Question

So given this input string:

=?ISO-8859-1?Q?TEST=2C_This_Is_A_Test_of_Some_Encoding=AE?=

And this function:

private string DecodeSubject(string input)
        {
            StringBuilder sb = new StringBuilder();
            MatchCollection matches = Regex.Matches(inputText.Text, @"=\?(?<encoding>[\S]+)\?.\?(?<data>[\S]+[=]*)\?=");
            foreach (Match m in matches)
            {
                string encoding = m.Groups["encoding"].Value;
                string data = m.Groups["data"].Value;

                Encoding enc = Encoding.GetEncoding(encoding.ToLower());
                if (enc == Encoding.UTF8)
                {
                    byte[] d = Convert.FromBase64String(data);
                    sb.Append(Encoding.ASCII.GetString(d));
                }
                else
                {                    
                    byte[] bytes = Encoding.Default.GetBytes(data);
                    string decoded = enc.GetString(bytes);
                    sb.Append(decoded);
                }
            }

            return sb.ToString();

        }

The result is the same as the data extracted from the input string. What am i doing wrong that this text is not getting decoded properly?

UPDATE

So i have this code for decoding quote-printable:

public string DecodeQuotedPrintable(string encoded)
        {
            byte[] buffer = new byte[1];
            return Regex.Replace(encoded, "=(\r\n?|\n)|=([A-F0-9]{2})", delegate(Match m)
            {
                if (byte.TryParse(m.Groups[2].Value, NumberStyles.HexNumber, CultureInfo.InvariantCulture, out buffer[0]))
                {
                    return Encoding.ASCII.GetString(buffer);
                }
                else
                {
                    return string.Empty;
                }
            });
        }

And that just leaves the underscores. Do i manually convert those to spaces (Replace(“_”,” “)), or is there something else i need to do to handle that?

Answer 1

Looks like you don’t fully understand format of input line. Check it here: http://www.ietf.org/rfc/rfc2047.txt
format is: encoded-word = "=?" charset "?" encoding "?" encoded-text "?="

so you have to

1. Extranct charset(encoding in terms of .net). Not just UTF8 or Default (Utf16)
2. Extract encoding: either B for base64 Q for quoted-printable (your case!)
3. Then perform decoding to bytes then to string