Parsing Markup

Use using Jint.HtmlParser; for the examples on this page. Public entry points accept an owned .NET string and an optional CancellationToken. Decode byte input before calling them; they do not expose a public stream, byte-encoding detector or incremental parser session.

HTML documents and fragments

MarkupParser.ParseHtml(source, options, cancellationToken) returns a Document. It recovers malformed HTML and inserts implied structure. HtmlParseOptions.ScriptingEnabled defaults to false; setting it changes scripting-dependent grammar, including noscript, without running JavaScript.

Use an existing element as the context for ParseHtmlFragment. The context controls grammar and namespaces: a table context can imply a tbody, and an SVG context preserves foreign namespaces.

var document = Document.CreateHtml();
var table = document.CreateElement("table");
document.AppendChild(table);
var fragment = MarkupParser.ParseHtmlFragment("<tr><td>Hello</td></tr>", table);
table.AppendChild(fragment);
Console.WriteLine(MarkupSerializer.ToHtml(table));

The returned fragment belongs to the context's owner document and is detached. Parsing leaves the context's existing children unchanged; appending the fragment moves its children into the context. Template context results also belong to the context owner and do not populate existing template content. Native parsing leaves declarative shadow markup as inert template elements.

XML and SVG

ParseXml parses a complete XML document; malformed XML raises MarkupParseException. It preserves namespaces and reports zero-based offsets in the original UTF-16 input. ParseSvg additionally requires an svg root in Namespaces.Svg. An unnamespaced <svg/> is not accepted as strict SVG.

ParseXmlFragment inherits namespace bindings from its element context:

var document = MarkupParser.ParseXml("<catalog xmlns='urn:catalog'><item/></catalog>");
var root = document.DocumentElement!;
var fragment = MarkupParser.ParseXmlFragment("<next/>", root);
root.AppendChild(fragment);
Console.WriteLine(((Element) root.LastChild!).NamespaceUri); // urn:catalog

The example produces <next> in urn:catalog. As with HTML, parsing does not replace the context's children. Use Document.CreateXml() or CreateHtml() when constructing a tree without parsing.

Entity handling and XML metadata

XML supports internal entity declarations and applies an entity-expansion ceiling of 10,000,000 UTF-16 replacement units by default. It does not fetch external subsets or expose an external resolver. The resulting Document provides immutable inventories:

Property	Meaning
SkippedXmlEntities	Entity content omitted by the parser's external-entity policy
XmlNotations	Retained notation declarations
XmlDtdProcessingInstructions	DTD instructions, with target, data and original-input offsets

These are parse metadata, rather than additional DOM children. Cloning a document preserves the inventories. Fragment parsing does not overwrite them; importing or adopting a node does not transfer a source document's inventories. DOM serialization and XPath do not relocate DTD instructions into the tree.

For input/token limits, cooperative cancellation and bounded HTML diagnostics, see Limits and threading.