Revision 0.3.4:
- Fixed Native encoder to handle SEQUENCE/SET objects without
  the componentType property
- Added missing component-less SEQUENCE/SET objects dict duck-typing support
- Fixed unnecessary duplicate tags detection at NamesType.tagMap
- Fixed crash at SEQUENCE and SEQUENCE OF CER encoder when running
  in schemaless mode
- Fixed Character types instantiation from OctetString type -- double
  unicode decoding may have scrambled the data

(adam)