Serialization

Use using Jint.HtmlParser;. Select the output format through the method you call; the input document kind does not choose it automatically.

var document = MarkupParser.ParseXml("<root><item/></root>");
string xml = MarkupSerializer.ToXml(document, requireWellFormed: true,
    limits: new SerializationLimits { MaxOutputCharacters = 100_000 });
Console.WriteLine(xml);

The example prints <root><item/></root>. XML serialization repairs namespace prefixes in the output without modifying the source tree. Strings include no added encoding declaration or byte-order mark; encode them explicitly when writing bytes.

Method	Content
ToHtml(node)	The node as HTML
ToHtmlChildren(parent)	Children of an element, document or fragment as HTML
ToXml(node, requireWellFormed)	The node as XML
ToXmlChildren(parent, requireWellFormed)	Children with self-contained XML namespace declarations
ToXml(attribute, requireWellFormed)	DOM Parsing's empty serialization of an attribute

Child serialization omits the container's tags and uses template contents where applicable. requireWellFormed applies DOM Parsing serialization checks; it does not perform general XML or DTD validation. Invalid trees can raise DomException. Pass SerializationLimits.MaxOutputCharacters to bound output and a cancellation token to cooperate with host cancellation. The output ceiling counts UTF-16 units and defaults to zero (disabled); a failure raises SerializationLimitException without publishing partial output.

Shadow roots

Normal HTML serialization omits attached shadow roots. Select explicit roots, including closed ones, or include reachable roots whose Serializable flag is set:

var document = Document.CreateHtml();
var host = document.CreateElement("div");
document.AppendChild(host);
var shadow = host.AttachShadow(new ShadowRootInit(ShadowRootMode.Closed, Serializable: true));
shadow.AppendChild(document.CreateTextNode("shadow content"));
string html = MarkupSerializer.ToHtml(host,
    new HtmlSerializationOptions(shadowRoots: new[] { shadow }));
Console.WriteLine(html);

HtmlSerializationOptions copies and deduplicates an explicit root sequence. Roots that are not reachable from the serialized subtree are not appended. Shadow output uses declarative shadow templates. serializableShadowRoots: true selects reachable serializable roots without supplying the sequence. XML serialization does not take HTML shadow-selection options.

The constructor's scriptingEnabled parameter controls noscript escaping; it never executes script. Keep it consistent with the scripting context used to parse or construct the tree. Serialization does not mutate nodes, execute script or fetch resources. Keep the tree unchanged for the entire operation; detected concurrent mutation raises InvalidOperationException and discards unpublished output.