原文： http://blog.csdn.net/wh62592855/article/details/4817920

?

?

ROLLUP，是GROUP BY子句的一種擴展，可以為每個分組返回小計記錄以及為所有分組返回總計記錄。

CUBE，也是GROUP BY子句的一種擴展，可以返回每一個列組合的小計記錄，同時在末尾加上總計記錄。

?

在文章的最后附上了相關表和記錄創建的腳本。

?

1、 向ROLLUP傳遞一列

SQL> select division_id,sum(salary)
? ? from employees2
? ? group by rollup(division_id)
? ? order by division_id;

?

DIV SUM(SALARY)
--- -----------
BUS???? 1610000
OPE???? 1320000
SAL???? 4936000
SUP???? 1015000
??????? 8881000

SQL>

再來看一下如果使用普通的GROUP BY，而沒有ROLLUP是個什么情況

SQL> select division_id,sum(salary)
? ? from employees2
? ? group by division_id
? ? order by division_id;

?

DIV SUM(SALARY)
--- -----------
BUS???? 1610000
OPE???? 1320000
SAL???? 4936000
SUP???? 1015000

可以看到，缺少了最后的統計信息。

?

2、向ROLLUP傳遞多列

SQL> select division_id,job_id,sum(salary)
? ? from employees2
? ? group by rollup(division_id,job_id)
? ? order by division_id,job_id;

?

DIV JOB SUM(SALARY)
--- --- -----------
BUS MGR????? 530000
BUS PRE????? 800000
BUS WOR????? 280000
BUS???????? 1610000
OPE ENG????? 245000
OPE MGR????? 805000
OPE WOR????? 270000
OPE???????? 1320000
SAL MGR???? 4446000
SAL WOR????? 490000
SAL???????? 4936000

DIV JOB SUM(SALARY)
--- --- -----------
SUP MGR????? 465000
SUP TEC????? 115000
SUP WOR????? 435000
SUP???????? 1015000
??????????? 8881000

16 rows selected.

可以看到，除了在最后有一個求和記錄外，每個division_id分組也會有一個求和記錄。

那么我們現在交換一下ROLLUP中數據列的順序，看看結果怎樣

SQL> select job_id,division_id,sum(salary)
? ? from employees2
? ? group by rollup(job_id,division_id)
? ? order by job_id,division_id;

?

JOB DIV SUM(SALARY)
--- --- -----------
ENG OPE????? 245000
ENG????????? 245000
MGR BUS????? 530000
MGR OPE????? 805000
MGR SAL???? 4446000
MGR SUP????? 465000
MGR???????? 6246000
PRE BUS????? 800000
PRE????????? 800000
TEC SUP????? 115000
TEC????????? 115000

JOB DIV SUM(SALARY)
--- --- -----------
WOR BUS????? 280000
WOR OPE????? 270000
WOR SAL????? 490000
WOR SUP????? 435000
WOR???????? 1475000
??????????? 8881000

17 rows selected.

結果類似，只不過是每一個job_id分組有一個求和記錄罷了。

?

3、向CUBE傳遞一列

SQL> select division_id,sum(salary)
? ? from employees2
? ? group by cube(division_id)
? ? order by division_id;

?

?

DIV SUM(SALARY)
--- -----------
BUS???? 1610000
OPE???? 1320000
SAL???? 4936000
SUP???? 1015000
??????? 8881000

好像和ROLLUP沒什么區別哦，呵呵，繼續往下看。

?

4、向CUBE傳遞多列

SQL> select job_id,division_id,sum(salary)
? ? from employees2
? ? group by cube(job_id,division_id)
? ? order by job_id,division_id;

?

JOB DIV SUM(SALARY)
--- --- -----------
ENG OPE????? 245000
ENG????????? 245000
MGR BUS????? 530000
MGR OPE????? 805000
MGR SAL???? 4446000
MGR SUP????? 465000
MGR???????? 6246000
PRE BUS????? 800000
PRE????????? 800000
TEC SUP????? 115000
TEC????????? 115000

JOB DIV SUM(SALARY)
--- --- -----------
WOR BUS????? 280000
WOR OPE????? 270000
WOR SAL????? 490000
WOR SUP????? 435000
WOR???????? 1475000
??? BUS???? 1610000
??? OPE???? 1320000
??? SAL???? 4936000
??? SUP???? 1015000
??????????? 8881000

21 rows selected.

可以看到工資是根據job_id和division_id求和的，CUBE在每一個job_id中都返回一條記錄，表示其中的工資總數，同時在接近末尾處顯示每一種division_id的工資總數，最后一條記錄顯示所有工資的總數。

把兩列的順序換換會怎樣？呵呵，真的有興趣那就自己動手試試吧。

?

