SQL XML解析存储过程手动运行,而不是通过SQL Server代理运行

时间:2013-04-11 11:02:35

标签: sql sql-server xml xml-parsing sql-server-agent

存储过程正在手动运行,但不会使用SQL Server代理运行。

我收到以下错误

  

以用户身份执行:XXX。 XML解析:第9行,第9个字符,意外   输入结束[SQLSTATE 42000](错误9400)。步骤失败了。

存储过程从Web查询中获取xml,将其放入临时表中的单个单元格中。解析临时表中的数据,并将新的易读信息放入永久表中。我需要每天运行查询并获取当天的数据。

我已经尝试过查找问题,但我唯一的线索就是XML正在某处被切断。我不知道为什么这只是为了工作而发生的。

你能给予的任何帮助都会很棒。

谢谢

EDIT1:我已将程序分为7个不同的步骤。这是它正在堕落的步骤。

INSERT XMLData(XMLD) SELECT CAST(HTML AS XML) As XMLData FROM TextData

EDIT2:这是代码

DECLARE 

@url varchar(2048),
@win integer,
@hr integer ,
@text varchar(MAX),
@XMLdata XML,
@Date date,
@SearchDate nvarchar(50) 


Set @Date = GETDATE()

set @SearchDate = CAST(@Date as nvarchar(50))

set @SearchDate = REPLACE(@SearchDate,'-','')


--set @SearchDate='20130405'

/*-- Create Temporary tables to be used to store the xml data--*/
/*-- We need 2 tables as the raw data from SEMO is Unicode 'UTF-8' but SQL only handles 'UTF-16'--*/
/*-- We must therefore store the data first as Text and then CAST it to XML datatype for easy querying later --*/
CREATE TABLE #TextData(HTML text NULL)
CREATE TABLE #XMLData(XMLD xml NULL)

/*-- This url will have to be dynamically generated each day based on GetDate() query or similiar - to be cracked out fully - random date picked for now--*/
Select @url = 'http://.......'

/* Use OLE Automation Objects to go out and get the data--*/    
EXEC @hr=sp_OACreate 'WinHttp.WinHttpRequest.5.1',@win OUT 
IF @hr <> 0 EXEC sp_OAGetErrorInfo @win 

EXEC @hr=sp_OAMethod @win, 'Open',NULL,'GET',@url,'false'
IF @hr <> 0 EXEC sp_OAGetErrorInfo @win 

EXEC @hr=sp_OAMethod @win,'Send'
IF @hr <> 0 EXEC sp_OAGetErrorInfo @win 

INSERT #TextData(HTML)
EXEC @hr=sp_OAGetProperty @win,'ResponseText'
IF @hr <> 0 EXEC sp_OAGetErrorInfo @win

EXEC @hr=sp_OADestroy @win 
IF @hr <> 0 EXEC sp_OAGetErrorInfo @win 

/*-- Now CAST the text data to XML data type which makes it much easier to query back--*/
INSERT #XMLData(XMLD)
SELECT CAST(HTML AS XML) As XMLData FROM #TextData



/*-- WE NEED TO ADD IN A SECTION HERE NOW TO PARSE THE XML DATA AND EXTRACT THE INFO WE NEED INTO A DEDICATED TABLED--*/

/*-- Replace the : in the #XMLData table with _ as the colons were causing errors --*/
UPDATE #XMLData
SET XMLD = REPLACE(CAST(XMLD as varchar(max)),':' ,'_')
FROM #XMLData
WHERE CHARINDEX(':' ,CAST(XMLD as varchar(max)))>0

UPDATE #XMLData
SET XMLD = REPLACE(CAST(XMLD as varchar(max)),'DataSet xmlns="http_//localhost/SemoReporting/SemoAutomatedDataCollection/Datasets"' ,'DataSet')
FROM #XMLData
WHERE CHARINDEX('DataSet xmlns="http_//localhost/SemoReporting/SemoAutomatedDataCollection/Datasets"' ,CAST(XMLD as varchar(max)))>0

Select @XMLdata = XMLD
FROM #XMLData

/*-- This is to show the colons have been replaced --*/
SELECT * FROM #XMLData

Declare @i as int
set @i = 1

While @i < 97
BEGIN
/*Insert Data into SEMO Historical Table*/

INSERT INTO XMLParsing.dbo.T1301_SEMO_HISTORICAL_DATA 
/*-- This selects the relevant data from table1 element --*/
SELECT
a.b.value('Table1[sql:variable("@i")][1]/CURRENCY_FLAG[1]','nvarchar(50)') AS [Currency],
a.b.value('Table1[sql:variable("@i")][1]/TRADE_DATE[1]','date') AS [Trade Date],
a.b.value('Table1[sql:variable("@i")][1]/DELIVERY_DATE[1]','date') AS [Delivery Date],
a.b.value('Table1[sql:variable("@i")][1]/DELIVERY_HOUR[1]','int') AS [Delivery Hour],
a.b.value('Table1[sql:variable("@i")][1]/DELIVERY_INTERVAL[1]','int') AS [Delivery Interval],
a.b.value('Table1[sql:variable("@i")][1]/RUN_TYPE[1]','nvarchar(50)') AS [Run Type],
a.b.value('Table1[sql:variable("@i")][1]/SMP[1]','decimal(10,4)') AS [SMP],
a.b.value('Table1[sql:variable("@i")][1]/LAMBDA[1]','decimal(10,4)') AS [Lambda],
a.b.value('Table1[sql:variable("@i")][1]/SYSTEM_LOAD[1]','decimal(10,4)') AS [System Load],
a.b.value('Table1[sql:variable("@i")][1]/CMS_TIME_STAMP[1]','nvarchar(150)') AS [CMS Time Stamp]
FROM @XMLdata.nodes('DataSet/diffgr_diffgram/EA_RESULTS') a(b)
--where 
--not exists (Select * from SEMO_Historical_Data where [Trade Date] = @Date)
set @i=@i+1
END

Drop Table #TextData
Drop Table #XMLData

1 个答案:

答案 0 :(得分:0)

只是一个更新。这不能正常工作的原因是从网站上获取的XML被截断。我从来没有完全弄清楚为什么会发生这种情况,但是我使用了wget for windows而不是WinHttp.WinHttpRequest.5.1,它从网站上获取了所有数据并将其直接传递到我可以解析数据的网页。