【转】rollup、cub、grouping sets、grouping、grouping_id在报表中的应用

摘自 http://blog.itpub.net/26977915/viewspace-734114/

在报表语句中经常要使用各种分组汇总,rollup和cube就是常用的分组汇总方式。

第一:group by rollup

1、如果使用诸如group by rollup(A,B,C)的方式分组,那么返回的分组结果是
(A,B,C) (A,B) (A) (NULL) 一共四种结果。即从右到左递减,最后来个合计。

例如:

SQL> select * from t;

YEARS     MONTHS PRODUCT_NA      SALES
---------- ---------- ---------- ----------
      2008          1 A                1000
      2008          1 B                1500
      2008          2 A                2000
      2008          2 B                3000
      2008          2 C                1000
      2008          3 A                3000

已选择6行。

SQL> select years,months,product_name,sum(sales) sum_sales from t
  2  group by rollup(years,months,product_name)
  3  /

YEARS     MONTHS PRODUCT_NA  SUM_SALES
---------- ---------- ---------- ----------
      2008          1 A                1000 ----------group by (years,months,product_name)
      2008          1 B                1500
      2008          1                  2500 ----------group by (years,months)
      2008          2 A                2000
      2008          2 B                3000
      2008          2 C                1000
      2008          2                  6000 ----------group by (years,months)
      2008          3 A                3000
      2008          3                  3000 ----------group by (years,months)
      2008                            11500 ----------group by (years)
                                      11500 ----------group by (NULL)

已选择11行。

2、如果使用诸如group by A,ROLLUP(B,C) 那么返回的分组方式是:
(A,B,C)  (A,B) (A,NULL)  及在这种情况下,先计算rollup里面的分组情况,再与A组合。

例如:
SQL> select years,months,product_name,sum(sales) sum_sales from t
  2  group by years,rollup(months,product_name)
  3  /

YEARS     MONTHS PRODUCT_NA  SUM_SALES
---------- ---------- ---------- ----------
      2008          1 A                1000 ----------group by (years,months,product_name)
      2008          1 B                1500
      2008          1                  2500 ----------group by (years,months)
      2008          2 A                2000
      2008          2 B                3000
      2008          2 C                1000
      2008          2                  6000
      2008          3 A                3000
      2008          3                  3000
      2008                            11500 ----------group by (years)

已选择10行。

第二:group by cube

1、如果使用诸如cube(A,B,C)的方式,那么返回的分组组合是
(A) (A,B) (A,C) (A,B,C) (B) (B,C) (C) (null) 共8种组合方式

例如:

SQL> select years,months,product_name,sum(sales) sum_sales from t
  2  group by cube(years,months,product_name)
  3  /

YEARS     MONTHS PRODUCT_NA  SUM_SALES
---------- ---------- ---------- ----------
                                      11500 ----------group by (null)
                      A                6000 ----------group by (product_name)
                      B                4500
                      C                1000
                    1                  2500
                    1 A                1000
                    1 B                1500
                    2                  6000
                    2 A                2000
                    2 B                3000
                    2 C                1000
                    3                  3000 ----------group by (months)
                    3 A                3000 ----------group by (months,product_name)
      2008                            11500 ----------group by (years)
      2008            A                6000
      2008            B                4500
      2008            C                1000 ----------group by (years,product_name)
      2008          1                  2500
      2008          1 A                1000
      2008          1 B                1500
      2008          2                  6000
      2008          2 A                2000
      2008          2 B                3000
      2008          2 C                1000
      2008          3                  3000 ----------group by (years,months)
      2008          3 A                3000 ----------group by (years,months,product_name)

已选择26行。

2、如果使用GROUP BY A,CUBE(B,C),那么返回的分组组合为:
(A,B) (A,B,C) (A,C) (A)

例如:
SQL> select years,months,product_name,sum(sales) sum_sales from t
  2  group by years,cube(months,product_name)
  3  /

YEARS     MONTHS PRODUCT_NA  SUM_SALES
---------- ---------- ---------- ----------
      2008                            11500 ----------group by (years)
      2008            A                6000 ----------group by (years,product_name)
      2008            B                4500
      2008            C                1000
      2008          1                  2500 ----------group by (years,months)
      2008          1 A                1000 ----------group by (years,months,product_name)
      2008          1 B                1500
      2008          2                  6000
      2008          2 A                2000
      2008          2 B                3000
      2008          2 C                1000
      2008          3                  3000
      2008          3 A                3000

已选择13行。

3、如果使用GROUP BY A,ROLLUP(B,C),CUBE(D,E),那么返回的分组组合为:

先分解cube:

a,rollup(b,c),d,e
a,rollup(b,c),d
a,rollup(b,c),e
a,rollup(b,c)

再分解ROLLUP而得到最终所有情况为:

a,b,c,d,e
a,b,d,e
a,d,e
a,b,c,d
a,b,d
a,d
a,b,c,e
a,b,e
a,e
a,b,c
a,b
a

例如:
SQL> select years,months,product_name,sum(sales) sum_sales from t
  2  group by years,rollup(months),cube(product_name)
  3  /

