char-encoding, input-encoding, output-encoding

Why are there 3 options? Why not just input-encoding and
output-encoding? If specified, do input-encoding and output-encoding
override char-encoding?

Also, in the quickref
(http://tidy.sourceforge.net/docs/quickref.html#char-encoding) it
states...

"For mac and win1252, Tidy will accept vendor specific character values,
but will use entities for all characters whose value > 127."

I assume this is referring to the output side. Does this mean that chars
like å will be converted to entities like \&\#\n\n\n\; (remove
backslashes) on output?

-- 

Dennis