/// <summary>Parse a fragment of HTML into a list of nodes.</summary>
/// <remarks>Parse a fragment of HTML into a list of nodes. 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 IList <iText.StyledXmlParser.Jsoup.Nodes.Node> ParseFragment(String fragmentHtml, iText.StyledXmlParser.Jsoup.Nodes.Element
context, String baseUri)
{
HtmlTreeBuilder treeBuilder = new HtmlTreeBuilder();
return(treeBuilder.ParseFragment(fragmentHtml, context, baseUri, ParseErrorList.NoTracking()));
}
/// <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 IList <iText.StyledXmlParser.Jsoup.Nodes.Node> ParseXmlFragment(String fragmentXml, String baseUri
)
{
XmlTreeBuilder treeBuilder = new XmlTreeBuilder();
return(treeBuilder.ParseFragment(fragmentXml, baseUri, ParseErrorList.NoTracking()));
}
internal override void InitialiseParse(String input, String baseUri, ParseErrorList errors)
{
base.InitialiseParse(input, baseUri, errors);
stack.Add(doc);
// place the document onto the stack. differs from HtmlTreeBuilder (not on stack)
doc.OutputSettings().Syntax(iText.StyledXmlParser.Jsoup.Nodes.Syntax.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
// start tag to process
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 List <iText.StyledXmlParser.Jsoup.Nodes.Element>(32);
this.baseUri = baseUri;
}
internal Tokeniser(CharacterReader reader, ParseErrorList errors)
{
// html input
// errors found while tokenising
// current tokenisation state
// the token we are about to emit on next read
// characters pending an emit. Will fall to charsBuilder if more than one
// buffers characters to output as one token, if more than one emit per read
// 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 virtual Document Parse(String input, String baseUri, ParseErrorList errors)
{
InitialiseParse(input, baseUri, errors);
RunParser();
return(doc);
}
internal virtual Document Parse(String input, String baseUri)
{
return(Parse(input, baseUri, ParseErrorList.NoTracking()));
}
public virtual Document ParseInput(String html, String baseUri)
{
errors = IsTrackErrors() ? ParseErrorList.Tracking(maxErrors) : ParseErrorList.NoTracking();
return(treeBuilder.Parse(html, baseUri, errors));
}
/// <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 XML into a Document.</summary>
/// <param name="xml">XML 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 ParseXml(String xml, String baseUri)
{
TreeBuilder treeBuilder = new XmlTreeBuilder();
return(treeBuilder.Parse(xml, baseUri, ParseErrorList.NoTracking()));
}
internal virtual IList <iText.StyledXmlParser.Jsoup.Nodes.Node> ParseFragment(String inputFragment, String
baseUri, ParseErrorList errors)
{
InitialiseParse(inputFragment, baseUri, errors);
RunParser();
return(doc.ChildNodes());
}
internal Tokeniser(CharacterReader reader, ParseErrorList errors)
{
this.reader = reader;
this.errors = errors;
}
internal virtual IList <iText.StyledXmlParser.Jsoup.Nodes.Node> ParseFragment(String inputFragment, iText.StyledXmlParser.Jsoup.Nodes.Element
context, String baseUri, ParseErrorList errors)
{
// context may be null
state = HtmlTreeBuilderState.Initial;
InitialiseParse(inputFragment, baseUri, errors);
contextElement = context;
fragmentParsing = true;
iText.StyledXmlParser.Jsoup.Nodes.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 (iText.StyledXmlParser.Jsoup.Helper.StringUtil.In(contextTag, "title", "textarea"))
{
tokeniser.Transition(TokeniserState.Rcdata);
}
else
{
if (iText.StyledXmlParser.Jsoup.Helper.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);
}
else
{
// if scripting enabled, rawtext
if (contextTag.Equals("plaintext"))
{
tokeniser.Transition(TokeniserState.Data);
}
else
{
tokeniser.Transition(TokeniserState.Data);
}
}
}
}
}
// default
root = new iText.StyledXmlParser.Jsoup.Nodes.Element(iText.StyledXmlParser.Jsoup.Parser.Tag.ValueOf("html"
), baseUri);
doc.AppendChild(root);
stack.Add(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.Add(0, context);
foreach (iText.StyledXmlParser.Jsoup.Nodes.Element parent in contextChain)
{
if (parent is FormElement)
{
formElement = (FormElement)parent;
break;
}
}
}
RunParser();
if (context != null && root != null)
{
return(root.ChildNodes());
}
else
{
return(doc.ChildNodes());
}
}
// tag searches
// 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
// reused empty end tag
// 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));
}
