Question

I have a database which has a table with an XML column. The XML data has a bunch of child nodes which look something like this:

<test>
  <result id="1234">
    <data elementname="Message">some error message</data>
    <data elementname="Cat">Cat01</data>
    <data elementname="Type">WARNING</data>
  </result>
  <result id="5678">
    <data elementname="Message">some error message</data>
    <data elementname="Cat">Cat01</data>
    <data elementname="Type">WARNING</data>
  </result>
</test>

The Cat element can have a number of different values. I’m trying to create reports on this data, so one thing I’d like to do is get a list of all the categories througout our data. This is my query:

Select Id, XmlData.query('/test/result/data[@elementname = ''Cat''] ')  AS  Message
From Table
WHERE XmlData.exist('/test/result/data[@elementname = ''Cat'']') = 1
ORDER BY FriendlyName

This correctly gets all the rows in my table with this type of categorization (there’ll be other results in the same table without that element), but the categories are all combined into one column for each table record:

Id1, <data elementname="Cat">Cat01</data><data elementname="Cat">Cat01</data>
Id2, <data elementname="Cat">Cat01</data><data elementname="Cat">Cat01</data>

I’m including the Id column so it’s easy to see where the data is coming from, the main problem is that I can only get it to concatenate the values for each row – I need each of those data elements to have its own row, then maybe do a Select Distinct on the result.

Is there a way I can do that?

Thanks

Answer 1

Always the Google after you post your question….

Think I found the answer here: http://blogs.msdn.com/b/simonince/archive/2009/04/24/flattening-xml-data-in-sql-server.aspx

SELECT  DISTINCT  cref.value('(text())[1]', 'varchar(50)') as Cat
FROM   
      SGIS CROSS APPLY 
      Data.nodes('/test/result') AS Results(rref) CROSS APPLY
      rref.nodes('data[@elementname = ''Cat'']') AS Categories(cref)  

Seems the key is the Cross Apply keywords