warnings in sql_trigger.cc and sql_view.cc".

According to the current version of C++ standard offsetof() macro
can't be used for non-POD types. So warnings were emitted when we
tried to use this macro for TABLE_LIST and Table_triggers_list
classes. Note that despite of these warnings it was probably safe
thing to do.

This fix tries to circumvent this limitation by implementing
custom version of offsetof() macro to be used with these
classes. This hack should go away once we will refactor
File_parser class.

Alternative approaches such as disabling this warning for
sql_trigger.cc/sql_view.cc or for the whole server were
considered less explicit. Also I was unable to find a way
to disable particular warning for particular _part_ of
file in GCC.


sql/parse_file.h:
  Introduced auxillary macro which can be used instead of offsetof()
  to get offsets of members in class for non-POD types without getting
  warnings (assuming that all instances of the class has same offsets
  for same members).
sql/sql_trigger.cc:
  Use my_offsetof() macro instead of standard offsetof() macro with
  Table_triggers_list class in order to avoid warnings (offsetof()
  cannot be used for non-POD types according to the standard).
sql/sql_view.cc:
  Use my_offsetof() macro instead of standard offsetof() macro with
  TABLE_LIST class in order to avoid warnings (offsetof() cannot
  be used for non-POD types according to the standard).

into  moonlight.intranet:/home/tomash/src/mysql_ab/mysql-5.0-bug21726


mysql-test/r/rpl_insert_id.result:
  Use local.
mysql-test/t/rpl_insert_id.test:
  Use local.
sql/item_func.cc:
  Use local.
sql/item_func.h:
  Use local.
sql/log_event.cc:
  Use local.
sql/set_var.cc:
  Use local.
sql/sql_class.h:
  Use local.
sql/sql_insert.cc:
  Use local.
sql/sql_load.cc:
  Use local.
sql/sql_parse.cc:
  Use local.
sql/sql_select.cc:
  Use local.
sql/sql_update.cc:
  Use local.
tests/mysql_client_test.c:
  Use local.

Note: bug#21726 does not directly apply to 4.1, as it doesn't have stored
procedures.  However, 4.1 had some bugs that were fixed in 5.0 by the
patch for bug#21726, and this patch is a backport of those fixes.
Namely, in 4.1 it fixes:

  - LAST_INSERT_ID(expr) didn't return value of expr (4.1 specific).

  - LAST_INSERT_ID() could return the value generated by current
    statement if the call happens after the generation, like in

      CREATE TABLE t1 (i INT AUTO_INCREMENT PRIMARY KEY, j INT);
      INSERT INTO t1 VALUES (NULL, 0), (NULL, LAST_INSERT_ID());

  - Redundant binary log LAST_INSERT_ID_EVENTs could be generated.


mysql-test/r/rpl_insert_id.result:
  Add result for bug#21726: Incorrect result with multiple invocations
  of LAST_INSERT_ID.
mysql-test/t/rpl_insert_id.test:
  Add test case for bug#21726: Incorrect result with multiple invocations
  of LAST_INSERT_ID.
sql/item_func.cc:
  Add implementation of Item_func_last_insert_id::fix_fields(), where we
  set THD::last_insert_id_used when statement calls LAST_INSERT_ID().
  In Item_func_last_insert_id::val_int(), return THD::current_insert_id
  if called like LAST_INSERT_ID(), otherwise return value of argument if
  called like LAST_INSERT_ID(expr).
sql/item_func.h:
  Add declaration of Item_func_last_insert_id::fix_fields().
sql/log_event.cc:
  Do not set THD::last_insert_id_used on LAST_INSERT_ID_EVENT.  Though we
  know the statement will call LAST_INSERT_ID(), it wasn't called yet.
sql/set_var.cc:
  In sys_var_last_insert_id::value_ptr(), set THD::last_insert_id_used,
  and return THD::current_insert_id for @@LAST_INSERT_ID.
sql/sql_class.h:
  Update comments.
  Remove THD::insert_id(), as it has lost its purpose now.
sql/sql_insert.cc:
  Now it is OK to read THD::last_insert_id directly.
sql/sql_load.cc:
  Now it is OK to read THD::last_insert_id directly.
sql/sql_parse.cc:
  In mysql_execute_command(), remember THD::last_insert_id (first
  generated value of the previous statement) in THD::current_insert_id,
  which then will be returned for LAST_INSERT_ID() and @@LAST_INSERT_ID.
sql/sql_select.cc:
  If "IS NULL" is replaced with "= <LAST_INSERT_ID>", use right value,
  which is THD::current_insert_id, and also set THD::last_insert_id_used
  to issue binary log LAST_INSERT_ID_EVENT.
sql/sql_update.cc:
  Now it is OK to read THD::last_insert_id directly.
tests/mysql_client_test.c:
  Add test case for bug#21726: Incorrect result with multiple invocations
  of LAST_INSERT_ID.

