Develop Reference

RemoveChild when node is empty

I want to remove some child elements from an XML structure using RemoveChild() if a node is empty. My XML:
<Customers>
<customer>
<Name>John</Name>
<Age>25</Age>
<Status></Status>
</Customer>
<customer>
<Name>John</Name>
<Age></Age>
<Status></Status>
</Customer>
</Customers>
Should become:
<Customers>
<customer>
<Name>John</Name>
<Age>25</Age>
</Customer>
<customer>
<Name>John</Name>
</Customer>
</Customers>
My code so far:
XmlElement element3 = xmlDocument.CreateElement("Age");
element3.InnerText = str3;
element1.AppendChild((XmlNode)element3);
XmlElement element4 = xmlDocument.CreateElement("Status");
element4.InnerText = str4;
element1.AppendChild((XmlNode)element4);
How can I remove the empty node resp. nodes if there is more than one empty node?

Add a condition to only create the element if you need to:
if (!string.IsNullOrEmpty(status))
{
XmlElement statusElement = xmlDocument.CreateElement("Status");
statusElement.InnerText = status;
customerElement.AppendChild((XmlNode)statusElement);
}
Or if you need to remove it after it's already created (e.g. because you get XML from an external source with an empty Status), use the same sort of condition to remove it:
if (!string.IsNullOrEmpty(status))
{
customerElement.RemoveChild(statusElement);
}

That is because you are trying to remove a child element that simply does not exist. You must understand that in this context, the RemoveChild() method expects a reference to an existing node within the parent's children.
If i were to use your code sample above and create a standalone representation of this, it would look like this:
XmlDocument doc = new XmlDocument();
XmlElement element1 = doc.CreateElement("Customer");
XmlElement element2 = doc.CreateElement("Name");
element2.InnerText = "John";
element1.AppendChild(element2);
XmlElement element3 = doc.CreateElement("Age");
element3.InnerText = "25";
element1.AppendChild(element3);
XmlElement element4 = doc.CreateElement("Status");
element4.InnerText="";
element1.AppendChild(element4);
doc.AppendChild(element1);
element1.RemoveChild(element1.SelectSingleNode("Status"));

Using Linq-to-XML:
If you want nodes like:
<a>
<b>
<c></c>
</b>
</a>
to become:
<a />
Use:
XElement root = XElement.Load(file);
root.Descendants().Reverse().ToList().ForEach(x =>
{
if (x.IsEmpty)
x.Remove();
});
But if you want it to be:
<a>
<b />
</a>
Remove the .Reverse()

Extracting the particular part of an XML

I have an XML in a string variable i want to check if this xml content contains
<xml>
<message display='yes'>
....
or
<xml>
<xdp:xdp>
...
is it possible?

Here is how you can check for it:
string example = #"<xml>
<message display='yes'></message>
</xml>";
XDocument doc = XDocument.Parse(example);
if (doc.Element("xml").Element("message") != null)
{
// node "message" exists within node "xml" which is located at the root of the document
}
You can use XDocument class to check for existence of any node at any location in the hierarchy of an XML document. You can load contents from string or file easily.