==================================================================================

CREATE TABLE divisions (
? division_id CHAR(3) CONSTRAINT divisions_pk PRIMARY KEY,
? name VARCHAR2(15) NOT NULL
);

CREATE TABLE jobs (
? job_id CHAR(3) CONSTRAINT jobs_pk PRIMARY KEY,
? name VARCHAR2(20) NOT NULL
);

CREATE TABLE employees2 (
? employee_id INTEGER CONSTRAINT employees2_pk PRIMARY KEY,
? division_id CHAR(3)
??? CONSTRAINT employees2_fk_divisions
??? REFERENCES divisions(division_id),
? job_id CHAR(3) REFERENCES jobs(job_id),
? first_name VARCHAR2(10) NOT NULL,
? last_name VARCHAR2(10) NOT NULL,
? salary NUMBER(6, 0)
);

INSERT INTO divisions (
? division_id, name
) VALUES (
? 'SAL', 'Sales'
);

INSERT INTO divisions (
? division_id, name
) VALUES (
? 'OPE', 'Operations'
);

INSERT INTO divisions (
? division_id, name
) VALUES (
? 'SUP', 'Support'
);

INSERT INTO divisions (
? division_id, name
) VALUES (
? 'BUS', 'Business'
);
INSERT INTO jobs (
? job_id, name
) VALUES (
? 'WOR', 'Worker'
);

INSERT INTO jobs (
? job_id, name
) VALUES (
? 'MGR', 'Manager'
);

INSERT INTO jobs (
? job_id, name
) VALUES (
? 'ENG', 'Engineer'
);

INSERT INTO jobs (
? job_id, name
) VALUES (
? 'TEC', 'Technologist'
);

INSERT INTO jobs (
? job_id, name
) VALUES (
? 'PRE', 'President'
);
INSERT INTO employees2 (
? employee_id, division_id, job_id, first_name, last_name, salary
) VALUES (
? 1, 'BUS', 'PRE', 'James', 'Smith', 800000
);

INSERT INTO employees2 (
? employee_id, division_id, job_id, first_name, last_name, salary
) VALUES (
? 2, 'SAL', 'MGR', 'Ron', 'Johnson', 350000
);

INSERT INTO employees2 (
? employee_id, division_id, job_id, first_name, last_name, salary
) VALUES (
? 3, 'SAL', 'WOR', 'Fred', 'Hobbs', 140000
);

INSERT INTO employees2 (
? employee_id, division_id, job_id, first_name, last_name, salary
) VALUES (
? 4, 'SUP', 'MGR', 'Susan', 'Jones', 200000
);

INSERT INTO employees2 (
? employee_id, division_id, job_id, first_name, last_name, salary
) VALUES (
? 5, 'SAL', 'WOR', 'Rob', 'Green', 350000
);

INSERT INTO employees2 (
? employee_id, division_id, job_id, first_name, last_name, salary
) VALUES (
? 6, 'SUP', 'WOR', 'Jane', 'Brown', 200000
);

INSERT INTO employees2 (
? employee_id, division_id, job_id, first_name, last_name, salary
) VALUES (
? 7, 'SUP', 'MGR', 'John', 'Grey', 265000
);

INSERT INTO employees2 (
? employee_id, division_id, job_id, first_name, last_name, salary
) VALUES (
? 8, 'SUP', 'WOR', 'Jean', 'Blue', 110000
);

INSERT INTO employees2 (
? employee_id, division_id, job_id, first_name, last_name, salary
) VALUES (
? 9, 'SUP', 'WOR', 'Henry', 'Heyson', 125000
);

INSERT INTO employees2 (
? employee_id, division_id, job_id, first_name, last_name, salary
) VALUES (
? 10, 'OPE', 'MGR', 'Kevin', 'Black', 225000
);

INSERT INTO employees2 (
? employee_id, division_id, job_id, first_name, last_name, salary
) VALUES (
? 11, 'OPE', 'MGR', 'Keith', 'Long', 165000
);

INSERT INTO employees2 (
? employee_id, division_id, job_id, first_name, last_name, salary
) VALUES (
? 12, 'OPE', 'WOR', 'Frank', 'Howard', 125000
);

INSERT INTO employees2 (
? employee_id, division_id, job_id, first_name, last_name, salary
) VALUES (
? 13, 'OPE', 'WOR', 'Doreen', 'Penn', 145000
);

INSERT INTO employees2 (
? employee_id, division_id, job_id, first_name, last_name, salary
) VALUES (
? 14, 'BUS', 'MGR', 'Mark', 'Smith', 155000
);

