Problem removing all nodes from an XML file using LINQ

Question

I am trying to remove all nodes from an XML file. But it also removes the open tag with the open root node. Using C # anf Linq

Input data:

<?xml version="1.0" encoding="utf-8" standalone="no"?> <!--Log the error count and error message--> <root> <ErrData> <Count>1</Count> <Timestamp>2011-11-21T11:57:12.3539044-05:00</Timestamp> </ErrData> <ErrData>max of 20 ErrData elements</ErrData> </root> 

Expected OP:

 <?xml version="1.0" encoding="utf-8" standalone="no"?> <!--Log the error count and error message--> <root> </root> 

Actual OP: EDITED

 <?xml version="1.0" encoding="utf-8" standalone="no"?> <!--Log the error count and error message--> <root /> 

Code:

 XDocument docs = XDocument.Load(path); try { docs.Descendants("ErrData").Remove(); } 

CODE:

Below is the code I'm using, the concept is the error count and timestamp are written to an XML file. After reaching the threshold value, e-mail will be sent by function and delete all nodes from xml. Then, when the following error occurs, it will begin to enter into the xml file, as shown below,

 XDocument doc = null; XElement el; if (!System.IO.File.Exists(path)) { doc = new XDocument(new XDeclaration("1.0", "utf-8", "no")); el = new XElement("root"); //el = new XElement("root"); XComment comment = new XComment("Log the error count and error message"); doc.Add(comment); } else { doc = XDocument.Load(path); } XElement p1 = new XElement("ErrData"); XElement p1Count = new XElement("Count", eventCount); XElement p1Windowsatrt = new XElement("Timestamp", windowStart); p1.Add(p1Count ); p1.Add(p1Windowsatrt ); if (doc.Root != null) { el = doc.Root; el.Add(p1); } else { el = new XElement("root"); el.Add(p1); } try { doc.Add(el);//Line throwing the exeception } catch (Exception e) { } finally { doc.Save(path); } 

Answer 1

Use docs.Root.Nodes().Remove() .

Answer 2

<root /> is valid XML for a tag without content (self-closing tags). If you absolutely need a shortcut to open and close, you need to put some content in the root directory of the node as a comment or text.

Answer 3

The confusion in your first sentence is: "I'm trying to remove all nodes / elements from an XML file." Which one? Do you want to delete all nodes or all elements?

There are five types of nodes in XML: elements, text, comments, processing instructions, and attributes. If you use "node" and "element" interchangeably since you are here, you will not have any problems working with XML.

What you got, <root/> , is the right way out for code that removes all descendant nodes: this is a single element named root with no content.

What do you expect

 <root> </root> 

is a single element named root that contains the child text node containing whitespace, possibly a new line. The code you wrote removes all descendant nodes, not just the descendant element nodes, and therefore it deletes this text node.