SQLSERVER 去重
--好久没有写博客了,发现记性越来越不好了,博客写起来,废话少说,直接上代码
--创建表 CREATE TABLE [dbo].[students]( [stu_id] [INT] IDENTITY(1,1) NOT NULL, [stu_name] [NVARCHAR](20) NULL, [stu_gender] [CHAR](1) NULL, [stu_phone] [VARCHAR](15) NULL, [stu_birthday] [DATE] NULL ) ON [PRIMARY] GO
--添加测试数据 INSERT INTO [dbo].[students]([stu_name], [stu_gender], [stu_phone], [stu_birthday]) VALUES('宋江', '1', '13142224589', '2002-05-01'); INSERT INTO [dbo].[students]([stu_name], [stu_gender], [stu_phone], [stu_birthday]) VALUES('宋江', '1', '13142224589', '2002-05-01'); INSERT INTO [dbo].[students]([stu_name], [stu_gender], [stu_phone], [stu_birthday]) VALUES('卢俊义', '1', '18942224222', '2000-04-11'); INSERT INTO [dbo].[students]([stu_name], [stu_gender], [stu_phone], [stu_birthday]) VALUES('吴用', '0', '13942224511', '2002-05-01'); INSERT INTO [dbo].[students]([stu_name], [stu_gender], [stu_phone], [stu_birthday]) VALUES('公孙胜', '1', '13442228589', '1902-05-09'); INSERT INTO [dbo].[students]([stu_name], [stu_gender], [stu_phone], [stu_birthday]) VALUES('关胜', '1', '17642224566', '2002-09-01'); INSERT INTO [dbo].[students]([stu_name], [stu_gender], [stu_phone], [stu_birthday]) VALUES('林冲', '1', '18642224989', '2011-08-07'); INSERT INTO [dbo].[students]([stu_name], [stu_gender], [stu_phone], [stu_birthday]) VALUES('关胜', '1', '17642224566', '2002-09-01'); INSERT INTO [dbo].[students]([stu_name], [stu_gender], [stu_phone], [stu_birthday]) VALUES('林冲', '1', '18642224989', '2011-08-07'); GO --查询重复数据 SELECT S.stu_name,S.stu_gender,S.stu_phone,S.stu_birthday FROM dbo.students S GROUP BY S.stu_name,S.stu_gender,S.stu_phone,S.stu_birthday HAVING COUNT(1)>1 --根据某个字段去重 SELECT stu_name,s.stu_phone FROM ( SELECT stu_name,stu_phone ,ROW_NUMBER() OVER (PARTITION BY stu_name ORDER BY stu_phone) rn FROM dbo.students )AS s WHERE rn=1 --查询去重,写法1 SELECT DISTINCT stu_name,stu_gender,stu_phone,stu_birthday FROM dbo.students --查询去重,写法2 SELECT stu_id, stu_name,stu_gender,stu_phone,stu_birthday FROM dbo.students WHERE stu_id IN(SELECT MAX(stu_id) FROM dbo.students GROUP BY stu_name,stu_gender,stu_phone,stu_birthday) --删除重复的数据,方法1 DELETE FROM dbo.students WHERE stu_id NOT IN(SELECT MIN(stu_id) FROM dbo.students GROUP BY stu_name,stu_gender,stu_phone,stu_birthday) --删除重复的数据,方法2 DELETE FROM dbo.students WHERE NOT EXISTS(SELECT 1 FROM (SELECT MIN(S.stu_id) AS id FROM dbo.students S GROUP BY S.stu_name,S.stu_gender,S.stu_phone,S.stu_birthday) AS G WHERE students.stu_id=g.id) --删除重复的数据,方法3 CREATE TABLE #TBL(ID INT PRIMARY KEY); INSERT INTO #TBL(ID) SELECT MIN(stu_id) FROM dbo.students GROUP BY stu_name,stu_gender,stu_phone,stu_birthday; DELETE FROM dbo.students WHERE stu_id NOT IN(SELECT T.ID FROM #TBL T) DROP TABLE #TBL --删除重复的数据,方法4 DELETE dbo.students FROM dbo.students AS aa LEFT JOIN ( SELECT MIN(stu_id) AS id FROM dbo.students GROUP BY stu_name,stu_gender,stu_phone,stu_birthday ) AS bb ON aa.stu_id = bb.id WHERE bb.id is null --无唯一列(或主键)去重 --ALTER TABLE dbo.students DROP CONSTRAINT PK_students ALTER TABLE dbo.students drop column stu_id;--创建数据,删除stu_id --无唯一列(或主键)去重,只保留一行,方法1:查询去重后的数据后放到另一个表,删除原表,然后重命名 --注意: 更改对象名的任一部分都可能会破坏脚本和存储过程。 SELECT DISTINCT * INTO dbo.students2 FROM dbo.students; DROP TABLE dbo.students; EXEC sp_rename 'students2', 'students'; --无唯一列(或主键)去重,只保留一行,方法2:增加唯一列 ALTER TABLE dbo.students ADD stu_id INT IDENTITY(1,1); DELETE FROM dbo.students WHERE stu_id NOT IN(select stu_id FROM (SELECT stu_id,ROW_NUMBER() OVER(PARTITION BY stu_name,stu_gender,stu_phone,stu_birthday ORDER BY stu_name) AS rn FROM dbo.students ) s WHERE rn=1); --无唯一列(或主键)去重,只保留一行,方法3:将重复以及不重复数据去重插入新表,删除旧表相应数据,再把新表数据插入旧表 SELECT DISTINCT * INTO students2 FROM dbo.students WHERE stu_birthday>='2002-09-01';--条件 DELETE FROM dbo.students WHERE stu_birthday>='2002-09-01'; INSERT INTO dbo.students(stu_name, stu_gender, stu_phone, stu_birthday) SELECT * FROM dbo.students2; DROP TABLE dbo.students2