ORACLE 通过连接查询更新 update select

注意: 关键的地方是where 语句的加入. 在11G中, 如果不加11G , 或造成除匹配的行数更新为相应的值之后, 其余的会变成负数.

所以,执行前需要测试, 普通办法就是: 先查看需要更新的数量即连接的数量究竟有多少行, 然后update 之后多少行, 才能确定条件是否正确.

最终完成的2个 SQL, 经过检验, 放心食用:

1 : 通过关联表查询获取需要更新的多列数据更新. 注意Exists子句的必要性

UPDATE SFISM4.R_DT_DIP_TRACEABILITY_T B
SET (B.CURRENT_QTY, B.UPDATE_TIME, B.UPDATE_USER)=
(SELECT DECODE(SIGN(B.CURRENT_QTY - A.KP_COUNT),
-1,
0,
B.CURRENT_QTY - A.KP_COUNT),
SYSDATE,
:VUSER
FROM SFIS1.C_DT_DIP_BOM_T A, SFISM4.R_MO_BASE_T C
WHERE A.BOM_NO = B.BOM_NO
AND B.BOM_NO = C.KEY_PART_NO
AND A.KEY_PART_NO = B.KEY_PART_NO
AND C.MO_NUMBER =:VBOM
AND B.TRAC_LOT =:VTRACELOT)
WHERE B.TRAC_LOT =:VTRACELOT
AND EXISTS (SELECT 1
FROM SFIS1.C_DT_DIP_BOM_T AA
WHERE B.BOM_NO = AA.BOM_NO
AND B.KEY_PART_NO = AA.KEY_PART_NO)
UPDATE SFISM4.R_REEL_TRACKING_T A
SET A.REMAIN_QTY =
(SELECT B.REEL_QTY
FROM SFISM4.R_MMS_STOCK_SMT_SA_DETAIL_T B
WHERE A.REEL_ID = B.REEL_ID
AND A.REMAIN_QTY <0
AND A.REMAIN_QTY <> B.REEL_QTY)
WHERE EXISTS (SELECT 1
FROM SFISM4.R_MMS_STOCK_SMT_SA_DETAIL_T B
WHERE A.REEL_ID = B.REEL_ID
AND A.REMAIN_QTY <0
AND A.REMAIN_QTY <> B.REEL_QTY);

2: 如上简化版.
UPDATE SFISM4.R_REEL_TRACKING_T A
SET A.REMAIN_QTY =
(SELECT B.REEL_QTY
FROM SFISM4.R_MMS_STOCK_SMT_SA_DETAIL_T B
WHERE A.REEL_ID = B.REEL_ID
AND A.REMAIN_QTY <0
AND A.REMAIN_QTY <> B.REEL_QTY)
WHERE EXISTS (SELECT 1
FROM SFISM4.R_MMS_STOCK_SMT_SA_DETAIL_T B
WHERE A.REEL_ID = B.REEL_ID
AND A.REMAIN_QTY <0
AND A.REMAIN_QTY <> B.REEL_QTY);

从一开始有了想法, 找到的相应one step by step实验过程.

SQL> select * from wwm2; --要更新的表 , 多一点.

TOWN                         ID
-------------------- ----------
222                         222
111                         111
ww'jj                       111
llll                       1111
dddd                       2222
lllldf                      111
lllldf                      111
dsafdf                      111
3435                        111
ljjjjj                      222
dsafdf                      111

TOWN                         ID
-------------------- ----------
3435                        111
ljjjjj                      222

SQL> select * from wwm5; --更新的条件表 , 少一点

TOWN                         ID
-------------------- ----------
lllldf                      111
test                       9984

SQL> select wwm2.* from wwm2,wwm5 where wwm2.id=wwm5.id
2 /

TOWN                         ID
-------------------- ----------
111                         111
ww'jj                       111
lllldf                      111
lllldf                      111
dsafdf                      111
3435                        111
dsafdf                      111
3435                        111

8 rows selected.

所以,每次需要更新8条数据就是正确的. 这一步是验证更新是不是对, 错的关键.!!!

相信程序员是通过以下类似的SQL更新的,这是错误的,因为没有加WHERE
SQL> update wwm2 set wwm2.town=(select wwm5.town from wwm5 where wwm5.id=wwm2.id)
2 /

13 rows updated.

SQL> select * from wwm2;

TOWN                         ID
-------------------- ----------
                            222
lllldf                      111
lllldf                      111
                           1111
                           2222
lllldf                      111
lllldf                      111
lllldf                      111
lllldf                      111
                            222
lllldf                      111

TOWN                         ID
-------------------- ----------
lllldf                      111
                            222
13 rows selected.

