Posted with : Studying

Working with XPath to query Xml document

Using XPath is an effecient way to query, filter in XML document. I will explain some practices using XPath.

Related post: [Xml document processing][(/Xml-processing)]

##Prepare a document

Suppose that the XML document will look like (simplified)

<Customers>
	<Customer FirstName="Juan" LastName="Ho"></Customer>
	<Customer FirstName="Tuân" LastName="Ha"></Customer>
</Customers>

##Code to produce the document

var document = new XmlDocument();
var customers = document.CreateElement("Customers");

var customer1 = document.CreateElement("Customer");
customer1.SetAttribute("FirstName", "Juan");
customer1.SetAttribute("LastName", "Ho");

var customer2 = document.CreateElement("Customer");
customer2.SetAttribute("FirstName", "Tuân");
customer2.SetAttribute("LastName", "Ha");

customers.AppendChild(customer1);
customers.AppendChild(customer2);
document.AppendChild(customers);

###Query the document using XPath

So the above code produce a simple document with no custom namespace.

How to search for the Customer with LastName is Ha ? Below is an example

// Build an XPath query
const string query = "//Customers/Customer[@LastName='Ha']";
var nodes = document.SelectNodes(query);

// We must do a check because result of the SelectNodes could be null
if (nodes != null)
{
	foreach (XmlNode node in nodes)
	{
		// We must do a check because Attributes could be null
		if (node.Attributes == null)
		{
			continue;
		}

		var firstName = node.Attributes["FirstName"].Value;
		var lastName = node.Attributes["LastName"].Value;
		Console.WriteLine("FirstName: {0}, LastName: {1}", firstName, lastName);
	}
}

Another method to retrive nodes is using XPathNavigator. From an XML document we can get navigator.

// Create navigator of the document
var navigator = document.CreateNavigator();
var iterator = navigator.Select(query);

Console.WriteLine("Results: " + iterator.Count);

while (iterator.MoveNext())
{
	var firstName = iterator.Current.GetAttribute("FirstName", string.Empty);
	var lastName = iterator.Current.GetAttribute("LastName", string.Empty);
	Console.WriteLine("FirstName: {0}, LastName: {1}", firstName, lastName);
}

Two methods produce the same result "FirstName: Tuân, LastName: Ha"

Querying XML document with custom namespaces

Suppose that we have below document, just a little different on namespace with prefix is rb

<?xml version="1.0" encoding="utf-8"?>
<rb:Customers xmlns:rb="http://rabbit/xml">
	<Customer rb:First_x0020_Name="Juan" LastName="Ho" />
	<Customer FirstName="Tuân" LastName="Ha" />
</rb:Customers>

The XPath’s query must be updated to have namespace prefix rb and use another version of SelectNodes with a namespace resolver

var resolver = new XmlNamespaceManager(document.NameTable);
resolver.AddNamespace("rb", "http://rabbit/xml");

const string query = "//rb:Customers/Customer[@LastName='Ha']";
var nodes = document.SelectNodes(query, resolver);

Similarly, querying through XPathNavigator should be updated as well

var navigator = document.CreateNavigator();
var iterator = navigator.Select(query, resolver);

References

  • XPathNavigator.Select Methods 1, 2
Written on September 28, 2015

Tags