/// <summary>
/// Parse a start tag
/// </summary>
protected ParsedToken ParseStartTag()
{
_StartTagPosition = _CurrentPosition;
_TagBuffer = SourceReader.OpenBuffer();
CharInfo c = ReadChar();
// Comments ?
if (c == '!')
{
// Expect '--' or 'DOCTYPE'
c = ReadChar();
if (Char.IsLetter(c.AsChar))
{
SaveChar(c);
return(ParseDoctype());
}
else if (c == '-')
{
if (ReadChar() != '-')
{
throw new ParseError("Comments need to start with '<!--'.", ReadPosition);
}
return(ParseComment());
}
throw new ParseError("Comment or DOCTYPE expected.", ReadPosition);
}
// Process instruction ?
if (c == '?')
{
_State = ParseState.ProcessInstruction;
c = ReadChar(false);
// Pass whitespace
while (c != CharInfo.EOF && Char.IsWhiteSpace(c.AsChar))
{
c = ReadChar(false);
}
}
else if (c == '/')
{
_State = ParseState.EndTag;
c = ReadChar(false);
if (c == CharInfo.EOF || !Char.IsLetterOrDigit(c.AsChar))
{
SaveChar(c);
throw new ParseError("Invalid tag name. Need to start with an alphanumeric", ReadPosition);
}
}
else
{
_State = ParseState.Tag;
}
// Tagname
if (c == CharInfo.EOF || !Char.IsLetterOrDigit(c.AsChar))
{
throw new ParseError("Invalid tag name. Need to start with an alphanumeric", ReadPosition);
}
// Loop tag name
_CurrentRead = null;
AddToCurrentRead(c);
while ((c = ReadChar(false)) != CharInfo.EOF && (Char.IsLetterOrDigit(c.AsChar) || c == '.' || c == ':' || c == '-'))
{
AddToCurrentRead(c);
}
// If EndTag
if (_State == ParseState.EndTag)
{
_CurrentToken = ParsedTag.EndTag(GetCurrentRead(true));
_CurrentToken.Position = _StartTagPosition;
// Pass whitespace
while (c != CharInfo.EOF && Char.IsWhiteSpace(c.AsChar))
{
c = ReadChar(false);
}
try
{
if (c == CharInfo.EOF)
{
throw new ParseError("Unexpected end of stream.", ReadPosition);
}
if (IsAttributeNameChar(c.AsChar))
{
throw new ParseError("End tag can't contains attribute.", ReadPosition);
}
if (c != '>')
{
throw new ParseError("Unexpected char. End tag not closed.", ReadPosition);
}
}
catch
{
// Reset steam
while (c != CharInfo.EOF && c != '<' && c != '>')
{
c = ReadChar(false);
}
if (c == '<')
{
SaveChar(c);
}
throw;
}
_State = ParseState.Content;
ResetTagBuffer();
var result = _CurrentToken;
_CurrentToken = null;
return(result);
}
// Create the tag
if (c != CharInfo.EOF)
{
SaveChar(c);
}
_CurrentToken = _State == ParseState.Tag ? ParsedTag.OpenTag(GetCurrentRead(true)) : ParsedTag.OpenProcessInstruction(GetCurrentRead(true));
_CurrentToken.Position = _StartTagPosition;
return(_CurrentToken);
}
/// <summary>
/// Parse a content as raw text.
/// </summary>
/// <remarks>
/// This method is used for parsing the content of the script, style tag content.
/// The parsing is continue until matching the end of the <paramref name="tag"/>.
/// If <paramref name="tag"/> is null or empty then we accept all endtag.
/// </remarks>
/// <exception cref="InvalidOperationException">
/// Raised when the parser is not in a normal state : all tag need to be closed.
/// </exception>
/// <param name="tag">Tag name for the end tag expected.</param>
/// <returns>Content</returns>
public ParsedText ParseContentText(String tag)
{
// Verify
if (this._State != ParseState.Content)
{
throw new InvalidOperationException("Can't read a content in a opened tag.");
}
// Read loop
var start = SourceReader.Position;
CharInfo c;
while ((c = ReadChar(false)) != CharInfo.EOF)
{
// End detected ?
if (c == '<')
{
var endTagPos = c.Position;
StringBuilder saveTag = new StringBuilder(15);
saveTag.Append(c);
while ((c = ReadChar(false)) != CharInfo.EOF && Char.IsWhiteSpace(c.AsChar))
{
saveTag.Append(c.AsChar);
}
if (c == '/')
{
// Pass '/'
saveTag.Append(c);
while ((c = ReadChar(false)) != CharInfo.EOF && Char.IsWhiteSpace(c.AsChar))
{
saveTag.Append(c.AsChar);
}
if (c != CharInfo.EOF)
{
// Pass tag name
StringBuilder tagName = new StringBuilder(10);
saveTag.Append(c.AsChar);
tagName.Append(c.AsChar);
while ((c = ReadChar(false)) != CharInfo.EOF && IsAttributeNameChar(c.AsChar))
{
saveTag.Append(c.AsChar);
tagName.Append(c.AsChar);
}
// We find the good end tag ?
if (c != CharInfo.EOF)
{
if (String.IsNullOrEmpty(tag) || String.Equals(tagName.ToString(), tag, StringComparison.OrdinalIgnoreCase))
{
SaveChar(c);
// Search the good end
while ((c = ReadChar(false)) != CharInfo.EOF && Char.IsWhiteSpace(c.AsChar))
{
saveTag.Append(c.AsChar);
}
if (c == '>')
{
// Save the end tag for the next parse
_CurrentToken = ParsedTag.EndTag(tagName.ToString());
_CurrentToken.Position = endTagPos;
c = CharInfo.EOF; // Don't save the char
// Exit the loop
break;
}
}
}
}
}
// If here then we don't find a good end tag we convert to 'text'
var etp = endTagPos;
foreach (var st in saveTag.ToString())
{
AddToCurrentRead(new CharInfo(st, etp++));
}
}
//
AddToCurrentRead(c);
}
if (c != CharInfo.EOF)
{
SaveChar(c);
}
// Returns parse result
LastParsed = new ParsedText()
{
Position = start,
Text = HEntity.HtmlDecode(GetCurrentRead(true), RemoveUnknownOrInvalidEntities)
};
return((ParsedText)LastParsed);
}
