本文整理自网络，侵删。
方法一. 使用Group子查询查出重复的条目 (这里用max得到大的ID)
DELETE FROM [user] WHERE ID NOT IN
(SELECT MAX(ID) FROM [user] GROUP BY [name])
方法二. 使用自连接
delete from table where ID = (
select ID from table t1 , table t2 where t1.name = t2.name and t1.ID > t2.ID )
方法三. 使用重新插入 Ignore dup key
/*四板斧――轻松消除重复记录*/
--第一板斧――建立一张具有相同结构的临时表
--建立临时表
select * into tempCard from cmc_contactCard where 1=2
--第二板斧――为该表加上索引，并使其忽略重复的值
--在临时表上建立忽略重复值的索引
CREATE UNIQUE
INDEX Index_Name ON tempCard (Enterprise_Name, Contact_Name)
WITH
IGNORE_DUP_KEY
--第三板斧――拷贝产品信息到临时表
--插入源表数据到临时表
insert into tempCard select * from cmc_contactCard
--第四板斧――将新的数据导入原表
--清空源表记录
truncate table cmc_contactCard
--回写所有消掉重复值的记录
insert into cmc_contactCard select * from tempCard
--删除临时表
drop table tempCard
方法四: 使用游标
--创建表并填充数据
create table b_dist(id int,name varchar(20), dept varchar(20))
insert into b_dist values(1,'abc','sales')
insert into b_dist values(2,'abc','sales')
insert into b_dist values(3,'abc','market')
insert into b_dist values(4,'abc','sales')
insert into b_dist values(5,'abc','infos')
insert into b_dist values(6,'abc','market')
insert into b_dist values(7,'abc','backup')
--使用游标逐个比较, 把关键的字段要排序
declare @id int
declare @name varchar(20)
declare @dept varchar(20)
declare @prevname varchar(20)
declare @prevdept varchar(20)
declare cur cursor for select id, name, dept from b_dist order by name, dept
open cur
fetch next from cur into @id, @name, @dept
while @@FETCH_STATUS = 0
begin
if (@dept=@prevdept) and (@name=@prevname)
delete from b_dist where id=@id
select @prevname=@name, @prevdept=@dept
fetch next from cur into @id, @name, @dept
end
close cur
deallocate cur
相关阅读 >>
Delphi android device information
更多相关阅读请进入《Delphi》频道 >>
