Select data from XML file as table in TSQL

Question

Could someone show me some TSQL to use to query an xml file as if it were a table?

The file is on the server, \"C:\\xmlfile.xml\"

And contains

&l         


        

Answer 1

In my case - the data I was interested in was contained in the node attributes rather than values. Below includes an example of how the attributes can be accessed.

DECLARE @xmlData XML
set @xmlData='<?xml version="1.0"?>
<ArrayOfSpangemansFilter xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xmlns:xsd="http://www.w3.org/2001/XMLSchema">
<SpangemansFilter FilterID="1219" Name="Fred" Code="510" Department="N" Number="305327">
</SpangemansFilter>
<SpangemansFilter FilterID="3578" Name="Gary" Code="001" Department="B" Number="0692690">
</SpangemansFilter>
<SpangemansFilter FilterID="3579" Name="George" Code="001" Department="X" Number="35933">
</SpangemansFilter>
</ArrayOfSpangemansFilter>'


SELECT 
  ref.value('@FilterID', 'int') AS FilterID ,
  ref.value('@Name', 'NVARCHAR (10)') AS Name ,
  ref.value('@Code', 'NVARCHAR (10)') AS Code ,
  ref.value('@Department', 'NVARCHAR (3)') AS Department,
  ref.value('@Number', 'int') AS Number      
FROM @xmlData.nodes('/ArrayOfSpangemansFilter/SpangemansFilter') 
xmlData( ref )

Answer 2

set @xmlData='<?xml version="1.0"?>
<ArrayOfSpangemansFilter xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xmlns:xsd="http://www.w3.org/2001/XMLSchema">
<SpangemansFilter>
<FilterID>1219</FilterID>
<Name>Fred</Name>
<Code>510</Code>
<Department>N</Department>
<Number>305327</Number>
</SpangemansFilter>
<SpangemansFilter>
<FilterID>3578</FilterID>
<Name>Gary</Name>
<Code>001</Code>
<Department>B</Department>
<Number>0692690</Number>
</SpangemansFilter>
<SpangemansFilter>
<FilterID>3579</FilterID>
<Name>George</Name>
<Code>001</Code>
<Department>X</Department>
<Number>35933</Number>
</SpangemansFilter>
</ArrayOfSpangemansFilter>'


SELECT 
  ref.value('FilterID[1]', 'int') AS FilterID ,
  ref.value('Name[1]', 'NVARCHAR (10)') AS Name ,
  ref.value('Code[1]', 'NVARCHAR (10)') AS Code ,
  ref.value('Department[1]', 'NVARCHAR (3)') AS Department,
  ref.value('Number[1]', 'int') AS Number      
FROM @xmlData.nodes('/ArrayOfSpangemansFilter/SpangemansFilter') 
xmlData( ref )

Produces:

FilterID    Name       Code       Department Number
----------- ---------- ---------- ---------- -----------
1219        Fred       510        N          305327
3578        Gary       001        B          692690
3579        George     001        X          35933

Note: The [1] is needed to indicate that you want to select the first value of the sequence since the query may return more than one matched value per row (imagine your XML containing several FilterIDs per SpangemansFilter).

I thought this was useful to know, so I Googled and read many posts until I found this one.

UPDATE To load from file:

DECLARE @xmlData XML
SET @xmlData = (
  SELECT * FROM OPENROWSET (
    BULK 'C:\yourfile.xml', SINGLE_CLOB
  ) AS xmlData
)

SELECT @xmlData