YEARS     MONTHS PRODUCT_NA  SUM_SALES
---------- ---------- ---------- ----------
      2008          1 A                1000 ----------group by (years,months,product_name)
      2008          2 A                2000
      2008          3 A                3000
      2008          1 B                1500
      2008          2 B                3000
      2008          2 C                1000
      2008            A                6000 ----------group by (years,product_name)
      2008            B                4500
      2008            C                1000
      2008          1                  2500 ----------group by (years,product_name)
      2008          2                  6000
      2008          3                  3000
      2008                            11500 ----------group by (years)

已选择13行。

第三:grouping sets
如果使用group by A,grouping sets(B,C) 那么相当于group by A,B UNION ALL group by A,C

例如:
SQL> select years,months,product_name,sum(sales) sum_sales from t
  2  group by years,grouping sets(months,product_name)
  3  /

YEARS     MONTHS PRODUCT_NA  SUM_SALES
---------- ---------- ---------- ----------
      2008          2                  6000 ----------group by (years,months)
      2008          1                  2500
      2008          3                  3000
      2008            B                4500 ----------group by (years,product_name)
      2008            C                1000
      2008            A                6000

已选择6行。

-----------------------------------------------------------------------------------------------------------------华丽的分割线!

现实中可能希望出现小计、合计等字样的报表,那么可以使用grouping函数来达到美化的效果!

第三:grouping(exp),当没有对exp分组汇总时,便返回1;

例如:
SQL> select months,product_name,sum(sales) sum_sales,grouping(product_name) from t
  2  group by rollup(months,product_name)
  3  /

MONTHS PRODUCT_NA  SUM_SALES GROUPING(PRODUCT_NAME)
---------- ---------- ---------- ----------------------
         1 A                1000                      0 ----------group by (months,product_name)
         1 B                1500                      0
         1                  2500                      1 ----------group by (months)
         2 A                2000                      0
         2 B                3000                      0
         2 C                1000                      0
         2                  6000                      1 ----------group by (months)
         3 A                3000                      0
         3                  3000                      1 ----------group by (months)
                           11500                      1 ----------group by (null)

已选择10行。

第四:GROUPING_ID(exp1,exp2,…,expN)={GROUPING(exp1)||GROUPING(exp2)||…||GROUPING(expN)}变成十进制数,如:
如果GROUPING(A)=1,GROUPING(B)=0,GROUPING(C)=1,那么
GROUPING_ID(A,B,C) = [101]二进制 = 5,
GROUPING_ID(B,A,C) = [011]二进制 = 3.

例如:
SQL> select years,months,product_name,sum(sales) sum_sales,grouping_id(years,months,product_name) g_id from t
  2  group by rollup(years,months,product_name)
  3  /

YEARS     MONTHS PRODUCT_NA  SUM_SALES       G_ID
---------- ---------- ---------- ---------- ----------
      2008          1 A                1000          0
      2008          1 B                1500          0
      2008          1                  2500          1 ----------group by (years,months) 001=1
      2008          2 A                2000          0
      2008          2 B                3000          0
      2008          2 C                1000          0
      2008          2                  6000          1
      2008          3 A                3000          0
      2008          3                  3000          1
      2008                            11500          3 ----------group by (years)   011=3
                                      11500          7 ----------group by (null)    111=7

已选择11行。

了解了grouping和grouping_id函数后,便可以结合decode函数来生成小计合计的效果了;

SQL> select decode(grouping(months)+grouping(product_name),1,‘月份小计‘,2,‘合计:‘,months) months,
  2  product_name,sum(sales) sum_sales from t
  3  group by rollup(months,product_name)
  4  /

MONTHS                                   PRODUCT_NA  SUM_SALES
---------------------------------------- ---------- ----------
1                                        A                1000
1                                        B                1500
月份小计                                                  2500
2                                        A                2000
2                                        B                3000
2                                        C                1000
月份小计                                                  6000
3                                        A                3000
月份小计                                                  3000
合计:                                                   11500

已选择10行。

SQL> select decode(grouping_id(months,product_name),1,‘月份小计:‘,2,‘产品小计:‘,3,‘合计:‘,months) months,
  2  product_name,sum(sales) sum_sales from t
  3  group by cube(months,product_name)
  4  order by 2
  5  /

MONTHS                                   PRODUCT_NA  SUM_SALES
---------------------------------------- ---------- ----------
1                                        A                1000
2                                        A                2000
3                                        A                3000
产品小计:                               A                6000
1                                        B                1500
2                                        B                3000
产品小计:                               B                4500
2                                        C                1000
产品小计:                               C                1000
月份小计:                                                2500
月份小计:                                                6000
月份小计:                                                3000
合计:                                                   11500

已选择13行。

点评:group by rollup、group by cube、grouping sets、grouping函数、grouping_id函数这些属于报表常用函数,要灵活运用!

