Question

我有两个共享一个公共密钥和几个不同列的表。共同密钥是CampaignID。其中一个表比另一个表略多CampaignID，我想找出这两个表之间的区别。目前，我使用LEFT OUTER JOIN和CTE首先合并这两个表，然后检查前面步骤中派生的NULL结果集中的CTE列来计算差异在CampaignID列中。例如，

WITH CTE_Results
     AS (SELECT t1.CampaignID AS cd_CampaignID,
                t2.CampaignID AS cod_CampaignID,
                t1.NAME,
                t2.Vendor
         FROM   CampaignDetails AS t1
                LEFT OUTER JOIN CampaignOnlineDetails AS t2
                             ON t1.CampaignID = t2.CampaignID)

-- Now that I have CTE result, I'll use another SELECT to find the difference
SELECT cd_CampaignID, cod_CampaignID
FROM CTE_Results
WHERE cod_CampaignID is NULL

但这对我来说似乎效率低下。是否有更有效/更快的方法来比较Microsoft SQL Server中两个表之间特定列的差异？谢谢你的回答！

注意：我一般都是Microsoft SQL Server和SQL的新手。

Answer 1

如果一个表比另一个表更多，那么bout：

SELECT C1.* FROM CAMPAIGN_WITH_MORE_DATA AS C1
WHERE NOT EXISTS(SELECT * FROM CAMPAIGN_WITH_LESS_DATA AS C2
                                WHERE C2.CAMPAIGN_ID = C1.CAMPAIGN_ID)

如果其中一个可能有一个广告系列，那么UNION这个：

SELECT C1.Name AS [Col1],
  C1.CAMPAIGN_ID,
  'More Campaigns' AS [Source]
FROM CAMPAIGN_WITH_MORE_DATA AS C1
WHERE NOT EXISTS(SELECT * FROM CAMPAIGN_WITH_LESS_DATA AS C2
                 WHERE C2.CAMPAIGN_ID = C1.CAMPAIGN_ID)
UNION ALL
SELECT C2.Vendor AS [Col1],
  C2.CAMPAIGN_ID,
  'Less Campaigns' AS [Source]
FROM CAMPAIGN_WITH_LESS_DATA AS C2
WHERE NOT EXISTS(SELECT * FROM CAMPAIGN_WITH_MORE_DATA AS C1
                 WHERE C1.CAMPAIGN_ID = C2.CAMPAIGN_ID)

Answer 2

不存在可能比左连接更有效。这也会起作用：

SELECT CampaignID FROM CampaignDetails
EXCEPT
SELECT CampaignID FROM CampaignOnlineDetails

只需按相反的顺序检查反方向。

根据公共密钥（Microsoft SQL Server）

2 个答案: