I wrote an algorithm based on Oded’s suggestions. Here it is. Works like a charm.
It removes all tags except strong
, em
, u
and raw text nodes.
internal static string RemoveUnwantedTags(string data)
{
if(string.IsNullOrEmpty(data)) return string.Empty;
var document = new HtmlDocument();
document.LoadHtml(data);
var acceptableTags = new String[] { "strong", "em", "u"};
var nodes = new Queue<HtmlNode>(document.DocumentNode.SelectNodes("./*|./text()"));
while(nodes.Count > 0)
{
var node = nodes.Dequeue();
var parentNode = node.ParentNode;
if(!acceptableTags.Contains(node.Name) && node.Name != "#text")
{
var childNodes = node.SelectNodes("./*|./text()");
if (childNodes != null)
{
foreach (var child in childNodes)
{
nodes.Enqueue(child);
parentNode.InsertBefore(child, node);
}
}
parentNode.RemoveChild(node);
}
}
return document.DocumentNode.InnerHtml;
}