快捷導(dǎo)航

多表關(guān)聯(lián)同時(shí)更新多條不同的記錄方法分享

更新時(shí)間：2011年10月31日 23:47:43 作者：

因?yàn)轫?xiàng)目要求實(shí)現(xiàn)一次性同時(shí)更新多條不同的記錄的需求，和同事討論了一個(gè)比較不錯(cuò)的方案，這里供大家參考下

以下為測(cè)試?yán)印?
1.首先創(chuàng)建兩張臨時(shí)表并錄入測(cè)試數(shù)據(jù)：

 
create table #temptest1 
( 
id int, 
name1 varchar(50), 
age int 
) 
create table #temptest2 
( 
id int, 
name1 varchar(50), 
age int 
) 

查詢出此時(shí)的表數(shù)據(jù)為：

#temptest1 #temptest2

2.現(xiàn)在要將#temptest2中的年齡更新到相應(yīng)的#temptest1中的年齡。

其實(shí)就是讓[表1]中ID為1的年齡改成19，同時(shí)ID為2的年齡改成20。

當(dāng)然這里的要求是只用一句SQL，不能用循環(huán)。

結(jié)果如下：

實(shí)現(xiàn)方法如下：

Update t1

Set t1 .age = t2.age

From #temptest1 t1

Join #temptest2 t2

On t1.id = t2.id

(補(bǔ)充)Sql Server 2008 Merge命令寫法：

merge into #temptest1 t1
using(select age,id from #temptest2) t2
on t1.id = t2.id
when matched then
update set t1.age = t2.age

是不是挺有趣的Sql。

如何一次性更新多條不同值的記錄
標(biāo)題可能沒說清楚，假設(shè)有這樣兩張表：

復(fù)制代碼代碼如下:

 
create table testA( 
id number, 
eng varchar2(3), 
chi varchar2(3) 
) 
create table testB( 
id number, 
eng varchar2(3), 
chi varchar2(3)， 
anythingother varchar2(1) 
) 

現(xiàn)有記錄
testA：
ID ENG CHI
===============
1 a 一
2 b 二
3 c 三
testB:
ID ENG CHI ANY....
=================
1 d 四
2 e 五
3 f 六
我想把testB中的記錄的ENG,CHI字段更新到testA中去，以ID來對(duì)應(yīng)。

CODE:

SQL> set autot on
SQL> update ta set ta.b=(select tb.b from tb where ta.a=tb.a) where exists (select 1 from tb where ta.a=tb.a);
已更新4行。
已用時(shí)間: 00: 00: 00.01
執(zhí)行計(jì)劃
----------------------------------------------------------
Plan hash value: 1137212925
--------------------------------------------------------------------------------
| Id | Operation | Name | Rows | Bytes | Cost (%CPU)| Time |
--------------------------------------------------------------------------------
| 0 | UPDATE STATEMENT | | 5 | 165 | 20 (30)| 00:00:01 |
| 1 | UPDATE | TA | | | | |
|* 2 | HASH JOIN SEMI | | 5 | 165 | 5 (20)| 00:00:01 |
| 3 | TABLE ACCESS FULL | TA | 5 | 100 | 2 (0)| 00:00:01 |
| 4 | VIEW | VW_SQ_1 | 4 | 52 | 2 (0)| 00:00:01 |
| 5 | TABLE ACCESS FULL| TB | 4 | 52 | 2 (0)| 00:00:01 |
|* 6 | TABLE ACCESS FULL | TB | 1 | 26 | 2 (0)| 00:00:01 |
--------------------------------------------------------------------------------
Predicate Information (identified by operation id):
---------------------------------------------------
2 - access("TA"."A"="ITEM_1")
6 - filter("TB"."A"=:B1)
Note
-----
- dynamic sampling used for this statement (level=2)
統(tǒng)計(jì)信息
----------------------------------------------------------
0 recursive calls
4 db block gets
23 consistent gets
0 physical reads
1004 redo size
840 bytes sent via SQL*Net to client
856 bytes received via SQL*Net from client
3 SQL*Net roundtrips to/from client
1 sorts (memory)
0 sorts (disk)
4 rows processed
SQL> update ta set ta.b=(select tb.b from tb where ta.a=tb.a) where ta.a= (select tb.a from tb where ta.a=tb.a);
已更新4行。
已用時(shí)間: 00: 00: 00.00
執(zhí)行計(jì)劃
----------------------------------------------------------
Plan hash value: 3571861550
----------------------------------------------------------------------------
| Id | Operation | Name | Rows | Bytes | Cost (%CPU)| Time |
----------------------------------------------------------------------------
| 0 | UPDATE STATEMENT | | 1 | 20 | 7 (15)| 00:00:01 |
| 1 | UPDATE | TA | | | | |
|* 2 | FILTER | | | | | |
| 3 | TABLE ACCESS FULL| TA | 5 | 100 | 2 (0)| 00:00:01 |
|* 4 | TABLE ACCESS FULL| TB | 1 | 13 | 2 (0)| 00:00:01 |
|* 5 | TABLE ACCESS FULL | TB | 1 | 26 | 2 (0)| 00:00:01 |
----------------------------------------------------------------------------
Predicate Information (identified by operation id):
---------------------------------------------------
2 - filter("TA"."A"= (SELECT "TB"."A" FROM "TB" "TB" WHERE
"TB"."A"=:B1))
4 - filter("TB"."A"=:B1)
5 - filter("TB"."A"=:B1)
Note
-----
- dynamic sampling used for this statement (level=2)
統(tǒng)計(jì)信息
----------------------------------------------------------
11 recursive calls
1 db block gets
53 consistent gets
0 physical reads
588 redo size
840 bytes sent via SQL*Net to client
858 bytes received via SQL*Net from client
3 SQL*Net roundtrips to/from client
1 sorts (memory)
0 sorts (disk)
4 rows processed

如果 create unique index tb_a_uidx on tb(a);

[Copy to clipboard] [ - ]

CODE:

SQL> update (select ta.b tab1 ,tb.b tbb from ta,tb where ta.a=tb.a) set tab1=tbb;
已更新4行。
已用時(shí)間: 00: 00: 00.01
執(zhí)行計(jì)劃
----------------------------------------------------------
Plan hash value: 1761655026
----------------------------------------------------------------------------
| Id | Operation | Name | Rows | Bytes | Cost (%CPU)| Time |
----------------------------------------------------------------------------
| 0 | UPDATE STATEMENT | | 4 | 184 | 5 (20)| 00:00:01 |
| 1 | UPDATE | TA | | | | |
|* 2 | HASH JOIN | | 4 | 184 | 5 (20)| 00:00:01 |
| 3 | TABLE ACCESS FULL| TB | 4 | 104 | 2 (0)| 00:00:01 |
| 4 | TABLE ACCESS FULL| TA | 5 | 100 | 2 (0)| 00:00:01 |
----------------------------------------------------------------------------
Predicate Information (identified by operation id):
---------------------------------------------------
2 - access("TA"."A"="TB"."A")
Note
-----
- dynamic sampling used for this statement (level=2)
統(tǒng)計(jì)信息
----------------------------------------------------------
8 recursive calls
4 db block gets
17 consistent gets
0 physical reads
1004 redo size
840 bytes sent via SQL*Net to client
827 bytes received via SQL*Net from client
3 SQL*Net roundtrips to/from client
3 sorts (memory)
0 sorts (disk)
4 rows processed