查询中的 TSQL 顺序/序列组

Posted 2023-03-29

【中文标题】查询中的 TSQL 顺序/序列组

【英文标题】：TSQL Order/Sequence groups in a query

【发布时间】：2021-12-26 09:43:05

【问题描述】：

我有以下数据...

uid	groupid
1	0
2	0
3	1
4	1
5	1
6	1
7	0
8	0
9	2
10	2
11	2

我想对组进行唯一排序，以便获得....

uid	groupid	newGroupId
1	0	0
2	0	0
3	1	1
4	1	1
5	1	1
6	1	1
7	0	2
8	0	2
9	2	3
10	2	3
11	2	3

谁能帮我在 SQL (SQL Server) 中做到这一点

【问题讨论】：

uniquely order the groups 是什么意思？

【参考方案1】：

with cte as 
(
    select * from (values 
        (1  , 0),
        (2  , 0),
        (3  , 1),
        (4  , 1),
        (5  , 1),
        (6  , 1),
        (7  , 0),
        (8  , 0),
        (9  , 2),
        (10 , 2),
        (11 , 2))
    t([uid], [groupid])
),
cte2 as
(
    select 
        [uid], [groupid],
        newGroupId = 
            row_number() over(order by [uid]) 
            - 
            row_number() over(partition by [groupid] order by [uid])    
    from 
        cte
)
select
    [uid], [groupid],
    newGroupId = dense_rank() over(order by newGroupId) - 1
from
    cte2;

【参考方案2】：

与 Vadium 略有不同的方法，但可以考虑另一种选择：

;WITH CTE_MAIN AS
(
 select * from (values 
        (1  , 0),
        (2  , 0),
        (3  , 1),
        (4  , 1),
        (5  , 1),
        (6  , 1),
        (7  , 0),
        (8  , 0),
        (9  , 2),
        (10 , 2),
        (11 , 2))
    t([uid], [groupid])
),
CTE_2 AS
(
    SELECT uid, groupid, LAG(groupid,1,-1) OVER(ORDER BY uid) LastGroupID
    FROM CTE_MAIN
)
SELECT uid, groupid, SUM(CASE WHEN GroupID <> LastGroupID THEN 1 ELSE 0 END) OVER(ORDER BY uid) - 1 NewGroup
FROM CTE_2