RFC: Add support for Character Encoding

[lestcape]

This is marked as RFC (Request For Comments and discussions from developers), because this add support for Character Encoding, but this solution implies read the body twice.

The CharacterEncoding data should not necessary be set at the beginning of the body. So, any string that is set before found the Character Encoding of the file body it is then not decode. This is why, currently we need to read twice the body, once to find the encode and another to decode the body.

The question then is: We can find a way to resolve this problematic without need to read the body twice?