NOT FOUND ...' in conditional handled incorrectly".

Whenever we remove an instruction during optimization, we need to
adjust instruction numbers (ip - instruction pointer) stored in all
instructions. In addition to that, sp_instr_hpush_jump, which 
corresponds to DECLARE CONTINUE HANDLER needs adjustment for m_handler,
which holds the number of instruction with the continue handler.
In the bug report, a wrong ip stored in m_handler was pointing at 
FETCH, which resulted in an error message and abnormal SP termination.
The fix is to just remove m_handler member from sp_instr_hpush_jump,
as it's always points to the instruction next to the DECLARE
statement itself (m_ip+1).


mysql-test/r/sp.result:
  Test results fixed (Bug#12168)
mysql-test/t/sp.test:
  A test case for Bug#12168 "'DECLARE CONTINUE HANDLER FOR NOT
  FOUND ...' in conditional handled incorrectly"
sql/sp_head.cc:
  Remove m_handler (the number of continue handler instruction)
  as it always equal to m_ip+1
sql/sp_head.h:
  Remove m_handler (the number of continue handler instruction)
  as it always equal to m_ip+1

  Add diagnostic code to track a crash in adaptive hash indexes; the crash may happen if block->index == NULL


innobase/btr/btr0sea.c:
  Add diagnostic code to track a crash in adaptive hash indexes; the crash may happen if block->index == NULL

into  sanja.is.com.ua:/home/bell/mysql/bk/work-bug2-5.0

  Temporarily remove error print if MySQL tries to do a SELECT even though trx->n_mysql_tables_in_use is 0. We must return the error print when the lock count of cursors is properly implemented (Bug #12456)


innobase/row/row0sel.c:
  Temporarily remove error print if MySQL tries to do a SELECT even though trx->n_mysql_tables_in_use is 0. We must return the error print when the lock count of cursors is properly implemented (Bug #12456)

into  sanja.is.com.ua:/home/bell/mysql/bk/work-bug2-5.0

into lmy004.:/work/mysql-5.0-clean

mysql-test/r/innodb.result:
  fix result for fix of 12591
mysql-test/t/innodb.test:
  fix test for fix of 12591

into  sanja.is.com.ua:/home/bell/mysql/bk/work-bug2-5.0


sql/item_cmpfunc.cc:
  Auto merged

into  hundin.mysql.fi:/home/jan/mysql-5.0

into lmy004.:/work/mysql-5.0-bug12591


sql/sql_show.cc:
  Auto merged

into moonbone.local:/work/10624-bug-5.0-mysql


sql/sql_lex.cc:
  Auto merged

into  mysql.com:/home/jimw/my/mysql-5.0-clean


sql/item_strfunc.cc:
  Auto merged

into rurik.mysql.com:/home/igor/mysql-5.0

into mysql.com:/home/psergey/mysql-5.0-aug18-fix

Undo the test cset that shouldn't have been pushed (ChangeSet@1.1988, 2005-08-16 00:03:50+04:00, aivanov@mysql.com
 )

into rurik.mysql.com:/home/igor/mysql-5.0

into  mysql.com:/home/stewart/Documents/MySQL/5.0/main

into rurik.mysql.com:/home/igor/dev/mysql-5.0-0

into mysql.com:/usersnfs/pchardin/mysql-5.0

    and Bug#12297 SP crashes the server if data inserted inside a lon loop
Third commit attempt. With fixes to the issues, showed up after full rebuild and
tests on other hosts.


mysql-test/r/rpl_sp.result:
  New warnings appeared in result file, as now we always create spcont in a stored routine.
  This is correct behaviour. We swallowed some warnings, as we used thd->spcont to check whether
  we are in the SP though we didn't set spcont in certain cases. This is fixed now.
mysql-test/r/sp.result:
  fixed result file to reflect new tests
mysql-test/t/sp.test:
  Added tests for bugs. Though one of them is disabled, as it fails because of the other bug.
  It should be enabled, when bug 12297 is fixed.
sql/sp_head.cc:
  Per-instruction arena is implemented
sql/sp_rcontext.cc:
   Now we should deal with callers_arena->free_list when we employ reuse mechanism with callers_arena
   switched during sp_eval_func_item
sql/sp_rcontext.h:
  Add new member to sp_rcontext class, in order to handle instructions with assignment
  and/or with nested SP processing properly.