INSERT INTO employees2 (
? employee_id, division_id, job_id, first_name, last_name, salary
) VALUES (
? 15, 'BUS', 'MGR', 'Jill', 'Jones', 175000
);

INSERT INTO employees2 (
? employee_id, division_id, job_id, first_name, last_name, salary
) VALUES (
? 16, 'OPE', 'ENG', 'Megan', 'Craig', 245000
);

INSERT INTO employees2 (
? employee_id, division_id, job_id, first_name, last_name, salary
) VALUES (
? 17, 'SUP', 'TEC', 'Matthew', 'Brant', 115000
);

INSERT INTO employees2 (
? employee_id, division_id, job_id, first_name, last_name, salary
) VALUES (
? 18, 'OPE', 'MGR', 'Tony', 'Clerke', 200000
);

INSERT INTO employees2 (
? employee_id, division_id, job_id, first_name, last_name, salary
) VALUES (
? 19, 'BUS', 'MGR', 'Tanya', 'Conway', 200000
);

INSERT INTO employees2 (
? employee_id, division_id, job_id, first_name, last_name, salary
) VALUES (
? 20, 'OPE', 'MGR', 'Terry', 'Cliff', 215000
);

INSERT INTO employees2 (
? employee_id, division_id, job_id, first_name, last_name, salary
) VALUES (
? 21, 'SAL', 'MGR', 'Steve', 'Green', 275000
);

INSERT INTO employees2 (
? employee_id, division_id, job_id, first_name, last_name, salary
) VALUES (
? 22, 'SAL', 'MGR', 'Roy', 'Red', 375000
);

INSERT INTO employees2 (
? employee_id, division_id, job_id, first_name, last_name, salary
) VALUES (
? 23, 'SAL', 'MGR', 'Sandra', 'Smith', 335000
);

INSERT INTO employees2 (
? employee_id, division_id, job_id, first_name, last_name, salary
) VALUES (
? 24, 'SAL', 'MGR', 'Gail', 'Silver', 225000
);

INSERT INTO employees2 (
? employee_id, division_id, job_id, first_name, last_name, salary
) VALUES (
? 25, 'SAL', 'MGR', 'Gerald', 'Gold', 245000
);

INSERT INTO employees2 (
? employee_id, division_id, job_id, first_name, last_name, salary
) VALUES (
? 26, 'SAL', 'MGR', 'Eileen', 'Lane', 235000
);

INSERT INTO employees2 (
? employee_id, division_id, job_id, first_name, last_name, salary
) VALUES (
? 27, 'SAL', 'MGR', 'Doreen', 'Upton', 235000
);

INSERT INTO employees2 (
? employee_id, division_id, job_id, first_name, last_name, salary
) VALUES (
? 28, 'SAL', 'MGR', 'Jack', 'Ewing', 235000
);

INSERT INTO employees2 (
? employee_id, division_id, job_id, first_name, last_name, salary
) VALUES (
? 29, 'SAL', 'MGR', 'Paul', 'Owens', 245000
);

INSERT INTO employees2 (
? employee_id, division_id, job_id, first_name, last_name, salary
) VALUES (
? 30, 'SAL', 'MGR', 'Melanie', 'York', 255000
);

INSERT INTO employees2 (
? employee_id, division_id, job_id, first_name, last_name, salary
) VALUES (
? 31, 'SAL', 'MGR', 'Tracy', 'Yellow', 225000
);

INSERT INTO employees2 (
? employee_id, division_id, job_id, first_name, last_name, salary
) VALUES (
? 32, 'SAL', 'MGR', 'Sarah', 'White', 235000
);

INSERT INTO employees2 (
? employee_id, division_id, job_id, first_name, last_name, salary
) VALUES (
? 33, 'SAL', 'MGR', 'Terry', 'Iron', 225000
);

INSERT INTO employees2 (
? employee_id, division_id, job_id, first_name, last_name, salary
) VALUES (
? 34, 'SAL', 'MGR', 'Christine', 'Brown', 247000
);

INSERT INTO employees2 (
? employee_id, division_id, job_id, first_name, last_name, salary
) VALUES (
? 35, 'SAL', 'MGR', 'John', 'Brown', 249000
);

INSERT INTO employees2 (
? employee_id, division_id, job_id, first_name, last_name, salary
) VALUES (
? 36, 'SAL', 'MGR', 'Kelvin', 'Trenton', 255000
);

INSERT INTO employees2 (
? employee_id, division_id, job_id, first_name, last_name, salary
) VALUES (
? 37, 'BUS', 'WOR', 'Damon', 'Jones', 280000
);

?

ORACLE ROLLUP和CUBE的使用