[Oracle] Group By 語句的擴展 - Rollup、Cube和Grouping Sets

經常寫SQL語句的人應該知道Group by語句的主要用法是進行分類彙總，下面是一種它最常見的用法（根據部門、職位分別統計業績）：

SELECT  a.dname,b.job,SUM(b.sal) sum_sal
FROM dept a,emp b
WHERE a.deptno = b.deptno
GROUP  BY a.dname,b.job;

DNAME          JOB          SUM_SAL
-------------- --------- ----------
SALES          MANAGER         2850
SALES          CLERK            950
SALES          SALESMAN        5600
ACCOUNTING     MANAGER         2450
ACCOUNTING     PRESIDENT       5000
ACCOUNTING     CLERK           1300
RESEARCH       MANAGER         2975
RESEARCH       ANALYST         6000
RESEARCH       CLERK           1900

這時候，如果有人跑過來跟你說：我除了以上數據之外，還要每個部門總的業績以及所有部門加起來的業績，這時候你很可能會想到如下的笨方法（union all）：

select * from (
SELECT  a.dname,b.job,SUM(b.sal) sum_sal
FROM dept a,emp b
WHERE a.deptno = b.deptno
GROUP  BY a.dname,b.job
UNION ALL
--實現了部門的小計
SELECT  a.dname,NULL, SUM(b.sal) sum_sal
FROM dept a,emp b
WHERE a.deptno = b.deptno
GROUP  BY a.dname
UNION ALL
--實現了所有部門總的合計
SELECT  NULL,NULL, SUM(b.sal) sum_sal
FROM dept a,emp b
WHERE a.deptno = b.deptno)
order by dname;

DNAME          JOB          SUM_SAL
-------------- --------- ----------
ACCOUNTING     CLERK           1300
ACCOUNTING     MANAGER         2450
ACCOUNTING     PRESIDENT       5000
ACCOUNTING                     8750
RESEARCH       CLERK           1900
RESEARCH       MANAGER         2975
RESEARCH       ANALYST         6000
RESEARCH                      10875
SALES          CLERK            950
SALES          MANAGER         2850
SALES          SALESMAN        5600
SALES                          9400
                              29025

union all 合併笨辦法產生的執行計劃
-------------------------------------------------------------------------------
Plan hash value: 2979078843
-------------------------------------------------------------------------------
| Id  | Operation              | Name | Rows  | Bytes | Cost (%CPU)| Time     |
-------------------------------------------------------------------------------
|   0 | SELECT STATEMENT       |      |    29 |   812 |    23  (22)| 00:00:01 |
|   1 |  SORT ORDER BY         |      |    29 |   812 |    23  (22)| 00:00:01 |
|   2 |   VIEW                 |      |    29 |   812 |    22  (19)| 00:00:01 |
|   3 |    UNION-ALL           |      |       |       |            |          |
|   4 |     HASH GROUP BY      |      |    14 |   756 |     8  (25)| 00:00:01 |
|*  5 |      HASH JOIN         |      |    14 |   756 |     7  (15)| 00:00:01 |
|   6 |       TABLE ACCESS FULL| DEPT |     4 |    88 |     3   (0)| 00:00:01 |
|   7 |       TABLE ACCESS FULL| EMP  |    14 |   448 |     3   (0)| 00:00:01 |
|   8 |     HASH GROUP BY      |      |    14 |   672 |     8  (25)| 00:00:01 |
|*  9 |      HASH JOIN         |      |    14 |   672 |     7  (15)| 00:00:01 |
|  10 |       TABLE ACCESS FULL| DEPT |     4 |    88 |     3   (0)| 00:00:01 |
|  11 |       TABLE ACCESS FULL| EMP  |    14 |   364 |     3   (0)| 00:00:01 |
|  12 |     SORT AGGREGATE     |      |     1 |    39 |            |          |
|* 13 |      HASH JOIN         |      |    14 |   546 |     7  (15)| 00:00:01 |
|  14 |       TABLE ACCESS FULL| DEPT |     4 |    52 |     3   (0)| 00:00:01 |
|  15 |       TABLE ACCESS FULL| EMP  |    14 |   364 |     3   (0)| 00:00:01 |
-------------------------------------------------------------------------------

其實，如果你知道Group By的Rollup擴展的話，這種需求只是小case：

SELECT  a.dname,b.job, SUM(b.sal) sum_sal
FROM dept a,emp b 
WHERE a.deptno = b.deptno
GROUP  BY ROLLUP(a.dname,b.job);

DNAME          JOB          SUM_SAL
-------------- --------- ----------
SALES          CLERK            950
SALES          MANAGER         2850
SALES          SALESMAN        5600
SALES                          9400
RESEARCH       CLERK           1900
RESEARCH       ANALYST         6000
RESEARCH       MANAGER         2975
RESEARCH                      10875
ACCOUNTING     CLERK           1300
ACCOUNTING     MANAGER         2450
ACCOUNTING     PRESIDENT       5000
ACCOUNTING                     8750
                              29025

rollup寫法產生的執行計劃
-----------------------------------------------------------------------------
Plan hash value: 1037965942
-----------------------------------------------------------------------------
| Id  | Operation            | Name | Rows  | Bytes | Cost (%CPU)| Time     |
-----------------------------------------------------------------------------
|   0 | SELECT STATEMENT     |      |    14 |   756 |     8  (25)| 00:00:01 |
|   1 |  SORT GROUP BY ROLLUP|      |    14 |   756 |     8  (25)| 00:00:01 |
|*  2 |   HASH JOIN          |      |    14 |   756 |     7  (15)| 00:00:01 |
|   3 |    TABLE ACCESS FULL | DEPT |     4 |    88 |     3   (0)| 00:00:01 |
|   4 |    TABLE ACCESS FULL | EMP  |    14 |   448 |     3   (0)| 00:00:01 |
-----------------------------------------------------------------------------

可以發現，這種方法不但SQL書寫方便，性能也能得到提高。

這時候，如果又有人跑過來說：除了以上數據，他還需要每個職位總的業績，你只要把rollup換成cube就可以了，如下所示：

-- CUBE分組
SELECT  a.dname,b.job, SUM(b.sal) sum_sal
FROM dept a,emp b 
WHERE a.deptno = b.deptno
GROUP  BY CUBE(a.dname,b.job);

DNAME          JOB          SUM_SAL
-------------- --------- ----------
                              29025
               CLERK           4150
               ANALYST         6000
               MANAGER         8275
               SALESMAN        5600
               PRESIDENT       5000
SALES                          9400
SALES          CLERK            950
SALES          MANAGER         2850
SALES          SALESMAN        5600
RESEARCH                      10875
RESEARCH       CLERK           1900
RESEARCH       ANALYST         6000
RESEARCH       MANAGER         2975
ACCOUNTING                     8750
ACCOUNTING     CLERK           1300
ACCOUNTING     MANAGER         2450
ACCOUNTING     PRESIDENT       5000

從上面可以看出：cube比rollup的展現的粒度更細一些。

這時候，如果又有人跑過來說：他不需要那麼細的數據，只需要彙總的數據，可以使用Grouping Sets：

---GROUPING SETS分組
SELECT to_char(b.hiredate,'yyyy') hire_year,a.dname,b.job, SUM(sal) sum_sal
FROM dept a,emp b 
WHERE a.deptno = b.deptno
GROUP BY GROUPING SETS(to_char(b.hiredate,'yyyy'),a.dname,b.job);

HIRE DNAME          JOB          SUM_SAL
---- -------------- --------- ----------
1987                                4100
1980                                 800
1982                                1300
1981                               22825
     ACCOUNTING                     8750
     RESEARCH                      10875
     SALES                          9400
                    CLERK           4150
                    SALESMAN        5600
                    PRESIDENT       5000
                    MANAGER         8275
                    ANALYST         6000