An example below
<Order>
<AmazonOrderID>000-1111111-2222222</AmazonOrderID>
<MerchantOrderID>111-3333333-4444444</MerchantOrderID>
<PurchaseDate>2012-03-02T13:28:53+00:00</PurchaseDate>
<LastUpdatedDate>2012-03-02T13:29:05+00:00</LastUpdatedDate>
<OrderStatus>Pending</OrderStatus>
<SalesChannel>Amazon.com</SalesChannel>
<URL>http://www.amazon.com</URL>
<FulfillmentData>
<FulfillmentChannel>Amazon</FulfillmentChannel>
<ShipServiceLevel>Standard</ShipServiceLevel>
<Address>
<City>Beverly Hills</City>
<State>CA</State>
<PostalCode>90210-1234</PostalCode>
<Country>US</Country>
</Address>
</FulfillmentData>
<OrderItem>
<ASIN>AmazonASIN </ASIN>
<SKU> Internal-SKU</SKU>
<ItemStatus>Pending</ItemStatus>
<ProductName> This is the name of the product </ProductName>
<Quantity>1</Quantity>
<ItemPrice>
<Component>
<Type>Principal</Type>
<Amount currency="USD">19.99</Amount>
</Component>
</ItemPrice>
</OrderItem>
</Order>
List<string> getNodes(string path, string nodeName) {
List<string> nodes = new List<string>();
XDocument xmlDoc = XDocument.Load(path); //Create the XML document type
foreach (var el in xmlDoc.Descendants(nodeName)) {
//for debugging
//nodes.Add(el.Name + " " + el.Value);
//for production
nodes.Add(el.Value);
}
return nodes;
} //end getNodes
List<string> skuNodes = xml.getNodes(#"AmazonSalesOrders.xml", "SKU");

Parsing (and keeping) XML structure into SQL Server

I'm looking to parse a relatively complex XML file through C# and store a selection of the data into a SQL Server '08 database. This is what I'm looking to extract from the XML file:
<educationSystem>
<school>
<name>Primary School</name>
<students>
<student id="123456789">
<name>Steve Jobs</name>
<other elements>More Data</other elements>
</student>
<student id="987654">
<name>Jony Ive</name>
<otherElements>More Data</otherElements>
</student>
</students>
</school>
<school>
<name>High School</name>
<students>
<student id="123456">
<name>Bill Gates</name>
<other elements>More Data</other elements>
</student>
<student id="987654">
<name>Steve Ballmer</name>
<otherElements>More Data</otherElements>
</student>
</students>
</school>
</educationSystem>
[Before you ask, no this isn't a school assignment - I'm using school/students as an example and because the original is a lot more sensitive.]
I'm able to (using XDocument/XElement) parse the XML file and get a list of all school names, student names and student ID's, but when this gets added to the database, I end up with the Bill Gates student entry being under a second school. It's all just line-by-line.
I'm looking to find a way to say, achieve this:
Foreach school
put it's name into an XElement
foreach student
grab the name and id put into XElements
Grab next school and repeat
I believe Linq would be the best way to achieve this, but I'm having trouble in how to get started with the process. Would anyone be able to point me in the right direction?
Edit: Here's the code I'm currently using to save data to the database. It processes a list at a time (hence things aren't related as they should be). I'll also be tidying up the SQL as well.
private void saveToDatabase (List<XElement> currentSet, String dataName)
{
SqlConnection connection = null;
try
{
string connectionString = ConfigurationManager.ConnectionStrings["connString"].ConnectionString + "; Asynchronous Processing=true";
connection = new SqlConnection(connectionString);
connection.Open();
foreach (XElement node in currentSet)
{
SqlCommand sqlCmd = new SqlCommand("INSERT INTO dbo.DatabaseName (" + dataName + ") VALUES ('" + node.Value + "')", connection);
sqlCmd.ExecuteNonQuery();
}
}

This LINQ will generate a Collection of Objects,with two properties
Name of the school
List of students(again a collection)
var result = XElement.Load("data.xml")
.Descendants("school")
.Select( x => new {
name = XElement.Parse(x.FirstNode.ToString()).Value,
students =x.Descendants("student")
.Select(stud => new {
id = stud.Attribute("id"),
name = XElement.Parse(stud.FirstNode.ToString()).Value})
.ToList()});
Note:The LINQ assumes <name> as the first node under <school> and <student> tags
Then you can use the foreach that you intended and it will work like a charm
foreach (var school in result)
{
var schoolName = school.name;
foreach (var student in school.students)
{
//Access student.id and student.name here
}
}

For this particular type of workings with XML data, you could use XML Serialization / Deserialization.
This will allow you to Deserialize your XML Data into a IEnumerable Class Object, Perform your LINQ Queries on this Class and then save to SQL.
Hope this helps.

