如何在SQL Server中消除重复和合并列值到单个文本

Posted 2021-03-28

我正在尝试连接三个表并获取结果，但是其中一个表具有相同CSO_Item_key的多个event_code，这导致重复记录。

表格1

Entry Date      Cso Item Key    Fail Code
8/1/2018 4:28   BLXB796201      CSL120
8/1/2018 4:40   BLXB799101      CLL250
8/1/2018 4:55   BLXB803001      CMS130
8/1/2018 5:08   BLXB806201      CNE100

表2

Cso Item Key    Event Code
BLXB796201      GTS
BLXB796201      LC28
BLXB796201      SDR4
BLXB799101      GTS
BLXB799101      LC28
BLXB799101      SDR4
BLXB803001      GTS
BLXB803001      LC28
BLXB803001      SDR4
BLXB806201      GTS
BLXB806201      LC28
BLXB806201      SDR4

表3

Fail Code  Desc
CSL120     Bad Part
CLL250     Unit Scrapped
CNE100     OS Reinstall
CBN101     NTF

预期结果：

Entry_Date     Cso_Item_Key Fail_Code   Desc         Event_Code
8/1/2018 4:28   BLXB796201   CSL120   Bad Part       GTS,LC28,SDR4
8/1/2018 4:40   BLXB799101   CLL250   Unit Scrapped  GTS,LC28,SDR4
8/1/2018 4:55   BLXB803001   CMS130   Null           GTS,LC28,SDR4
8/1/2018 5:08   BLXB806201   CNE100   OS Reinstall   GTS,LC28,SDR4

enter image description here

答案

这是使用sql-server函数STUFF和FOR XML PATH()的另一种可能的解决方案：

--Declare temp tables to simulate data given in post
DECLARE @tempTable1 TABLE([EntryDate] DateTime, [Cso Item Key] varchar(100), [Fail Code] varchar(50))
DECLARE @tempTable2 TABLE([Cso Item Key] varchar(100), [Event Code] varchar(50))
DECLARE @tempTable3 TABLE([Fail Code] varchar(50), [Desc] varchar(MAX))

--Insert the sample data
INSERT INTO @tempTable1
([EntryDate], [Cso Item Key], [Fail Code])
VALUES('2018-08-01 04:28:00', 'BLXB796201', 'CSL120'), ('2018-08-01 04:40:00', 'BLXB799101', 'CLL250'), ('2018-08-01 04:55:00', 'BLXB803001', 'CMS130'),
        ('2018-08-01 05:08:00', 'BLXB806201', 'CNE100')

INSERT INTO @tempTable2
([Cso Item Key], [Event Code])
VALUES('BLXB796201', 'GTS'), ('BLXB796201', 'LC28'), ('BLXB796201', 'SDR4'), ('BLXB799101', 'GTS'), ('BLXB799101', 'LC28'), 
('BLXB799101', 'SDR4'), ('BLXB803001', 'GTS'), ('BLXB803001', 'LC28'), ('BLXB803001', 'SDR4'), ('BLXB806201', 'GTS'), ('BLXB806201', 'LC28'), ('BLXB806201', 'SDR4')


INSERT INTO @tempTable3
([Fail Code], [Desc])
VALUES('CSL120', 'Bad Part'), ('CLL250', 'Unit Scrapped'), ('CNE100', 'OS Reinstall'), ('CBN101', 'NTF')

--The query, you can LEFT Join table 1 and table 3 on the Fail Code
--The sub query uses Group By and STUFF to produce a single row with all Event_Codes
--For each unique Cso Item Key, which can then be Inner Joined with table 1's Cso Item Key column
SELECT t1.[EntryDate], t1.[Cso Item Key], t3.[Fail Code], t3.[Desc], x.[Event_Code]
FROM @tempTable1 AS t1
LEFT JOIN @tempTable3 AS t3
ON t3.[Fail Code] = t1.[Fail Code]
INNER JOIN
(
    SELECT tt2.[Cso Item Key],
       STUFF((SELECT ',' + t2.[Event Code]
       FROM @tempTable2 AS t2
       WHERE t2.[Cso Item Key] = tt2.[Cso Item Key]
       FOR XML PATH('')), 1, 1, '') AS [Event_Code]
    FROM @tempTable2 AS tt2
    GROUP BY tt2.[Cso Item Key] 
) AS x
ON x.[Cso Item Key] = t1.[Cso Item Key]

另一答案

-- @-tables 1..3 are defined with your data
declare @table_1 table([Entry Date] datetime, [Cso Item Key] varchar(20), [Fail Code] varchar(10))
Insert Into @table_1 Values('8/1/2018 4:28','BLXB796201','CSL120'),('8/1/2018 4:40','BLXB799101','CLL250'),
('8/1/2018 4:55','BLXB803001','CMS130'),('8/1/2018 5:08','BLXB806201','CNE100')

declare @table_2 table([Cso Item Key] varchar(20), [Event Code] varchar(10))
Insert into @table_2 Values('BLXB796201','GTS'),('BLXB796201','LC28'),('BLXB796201','SDR4'),('BLXB799101','GTS' ),
('BLXB799101','LC28'),('BLXB799101','SDR4'),('BLXB803001','GTS' ),('BLXB803001','LC28'),('BLXB803001','SDR4'),
('BLXB806201','GTS' ),('BLXB806201','LC28'),('BLXB806201','SDR4')

declare @table_3 table([Fail Code] varchar(10), [Desc] varchar(50))
Insert Into @table_3 Values('CSL120','Bad Part'),('CLL250','Unit Scrapped'),('CNE100','OS Reinstall'),('CBN101','NTF')

--Here is the FINAL query (what u ask for)
Select T1.[Entry Date], T1.[Cso Item Key], T1.[Fail Code], T3.[Desc],
Replace((Select T2.[Event Code] + ',' From @table_2 T2 Where T2.[Cso Item Key] = T1.[Cso Item Key] For XML Path('')) + ',', ',,', '')
From @table_1 T1
Left Outer Join @table_3 T3 On T3.[Fail Code] = T1.[Fail Code]