},
10,
([Measures].[Reseller Sales Amount], [Date].[Calendar Year].[CY 2004])
) ON ROWS
FROM [Step-by-Step]
;
看看Union的效果:
例6-20:
SELECT
{([Measures].[Reseller Sales Amount])} ON COLUMNS,
Union(
TopCount(
{[Product].[Product].[Product].Members},
10,
([Measures].[Reseller Sales Amount], [Date].[Calendar Year].[CY 2003])
),
TopCount(
{[Product].[Product].[Product].Members},
10,
([Measures].[Reseller Sales Amount], [Date].[Calendar Year].[CY 2004])
)
) ON ROWS
FROM [Step-by-Step]
;
使用Exception的效果:
例6-21:
SELECT
{([Measures].[Reseller Sales Amount])} ON COLUMNS,
Except(
TopCount(
{[Product].[Product].[Product].Members},
10,
([Measures].[Reseller Sales Amount], [Date].[Calendar Year].[CY 2003])
),
TopCount(
{[Product].[Product].[Product].Members},
10,
([Measures].[Reseller Sales Amount], [Date].[Calendar Year].[CY 2004])
)
) ON ROWS
FROM [Step-by-Step]
;
InterSection的效果:
例6-22:
SELECT
{([Measures].[Reseller Sales Amount])} ON COLUMNS,
Intersect(
TopCount(
{[Product].[Product].[Product].Members},
10,
([Measures].[Reseller Sales Amount], [Date].[Calendar Year].[CY 2004])
),
TopCount(
{[Product].[Product].[Product].Members},
10,
([Measures].[Reseller Sales Amount], [Date].[Calendar Year].[CY 2003])
)
) ON ROWS
FROM [Step-by-Step]
;
4、Generate和Extract函数
MDX语言提供了两个强有力的“深奥”函数
Generate(http://msdn.microsoft.com/zh-cn/library/ms145526%28v=SQL.105%29.aspx)
Extract(http://msdn.microsoft.com/zh-cn/library/ms145980.aspx)
看原始语句:
例6-23:
SELECT
{([Measures].[Reseller Sales Amount])} ON COLUMNS,
TopCount(
{[Product].[Product].[Product].Members},
5,
([Measures].[Reseller Sales Amount])
) ON ROWS
FROM [Step-by-Step]
;
加上Generate
例6-24:
SELECT
{([Measures].[Reseller Sales Amount])} ON COLUMNS,
Generate(
{[Product].[Category].[Category].Members},
TopCount(
{[Product].[Product].[Product].Members},
5,
([Measures].[Reseller Sales Amount])
)
) ON ROWS
FROM [Step-by-Step]
;
所得的结果与上例相同,
那么,generate起什么作用呢?我们再加上All标志符
例6-25:
SELECT
{([Measures].[Reseller Sales Amount])} ON COLUMNS,
Generate(
{[Product].[Category].[Category].Members},
TopCount(
{[Product].[Product].[Product].Members},
5,
([Measures].[Reseller Sales Amount])
),
ALL
) ON ROWS
FROM [Step-by-Step]
;
换句话说,Generate其实是实现了遍历,按每个分类进行TopCount,这个在统计中非常棒。如果老板要统计今年每个月工作量最大的十个员工,那么你应该想到Generate,更进一步,你可以使用Category的hierarchy属性,得到如下结果:
例6-26
SELECT
{([Measures].[Reseller Sales Amount])} ON COLUMNS,
Generate(
{[Product].[Category].[Category].Members},
{([Product].[Category].CurrentMember)} *
TopCount(
EXISTING {[Product].[Product].[Product].Members},
5,
([Measures].[Reseller Sales Amount])
),
ALL
) ON ROWS
FROM [Step-by-Step]
;
结果:
Extract的用途,先看下面的例子,使用Filter
例6-27
SELECT
{([Measures].[Reseller Sales Amount])} ON COLUMNS,
Filter(
{[Product].[Product].[Product].Members} *
{[Date].[Calendar].[Month].Members},
([Measures].[Reseller Sales Amount])>160000
) ON ROWS
FROM [Step-by-Step]
;
使用Extract的例子
例6-28
SELECT
{([Measures].[Reseller Sales Amount])} ON COLUMNS,
Extract(
Filter(
{[Product].[Product].[Product].Members} *
{[Date].[Calendar].[Month].Members},
([Measures].[Reseller Sales Amount])>160000
),
[Product].[Product]
) ON ROWS
FROM [Step-by-Step]
;
请注意:Extract在此处相当于使用了distinct。
小结:本文是集合(Sets)的进阶,介绍了几个常用的函数如Filter和NonEmpty,Generate和Extract,集的组合方式如Union、InterSection和Exception。
下文将继续学习执行Aggregation的相关内容。
参考资源:
1、MDX官方教程(http://msdn.microsoft.com/zh-cn/library/ms145506.aspx)