时间: 2024-08-01 21:53:07

【转】rollup、cub、grouping sets、grouping、grouping_id在报表中的应用的相关文章

Oracle中group by 的扩展函数rollup、cube、grouping sets

Oracle的group by除了基本用法以外,还有3种扩展用法,分别是rollup.cube.grouping sets,分别介绍如下: 1.rollup 对数据库表emp,假设其中两个字段名为a,b,c. 如果使用group by rollup(a,b),首先会对(a,b)进行group by ,然后对 a 进行 group by ,最后对全表进行 group by 操作. 如下查询结果: 查询语句 Select deptno,job,sum(sal) from emp group by r

T-Sql语法:GROUP BY子句GROUPING SETS、CUBE、ROLLUP

#cnblogs_post_body h1 { background-color: #A5A5A5; color: white; padding: 5px } GROUP BY子句 1.ROLLUP() 生成某一维度的分组的小计行,还生成一个总计行. 示例表: select * from student 我们来看一下具体示例: select sex,sclass,sum(score) from student group by rollup(sex,sclass) 如图中所示,ROLLUP()为

Hive分析窗口函数(五) GROUPING SETS,GROUPING__ID,CUBE,ROLLUP

1.GROUPING SETS与另外哪种方式等价? 2.根据GROUP BY的维度的所有组合进行聚合由哪个关键字完成? 3.ROLLUP与ROLLUP关系是什么? GROUPING SETS,GROUPING__ID,CUBE,ROLLUP 这几个分析函数通常用于OLAP中,不能累加,而且需要根据不同维度上钻和下钻的指标统计,比如,分小时.天.月的UV数. Hive版本为 apache-hive-0.13.1 数据准备: 2015-03,2015-03-10,cookie1 2015-03,20

hive grouping sets 等聚合函数

函数说明: grouping sets 在一个 group by 查询中,根据不同的维度组合进行聚合,等价于将不同维度的 group by 结果集进行 union allcube 根据 group by 的维度的所有组合进行聚合rollup 是 cube 的子集,以最左侧的维度为主,从该维度进行层级聚合. -- grouping sets select order_id, departure_date, count(*) as cnt from ord_test where order_id=4

oracle group by rollup,decode,grouping,nvl,nvl2,nullif,grouping_id,group_id,grouping sets,RATIO_TO

干oracle 047文章12当问题,经验group by 声明.因此邂逅group by  rollup,decode,grouping,nvl,nvl2,nullif,RATIO_TO_REPORT等一下. 1. decode  与if...then,case...when...这类流数据语句功能差点儿相同 decode(条件,值1,返回值1,值2,返回值2,...值n,返回值n,缺省值) 该函数的含义例如以下: IF 条件=值1 THEN RETURN(值1) ELSIF 条件=值2 TH

Group By 多个分组集小结 --GROUPING SETS,GROUP BY CUBE,GROUP BY ROLLUP,GROUPING(),GROUPING_ID()

T-SQL 多个分组集共有三种 GROUPING SETS, CUBE, 以及ROLLUP, 其中 CUBE和ROLLUP可以当做是GROUPING SETS的简写版 示例数据库下载: http://files.cnblogs.com/files/haseo/TSQL2012.rar GROUPING SETS 列出所有你设置的分组集 SELECT shipperid, YEAR(shippeddate) AS shipyear, COUNT(*) AS numorders FROM Sales

SQL Server ->> GROUPING SETS, CUBE, ROLLUP, GROUPING, GROUPING_ID

在我们制作报表的时候常常需要分组聚合.多组聚合和总合.如果通过另外的T-SQL语句来聚合难免性能太差.如果通过报表工具的聚合功能虽说比使用额外的T-SQL语句性能上要好很多,不过不够干脆,还是需要先生成整个结果集然后再聚合,而且最最重要的时很多情况下报表的聚合功能可能没办法达到我们需要的效果.GROUPING SETS, CUBE, ROLLUP, GROUPING, GROUPING_ID这几个聚合函数的作用就是在原始语句的基础上完成很多像财务报表需要的聚合功能. GROUPING SETS相

GROUPING SETS、CUBE、ROLLUP

其实还是写一个Demo 比较好 USE tempdb IF OBJECT_ID( 'dbo.T1' , 'U' )IS NOT NULL BEGIN DROP TABLE dbo.T1; END; GO CREATE TABLE dbo.T1 ( id INT , productName VARCHAR(200) , price MONEY , num INT , amount INT , operatedate DATETIME ) GO DECLARE @i INT DECLARE @ran

grouping sets,cube,rollup,grouping__id,group by

例1: hive -e" select type ,status ,count(1) from usr_info where pt='2015-09-14' group by type,status grouping sets ((type,status),( type),()); ">one.txt Grouping sets按照各种指定聚类汇总方式,如group by type,status grouping sets ((type,status),( type),()) 表