///|
/// # MoonBit HTML5 Parser - API Specification
///
/// A WHATWG-compliant HTML5 parser with tokenizer, tree builder, and serializer.
// =============================================================================
// CORE PARSING API
// =============================================================================
///|
/// Parse an HTML string into a Document.
///
/// This is the main entry point for parsing HTML. The parser follows the
/// WHATWG HTML5 specification and handles malformed HTML gracefully.
///
/// Example:
/// ```
/// let doc = parse("Hello
")
/// ```
#declaration_only
pub fn parse(input : String) -> Document {
...
}
///|
/// Parse HTML and return both the document and any parse errors encountered.
///
/// Use this when you need to report or log parsing errors for diagnostics.
///
/// Example:
/// ```
/// let (doc, errors) = parse_with_errors("Test
")
/// // errors contains parse error information
/// ```
#declaration_only
pub fn parse_with_errors(input : String) -> (Document, Array[ParseError]) {
...
}
///|
/// Parse HTML with scripting enabled.
///
/// When scripting is enabled, the content inside `` tags is treated
/// as raw text rather than being parsed as HTML.
#declaration_only
pub fn parse_with_scripting(input : String) -> Document {
...
}
///|
/// Tokenize HTML without building a tree.
///
/// Returns an array of tokens and any parse errors. Useful for syntax
/// highlighting, linting, or custom tree construction.
///
/// Example:
/// ```
/// let (tokens, errors) = tokenize("Hello
")
/// // tokens[0] = StartTag(name="div", ...)
/// ```
#declaration_only
pub fn tokenize(input : String) -> (Array[Token], Array[ParseError]) {
...
}
// =============================================================================
// DOCUMENT METHODS
// =============================================================================
///|
/// Create a new empty document.
#declaration_only
pub fn Document::new() -> Document {
...
}
///|
/// Serialize the document to an HTML string.
///
/// Produces valid HTML5 output with proper escaping of text content
/// and attribute values.
///
/// Example:
/// ```
/// let doc = parse("Hello & World
")
/// doc.to_html() // "Hello & World
"
/// ```
#declaration_only
pub fn Document::to_html(self : Document) -> String {
...
}
///|
/// Pretty-print the document tree for debugging.
///
/// Returns a human-readable indented representation of the DOM tree.
#declaration_only
pub fn Document::dump(self : Document) -> String {
...
}
///|
/// Get child node IDs for a given node.
#declaration_only
pub fn Document::get_children(self : Document, node_id : Int) -> Array[Int] {
...
}
///|
/// Get the parent node ID. Returns NO_NODE (-1) if no parent.
#declaration_only
pub fn Document::get_parent(self : Document, node_id : Int) -> Int {
...
}
///|
/// Get the tag name of an element node.
#declaration_only
pub fn Document::get_tag_name(self : Document, node_id : Int) -> String? {
...
}
///|
/// Get an attribute value by name.
#declaration_only
pub fn Document::get_attribute(
self : Document,
node_id : Int,
name : String,
) -> String? {
...
}
///|
/// Check if an element has a specific attribute.
#declaration_only
pub fn Document::has_attribute(
self : Document,
node_id : Int,
name : String,
) -> Bool {
...
}
///|
/// Set an attribute on an element.
#declaration_only
pub fn Document::set_attribute(
self : Document,
node_id : Int,
name : String,
value : String,
) -> Unit {
...
}
///|
/// Get the concatenated text content of a node and its descendants.
#declaration_only
pub fn Document::get_text_content(self : Document, node_id : Int) -> String {
...
}
///|
/// Get the namespace of an element node.
#declaration_only
pub fn Document::get_ns(self : Document, node_id : Int) -> Namespace? {
...
}
///|
/// Get the Element struct for a node, if it is an element.
#declaration_only
pub fn Document::get_element(self : Document, node_id : Int) -> Element? {
...
}
///|
/// Get the raw Node for a given node ID.
#declaration_only
pub fn Document::get_node(self : Document, node_id : Int) -> Node? {
...
}
///|
/// Check if a node is an element with the given tag name and namespace.
#declaration_only
pub fn Document::is_element(
self : Document,
node_id : Int,
tag_name : String,
ns : Namespace,
) -> Bool {
...
}
///|
/// Check if a node is an HTML element with the given tag name.
#declaration_only
pub fn Document::is_html_element(
self : Document,
node_id : Int,
tag_name : String,
) -> Bool {
...
}
// =============================================================================
// DOM MANIPULATION
// =============================================================================
///|
/// Add a new node to the document. Returns the node ID.
#declaration_only
pub fn Document::add_node(self : Document, node : Node) -> Int {
...
}
///|
/// Append a child node to a parent node.
#declaration_only
pub fn Document::append_child(
self : Document,
parent_id : Int,
child_id : Int,
) -> Unit {
...
}
///|
/// Insert a node before a reference node.
#declaration_only
pub fn Document::insert_before(
self : Document,
parent_id : Int,
new_node_id : Int,
ref_node_id : Int,
) -> Unit {
...
