/// <summary>
/// Parse HTML into a Document
/// </summary>
/// <param name="html"></param>
/// <param name="baseUri"></param>
/// <returns></returns>
public Document ParseInput(string html, string baseUri)
{
errors = CanTrackErrors ? ParseErrorList.Tracking(maxErrors) : ParseErrorList.NoTracking();
Document doc = treeBuilder.Parse(html, baseUri, errors);
return(doc);
}
internal override void InitialiseParse(string input, string baseUri, ParseErrorList errors)
{
base.InitialiseParse(input, baseUri, errors);
stack.AddLast(doc);
// place the document onto the stack. differs from HtmlTreeBuilder (not on stack)
doc.OutputSettings.Syntax = DocumentSyntax.Xml;
}
// current doc we are building into
// the stack of open elements
// current base uri, for creating new elements
// currentToken is used only for error tracking.
// null when not tracking errors
internal virtual void InitialiseParse(string input, string baseUri, ParseErrorList errors)
{
Validate.NotNull(input, "String input must not be null");
Validate.NotNull(baseUri, "BaseURI must not be null");
doc = new Document(baseUri);
reader = new CharacterReader(input);
this.errors = errors;
tokeniser = new Tokeniser(reader, errors);
stack = new DescendableLinkedList <Element>();
this.baseUri = baseUri;
}
internal Tokeniser(CharacterReader reader, ParseErrorList errors)
{
// replaces null character
// html input
// errors found while tokenising
// current tokenisation state
// the token we are about to emit on next read
// buffers characters to output as one token
// buffers data looking for </script>
// tag we are building up
// doctype building up
// comment building up
// the last start tag emitted, to test appropriate end tag
this.reader = reader;
this.errors = errors;
}
internal IReadOnlyList <Node> ParseFragment(string inputFragment, Element context, string baseUri, ParseErrorList errors)
{
// context may be null
state = HtmlTreeBuilderState.Initial;
InitialiseParse(inputFragment, baseUri, errors);
contextElement = context;
fragmentParsing = true;
Element root = null;
if (context != null)
{
if (context.OwnerDocument != null)
{
// quirks setup:
doc.QuirksMode = context.OwnerDocument.QuirksMode;
}
// initialise the tokeniser state:
string contextTag = context.TagName;
if (StringUtil.In(contextTag, "title", "textarea"))
{
tokeniser.Transition(TokeniserState.Rcdata);
}
else if (StringUtil.In(contextTag, "iframe", "noembed", "noframes", "style", "xmp"))
{
tokeniser.Transition(TokeniserState.Rawtext);
}
else if (contextTag.Equals("script"))
{
tokeniser.Transition(TokeniserState.ScriptData);
}
else if (contextTag.Equals(("noscript")))
{
tokeniser.Transition(TokeniserState.Data); // if scripting enabled, rawtext
}
else if (contextTag.Equals("plaintext"))
{
tokeniser.Transition(TokeniserState.Data);
}
else
{
tokeniser.Transition(TokeniserState.Data);
}
// default
root = new Element(Tag.ValueOf("html"), baseUri);
doc.AppendChild(root);
stack.Push(root);
ResetInsertionMode();
// setup form element to nearest form on context (up ancestor chain). ensures form controls are associated
// with form correctly
Elements contextChain = context.Parents;
contextChain.Insert(0, context);
foreach (Element parent in contextChain)
{
if (parent is FormElement)
{
formElement = (FormElement)parent;
break;
}
}
}
RunParser();
if (context != null)
{
return(root.ChildNodes);
}
else
{
return(doc.ChildNodes);
}
}
// tag searches
//private static final String[] TagsScriptStyle = new String[]{"script", "style"};
// the current state
// original / marked state
// the current head element
// the current form element
// fragment parse context -- could be null even if fragment parsing
// active (open) formatting elements
// chars in table to be shifted out
// if ok to go into frameset
// if next inserts should be fostered
// if parsing a fragment of html
internal override Document Parse(string input, string baseUri, ParseErrorList errors)
{
state = HtmlTreeBuilderState.Initial;
baseUriSetFromDoc = false;
return(base.Parse(input, baseUri, errors));
}
internal IReadOnlyList <Node> ParseFragment(string inputFragment, string baseUri, ParseErrorList errors)
{
InitialiseParse(inputFragment, baseUri, errors);
RunParser();
return(doc.ChildNodes);
}
/// <summary>
/// Utility method to unescape HTML entities from a string
/// </summary>
/// <param name="string">HTML escaped string</param>
/// <param name="inAttribute">if the string is to be escaped in strict mode (as attributes are)</param>
/// <returns>an unescaped string</returns>
public static string UnescapeEntities(string @string, bool inAttribute)
{
Tokeniser tokeniser = new Tokeniser(new CharacterReader(@string), ParseErrorList.NoTracking());
return(tokeniser.UnescapeEntities(inAttribute));
}
/// <summary>
/// Parse a fragment of XML into a list of nodes.
/// </summary>
/// <param name="fragmentXml">the fragment of XML to parse</param>
/// <param name="baseUri">base URI of document (i.e. original fetch location), for resolving relative URLs.</param>
/// <returns>list of nodes parsed from the input XML.</returns>
public static IReadOnlyList <Node> ParseXmlFragment(string fragmentXml, string baseUri)
{
XmlTreeBuilder treeBuilder = new XmlTreeBuilder();
return(treeBuilder.ParseFragment(fragmentXml, baseUri, ParseErrorList.NoTracking()));
}
/// <summary>
/// Parse a fragment of HTML into a list of nodes.
/// </summary>
/// <remarks>
/// The context element, if supplied, supplies parsing context.
/// </remarks>
/// <param name="fragmentHtml">the fragment of HTML to parse</param>
/// <param name="context">
/// (optional) the element that this HTML fragment is being parsed for (i.e. for inner HTML). This
/// provides stack context (for implicit element creation).
/// </param>
/// <param name="baseUri">
/// base URI of document (i.e. original fetch location), for resolving relative URLs.
/// </param>
/// <returns>
/// list of nodes parsed from the input HTML. Note that the context element, if supplied, is not modified.
/// </returns>
public static IReadOnlyList <Node> ParseFragment(string fragmentHtml, Element context, string baseUri)
{
HtmlTreeBuilder treeBuilder = new HtmlTreeBuilder();
return(treeBuilder.ParseFragment(fragmentHtml, context, baseUri, ParseErrorList.NoTracking()));
}
// utility methods
/// <summary>
/// Parse HTML into a Document.
/// </summary>
/// <param name="html">HTML to parse</param>
/// <param name="baseUri">base URI of document (i.e. original fetch location), for resolving relative URLs.</param>
/// <returns>parsed Document</returns>
public static Document Parse(string html, string baseUri)
{
TreeBuilder treeBuilder = new HtmlTreeBuilder();
return(treeBuilder.Parse(html, baseUri, ParseErrorList.NoTracking()));
}
internal virtual Document Parse(string input, string baseUri, ParseErrorList errors)
{
InitialiseParse(input, baseUri, errors);
RunParser();
return(doc);
}
internal Document Parse(string input, string baseUri)
{
return(Parse(input, baseUri, ParseErrorList.NoTracking()));
}