Field_varstring::store
      
The code that temporary saved the bitmaps of the read set and the write set so that
it can set it to all columns for debug purposes was not expecting that the
table->read_set and table->write_set can be the same. And was always saving both in 
sequence.
As a result the original value was never restored.
Fixed by saving & restoring the original value only once if the two sets are the
same (in a special set of functions).

mysql-test/r/select.result:
  Bug #37936: test case
mysql-test/t/select.test:
  Bug #37936: test case
sql/item_cmpfunc.cc:
  Bug #37936: don't save/restore twice if the read and write sets are the same
sql/opt_range.cc:
  Bug #37936: don't save/restore twice if the read and write sets are the same
sql/table.h:
  Bug #37936: Make a designated set of functions that save/restore
  both the read and the write sets in a single call.

no conflicts

mysql-test/suite/parts/r/partition_bit_innodb.result:
  updated test result
mysql-test/suite/parts/r/partition_bit_myisam.result:
  updated test result

if table has bit fields then uneven bits(if exist) are stored into null bits place.
So we need to copy null bits in case of uneven bit field presence.


mysql-test/r/type_bit.result:
  test result
mysql-test/t/type_bit.test:
  test case
sql/sql_select.cc:
  if table has bit fields then uneven bits(if exist) are stored into null bits place.
  So we need to copy null bits in case of uneven bit field presence.

added ability for TINY[MEDIUM] text fields 
to be converted to greater subtype during
alter if necessary(altered charset)


mysql-test/r/alter_table.result:
  test result
mysql-test/t/alter_table.test:
  test case
sql/sql_table.cc:
  added ability for TINY[MEDIUM] text fields 
  to be converted to greater subtype during
  alter if necessary(altered charset)

mysql-test/r/myisam_data_pointer_size_func.result:
  restore default value
mysql-test/r/partition_datatype.result:
  test case fix
mysql-test/t/myisam_data_pointer_size_func.test:
  restore default value

the patch manually to another tree. 

the problem: FORMAT func max_length value was calculated incorrectly
the fix: correct calculation of max_length


mysql-test/r/func_str.result:
  test result
mysql-test/t/func_str.test:
  test case
sql/item_strfunc.h:
  the problem: FORMAT func max_length value was calculated incorrectly
  the fix: correct calculation of max_length

show default value for BIT field in printable format


mysql-test/r/type_bit.result:
  Bug#35796 SHOW CREATE TABLE and default value for BIT field
  test result
mysql-test/t/type_bit.test:
  Bug#35796 SHOW CREATE TABLE and default value for BIT field
  test case
sql/item.cc:
  Bug#35796 SHOW CREATE TABLE and default value for BIT field
  issue an error if BIN|HEX deafult value is empty value
  (behaviour like for INT fields)

reset diagnostics area state after error message is sent


mysql-test/r/myisampack.result:
  test result
mysql-test/t/myisampack.test:
  test case
sql/sql_table.cc:
  reset diagnostics area state after error message is sent

    modifying the original fix.
    As it turned out --fix-db-names option of the mysqlcheck suppress
    the --check_upgrade option, so we have to call the mysqlcheck twice
    from the mysql_upgrade.

per-file comments:
  client/mysql_upgrade.c
Bug#35934      mysql_upgrade calls mysqlcheck with insufficient parameters

The reason for the failures was bug #21205 (fixed in 6.0 by dtoa, but still present in 5.0/5.1).

mysql-test/r/type_float.result:
  Modified the failing test case so that bug #21205 does not occur.
mysql-test/t/type_float.test:
  Modified the failing test case so that bug #21205 does not occur.

Nothing to do, was already in 5.1

binlog_row_mix_innodb_myisam resutls are corrected.
The last operations prior the dup error is
TRUNCATE table t2;
Therefore after
--error ER_DUP_ENTRY
INSERT INTO t2 select * from t1;
table t2 must be empty, and that is what the updated results confirm.


mysql-test/suite/binlog/r/binlog_row_mix_innodb_myisam.result:
  results changed due to Bug #33420 fixing.