Added put_length() to get_length() and unpack_key() to pack_key().
Keys were packed with the minimum size of the length field for the key part and 
unpacked with length size of the base column. 
For the purpose of optimal key packing we have the method pack_key(), while rows are 
packed with pack(). Now keys are unpacked with unpack_key() and no longer with 
unpack() which is used for rows.


mysql-test/r/bdb.result:
  bug#2686 - index_merge select on BerkeleyDB table with varchar PK causes mysqld to crash.
  Added the test case results.
mysql-test/t/bdb.test:
  bug#2686 - index_merge select on BerkeleyDB table with varchar PK causes mysqld to crash.
  Added the test case.
sql/field.cc:
  bug#2686 - index_merge select on BerkeleyDB table with varchar PK causes mysqld to crash.
  Added put_length() to get_length() and unpack_key() to pack_key().
  Keys were packed with the minimum size of the length field for the key part and 
  unpacked with length size of the base column. 
  For the purpose of optimal key packing we have the method pack_key(), while rows are packed 
  with pack(). Now keys are unpacked with unpack_key() and no longer with unpack() which is 
  used for rows.
sql/field.h:
  bug#2686 - index_merge select on BerkeleyDB table with varchar PK causes mysqld to crash.
  Added put_length() to get_length() and unpack_key() to pack_key().
  The default implementation simply calls unpack() for those field types that don't need 
  a special key unpacking.
sql/ha_berkeley.cc:
  bug#2686 - index_merge select on BerkeleyDB table with varchar PK causes mysqld to crash.
  Now keys are unpacked with unpack_key() and no longer with unpack() which is used for rows.
  For most field types, however, this simply calls unpack().