XML / XPath Query Examples

Here are a few little XPath snippets of how to query for a specific node or node list in an XML document looking for specific attribute values.

The object model shown here is for the COM xml parser.  If you are using .NET's XmlDocument class instead of MSXML2.DOMDocument.4.0, the XPATH query syntax is the same.  In fact, much of the object model is the same.

"xml" string equals the following:

<?xml version='1.0'?>
  <EXAMPLE>
      <CUSTOMER id="1" type="B">Mr.  Jones</CUSTOMER>
      <CUSTOMER id="2" type="C">Mr.  Johnson</CUSTOMER>
  </EXAMPLE>

<%

  Dim doc
  Dim node
  Dim nodeList

   Set doc =  Server.CreateObject("MSXML2.DOMDocument.4.0")

          doc.LoadXML(xml)

  Set node = doc.selectSingleNode("//EXAMPLE/CUSTOMER[@id='2' or @type='C']")

         ' node.Text will show "Mr Johnson"

  Set node = doc.selectSingleNode("//EXAMPLE/CUSTOMER[@id='1' and @type='B']")

         ' node.Text will show "Mr Jones"
         ' node.Attributes.getNamedItem("id").Text will show "1"
         ' node.Attributes.getNamedItem("type").Text will show "B"
      
  ' Select a node list where id equals 1 and the type equals B or C.  
  ' In our example, only Mr.  Jones will be returned.

   Set nodeList = doc.selectNodes("//EXAMPLE/CUSTOMER[@id='1' and (@type='B' or @type='C')]")

     ' node.Text will show "Mr Jones"

  ' Select a node with a specific text value

   Set node = doc.selectSingleNode("//EXAMPLE/CUSTOMER[.  ='Mr.  Johnson']")

         ' node.Text will show "Mr.  Johnson"

  ' Select node list of all nodes with CUSTOMER not equal to "EggHeadCafe"

   nodeList.selectNodes("//EXAMPLE/CUSTOMER[.  !='EggHeadCafe']")      

' Start of MSXML4 or higher features.   If you want to use MSXML3, then
'  you'll need to set the following:  

    doc.setProperty "SelectionLanguage", "XPath"

' Search for a substring in an attribute

Set node = doc.selectSingleNode("//EXAMPLE/CUSTOMER[substring(@type,1,2) ='DE']")      

' Search for a substring in a node value

Set node = doc.selectSingleNode("//EXAMPLE/CUSTOMER[substring(.,1,3) ='Mr.']")    

' Search for a value contained in an attribute

Set node = doc.selectSingleNode("//EXAMPLE/CUSTOMER[contains(@type,'DECEA')]")      

' Search for a contained in a node value

Set node =  doc.selectSingleNode("//EXAMPLE/CUSTOMER[contains(.,'Smith')]")    

' Search only from root of selected node.   This examples gets a list of nodes
' from a specific node not the whole XML document.   This XML document
' is too small to be a reasonable example but you get the idea.  
' The ".//" tells the parser to make the current node the "root" node of the
' document for this specific XPath query.   Without the ".", the XPath query
' would return all CUSTOMER nodes in the entire document even though
' they may be above or below the current node.

  Set node = doc.selectSingleNode("//EXAMPLE")

  Set treeList = node.selectNodes(".//CUSTOMER")        

         nTot = treeList.length - 1
                
         For nCnt = 0 to nTot
                          
                 Set node = treeList.nextNode()
                 msgbox  node.Attributes.getNamedItem("id").Text  
      
         Next



'  End of MSXML4 or higher features    
        

Set node = nothing
Set doc = nothing

%>