MixedUtf8.cs

static string PreProcessMixedEntities(string xml)
{
    // Handle standard Unicode character references (4-6 hex digits)
    string processed = Regex.Replace(xml, 
        @"&#x([0-9A-Fa-f]{4,6});", 
        match => {
            int codePoint = Convert.ToInt32(match.Groups[1].Value, 16);
            return char.ConvertFromUtf32(codePoint);
        });
    
    // Find sequences of byte entities (patterns like ●)
    processed = Regex.Replace(processed, 
        @"(&#x[0-9A-Fa-f]{2};)+", 
        match => {
            // Extract all hex values from the sequence
            var hexValues = Regex.Matches(match.Value, @"&#x([0-9A-Fa-f]{2});")
                .Cast<Match>()
                .Select(m => Convert.ToByte(m.Groups[1].Value, 16))
                .ToArray();
            
            // Only process if we have a valid UTF-8 sequence
            try {
                return Encoding.UTF8.GetString(hexValues);
            }
            catch {
                // If not a valid UTF-8 sequence, return the original
                return match.Value;
            }
        });
    
    return processed;
}