本文通过例子展示sum, rollup, cube, grouping的用法。

//首先建score表

create table score(
class  nvarchar2(20),
course   nvarchar2(20),
stu_no  number(5),
stu_name nvarchar2(20),
score   number(2));

//插入数据

insert into score values ('Class_A','Math',10001,'Tough1',95);
insert into score values ('Class_A','Math',10002,'Tough2',93);
insert into score values ('Class_B','Math',10003,'Tough3',94);
insert into score values ('Class_B','Math',10004,'Tough4',88);
insert into score values ('Class_A','Computer',10001,'Tough1',89);
insert into score values ('Class_A','Computer',10002,'Tough2',98);
insert into score values ('Class_B','Computer',10003,'Tough3',89);
insert into score values ('Class_B','Computer',10004,'Tough4',87);

SELECT * FROM score;

CLASS	COURSE	STU_NO	STU_NAME	SCORE
Class_A	Math	10001	Tough1	95
Class_A	Math	10002	Tough2	93
Class_B	Math	10003	Tough3	94
Class_B	Math	10004	Tough4	88
Class_A	Computer	10001	Tough1	89
Class_A	Computer	10002	Tough2	98
Class_B	Computer	10003	Tough3	89
Class_B	Computer	10004	Tough4	87

• sum函数

按class,course分组，然后对score统计总合

select class,course,sum(score)/count(*) "AVG_SCORE"
from score
group by class,course;

CLASS	COURSE	AVG_SCORE
Class_B	Math	91
Class_A	Computer	93.5
Class_A	Math	94
Class_B	Computer	88

• rollup函数

先按class,course分组汇总;再按class分组汇总;最后全表分组汇总

select class,course,sum(score)/count(*) "AVG_SCORE"
from score
group by rollup(class,course);

CLASS	COURSE	AVG_SCORE
Class_A	Math	94
Class_A	Computer	93.5
Class_A		93.75
Class_B	Math	91
Class_B	Computer	88
Class_B		89.5
		91.625

• cube函数

先按class,course分组汇总;再按class分组汇总;再按course分组汇总;最后全表分组汇总

select class,course,sum(score)/count(*) "AVG_SCORE"
from score
group by cube(class,course)
order by class,course nulls last;

CLASS	COURSE	AVG_SCORE
Class_A	Computer	93.5
Class_A	Math	94
Class_A		93.75
Class_B	Computer	88
Class_B	Math	91
Class_B		89.5
	Computer	90.75
	Math	92.5
		91.625

rollup和cube区别：

ROLLUP(A,B,C)的话，GROUP BY顺序
(A,B,C)
(A,B)
(A)
最后对全表进行GROUP BY操作。

GROUP BY CUBE(A, B, C)，GROUP BY顺序
(A,B,C)
(A,B)
(A,C)
(A)，
(B,C)
(B)
(C)，
最后对全表进行GROUP BY操作。

• grouping函数

rollup和cube函数都会对结果集产生null，这时候可用grouping函数来确认该记录是由哪个字段得出来的。grouping函数用法，带一个参数，参数为字段名，结果是根据该字段得出来的就返回1，反之返回0

select decode(grouping(class), 1, 'All Class', class) "CLASS",
         decode(grouping(course), 1, 'All Course', course) "COURSE",
         sum(score) / count(*) "AVG_SCORE"
from   score
group  by cube(class, course)
order  by class, course nulls last;

CLASS	COURSE	AVG_SCORE
All Class	All Course	91.625
All Class	Computer	90.75
All Class	Math	92.5
Class_A	All Course	93.75
Class_A	Computer	93.5
Class_A	Math	94
Class_B	All Course	89.5
Class_B	Computer	88
Class_B	Math	91

---