Update: The original code example did not mention a namespace. Namespaces need to be either accounted for when searching for elements by XName or one needs to to search using the XName.LocalName property. Updated the example to show how to handle selecting elements in such a case.
namespace Stackover
{
using System;
using System.Xml.Linq;
class Program
{
private const string Xml = #"<?xml version=""1.0"" encoding=""UTF-8""?>
<namespaceDocument xmlns=""http://www.namedspace/schemas"" xmlns:xsi=""http://www.w3.org/2001/XMLSchema-instance"" xsi:schemaLocation=""http://www.namedspace/schemas.xsd"">
<educationSystem>
<school>
<name>Primary School</name>
<students>
<student id=""123456789"">
<name>Steve Jobs</name>
<otherElements>
<dataA>data</dataA>
</otherElements>
</student>
<student id=""987654"">
<name>Jony Ive</name>
<otherElements>
<dataB>data</dataB>
</otherElements>
</student>
</students>
</school>
<school>
<name>High School</name>
<students>
<student id=""123456"">
<name>Bill Gates</name>
<otherElements>
<dataC>data</dataC>
</otherElements>
</student>
<student id=""987654"">
<name>Steve Ballmer</name>
<otherElements>
<dataD>data</dataD>
</otherElements>
</student>
</students>
</school>
</educationSystem>
</namespaceDocument>";
static void Main(string[] args)
{
var root = XElement.Parse(Xml);
XNamespace ns = "http://www.namedspace/schemas";
foreach(var school in root.Descendants(ns + "school")) // or root.Descendants().Where(e => e.Name.LocalName.Equals("school"));
{
Console.WriteLine(school.Element(ns + "name").Value);
foreach (var students in school.Elements(ns+ "students"))
{
foreach (var student in students.Elements())
{
Console.WriteLine(student.Attribute("id"));
Console.WriteLine(student.Name); // Name = namespace + XName
Console.WriteLine(student.Name.LocalName); // no namespace
}
}
}
}
}
}

How do I differentiate between different descendents with the same name?

I've got some XML I'm trying to import with c#, which looks something like this:
<root>
<run>
<name = "bob"/>
<date = "1958"/>
</run>
<run>
<name = "alice"/>
<date = "1969"/>
</run>
</root>
I load my xml using
XElement xDoc=XElement.Load(filename);
What I want to do is have a class for "run", under which I can store names and dates:
public class RunDetails
{
public RunDetails(XElement xDoc, XNamespace xmlns)
{
var query = from c in xDoc.Descendants(xmlns + "run").Descendants(xmlns + "name") select c;
int i=0;
foreach (XElement a in query)
{
this.name= new NameStr(a, xmlns); // a class for names
Name.Add(this.name); //Name is a List<NameStr>
i++;
}
// Here, i=2, but what I want is a new instance of the RunDetails class for each <run>
}
}
How can I set up my code to create a new instance of the RunDetails class for every < run>, and to only select the < name> and < date> inside a given < run>?

You can just LINQ to XML to create an IEnumerable from your XML.
IEnumerable<RunDetail> runDetails = from run in xdocument.Descendants("run")
select new RunDetail
{
Name = run.Element("name").Value,
Date = int.Parse(run.Element("date").Value)
};
This, of course, suggests there's a class named RunDetail with public Name and Date (int for the year) properties. You can iterate over the enumerable as it is, or if you need more explicit access to the individual members, you can use .ToList() or .ToArray() to convert the query.

