Question

SELECT * 
FROM (SELECT *, 
             ROW_NUMBER() 
             OVER (
                  PARTITION BY ResidentID, PostedON, PayerTypeID    
                  ORDER BY StatementNumber
                  ) seq
    FROM Statements
    ) s
        INNER JOIN person p ON s.ResidentID = p.ID
WHERE seq > 1
AND p.FacilityID = 275
ORDER BY s.ResidentID

有一段时间的腌制将其转换为DELETE。有人知道为什么吗？

这是原始问题：

好的，所以我的代码中的一个缺陷创建了许多重复的语句，但费用相同。它们具有相同的ResidentID和相同的发布日期，但是ID和StatementID不同

ID  |  ResidentID  |  StatementID  |  PostedON
1         4039            10          06-15-18
2         4039            11          06-15-18
3         4039            12          06-15-18

4         4039            20          06-20-18
5         4039            21          06-20-18
6         4039            22          06-20-18

7         3456            13          06-15-18
8         3456            14          06-15-18
9         3456            15          06-15-18

10        3456            23          06-21-18
11        3456            24          06-21-18

我有一长串的很多ResidentID，这些ID在许多日期都有重复。如何删除除每个residentID每个唯一日期的StatementID最低的行以外的所有行

除了与该日期/ residentID关联的最小StatementID，我想删除所有重复项

编辑：

实际上，我可以使用更新命令，因为此表中有一个软删除列，但仍然无法使其正常工作。我很可悲。

Answer 1

假设您需要从状态菜单中删除，而不是个人那个StatementNumber是PRIMARY KEY

DELETE
FROM
Statements SX
WHERE
SX.StatementNumber IN 
(SELECT s.StatementNumber  
FROM (SELECT *, 
             ROW_NUMBER() 
             OVER (
                  PARTITION BY ResidentID, PostedON, PayerTypeID    
                  ORDER BY StatementNumber
                  ) seq
    FROM Statements
    ) s
        INNER JOIN person p ON s.ResidentID = p.ID
WHERE seq > 1
AND p.FacilityID = 275
ORDER BY s.ResidentID)

Answer 2

将查询写为：

SELECT s.* 
FROM (SELECT s.*, 
             ROW_NUMBER() OVER (PARTITION BY ResidentID, PostedON, PayerTypeID    
                                ORDER BY StatementNumber
                               ) as seqnum
      FROM Statements s
      WHERE EXISTS (SELECT 1
                    FROM person p 
                    WHERE s.ResidentID = p.ID AND p.FacilityID = 275
     ) s
WHERE seq > 1;

Voila！这是可更新的CTE。您可以将其删除。我可以这样说：

with todelete as (
          SELECT s.*, 
                 ROW_NUMBER() OVER (PARTITION BY ResidentID, PostedON, PayerTypeID    
                                    ORDER BY StatementNumber
                                   ) as seqnum
          FROM Statements s
          WHERE EXISTS (SELECT 1
                        FROM person p 
                        WHERE s.ResidentID = p.ID AND p.FacilityID = 275
         )
delete todelete
    where seqnum > 1;

您可以随时将其更改为update：

with toupdate as (
          SELECT s.*, 
                 ROW_NUMBER() OVER (PARTITION BY ResidentID, PostedON, PayerTypeID    
                                    ORDER BY StatementNumber
                                   ) as seqnum
          FROM Statements s
          WHERE EXISTS (SELECT 1
                        FROM person p 
                        WHERE s.ResidentID = p.ID AND p.FacilityID = 275
         )
update toupdate
    set softdelete = 1
    where seqnum > 1;

如何将该SELECT语句转换为DELETE？

2 个答案: