问题描述
我有一个以下格式的表(我正在使用MS SQL Server 2008):
I have a table in the following format (I'm using MS SQL Server 2008):
car_id | trace
1 1300275738;57.72588;11.84981;0.00026388888888888886;1300275793;57.72596;11.8529;0.001055...
跟踪值是一个以分号作为分隔符的csv字符串.跟踪字符串中的值分为四个和四个分组,如下所示(除了没有换行符):
The trace value is a csv string with semicolon as deliminator. The values in the trace string are grouped four and four, like this (except no linebreaks):
1300275738;57.72588;11.84981;0.00026388888888888886;
1300275793;57.72596;11.8529;0.0010555555555555555;
1300275785;57.72645;11.85242;0.007416666666666665;
1300275780;57.72645;11.85242;0.0010138888888888888;
我想做的是在插入时创建一个触发器,该触发器基于四个组中的第一个值对跟踪字符串进行排序.因此,以上结果将变为
What I want to do is to create a trigger on insert that sorts the trace string for based on the first value in the groups of four. So the result of the above would become
1300275738;57.72588;11.84981;0.00026388888888888886;1300275780;57.72645;11.85242;0.0010138888888888888;1300275785;57.72645;11.85242;0.007416666666666665;1300275793;57.72596;11.8529;0.0010555555555555555;
我试图做的就是将值分成这样的临时表中的单独行:
What I have tried to do is to split the value into separate rows in a temporary table like this:
USE tempdb
GO
checkpoint
dbcc dropcleanbuffers
dbcc freeproccache
GO
--declare a variable and populate it with a comma separated string
DECLARE @SQLString VARCHAR(MAX)
SET @SQLString = (SELECT trace FROM mypev_trips.dbo.trips)
--append a comma to the string to get correct results with empty strings or strings with a single value (no commas)
SET @SQLString = @SQLString + ';';
DECLARE @X XML
SET @X = CAST('<A>' + REPLACE(@SQLString, ';', '</A><A>') + '</A>' AS XML)
SELECT t.value('.', 'nvarchar(20)')
FROM @x.nodes('/A') as x(t)
得出以下结果:
(No column name)
1300275738
57.72588
11.84981
0.000263888888888888
1300275780
57.72645
11.85242
0.001013888888888888
.
.
有人知道我如何将我的临时表转换回以逗号分隔的字符串,该字符串按每四个一组的第一个值排序?
Does anyone know how I can transform my temporary table back to a comma separated string sorted on the first value in each group of four?
推荐答案
;WITH cte(car_Id, traceXML) AS
(
SELECT car_Id, CAST('<A>' + REPLACE(trace, ';', '</A><A>') + '</A>' AS XML)
FROM dbo.cars
), cte2 AS
(
SELECT car_Id, trace, NTILE(4) OVER(ORDER BY (SELECT 1)) AS grId
FROM cte
CROSS APPLY (SELECT Tbl.Col.value('.', 'nvarchar(250)') AS trace
FROM traceXML.nodes('/A') Tbl(Col)) AS List
), cte3 AS
(
SELECT DISTINCT a.car_Id,(
SELECT ISNULL(b.trace, '') + ';'
FROM cte2 b
WHERE b.grId = a.grId
FOR XML PATH('')) AS trace
FROM cte2 a
)
SELECT DISTINCT a.car_Id,(
SELECT ISNULL(b.trace, '')
FROM cte3 b
WHERE b.car_Id = a.car_Id
FOR XML PATH('')) AS trace
FROM cte3 a
这篇关于SQL Server 2008拆分,排序和合并值的文章就介绍到这了,希望我们推荐的答案对大家有所帮助,也希望大家多多支持!