Question

Below is my code.

[XmlRootAttribute("book")]
public class BookHtml
{
    [XmlElement("book-id")]
    public string BookId { get; set; }

    [XmlElement("book-xhtml")]
    public BookHtmlMetadata BookXhtml { get; set; }

    public String ToHtml()
    {
        return this.BookXhtml.Xhtml.ToString();
    }
}

public class BookHtmlMetadata
{
    [XmlElement("xhtml")]
    public XElement Xhtml { get; set; }
}

public class Program
{
    private static string GetXhtmlWithNoTags()
    {
        return "<content>" +
                 "<book>" +
                       "<book-id label=\"Book Id\">2</book-id>" +
                       "<book-xhtml label=\"Book Xhtml\">" +
                            "<xhtml>" +
                                   "Copyright © 2010 . All rights reserved.<a href=\"/Home/Book.asp\">Best book ever</a>. " +
                            "</xhtml>" +
                        "</book-xhtml>" +
                    "</book>" +
                "</content>";
    }

    static void Main(string[] args)
    {
        string xml = GetXhtmlWithNoTags();

        XElement contentXml = XElement.Parse(xml);

        var xmlSerializer = new XmlSerializer(typeof(BookHtml));
        var list = new List<BookHtml>();

        foreach (var child in contentXml.Elements())
        {
            list.Add((BookHtml)xmlSerializer.Deserialize(child.CreateReader()));
        }

        string contentToRender = list.Single().BookXhtml.Xhtml;
   }

When I run this code I get an error on:

xmlSerializer.Deserialize(child.CreateReader());

The XmlReader must be on a node of type Element instead of a node of type Text.

How can I deserialize the content within <xhtml/> tags without losing any of the html tags such as <a href="/Home/Book.asp"> ? I should be able to use the xhtml and render the html tags/links in the browser.

Any ideas, suggessions greatly appreciated.

Answer 1

Since the interior of the xhtml tags could be a free form, you should change the type from XElement to string. Then, to escape the html/xhtml block from being processed you can use the CDATA to tell the deserialization routine that this is not quite valid xml, and you don’t have a typed structure to use.

In Code:

public class BookHtmlMetadata
{
    [XmlElement("xhtml")]
    public string Xhtml { get; set; }
}

and

private static string GetXhtmlWithNoTags()
{
    return "<?xml version=\"1.0\" encoding=\"UTF-8\"?><content>" +
                "<book>" +
                    "<book-id label=\"Book Id\">2</book-id>" +
                    "<book-xhtml label=\"Book Xhtml\">" +
                        "<xhtml><![CDATA[" +
                                "Copyright © 2010 . All rights reserved.<a href=\"/Home/Book.asp\">Best book ever</a>. " +
                        "!]]></xhtml>" +
                    "</book-xhtml>" +
                "</book>" +
            "</content>";
}

Now, if the xhtml block is valid xml, then you can use an XMLDocument to load the xml and traverse the tree.