Convert to XML having special character name in SQL server - sql

Question

There is a XML in the below format present in a column of datatype VARCHAR(MAX) in sql table
<?xml version="1.0" encoding="UTF-8"?>
<APIDATA xmlns="mynamespace">
<TRADE Action="Insert" CompanyID="33" Trader="Aleš Holubec"
</TRADE>
</APIDATA>
I need to fetch attribute values of Action and CompanyID . i.e, Insert and 33
In SQL, i used the below query
;WITH XMLNAMESPACES(DEFAULT 'mynamespace')
SELECT CONVERT(XML,column_name).value ('(APIDATA/TRADE/#Action)[1]', 'varchar(100)')
But i get the below error
XML parsing: line 1, character 537, illegal xml character
The reason is Aleš Holubec is the name of the trader in the XML which we cannot change.
Please help how to resolve this

Answer 1

You can fix this by converting your string to nvarchar(max) and then replace the encoding to UTF-16.
;WITH XMLNAMESPACES(DEFAULT 'mynamespace')
SELECT CONVERT(XML, replace(cast(column_name as nvarchar(max)), '<?xml version="1.0" encoding="UTF-8"?>', '<?xml version="1.0" encoding="UTF-16"?>')).value ('(APIDATA/TRADE/#Action)[1]', 'varchar(100)')
Or you can just remove the XML declaration.
;WITH XMLNAMESPACES(DEFAULT 'mynamespace')
SELECT CONVERT(XML, replace(column_name, '<?xml version="1.0" encoding="UTF-8"?>', '')).value ('(APIDATA/TRADE/#Action)[1]', 'varchar(100)')

Answer 2

I ran into the same issue with special characters and XML. Below is an example of what I did to resolve the problem with working code that you will need to modify for your DB.
The only thing I needed C# to handle for me was the single quote. C# will need to change the single quote to ' before sending it as a CSV list, if not it will fail.
Hope this helps..
/**************************************************************************************
**************************************************************************************
***
*** This is to demonstrate how to use XML in SQL statements.
*** C# must send the the list of items in comma seperated values and
*** replace the character Single Quote with ' or the SQL statement
*** will fail. This is the only character that SQL can not handle when sent as a
*** string.
***
*** The steps below will explain each part and how it handles special
*** characters in XML and turns them into a temp table.
***
***
**************************************************************************************
**************************************************************************************/
SET ARITHABORT ON
DECLARE #CUST_REF3 AS XML
DECLARE #CSharpString AS NVARCHAR(MAX)
--This is used for when the CSV string is passed in normally.
SET #CSharpString = ISNULL('{0}', '')
--Example of when the CSV string is passed in and it having a special characters.
SET #CSharpString = 'SHOP A, LA WHEEL & BRAKE SHOP, SHOP's B'
--Example of when nothing is passed in from C#
--SET #CSharpString = ''
--Replaces any special character for preperation of XML formating
SET #CSharpString = REPLACE(REPLACE(REPLACE(REPLACE(REPLACE(#CSharpString,'&', '&'),'<', '<'),'>', '>'),'"', '"'),'''', '&apos;')
-- Turn the #CSharpString (CSV) into XML
SET #CUST_REF3 = CAST('<A>'+ NULLIF(REPLACE(#CSharpString, N',', '</A><A>'), '') + '</A>' AS XML)
--Display the converted XML list
SELECT #CUST_REF3
--Converts the XML nodes into a temp table. Make sure the data type is set correctly for the data being received.
SELECT dbo.Trim(CUST_REF3.value('.','nchar(22)')) AS [CUST_REF3]
INTO #XML_List
FROM #CUST_REF3.nodes('/A') AS x(CUST_REF3)
--Display what Converted list
SELECT * FROM #XML_List
--example of output from the C# String
SELECT USER_DOC, CUST_REF3
FROM dbo.WO_HDR
WHERE
--This matches the list of items sent from C#
((SELECT TOP(1) x.CUST_REF3
FROM #XML_List x
WHERE x.CUST_REF3 = dbo.WO_HDR.CUST_REF3) IS NOT NULL OR
--If the user did not select anything, do not filter
#CSharpString = '')
--Dispose of temp table
IF OBJECT_ID('tempdb..#XML_List') IS NOT NULL
DROP TABLE #XML_List
/**********************************************************************************************************************************************************
--Summerized code to use:
SET ARITHABORT ON
DECLARE #CUST_REF3 AS XML
DECLARE #CSharpString AS NVARCHAR(MAX)
SET #CSharpString = ISNULL('{0}', '')
SET #CSharpString = REPLACE(REPLACE(REPLACE(REPLACE(REPLACE(#CSharpString,'&', '&'),'<', '<'),'>', '>'),'"', '"'),'''', '&apos;')
SET #CUST_REF3 = CAST('<A>'+ NULLIF(REPLACE(#CSharpString, N',', '</A><A>'), '') + '</A>' AS XML)
SELECT dbo.Trim(CUST_REF3.value('.','nchar(22)')) AS [CUST_REF3] INTO #XML_List FROM #CUST_REF3.nodes('/A') AS x(CUST_REF3)
--Add this scring to every WHERE statment you want this filter to be used.
--((SELECT TOP(1) x.CUST_REF3 FROM #XML_List x WHERE x.CUST_REF3 = dbo.WO_HDR.CUST_REF3) IS NOT NULL OR #CSharpString = '')
IF OBJECT_ID('tempdb..#XML_List') IS NOT NULL DROP TABLE #XML_List
**********************************************************************************************************************************************************/