internal void GetNextContentToken()
{
Debug.Assert(_nextTokenType != HtmlTokenType.EOF);
_nextToken.Length = 0;
if (this.IsAtEndOfStream)
{
_nextTokenType = HtmlTokenType.EOF;
return;
}
if (this.IsAtTagStart)
{
this.GetNextCharacter();
if (this.NextCharacter == '/')
{
_nextToken.Append("</");
_nextTokenType = HtmlTokenType.ClosingTagStart;
// advance
this.GetNextCharacter();
_ignoreNextWhitespace = false; // Whitespaces after closing tags are significant
}
else
{
_nextTokenType = HtmlTokenType.OpeningTagStart;
_nextToken.Append("<");
_ignoreNextWhitespace = true; // Whitespaces after opening tags are insignificant
}
}
else if (this.IsAtDirectiveStart)
{
// either a comment or CDATA
this.GetNextCharacter();
if (_lookAheadCharacter == '[')
{
// cdata
this.ReadDynamicContent();
}
else if (_lookAheadCharacter == '-')
{
this.ReadComment();
}
else
{
// neither a comment nor cdata, should be something like DOCTYPE
// skip till the next tag ender
this.ReadUnknownDirective();
}
}
else
{
// read text content, unless you encounter a tag
_nextTokenType = HtmlTokenType.Text;
while (!this.IsAtTagStart && !this.IsAtEndOfStream && !this.IsAtDirectiveStart)
{
if (this.NextCharacter == '<' && !this.IsNextCharacterEntity && _lookAheadCharacter == '?')
{
// ignore processing directive
this.SkipProcessingDirective();
}
else
{
if (this.NextCharacter <= ' ')
{
// Respect xml:preserve or its equivalents for whitespace processing
if (_ignoreNextWhitespace)
{
// Ignore repeated whitespaces
}
else
{
// Treat any control character sequence as one whitespace
_nextToken.Append(' ');
}
_ignoreNextWhitespace = true; // and keep ignoring the following whitespaces
}
else
{
_nextToken.Append(this.NextCharacter);
_ignoreNextWhitespace = false;
}
this.GetNextCharacter();
}
}
}
}
// ---------------------------------------------------------------------
//
// Constructors
//
// ---------------------------------------------------------------------
#region Constructors
/// <summary>
/// Constructor. Initializes the _htmlLexicalAnalayzer element with the given input string
/// </summary>
/// <param name="inputString">
/// string to parsed into well-formed Html
/// </param>
private HtmlParser(string inputString)
{
// Create an output xml document
_document = new XmlDocument();
// initialize open tag stack
_openedElements = new Stack <XmlElement>();
_pendingInlineElements = new Stack <XmlElement>();
// initialize lexical analyzer
_htmlLexicalAnalyzer = new HtmlLexicalAnalyzer(inputString);
// get first token from input, expecting text
_htmlLexicalAnalyzer.GetNextContentToken();
}
