I would like to query an XDocument
object for a given path, (e.g. "/path/to/element/I/want") but I don't know how to proceed.
Asked
Active
Viewed 2.1k times
25
-
possible duplicate of [how to use XPath with XDocument?](http://stackoverflow.com/questions/6209841/how-to-use-xpath-with-xdocument) – Fyodor Soikin Jun 27 '12 at 10:53
-
You can see an example selecting different paths, with / without namespace definitions, etc. here: http://stackoverflow.com/a/38272604/5838538. – Jelgab Jul 08 '16 at 20:31
4 Answers
62
You can use methods from System.Xml.XPath.Extensions
to do this.
For example, if you want to select a single element, you would use XPathSelectElement()
:
var element = doc.XPathSelectElement("/path/to/element/I/want");
The queries don't have to be simple paths like what you described, they use the XPath language.

svick
- 236,525
- 50
- 385
- 514
-
6
-
7@lionheart Isn't that exactly what I suggested? And the `System.Xml.XPath` namespace certainly isn't new in .Net 4.5. – svick Jun 09 '13 at 10:39
7
Even though this is a somewhat older post, it should be noted that LINQ-to-XML
can be used as an alternative to System.XML.XPath
to find elements based on a path within an XDocument
Example:
var results = x.Elements("path").Elements("to").Elements("element").Elements("I").Elements("want").FirstOrDefault();
Note: The LINQ to XML command may need to be altered to accommodate for the actual structure and/or cardinality of the XML.

Seymour
- 7,043
- 12
- 44
- 51
-
This makes no sense. What is x ? If x is an XContainer then Elements is an IEnumerable of XElement and you cannot chain. – user487779 Mar 15 '22 at 12:47
0
Similar to DaveDev but does work, if you do not want to use XPath.
public static class XElementExtensions
{
public static XElement GetStrictDescendant(this XContainer element, string path)
{
// todo guard against null path
var names = path.Split(new[] { '/' }, StringSplitOptions.RemoveEmptyEntries);
if (names.Length == 0)
throw new ArgumentException("Empty path", nameof(path));
var result = element;
foreach (var name in names)
{
result = result.Element(name);
if (result == null)
{
return null;
}
}
return (XElement)result;
}
}

user487779
- 550
- 5
- 12
-2
Something similar to this might work:
var path = "/path/to/element/I/want";
var route = path.Split(new []{'/'}, StringSplitOptions.RemoveEmptyEntries);
XElement result = null;
foreach (var node in route)
{
if (result == null)
{
result = _xmlDocument.Element(node);
}
else
{
result = result.Element(node);
}
}
return result;
-
1`RemoveEmptyEntries` is not a good idea, `//` has a specific meaning in XPath and it's to search all descendants. – svick Jun 27 '12 at 10:55
-
-
tried to edit. Nothing wrong there with the Split given that has nothing to do with XPath – user487779 Mar 15 '22 at 12:44