into  mockturtle.local:/home/dlenev/src/mysql-5.0-rt-merge


mysql-test/r/temp_table.result:
  Auto merged
mysql-test/t/temp_table.test:
  Auto merged
sql/sql_select.cc:
  Manual merge.
sql/sql_table.cc:
  Manual merge.

into  mockturtle.local:/home/dlenev/src/mysql-5.0-rt-merge


BitKeeper/etc/collapsed:
  auto-union
mysql-test/mysql-test-run.pl:
  Auto merged
sql/mysql_priv.h:
  Auto merged
sql/sp.cc:
  Auto merged
sql/sql_acl.cc:
  Auto merged
mysql-test/r/sp.result:
  Manual merge.
mysql-test/t/sp.test:
  Manual merge.

create_tmp_table()".

The fix for bug 21787 "COUNT(*) + ORDER BY + LIMIT returns wrong
result" introduced valgrind warnings which occured during execution
of information_schema.test and sp-prelocking.test in version 5.0.
There were no user visible effects.

The latter fix made create_tmp_table() dependant on
THD::lex::current_select value. Valgrind warnings occured when this
function was executed and THD::lex::current_select member pointed
to uninitialized SELECT_LEX instance.

This fix tries to remove this dependancy by moving some logic
outside of create_tmp_table() function.


sql/sql_select.cc:
  create_tmp_table():
    Moved code which is responsible for determining if optimization
    which pushes down LIMIT clause to temporary table creation is
    applicable out of this function.
    Such move made this function independant of THD::lex::current_select
    value and removed valgrind warnings which occured in cases when this
    member pointed to uninitialized SELECT_LEX object (particularly these
    warnings occured in sp-prelocking.test and information_schema.test
    in 5.0). This seems like a better solution than trying to force this
    pointer always to point to relevant select because:
    - In some cases when we use create_tmp_table() there are no relevant
      SELECT_LEX object (we use it just to create temporary table/object).
    - There is only one place in code where we call this funciton and
      where this optimization can be enabled. And in this place we
      already have some logic which tries to determine if it is applicable.

into  mockturtle.local:/home/dlenev/src/mysql-4.1-runtime

into  example.com:/work/mysql-5.0-runtime


mysql-test/t/sp.test:
  Auto merged
sql/sp.cc:
  Auto merged
mysql-test/r/sp.result:
  manual merge

  
There was possible stack overrun in an edge case which handles invalid body of
a SP in mysql.proc . That should be case when mysql.proc has been changed
manually. Though, due to bug 21513, it can be exploited without having access
to mysql.proc only being able to create a stored routine.


mysql-test/r/sp.result:
  update result
mysql-test/t/sp.test:
  add a test case for the bug
sql/sp.cc:
  Fix stack overrun. This happen mostly when mysql.proc is damaged, though
  it's possible due to another bug which creates invalid SP body in mysql.proc
  (leading quote from a label being cut) to create stack overrun even without
  having direct access to mysql.proc

into  mysql.com:/home/gluh/MySQL/Merge/5.0


sql/sql_parse.cc:
  Auto merged
sql-common/client.c:
  Auto merged
sql/table.cc:
  Auto merged
include/mysql_com.h:
  manual merge
mysql-test/r/ctype_utf8.result:
  manual merge
mysql-test/t/ctype_utf8.test:
  manual merge
sql/sql_acl.cc:
  manual merge

