Select all nodes with attribute using xpath

Question

I am trying to select all nodes using root.SelectNodes() with XPath. For reference, see msdn-documentation.

In the following document explained, you can also search for nodes that contain attributes (correct me if this is actually a wrong understanding).

So I used the following line of code:

XmlNodeList nodes = projectDoc.DocumentElement.SelectNodes("descendant::Compile[attribute::Include]");

And I am trying to read the following data:

<Project DefaultTargets="Build" xmlns="http://schemas.microsoft.com/developer/msbuild/2003" ToolsVersion="4.0">
    <ItemGroup>
        <Compile Include="ArrayExtensions.cs" />
        <Compile Include="ArrayIterator.cs" />
        <Compile Include="AutoInitializeAttribute.cs" />
        <Compile Include="AutoInitializePriority.cs" />
        <Compile Include="AutoRegisterAttribute.cs" />
        <Compile Include="FormattableExtensions.cs" />
        <Compile Include="Mathematics\PrimeNumbers.cs" />
    </ItemGroup>
</Project>

As shown in the code-sample above, I want to get all the XmlNodes that contain the Include-attribute. However, when I execute my code, nodes contains 0 elements.

What am I doing wrong here?

I strongly suspect that the problem is the namespace - you should actually be looking for Compile elements in the http://schemas.microsoft.com/developer/msbuild/2003 namespace. Is there any reason you need to do this in XPath? With LINQ to XML this would be trivial. — Jon Skeet
– Jon Skeet, Commented Jun 11, 2014 at 8:31
No particular reason no, if you have a better solution that would be much appreciated! Also a solution to the question I provided would be awesome too, just for learning purposes. — Matthijs
– Matthijs, Commented Jun 11, 2014 at 8:32

Jon Skeet · Accepted Answer · 2014-06-11 08:39:40Z

1

I suspect the reason it's failing has nothing to do with the attribute part - it's failing to find the elements at all, as you've asked for just Compile elements, whereas there are only actually Compile elements in the namespace with the URI http://schemas.microsoft.com/developer/msbuild/2003.

Doing this with XPath probably requires the use of an XmlNamespaceManager which you'd then pass into another overload of SelectNodes. Personally I would use LINQ to XML instead though:

XDocument doc = XDocument.Load("myfile.xml");
XNamespace ns = "http://schemas.microsoft.com/developer/msbuild/2003";
var elements = doc.Descendants(ns + "Compile")
                  .Where(x => x.Attribute("Include") != null);

In general, I find LINQ to XML to be a much cleaner API than the "old" XmlDocument-based API.

answered Jun 11, 2014 at 8:39

Jon Skeet

1.5m893 gold badges9.3k silver badges9.3k bronze badges

Sign up to request clarification or add additional context in comments.

2 Comments

Matthijs Over a year ago

How can I use a Count on this? I don't see any XNodeList (or similar) in the XDocument-API. What I am trying to say is, what does this return? A list of elements or?

Jon Skeet Over a year ago

@Matthijs: It returns an IEnumerable<XElement>. You can use the Count() extension method just as with any other sequence. If you want a List<XElement>, just add a call to .ToList().

Collectives™ on Stack Overflow

Select all nodes with attribute using xpath

1 Answer 1

2 Comments

Your Answer

Linked

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

2 Comments

Your Answer

Sign up or log in

Post as a guest

Linked

Related