最近在项目中遇到这样一个问题
运行sql语句：select a.contno from LIS_BUSI_TRANSACTIONS a where 1=1 and a.ruleid='16' group by a.contno having count(a.contno)>1
这条sql语句的意思是能够查询出来有多少张保单是存在重复数据。
现在的需求是将有重复的数据按照保单号进行分组，只要最先产生的那一条数据。
现在要做的事，我要写一个sql脚本对这张表中的数据进行维护：
思路：想：一条sql是肯定是办不成事的，需要借助于中间表，基本做法是：先将自己要保留的数据保存到中间表中，然后后将将有问题的数据删除掉，将中间表的数据保存到原表中，最后，再将借助的中间表删除掉就可以了，（其实中间表不删掉也是可以的，留一个备份也是不错的选择的，毕竟在实际项目中数据是不能随随便便的删除）。
sql脚本如下：
create table temp_lis_busi_transactions_a as
( SELECT min(sid) as sid,contno FROM lis_busi_transactions where invoiceflag = '00' and successflag = '1' and sum_total > 0 and contno in (select distinct contno from lis_busi_transactions where 1 = 1 and invoiceflag = '00' and successflag = '1' and sum_total > 0 group by contno having count(contno) > 1) /* and rownum=1*/ group by contno );
create table temp_lis_busi_transactions_b as select * from lis_busi_transactions where 1=2;
insert into temp_lis_busi_transactions_b
select * from lis_busi_transactions where sid in(
select sid from temp_lis_busi_transactions_a);
delete from lis_busi_transactions where invoiceflag='00' and successflag='1' and sum_total>0;
insert into lis_busi_transactions select * from temp_lis_busi_transactions_b;
drop table temp_lis_busi_transactions_a;
drop table temp_lis_busi_transactions_b;