You need to have some parent element in your xml cause your one is not valid. If you have following xml
<root>
<run>
<name = "bob"/>
<date = "1958"/>
</run>
<run>
<name = "alice"/>
<date = "1969"/>
</run>
</root>
you can load it to XDocument and iterate through children of the root element. For each run child element you can create RunDetails.

how to get root node attribute value using linq

I have the following XML. How to read the root node attribite value and it's decendents using LINQ? I am trying to read "dId" and "dTime" from root node, "id" from Customer element and Order number.
<?xml version="1.0" encoding="utf-8" ?>
<Customers dId="wqwx" dTime="10-9-09 11:23">
<Customer id="1">
<Orders>
<Order number="22" status="ok">
</Orders>
</Customer>
</Customers>
I tried the following code but it doesn't work.
XDocument doc= XDocument.Load(#"C:\Customers.xml");
var q = from c in doc.Descendants("Customers")
select new
{
dID = c.Attribute("dId"),
dTime = c.Attribute("dTime");
}

first, fix your xml (<Order .... />)
then, your linq should look like this....
// .Elements(...) selects all elements of type "Customer"
var q = from c in xDoc.Elements("Customers")
select new
{
dID = c.Attribute("dId"),
dTime = c.Attribute("dTime")
};
you should dl LinqPad... it lets you do Linq queries on the fly, even agains SQL databases. Then, once you get the results you want, copy and past your linq into your source code.

You have to end the order tag with: />
xDoc.Descendants("Customers") should work as well as xDoc.Elements("Customers").
Chris, is there a specific advantage to using .Elements?

You can't use LINQ to access the root tag.
The code below does what you want (I included a well formed xml file as well):
using System;
using System.Linq;
using System.Xml.Linq;
namespace ReadXmlSpike
{
class Program
{
static void Main(string[] args)
{
Console.WriteLine("Reading file...");
XDocument doc = XDocument.Load("Customers.xml");
var customers =
new
{
DID = (string) doc.Element("Customers").Attribute("did"),
DTime = (DateTime) doc.Element("Customers").Attribute("dTime"),
Customers = from customerxml in doc.Descendants("Customer")
select
new
{
ID = (string)customerxml.Attribute("id"),
Orders = from orderxml in customerxml.Descendants("Order")
select
new
{
Number =(string) orderxml.Attribute("number")
}
}
};
Console.WriteLine("Customersfile with id: {0} and time {1}",customers.DID,customers.DTime);
foreach (var customer in customers.Customers)
{
Console.WriteLine("Customer with id {0} has the following orders:",customer.ID);
foreach (var order in customer.Orders)
{
Console.WriteLine("Order with number {0}",order.Number);
}
}
Console.ReadLine();
}
}
}
and the xml file:
<?xml version="1.0" encoding="utf-8" ?>
<Customers did="www" dTime="10-09-09 11:23">
<Customer id="1">
<Orders>
<Order number="22" status="ok"/>
<Order number="23" status="bad"/>
</Orders>
</Customer>
<Customer id="2">
<Orders>
<Order number="24" status="ok"/>
<Order number="25" status="bad"/>
</Orders>
</Customer>
</Customers>

XDocument d = XDocument.Parse(#"<?xml version='1.0' encoding='utf-8' ?>
<Customers dId='wqwx' dTime='10-9-09 11:23'>
<Customer id='1'>
<Orders>
<Order number='22' status='ok'/>
</Orders>
</Customer>
</Customers>");
var cu = d.Root.Elements().Where(n => n.Name == "Customer");
var c = from cc in cu
select new
{
dId = cc.Document.Root.Attribute("dId").Value,
dTime = cc.Document.Root.Attribute("dTime").Value,
ID = cc.Attribute("id").Value,
number = cc.Element("Orders").Element("Order").Attribute("number").Value
};
foreach (var v in c)
{
Console.WriteLine("dId \t\t= {0}", v.dId);
Console.WriteLine("dTime \t\t= {0}", v.dTime);
Console.WriteLine("CustomerID \t= {0}", v.ID);
Console.WriteLine("OrderCount \t= {0}", v.number);
}
Console Output:
================================
dId = wqwx
dTime = 10-9-09 11:23
CustomerID = 1
OrderCount = 22
请按任意键继续. . .

It does not work the way you wrote it: while printing the above code would complain about anonymous type.
However, with this simple modified version d.Document.Root.Attribute("dId").Value; you can assign it to a string.