Updated to take character encoding into accout:
- Correct character encoding labeling and detection is a prerequesite for
validation (and all other operations!)
- Works now for encodings such as iso-2022-jp,
Shift_JIS,... (the "<" byte is used for the '<'
character and other things)
- Follows standard priorities (example)
- Converts bytes to characters and fails
if this is not possible (more examples)
- Cannot catch all problems automatically, because some encodings are
very similar, but allows trial-and-error
with visual
(or audio) checking
- To be done: Support UTF-16
