public override void VisitHtml(HtmlContentIntermediateNode node)
{
for (var i = 0; i < node.Children.Count; i++)
{
var child = node.Children[i];
if (!(child is IntermediateToken token) || !token.IsHtml || string.IsNullOrEmpty(token.Content))
{
// We only care about Html tokens.
continue;
}
for (var j = 0; j < token.Content.Length; j++)
{
var ch = token.Content[j];
// ASCII range is 0 - 127
if (ch > 127 || EncodedCharacters.Contains(ch))
{
node.SetEncoded();
return;
}
}
}
// If we reach here, we don't have newlines, tabs or non-ascii characters in this node.
// If we can successfully decode all HTML entities(if any) in this node, we can safely let it call AddContent.
var decodedContent = new string[node.Children.Count];
for (var i = 0; i < node.Children.Count; i++)
{
var child = node.Children[i];
if (!(child is IntermediateToken token) || !token.IsHtml || string.IsNullOrEmpty(token.Content))
{
// We only care about Html tokens.
continue;
}
if (TryDecodeHtmlEntities(token.Content, out var decoded))
{
decodedContent[i] = decoded;
}
else
{
node.SetEncoded();
return;
}
}
// If we reach here, it means we have successfully decoded all content.
// Replace all token content with the decoded value.
for (var i = 0; i < node.Children.Count; i++)
{
var child = node.Children[i];
if (!(child is IntermediateToken token) || !token.IsHtml || string.IsNullOrEmpty(token.Content))
{
// We only care about Html tokens.
continue;
}
token.Content = decodedContent[i];
}
}
