[英]Combining records from query results
在下面我的查询,在数据customfielddata.fielddata
保存独立基础上的价值customfielddefinitions.customfieldname
。
例如:
表CustomFieldDefinitions
:
CustomFieldID CustomFieldName
------------------------------
1234 Business Owner
5678 Client
9012 Cost Center
表CustomFieldData
:
CustomFieldID FieldData Updated
--------------------------------------
1234 barb 1/1/2018
5678 health plan 1/1/2018
9012 68121 1/1/2018
我希望能够提取具有所有FieldData
值组合的1条记录。 但是,查询将返回不适用于CASE
语句的CustomFieldIDs
NULL
。
例如:
IssueID Business_Owner Client Cost_Center
-------------------------------------------------
176367 NULL NULL 68121
176367 Barb S NULL NULL
176367 NULL Health Plan NULL
我知道为什么会这样,但是我不确定是否有解决方案。 我正在使用SSMS。 感谢您对初学者的任何帮助,谢谢!
SELECT
GI.issueid,
CASE
WHEN GCD.customfieldname = 'Business Owner'
THEN GC1.fielddata
END AS Business_Owner,
CASE
WHEN GCD.customfieldname = 'Client'
THEN GC1.fielddata
END AS Client,
CASE
WHEN GCD.customfieldname = 'Cost Center'
THEN GC1.fielddata
END AS Cost_Center,
GI.closeddate AS Closed
FROM
dbo.gemini_issues GI
INNER JOIN
(SELECT
MAX(created) AS Created,
fielddata, issueid, customfieldid
FROM
gemini_customfielddata
GROUP BY
fielddata, issueid, customfieldid) GC1 ON GI.issueid = GC1.issueid
INNER JOIN
dbo.gemini_customfielddefinitions GCD ON GC1.customfieldid = GCD.customfieldid
AND GCD.customfieldname IN ('cost center', 'business owner', 'client')
WHERE
GI.projectid IN (193, 194, 195)
ORDER BY
issueid
我认为您只需要在此处设置一个最大值:
SELECT GI.issueid
,max(CASE
WHEN GCD.customfieldname = 'Business Owner'
THEN GC1.fielddata
END) AS Business_Owner
,max(CASE
WHEN GCD.customfieldname = 'Client'
THEN GC1.fielddata
END) AS Client
,max(CASE
WHEN GCD.customfieldname = 'Cost Center'
THEN GC1.fielddata
END) AS Cost_Center
,GI.closeddate AS Closed
FROM dbo.gemini_issues GI
INNER JOIN (
SELECT max(created) AS Created
,fielddata
,issueid
,customfieldid
FROM gemini_customfielddata
GROUP BY fielddata
,issueid
,customfieldid
) GC1 ON GI.issueid = GC1.issueid
INNER JOIN dbo.gemini_customfielddefinitions GCD ON GC1.customfieldid = GCD.customfieldid AND GCD.customfieldname IN ('cost center', 'business owner', 'client')
WHERE GI.projectid IN (193, 194, 195)
group by GI.issueid
ORDER BY GI.issueid
使用PIVOT运算符将行旋转到列是解决此问题的最佳方法,并且在大多数情况下还可以提高性能。 请参阅以下示例:
-- Sample data
CREATE TABLE Issues (
IssueID int NOT NULL,
CustomFieldID int NOT NULL,
FieldData nvarchar(max) NOT NULL,
Created datetime default getutcdate()
)
INSERT Issues (IssueID, CustomFieldID, FieldData)
VALUES
(176367, 1234, 'barb'),
(176367, 5678, 'health plan'),
(176367, 9012, '68121'),
(176368, 1234, 'don'),
(176368, 5678, 'health plan2'),
(176368, 9012, '12345')
CREATE TABLE CustomFieldDefinitions (
CustomFieldID int NOT NULL PRIMARY KEY,
CustomFieldName nvarchar(max) NOT NULL)
INSERT CustomFieldDefinitions VALUES
(1234, 'Business Owner'),
(5678, 'Client'),
(9012, 'Cost Center')
-- Query
SELECT IssueID, [Business Owner], [Client], [Cost Center]
FROM
(
SELECT i.IssueID, c.CustomFieldName, i.FieldData
FROM Issues i
INNER JOIN CustomFieldDefinitions c ON i.CustomFieldID = c.CustomFieldID
-- add other conditions to filter Issues
) AS SourceTable
PIVOT
(
MAX(FieldData)
FOR CustomFieldName IN ([Business Owner], [Client], [Cost Center])
) AS PivotTable;
-- Results
| IssueID | Business Owner | Client | Cost Center |
|---------|----------------|--------------|-------------|
| 176367 | barb | health plan | 68121 |
| 176368 | don | health plan2 | 12345 |
有关工作示例,请参见此sql小提琴: http : //sqlfiddle.com/#!18/5cd01/4
编辑:我将Issues和CustomFieldData合并到示例中的单个Issues表中。 对于您的情况,您只需要在内部查询(SourceTable)中进行其他联接即可
声明:本站的技术帖子网页,遵循CC BY-SA 4.0协议,如果您需要转载,请注明本站网址或者原文地址。任何问题请咨询:yoyou2525@163.com.