使用ReplaceWith方法时XDocument XElement的编码问题

Question

I have the following code:

XDocument doc = XDocument.Load(file);
var x = doc.Descendants("span");

XElement xelm = x.FirstOrDefault(xm => xm.Attribute("class").Value=="screenitems");

Regex rgx = new Regex("^<span class=\"screenitems\">(.*)</span>$");
Match mtc = rgx.Match(xelm.Value);
if (mtc.Success)
{
    xelm.ReplaceWith(mtc.Groups[1].Value);
}
doc.Save(file);

When I get a match and replace the content of the XML file loaded into variable doc using the ReplaceWith method for the XElement variable xelm , the content of the XML file is getting encoded, so instead of having a tag like <p> I get <p&gt .

So How can I prevent that it encodes into html but actually replaces with the matched regex expression?

I have looked at some of the solutions here, like using XElement.Parse method or HTTPUtility.HtmlDecode , but I couldn't get it to work. It still encodes like html.

Answer 1

While you could try and parse your RegEx match into XElement s to solve the problem, I think you're going about this the wrong way.

As I understand it, your requirement is to replace a span element with the screenItems class with its contents. Rather than doing this using a combination of LINQ to XML and RegEx, you should stick to LINQ to XML.

Find all your span elements with the screenItems class:

var spans = doc.Descendants("span")
    .Where(e => (string)e.Attribute("class") == "screenItems")
    .ToList();

Then replace each of these with their own content:

foreach (var span in spans)
{
    span.ReplaceWith(span.Nodes());
}

See this fiddle for a working example.