如何有条件的分步删除数据表中的记录
作者:eygle
日期:February 22, 2005
有时候我们需要分配删除数据表的一些记录，分批提交以减少对于Undo的使用，本文提供一个简单的存储过程用于实现该逻辑。
你可以根据你的需要进行适当调整，本例仅供参考:
SQL> create table test as select * from dba_objects;
Table created.
SQL> create or replace procedure deleteTab
2 /**
3 ** Usage: run the script to create the proc deleteTab
4 ** in SQL*PLUS, type "exec deleteTab('Foo','ID>=1000000','3000');"
5 ** to delete the records in the table "Foo", commit per 3000 records.
6 **
7 **/
8 (
9 p_TableName in varchar2, -- The TableName which you want to delete from
10 p_Condition in varchar2, -- Delete condition, such as "id>=100000"
11 p_Count in varchar2 -- Commit after delete How many records
12 )
13 as
14 pragma autonomous_transaction;
15 n_delete number:=0;
16 begin
17 while 1=1 loop
18 EXECUTE IMMEDIATE
19 'delete from '||p_TableName||' where '||p_Condition||' and rownum <= :rn'
20 USING p_Count;
21 if SQL%NOTFOUND then
22 exit;
23 else
24 n_delete:=n_delete + SQL%ROWCOUNT;
25 end if;
26 commit;
27 end loop;
28 commit;
29 DBMS_OUTPUT.PUT_LINE('Finished!');
30 DBMS_OUTPUT.PUT_LINE('Totally '||to_char(n_delete)||' records deleted!');
31 end;
32 /
Procedure created.
SQL> insert into test select * from dba_objects;
6374 rows created.
SQL> /
6374 rows created.
SQL> /
6374 rows created.
SQL> commit;
Commit complete.
SQL> exec deleteTab('TEST','object_id >0','3000')
Finished!
Totally 19107 records deleted!
PL/SQL procedure successfully completed.
很简单，但是想来也有人可以用到。