mysql-test/r/row.result:
  Correct NULL handling in row comporison
mysql-test/t/row.test:
  Correct NULL handling in row comporison (BUG#12509)
sql/item_cmpfunc.cc:
  stop on NULL comparison only if it is allowed
sql/item_cmpfunc.h:
  support optimisation in case of processing WHERE/HAVING, where NULL and FALSE are equal

- Fixed minor error after Monty's review.


sql/table.cc:
  cur_table_ref is used in the first pass of the loop *before* it is set later in the loop.

into  mysql.com:/home/stewart/Documents/MySQL/5.0/main

config/ac-macros/ha_ndbcluster.m4:
  Merge fix for BUG#11538 fix.

into rurik.mysql.com:/home/igor/dev/mysql-5.0-0


sql/sql_view.cc:
  Auto merged

  Added a test case for bug #10970.
view.result:
  Added a test case for bug #10970.
  Modified the error messages for error ER_VIEW_SELECT_TMPTABLE.
sql_view.cc:
  Fixed bug #10970.
  In the function mysql_create_view if a view does not refer
  any tables directly the variable table must be updated
  after the call of open_and_lock_tables.
errmsg.txt:
  Modified the error messages for error ER_VIEW_SELECT_TMPTABLE
  (when fixing bug #10970).


sql/share/errmsg.txt:
  Modified the error messages for error ER_VIEW_SELECT_TMPTABLE
  (when fixing bug #10970).
sql/sql_view.cc:
  Fixed bug #10970.
  In the function mysql_create_view if a view does not refer
  any tables directly the variable table must be updated
  after the call of open_and_lock_tables.
mysql-test/r/view.result:
  Fixed bug #10970.
  Modified the error messages for error ER_VIEW_SELECT_TMPTABLE.
mysql-test/t/view.test:
  Added a test case for bug #10970.

This is quick fix and needs to be fixed when semi-consistent 
high-granularity read view is implemented (Bug #12456).

sql/item_func.cc:
  Remove unnecessary initialization, fix parens placement

mysql-test/r/select.result:
  Update results
mysql-test/t/select.test:
  Remove regression test
sql/item_cmpfunc.cc:
  Revert patch for checking length of ESCAPE, it is not correct.

into  mysql.com:/home/jimw/my/mysql-5.0-clean

sql/item_strfunc.cc:
  Fix bad merge

include/my_global.h:
  Add distinct set_timespec_nsec() macro for when HAVE_TIMESPEC_TS_SEC
  is defined.

into  mysql.com:/home/my/mysql-5.0


sql/sql_parse.cc:
  Auto merged

mysql-test/t/kill.test:
  Moved --disable_reconnect earlier to avoid race condition
sql/sql_help.cc:
  Cleanup during review of new code
  (Moved variable definitions first in function
sql/sql_insert.cc:
  Cleanup during review of new code
sql/sql_lex.cc:
  Cleanup during review of new code
sql/sql_parse.cc:
  Cleanup during review of new code
  Changed potential problem with previous_table_ref where it dependent that TABLE_LIST first element would be next_local
  Rearanged code in add_table_to_list() to remove extra if
  Combined 2 calls to calloc() to one
sql/sql_view.cc:
  Remove extra indentation level
  Combined common 'on error' exit
sql/sql_yacc.yy:
  Fixed comment style
sql/table.cc:
  Cleanup during review of new code
  - Changed while() loops to for() loop (to make code more readable)
  - Removed not needed initialization of variables
  - Removed not needed 'else' cases
  - Removed trivial ASSERT's that was checked by previous code
  - Moved comment setting last in Natural_join_column::check_grants()

into  mysql.com:/home/alexi/mysql-5.0

mysql-test/r/show_check.result:
  results for test for bug #12591
mysql-test/t/show_check.test:
  test for bug #12591
sql/sql_show.cc:
  don't let my_dir() emit a message bug catch the errno (my_errno - a thread specific variable) and check it for ENOENT - not existing.
  For other errors emulate the message used to be emitted by my_dir()