views:

1330

answers:

1

I'm trying to parse an Atom feed programmatically. I have the atom XML downloaded as a string. I can load the XML into an XmlDocument. However, I can't traverse the document using XPath. Whenever I try, I get null.

I've been using this Atom feed as a test: http://steve-yegge.blogspot.com/feeds/posts/default

Calling SelectSingleNode() always returns null, except for when I use "/". Here is what I'm trying right now:

using (WebClient wc = new WebClient())
{
    string xml = wc.DownloadString("http://steve-yegge.blogspot.com/feeds/posts/default");
    XmlNamespaceManager nsMngr = new XmlNamespaceManager(new NameTable());
    nsMngr.AddNamespace(string.Empty, "http://www.w3.org/2005/Atom");
    nsMngr.AddNamespace("app", "http://purl.org/atom/app#");
    XmlDocument atom = new XmlDocument();
    atom.LoadXml(xml);
    XmlNode node = atom.SelectSingleNode("//entry/link/app:edited", nsMngr);
}

I thought it might have been because of my XPath, so I've also tried a simple query of the root node since I knew the root should work:

// I've tried both with & without the nsMngr declared above
XmlNode node = atom.SelectSingleNode("/feed");

No matter what I do, it seems like it can't select anything. Obviously I'm missing something, I just can't figure out what. What is it that I need to do in order to make XPath work on this Atom feed?

EDIT

Although this question has an answer, I found out this question has an almost exact duplicate: http://stackoverflow.com/questions/24734/selectnodes-not-working-on-stackoverflow-feed

+7  A: 

While the C# implementation may allow default namespaces (I don't know), the XPath 1.0 spec doesn't. So, give "Atom" its own prefix:

nsMngr.AddNamespace("atom", "http://www.w3.org/2005/Atom");

And change your XPath appropriately:

XmlNode node = atom.SelectSingleNode("//atom:entry/atom:link/app:edited", nsMngr);
kdgregory