共计 1468 个字符,预计需要花费 4 分钟才能阅读完成。
利用 SQL 对数据库实行数据拆分与组合实现提供以下几种方案:
方法一:
WITH CTE AS (
SELECT A.Id,A.[Uid],UserName FROM (
SELECT A.[id], REPLACE(Split.a.value(‘.’, ‘NVARCHAR(max)’),””,”) AS [Uid]
FROM
(
SELECT [id], CAST (‘<M>’ + REPLACE(Uid, ‘,’, ‘</M><M>’) + ‘</M>’ AS XML) AS [Uid] FROM A
) AS A CROSS APPLY [Uid].nodes (‘/M’) AS Split(a)
) A INNER JOIN dbo.B B ON A.[Uid]=B.[Uid]
)
SELECT DISTINCT classList=stuff((select ‘,’ +UserName from CTE B WHERE A.Id=B.Id for xml path(”)),1,1,”) FROM CTE A
方法二:
SELECT t.id,classList=STUFF((
SELECT ‘,’+tt1.UserName
FROM A tt left join B tt1 ON ”+tt.Uid+’,’ like ‘%’+tt1.Uid+’,%’
WHERE tt.id=t.id FOR XML PATH(”)),1,1,”)
FROM A t
ORDER BY t.id
方法三:
select id, (select UserName from B where Uid=t02.item) c into #ttt from dbo.A as t01 cross apply dbo.split(t01.Uid,’,’) as t02
select id,stuff((select ‘,’+c from dbo.#ttt as tb01 where tb01.id=tb02.id for xml path(”)),1,1,”) as classList
from dbo.#ttt as tb02 group by id
drop table #ttt
方法三需要添加内置方法:
create function split(
@string varchar(255),
@separator varchar(255)
)returns @array table(item varchar(255))
as
begin
declare @begin int,@end int,@item varchar(255)
set @begin = 1
set @end=charindex(@separator,@string,@begin)
while(@end<>0)
begin
set @item = substring(@string,@begin,@end-@begin)
insert into @array(item) values(@item)
set @begin = @end+1
set @end=charindex(@separator,@string,@begin)
end
set @item = substring(@string,@begin,len(@string)+1-@begin)
if (len(@item)>0)
insert into @array(item) values(substring(@string,@begin,len(@string)+1-@begin))
return
end
下面们直接看结果吧:
表 A:
表 B:
最终结果:
跟根据讨论给出的方法,我个人给出这样的三种,如果有不对之处,请多多指教。
本文永久更新链接地址 :http://www.linuxidc.com/Linux/2016-08/133779.htm