}
///|
/// Remove a child node from its parent.
#declaration_only
pub fn Document::remove_child(
self : Document,
parent_id : Int,
child_id : Int,
) -> Unit {
...
}
///|
/// Get the template content node ID for a `` element.
#declaration_only
pub fn Document::get_template_content(self : Document, node_id : Int) -> Int {
...
}
// =============================================================================
// ELEMENT CREATION
// =============================================================================
///|
/// Create a new Element with tag name, namespace, and attributes.
#declaration_only
pub fn Element::new(
tag_name : String,
ns : Namespace,
attributes : Array[(String, String)],
) -> Element {
...
}
// =============================================================================
// TOKENIZER API
// =============================================================================
///|
/// Create a new tokenizer for the given input string.
#declaration_only
pub fn Tokenizer::new(input : String) -> Tokenizer {
...
}
///|
/// Get the next token from the input.
#declaration_only
pub fn Tokenizer::next_token(self : Tokenizer) -> Token {
...
}
///|
/// Get all parse errors accumulated during tokenization.
#declaration_only
pub fn Tokenizer::get_errors(self : Tokenizer) -> Array[ParseError] {
...
}
///|
/// Get the current source position.
#declaration_only
pub fn Tokenizer::get_position(self : Tokenizer) -> SourcePosition {
...
}
///|
/// Switch tokenizer to RCDATA state (for ``, `<textarea>`).
#declaration_only
pub fn Tokenizer::switch_to_rcdata(self : Tokenizer) -> Unit {
...
}
///|
/// Switch tokenizer to RAWTEXT state (for `<style>`, `<xmp>`, etc.).
#declaration_only
pub fn Tokenizer::switch_to_rawtext(self : Tokenizer) -> Unit {
...
}
///|
/// Switch tokenizer to Script Data state (for `<script>`).
#declaration_only
pub fn Tokenizer::switch_to_script_data(self : Tokenizer) -> Unit {
...
}
///|
/// Switch tokenizer to PLAINTEXT state.
#declaration_only
pub fn Tokenizer::switch_to_plaintext(self : Tokenizer) -> Unit {
...
}
// =============================================================================
// TREE BUILDER API
// =============================================================================
///|
/// Create a new tree builder for the given input.
#declaration_only
pub fn TreeBuilder::new(input : String) -> TreeBuilder {
...
}
///|
/// Create a tree builder with options.
#declaration_only
pub fn TreeBuilder::new_with_options(
input : String,
scripting~ : Bool,
) -> TreeBuilder {
...
}
///|
/// Run the tree construction algorithm and return the document.
#declaration_only
pub fn TreeBuilder::parse(self : TreeBuilder) -> Document {
...
}
// =============================================================================
// UTILITY FUNCTIONS
// =============================================================================
///|
/// Look up a named character reference (e.g., "amp" -> [38]).
///
/// Returns the code points for the character reference, or None if not found.
#declaration_only
pub fn lookup_entity(name : String) -> Array[Int]? {
...
}
///|
/// Check if a tag name is a formatting element (b, big, code, em, font, etc.).
#declaration_only
pub fn is_formatting_element(tag_name : String) -> Bool {
...
}
</code></pre>
<script>
let moonbitLanguageFn = hljs => {
return {
case_insensitive: true,
keywords: {
keyword: 'func fn enum struct type if else match return continue break while let var interface pub priv readonly',
literal: 'true false',
type: "Int Int64 Double String Bool Char Bytes Option Array Result",
built_in: 'lsl lsr asr shl shr land lor lxor Show Debug Hash Eq Compare Some None'
},
contains: [
{
scope: "char",
begin: "'", end: "'"
},
{
scope: "string",
begin: "\"", end: "\""
},
{
scope: "number",
begin: "\\b\\d+(\\.\\d+)?\\b"
},
{
scope: "codelink",
match: /\<a href\="(?<link>[^<>]+?)"\>(?<code>[^\/<>]+?)\<\/a\>/g
},
hljs.COMMENT(
'//', // begin
'\n', // end
)
]
}
}
hljs.registerLanguage('moonbit', moonbitLanguageFn);
hljs.highlightAll();
hljs.initLineNumbersOnLoad();
const number = window.location.href.split('#')[1];
function waitForLineNumbers() {
setTimeout(function () {
const target = document.querySelector(`.hljs-ln-line[data-line-number="${number}"]`);
if (target == null) waitForLineNumbers();
else target.scrollIntoView();
}, 50);
}
waitForLineNumbers()
</script>
<style>
.hljs-ln-numbers {
-webkit-touch-callout: none;
-webkit-user-select: none;
-khtml-user-select: none;
-moz-user-select: none;
-ms-user-select: none;
user-select: none;
}
.hljs-ln-n {
color: #ccc;
border-right: 1px solid #dfdddd;
margin-right: 1em;
text-align: center;
vertical-align: top;
padding-right: 0.5em;
}
.hljs {
background: none;
}
body {
background-color: #fafafa;
}
</style>
</body>
</html>