可以看到13条记录被更新,符合条件的更新正确,不符合条件的也更新为NULL.以下是正确的方法

方法一:
SQL> update wwm2
2 set town=(select town from wwm5 where wwm5.id=wwm2.id)
3 where id=(select wwm5.id from wwm5 where wwm5.id=wwm2.id)
4 /

8 rows updated.

方法二:    与方法一道理相同,这里需要掌握EXIST的相关用法.
SQL> update wwm2
   set town=(select town from wwm5 where wwm5.id=wwm2.id)
   where exists (select 1 from wwm5 where wwm5.id=wwm2.id)
8 rows updated.

方法三:
SQL> update (select a.town atown,a.id aid,b.town btown,b.id bid from wwm2 a,wwm5 b where a.id=b.id)
2 set atown=btown
3 /
set atown=btown
*
ERROR at line 2:
ORA-01779: cannot modify a column which maps to a non key-preserved table

1* alter table wwm5 add primary key (id)
SQL> /

Table altered.

1 update (select a.town atown,a.id aid,b.town btown,b.id bid from wwm2 a,wwm5 b where a.id=b.id) -- 表连接之后, 更新视图的方式很好理解. 但是需要primary key.
2* set atown=btown
SQL> /

8 rows updated.

这种方法的局限性就是需要PRIMARY 的支持.! 需要更新被更新表有主键, 连接栏位是不是需要呢>?

方法四: --万金油. 但是数据库顾问说效能没有使用表连接好.
1 declare
2 cursor cur_wwm is select town,id from wwm5; -- 驱动表
3 begin
4     for my_wwm in cur_wwm loop
5     update wwm2 set town=my_wwm.town -- 被驱动表.
6     where id=my_wwm.id;
7     end loop;
8* end;
SQL> /

PL/SQL procedure successfully completed.

SQL> select * from wwm2;

TOWN                         ID
-------------------- ----------
222                         222
lllldf                      111
lllldf                      111
llll                       1111
dddd                       2222
lllldf                      111
lllldf                      111
lllldf                      111
lllldf                      111
ljjjjj                      222
lllldf                      111

TOWN                         ID
-------------------- ----------
lllldf                      111
ljjjjj                      222

这个方法是最灵活的了.

方法五:

注意,方法五只能适用于WWM5是WWM2的子集的时候. (新特性 merge Into,哎呀! 不想学.....)
1   merge into wwm2
2   using (select town,id from wwm5) b
3   on (wwm2.id=b.id)
4   when matched then update set town=b.town
5* when not matched then insert (town,id) values (null,null)
SQL> /

9 rows merged.

SQL> select * from wwm2;

TOWN                         ID
-------------------- ----------
                                  ---注意这个地方,被插入了一个空值.因为WWM5的ID=9984在WWM2中不能匹配,根本原因是ORACLE9必须有WHEN NOT MATCHED子句,但是ORACLE10可以不许要,也就是ORACLE10可以不写WHEN NOT MATCHED ,就不必插入NULL值了,为解决这个问题,下一步会DELETE WWM5的ID=9984,这样一来就不会执行WHEN NOT MATCHED
222                         222
lllldf                      111
lllldf                      111
llll                       1111
dddd                       2222
lllldf                      111
lllldf                      111
lllldf                      111
lllldf                      111
ljjjjj                      222

TOWN                         ID
-------------------- ----------
lllldf                      111
lllldf                      111
ljjjjj                      222

14 rows selected.

SQL> delete from wwm5 where id=9984;

1 row deleted.

SQL> 1   merge into wwm2
SQL>   2   using (select town,id from wwm5) b
SQL>   3   on (wwm2.id=b.id)
SQL>   4   when matched then update set town=b.town
SQL>   5* when not matched then insert (town,id) values (null,null)
SQL> /

8 rows merged.

以上就是5种关连更新的例子了,希望能给开发人员解惑.

说明：如果select 子句可以返回多行记录，但返回适合where条件的记录只能是唯一的，否则将会报返回单行的select子句返回多行的错误，因为update只能跟据此处的where子句（内层where)进行相应记录的匹配更新，一次只能是一条。

参考网址: http://www.blogjava.net/Jhonney/archive/2010/06/25/324503.html

来自为知笔记(Wiz)

posted @ 2015-02-05 09:25 hijushen 阅读(29380) 评论(0) 编辑收藏举报

刷新页面返回顶部

hijushen

ORACLE 通过连接查询更新 update select

公告