TODO:
Add a unicode-mappings.hrl file containing a list
[{CHARSET,BitMap},...]
If no such definition is given, Erlang "integer" is chosen.


KNOWN ISSUES
- All mappings are made against an integer. There is currently no support
  for different byte orders and e.g. 7 bit encoding/decoding.
- Mappings between Unicode and some character sets has moved to a separate file
  in the Unicide standard (v3.1). 


Changes  0.2 -> 0.3 
-------------------
- Added support for non-IANA defined character sets (e.g. GSM 03.38)
- Made the "preferred MIME name", if given, the defined name of a character set.
- Improved performance for character sets known to Unicode subsets

Changes  0.1 -> 0.2 
-------------------
- Added support for IANA character set names. Includes functions to lookup
  IANA defined name/alias, given MIB number and vice versa.
- Added a "first try" for generic transformation between Unicode and character
  sets to where there are "mappings" defined.
- Added function to test wheter a given character code has a character mapping
  within a given character set.
