Thank you to
Avensen for letting me know that some URLs in certain character sets were still returning garbage characters.
This lead me to review the code and as well as a bug in the conversion of the encoding, I added some additional code to better detect the original encoding and some improvements to the code in general.