字段1,字段2.....字段N,Status,ParentID
1,Name1....test1,1,99
1,Name1....test1,3,99
1,Name2....test2,1,101
1,Name2....test2,3,101
1,Name3....test3,2,101
1,Name1....test1,4,101
想要的结果是:
1,Name1....test1,3,99
1,Name1....test1,4,101
其实就是先按ParentID分组,再取出Status最大的记录
请大家帮忙一下啊.
字段1,字段2.....字段N,Status,ParentID
1,Name1....test1,1,99
2,Name1....test1,3,99
3,Name2....test2,1,101
4,Name2....test2,3,101
5,Name3....test3,2,101
6,Name7....test9,4,101
想要的结果是:
2,Name1....test1,3,99
6,Name7....test9,4,101
这样更清楚一点
SQL code:
---测试数据---
if object_id('[tb]') is not null drop table [tb]
go
create table [tb]([字段1] int,[字段2] varchar(5),[段N] varchar(5),[Status] int,[ParentID] int)
insert [tb]
select 1,'Name1','test1',1,99 union all
select 2,'Name1','test1',3,99 union all
select 3,'Name2','test2',1,101 union all
select 4,'Name2','test2',3,101 union all
select 5,'Name3','test3',2,101 union all
select 6,'Name7','test9',4,101
---查询---
select *
from tb t
where not exists(select 1 from tb where ParentID=t.ParentID and Status>t.Status)
---结果---
字段1
