Question

我每周需要更新2个表。我将在文件中复制一个唯一标识符，我需要根据另一个字段的Max值删除重复项，但是，可能有最大字段的重复。

TrackID  |  GrossOpens    (Amoung other fields...)  
1     |     2  
1     |     1  
2     |     5  
2     |     10  
2     |     10  
3     |     3

我能够获取我的数据以获得以下结果：

TrackID  |  GrossOpens  
1    |      2  
2    |      10  
2    |      10  
3    |      3

我需要具有匹配＆＃34; GrossOpens＆＃34;的重复项。在每组中删除。我使用以下代码来获得adove结果：

DELETE *  
FROM NL_2015 AS NL  
WHERE EXISTS ( SELECT * FROM NL_2015 DL WHERE DL.TrackID = NL.TrackID AND   (DL.GrossOpens > NL.GrossOpens OR (DL.GrossOpens = NL.GrossOpens AND DL.TrackID   > NL.TrackID)));

Answer 1

使用此查询创建行号，所有重复项的行号都优于1.

WITH dupes
AS (
  SELECT 
    ROW_NUMBER() OVER (PARTITION BY TrackID ORDER BY grossopens DESC) row
  FROM NL_2015)
DELETE FROM dupes
WHERE  row > 1

编辑没有WITH的查询：

DELETE FROM  (
  SELECT 
    ROW_NUMBER() OVER (PARTITION BY TrackID ORDER BY grossopens DESC) row
  FROM NL_2015
)
WHERE 
row > 1

删除组中的重复记录 - SQL

1 个答案: