oracle merge 剖析(剑破冰山)

最近在读剑破冰山这本书，写写博文，当做做笔记。感谢那些出书的人，感谢他们分享的知识。

1、构造测试表 d1、d2

CREATE TABLE d1 AS SELECT * FROM scott.dept;
CREATE TABLE d2 AS SELECT * FROM scott.dept;
delete from d2 where deptno in (40, 20);

SQL> select * from d1;
    DEPTNO DNAME          LOC
---------- -------------- -------------
        10 ACCOUNTING     NEW YORK
        20 RESEARCH       DALLAS
        30 SALES          CHICAGO
        40 OPERATIONS     BOSTON

SQL> select * from d2;
    DEPTNO DNAME          LOC
---------- -------------- -------------
        10 ACCOUNTING     FU JIAN
        30 SALES          CHICAGO

2、使用d1表的记录去更新d2表的记录

MERGE INTO d2
USING d1
ON (d1.deptno = d2.deptno)
WHEN MATCHED THEN
        UPDATE SET d2.loc = d1.loc || '...'
WHEN NOT MATCHED THEN
        INSERT VALUES(d1.deptno, d1.dname, d1.loc)
		
SQL> select * from d2;

    DEPTNO DNAME          LOC
---------- -------------- -------------
        10 ACCOUNTING     NEW YORK...
        30 SALES          CHICAGO...
        40 OPERATIONS     BOSTON
        20 RESEARCH       DALLAS

3、查看 merge 语句的执行计划

--将环境变到原来
SQL> rollback;
Rollback complete.

SQL> select * from d1;
    DEPTNO DNAME          LOC
---------- -------------- -------------
        10 ACCOUNTING     NEW YORK
        20 RESEARCH       DALLAS
        30 SALES          CHICAGO
        40 OPERATIONS     BOSTON

SQL> select * from d2;
    DEPTNO DNAME          LOC
---------- -------------- -------------
        10 ACCOUNTING     FU JIAN
        30 SALES          CHICAGO

--实行merge操作并查看执行计划
EXPLAIN PLAN FOR
MERGE INTO d2
USING d1
ON (d1.deptno = d2.deptno)
WHEN MATCHED THEN
        UPDATE SET d2.loc = d1.loc || '...'
WHEN NOT MATCHED THEN
        INSERT VALUES(d1.deptno, d1.dname, d1.loc);

SQL> SELECT * FROM table(DBMS_XPLAN.DISPLAY);

PLAN_TABLE_OUTPUT
-----------------------------------------------------------------------------
Plan hash value: 872405927

-----------------------------------------------------------------------------
| Id  | Operation            | Name | Rows  | Bytes | Cost (%CPU)| Time     |
-----------------------------------------------------------------------------
|   0 | MERGE STATEMENT      |      |     4 |   240 |     8  (13)| 00:00:01 |
|   1 |  MERGE               | D2   |       |       |            |          |
|   2 |   VIEW               |      |       |       |            |          |
|*  3 |    HASH JOIN OUTER   |      |     4 |   288 |     8  (13)| 00:00:01 |
|   4 |     TABLE ACCESS FULL| D1   |     4 |   120 |     3   (0)| 00:00:01 |
|   5 |     TABLE ACCESS FULL| D2   |     2 |    84 |     4   (0)| 00:00:01 |
-----------------------------------------------------------------------------

Predicate Information (identified by operation id):
---------------------------------------------------

   3 - access("D1"."DEPTNO"="D2"."DEPTNO"(+))

Note
-----
   - dynamic sampling used for this statement (level=2)

21 rows selected.

4、merge只执行update 或 只执行insert

SQL> rollback;

Rollback complete.

SQL> select * from d1;

    DEPTNO DNAME          LOC
---------- -------------- -------------
        10 ACCOUNTING     NEW YORK
        20 RESEARCH       DALLAS
        30 SALES          CHICAGO
        40 OPERATIONS     BOSTON

SQL> select * from d2;

    DEPTNO DNAME          LOC
---------- -------------- -------------
        10 ACCOUNTING     FU JIAN
        30 SALES          CHICAGO
		
MERGE INTO d2
USING d1
ON (d1.deptno = d2.deptno)
WHEN MATCHED THEN
UPDATE SET d2.loc = d1.loc || '...';

SQL> select * from d2;

    DEPTNO DNAME          LOC
---------- -------------- -------------
        10 ACCOUNTING     NEW YORK...
        30 SALES          CHICAGO...
		
SQL> rollback;
Rollback complete.

Wrote file afiedt.buf
select * from d2
MERGE INTO d2
USING d1
ON (d1.deptno = d2.deptno)
WHEN NOT MATCHED THEN
        INSERT VALUES(d1.deptno, d1.dname, d1.loc);

SQL> select * from d2;

    DEPTNO DNAME          LOC
---------- -------------- -------------
        10 ACCOUNTING     FU JIAN
        30 SALES          CHICAGO
        40 OPERATIONS     BOSTON
        20 RESEARCH       DALLAS

5、merge语句添加条件更新，只更新deptno = 10的行（提示：下面的操作都是基于开始的d1、和d2表数据）

MERGE INTO d2
USING d1
ON (d1.deptno = d2.deptno)
WHEN MATCHED THEN
        UPDATE SET d2.loc = d1.loc || '...'
        WHERE d1.deptno = 10

SQL> select * from d2;
    DEPTNO DNAME          LOC
---------- -------------- -------------
        10 ACCOUNTING     NEW YORK...
        30 SALES          CHICAGO

6、在merge中使用delete子句清除行，只删除deptno = 10的行

MERGE INTO d2
USING d1
ON (d1.deptno = d2.deptno)
WHEN MATCHED THEN
        UPDATE SET d2.loc = d1.loc || '...'
        DELETE WHERE (d2.deptno = 10)
		
SQL> select * from d2;

    DEPTNO DNAME          LOC
---------- -------------- -------------
        30 SALES          CHICAGO...

7、merge无条件insert

MERGE INTO d2
USING d1
ON (1 = 2)
WHEN NOT MATCHED THEN
         INSERT VALUES (d1.deptno, d1.dname, d1.loc)；

SQL> select * from d2;
    DEPTNO DNAME          LOC
---------- -------------- -------------
        30 SALES          CHICAGO...
        10 ACCOUNTING     NEW YORK
        20 RESEARCH       DALLAS
        30 SALES          CHICAGO
        40 OPERATIONS     BOSTON