重復數據,通常有兩種:一是完全重復的記錄,也就是所有字段的值都一樣;二是部分字段值重復的記錄。
一. 刪除完全重復的記錄
完全重復的數據,通常是由於沒有設置主鍵/唯一鍵約束導致的。
測試數據:
if OBJECT_ID('duplicate_all') is not null
drop table duplicate_all
GO
create table duplicate_all
(
c1 int,
c2 int,
c3 varchar(100)
)
GO
insert into duplicate_all
select 1,100,'aaa' union all
select 1,100,'aaa' union all
select 1,100,'aaa' union all
select 1,100,'aaa' union all
select 1,100,'aaa' union all
select 2,200,'bbb' union all
select 3,300,'ccc' union all
select 4,400,'ddd' union all
select 5,500,'eee'
GO
(1) 借助臨時表
利用DISTINCT得到單條記錄,刪除源數據,然後導回不重復記錄。
如果表不大的話,可以把所有記錄導出一次,然後truncate表後再導回,這樣可以避免delete的日志操作。
if OBJECT_ID('tempdb..#tmp') is not null
drop table #tmp
GO
select distinct * into #tmp
from duplicate_all
where c1 = 1
GO
delete duplicate_all where c1 = 1
GO
insert into duplicate_all
select * from #tmp
(2) 使用ROW_NUMBER
with tmp
as
(
select *,ROW_NUMBER() OVER(PARTITION BY c1,c2,c3 ORDER BY(getdate())) as num
from duplicate_all
where c1 = 1
)
delete tmp where num > 1
如果多個表有完全重復的行,可以考慮通過UNION將多個表聯合,插到一個新的同結構的表,SQL Server會幫助去掉表和表之間的重復行。
二. 刪除部分重復的記錄
部分列重復的數據,通常表上是有主鍵的,可能是程序邏輯造成了多行數據列值的重復。
測試數據:
if OBJECT_ID('duplicate_col') is not null
drop table duplicate_col
GO
create table duplicate_col
(
c1 int primary key,
c2 int,
c3 varchar(100)
)
GO
insert into duplicate_col
select 1,100,'aaa' union all
select 2,100,'aaa' union all
select 3,100,'aaa' union all
select 4,100,'aaa' union all
select 5,500,'eee'
GO
(1) 唯一索引
唯一索引有個忽略重復建的選項,在創建主鍵約束/唯一鍵約束時都可以使用這個索引選項。
if OBJECT_ID('tmp') is not null
drop table tmp
GO
create table tmp
(
c1 int,
c2 int,
c3 varchar(100),
constraint UQ_01 unique(c2,c3) with(IGNORE_DUP_KEY = ON)
)
GO
insert into tmp
select * from duplicate_col
select * from tmp
(2) 借助主鍵/唯一鍵來刪除
通常會選擇主鍵/唯一鍵的最大/最小值保留,其他行刪除。以下只保留重復記錄中c1最小的行。
delete from duplicate_col
where exists(select 1 from duplicate_col b where duplicate_col.c1 > b.c1 and (duplicate_col.c2 = b.c2 and duplicate_col.c3 = b.c3))
--或者
delete from duplicate_col
where c1 not in (select min(c1) from duplicate_col group by c2,c3)
如果要保留重復記錄中的第N行,可以參考05.取分組中的某幾行。
(3) ROW_NUMBER
和刪除完全重復記錄的寫法基本一樣。
with tmp
as
(
select *,ROW_NUMBER() OVER(PARTITION BY c2,c3 ORDER BY(getdate())) as num
from duplicate_col
)
delete tmp where num > 1
select * from duplicate_col