Antix Software

I thought XLinq would be a convenient way to load and manipulate XHTML data but was scuppered as it wouldn't parse due to some undeclared entities   for example.

In order to load/parse an XHTML document using XLinq you must do two things.

1. Declare the DOCTYPE in your source
2. Tell XLinq what entities to expect, like   and £ and a few more besides, and for this you need to create an XmlResolver

/// <summary>
/// <para>XmlResolver for XHTML</para>
/// </summary>
public class XhtmlResolver : XmlUrlResolver {

    /// <summary>
    /// <para>URN string to identify all the Entities</para>
    /// </summary>
    private const string ENTITIES_URN = "urn:Entities";

    /// <summary>
    /// <para>Get Entity</para>
    /// </summary>
    public override object GetEntity(
            Uri absoluteUri, string role, 
            Type ofObjectToReturn) {

        if (absoluteUri.AbsoluteUri == ENTITIES_URN) {

            return "<!ENTITY nbsp \"&#x000A0;\" >";
        }

        return null;
    }

    /// <summary>
    /// <para>Resolves XHTML DOCTYPE</para>
    /// </summary>
    public override Uri ResolveUri(
            Uri baseUri, string relativeUri) {

        // make sure the doc is declared as XHTML
        if (relativeUri.Equals("-//W3C//DTD XHTML 1.0 Transitional//EN", StringComparison.OrdinalIgnoreCase)
            || relativeUri.Equals("-//W3C//DTD XHTML 1.0 Strict//EN", StringComparison.OrdinalIgnoreCase)
            || relativeUri.Equals("-//W3C//DTD XHTML 1.0 Frameset//EN", StringComparison.OrdinalIgnoreCase)
            || relativeUri.Equals("-//W3C//DTD XHTML 1.1//EN", StringComparison.OrdinalIgnoreCase)) {

            return new Uri(ENTITIES_URN);
        }

        return base.ResolveUri(baseUri, relativeUri);
    }
}

This will take care to the   entities, to take care of all of them, either list all the entities in the GetEntity method or create an embedded resource file with the contents of http://www.w3.org/2003/entities/2007/isonum.ent.

Now to find out if XLinq is a convenient way to play with XHTML or not ...

References:

http://www.w3.org/2003/entities/2007/isonum.ent
http://www.martinwilley.com/net/code/xhtmldoc.html