result set when SQLEXCEPTION is active.

The problem was in a hackish THD::no_warnings_for_error attribute.
When it was set, an error was not written to Warning_info -- only
Diagnostics_area state was changed. That means, Diagnostics_area
might contain error state, which is not present in Warning_info.

The user-visible problem was that in some cases SHOW WARNINGS
returned empty result set (i.e. there were no warnings) while
the previous SQL statement failed. According to the MySQL
protocol errors must be presented in warning list.

The main idea of this patch is to remove THD::no_warnings_for_error.
There were few places where it was used:
  - sql_admin.cc, handling of REPAIR TABLE USE_FRM.
  - sql_show.cc, when calling fill_schema_table_from_frm().
  - sql_show.cc, when calling fill_table().
The fix is to either use internal-error-handlers, or to use
temporary Warning_info storing warnings, which might be ignored.

This patch is needed to fix Bug 11763162 (55843).

Some multibyte sequences could be considered by my_mbcharlen() functions
as multibyte character but more exact my_ismbchar() does not think so.
In such a case this multibyte sequences is pushed into 'stack' buffer which
is too small to accommodate the sequence.
The fix is to allocate stack buffer in
compliance with max character length.

Valgrind warnings were caused by comparing index values to an un-initialized field.

               ASSERTION THD->TRANSACTION.XID_STATE.XID.IS_NULL()
               FAILED

The triggered assert checks that the previous XA transaction has
done proper cleanup before a new XA transaction is started.
The bug that caused it to be triggered was that XA COMMIT did not
clean up error state if XA COMMIT discovered that the current XA
transaction had to be rolled back.

This patch fixes the problem by resetting the XA error state
before XA COMMIT calls ha_rollback_trans(). This allows following
XA transactions to be started without triggering the assert.

Test case added to xa.test.

There are two problems with ANALYSE():

1. Memory leak 
   it happens because do_select() can overwrite
   JOIN::procedure field(with zero value in our case) and
   JOIN destructor don't free the memory allocated for
   JOIN::procedure. The fix is to save original JOIN::procedure
   before do_select() call and restore it after do_select
   execution.

2. Wrong result
   If ANALYSE() procedure is used for the statement with LIMIT clause
   it could retrun empty result set. It happens because of missing 
   analyse::end_of_records() call. First end_send() function call
   returns NESTED_LOOP_QUERY_LIMIT and second call of end_send() with
   end_of_records flag enabled does not happen. The fix is to return
   NESTED_LOOP_OK from end_send() if procedure is active.

When MySQL converted from autotools to CMake, the
preprocessor symbol USE_SYMDIR was omitted by mistake.

Without this symbol, the code for checking .sym files
is not built.

This patch defines USE_SYMDIR when built on MS Windows.

the list. Previously, the counter would only be incremented if the
insertion method push_front() was used, in which case the counter
wouldn't be incremented if a element was inserted using the push_back()
and/or insert_after() methods.

Currently this does not affect the code base because there isn't any
code that uses a counted list with the push_back() or insert_after()
methods.

CLAUSE FAILS OR ABORTS SERVER".
 
Attempt to re-execute prepared ALTER TABLE statement which 
involves .FRM-only changes and also have RENAME clause led
to unwarranted 'Table doesn't exist' error in production
builds and assertion failure for debug builds.
 
This problem stemmed from the fact that for such ALTER TABLE
mysql_alter_table() code changed table list element for table 
to be altered when it tried to re-open table under new name. 
Since this change was not reverted back before next 
re-execution, it made this statement re-execution unsafe.
 
This fix addresses this problem by avoiding changing table list
element from the main table list in such a situation. Instead 
temporary TABLE_LIST object is used.

               TRX->CONC_STATE == 0 || TRX->CONC_STATE == 1

This bug was a different manifestation of Bug#11766752,
which was previously only fixed on mysql-trunk.

This patch backports the fix for Bug#11766752 to mysql-5.5,
which fixes the problem. The patch also adds some extra test
coverage.

When we create temporary result table for UNION
incorrect max_length for YEAR field is used and
it leads to incorrect field value and incorrect
result string length as YEAR field value calculation
depends on field length.
The fix is to use underlying item max_length for
Item_sum_hybrid::max_length intialization.

Valgrind warning happens due to early null values check
in Item_func_in::fix_length_and_dec(before item evaluation).
As result null value items with uninitialized values are
placed into array and it leads to valgrind warnings during
value array sorting.
The fix is to check null value after item evaluation, item
is evaluated in in_array::set() method.

to a bug that has been fixed. made these tests non-experimental.

for code other then in mdl.h. Since the adapter is generic, it is
better located in sql_plist.h.