/// <summary>
/// Gets a named entity starting at the currrent
/// character ('&') and proceeding until ';'.
/// Returns null if the substring does not match
/// the expected entity sequence.
/// </summary>
private static string GetNamedEntity(ITextParser tp)
{
const string kValidCharacters = "abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ";
var startPos = tp.Position + 1;
var offset = 1;
char c;
while (true)
{
c = tp.Peek(offset);
if (kValidCharacters.IndexOf(c) == -1)
{
break;
}
offset++;
}
if (c != ';')
{
return(null);
}
var length = offset - 1;
if (length < 1)
{
return(null);
}
return(tp.Substring(startPos, length));
}
/// <summary>
/// Skips all characters until the end tag for the given
/// <paramref name="tagName"/>.
/// </summary>
/// <remarks>
/// Uses a simple rule: the CData ends at the next
/// instance of the end tag. An end that appears in
/// the text of the tag--not sure if that can
/// happen, but maybe--will be treated as the end.
/// </remarks>
private void SkipToEndOfScriptCData(string tagName)
{
var endOffset = tagName.Length + 2;
var haveCData = false;
while (!tp.EndOfText)
{
if (tp.Peek() == '<' && tp.Peek(1) == '/' && tp.Peek(endOffset) == '>')
{
var endTag = tp.Substring(tp.Position + 2, tagName.Length);
if (endTag.Equals(tagName, StringComparison.OrdinalIgnoreCase))
{
break;
}
}
if (!haveCData)
{
haveCData = true;
xml.Append(kScriptCDataStart);
}
xml.Append(tp.Peek());
tp.MoveAhead();
}
if (haveCData)
{
xml.Append(kScriptCDataEnd);
}
return;
}
/// <summary>
/// Parses the element name from the tag using the given
/// <see cref="ITextParser"/> <paramref name="tp"/>. The
/// <see cref="ITextParser"/> position should be set to
/// the first character of the tag following the "<",
/// or following the "</" for end tags.
/// Returns true if a syntactically valid name is found.
/// </summary>
private bool ParseName(ITextParser tp)
{
const string nameCharacters = "abcdefghijklmnopqrstuvwxyz"
+ "ABCDEFGHIJKLMNOPQRSTUVWXYZ0123456789:";
var startPos = tp.Position;
var offset = 0;
char c;
while ((c = tp.Peek(offset)) != TextParser.NullChar)
{
if (nameCharacters.IndexOf(c) == -1)
{
break;
}
offset++;
}
// Did we get any valid characters?
if (offset < 1)
{
return(false);
}
// Does the tag name end properly?
if (c != '>' && c != '/' && !Char.IsWhiteSpace(c))
{
return(false);
}
if (tp.Peek(offset - 1) == ':')
{
return(false);
}
// Our minimal validation has passed...
tp.MoveAhead(offset);
var length = tp.Position - startPos;
Name = tp.Substring(startPos, length).ToLower();
// Force void elements to be self-closing tags.
IsSelfClosingTag = IsVoidElement(Name);
return(true);
}