into  moonlight.intranet:/home/tomash/src/mysql_ab/mysql-5.0-bug21414


mysql-test/r/sp.result:
  Auto merged
mysql-test/t/sp.test:
  Auto merged
sql/mysql_priv.h:
  Auto merged
sql/sp.cc:
  Auto merged
sql/sql_acl.cc:
  Auto merged

into  neptunus.(none):/home/msvensson/mysql/mysql-5.0-runtime


server-tools/instance-manager/instance.cc:
  Auto merged

into  trift2.:/MySQL/M50/push-5.0


mysql-test/r/func_time.result:
  Null-merge, the change in the 5.0 general tree supersedes the 5.0.25 patch.
mysql-test/r/query_cache.result:
  Auto merged
mysql-test/r/type_date.result:
  Auto merged
mysql-test/r/view.result:
  Null-merge, the change in the 5.0 general tree supersedes the 5.0.25 patch.
mysql-test/t/func_time.test:
  Null-merge, the change in the 5.0 general tree supersedes the 5.0.25 patch.
mysql-test/t/type_date.test:
  Auto merged
mysql-test/t/view.test:
  Null-merge, the change in the 5.0 general tree supersedes the 5.0.25 patch.
sql/item_cmpfunc.cc:
  Null-merge, the change in the 5.0 general tree supersedes the 5.0.25 patch.

Remove race situations that occur when removing pidfiles. Primarily each process should remove its own
pidfile, secondly it should be removed by the process that created it and _only_ if it's
certain the process is dead. Third, mysql-test-run.pl will remove the pidfile when process has been killed.
- Set state of an instance to STARTING _before_ calling instance->start()
- Check that pidfile of instance has been created before changing STARTING => STARTED
- Only remove the pidfile if IM kills an instance with SIGKILL, otherwise the instance will remove it itself


server-tools/instance-manager/guardian.cc:
  If state of an instance is STARTING, chech that the instance pidfile has been created
  before changing state to STARTED
  Set state to STARTING before calling instance->start(), it can take some time
  before it is fully started and during that time it should be in state STARTING
server-tools/instance-manager/instance.cc:
  Only remove the pid file of instance manager when a SIGKILL has 
  been performed sucessfully
server-tools/instance-manager/instance_options.cc:
  Check that fscanf returns 1 which is the number of args that should be scanned from
  the pid file

into  mysql.com:/Users/kent/mysql/bk/mysql-5.0

  Filter out plain -O and Sun C/C++ style optimization flags, -xO<level>
  Filter out icc specific options from cflags/libs(_r)


scripts/mysql_config.sh:
  Filter out plain -O and Sun C/C++ style optimization flags, -xO<level>

into  mysql.com:/users/lthalmann/bk/MERGE/mysql-5.0-merge


BitKeeper/etc/ignore:
  auto-union
mysql-test/r/myisam.result:
  Auto merged
mysql-test/t/myisam.test:
  Auto merged
sql/sql_select.cc:
  Auto merged

into  mysql.com:/users/lthalmann/bk/MERGE/mysql-5.0-merge

into  mysql.com:/users/lthalmann/bk/MERGE/mysql-4.1-merge


BitKeeper/etc/ignore:
  auto-union
mysql-test/r/myisam.result:
  Auto merged
mysql-test/t/myisam.test:
  Auto merged
sql/sql_select.cc:
  Auto merged

into  mysql.com:/users/lthalmann/bk/MERGE/mysql-5.0-merge


sql/ha_ndbcluster.cc:
  Auto merged
sql/sql_parse.cc:
  Auto merged

into  mysql.com:/users/lthalmann/bk/MERGE/mysql-4.1-merge


BitKeeper/etc/ignore:
  auto-union
mysql-test/r/myisam.result:
  Auto merged
mysql-test/t/myisam.test:
  Auto merged
sql/sql_select.cc:
  Auto merged

mysql-test/mysql-test-run.pl:
  fixed path for udf_example.so when running testsuite on Windows
sql/CMakeLists.txt:
  Added missing udf_example
sql/Makefile.am:
  Added udf_example files for make dist
sql/udf_example.c:
  fix for Windows: Windows doesn't have socket.h and friends
  
  Added replacements for strmov,bzero and memcpy_fixed when compiling standalone.
sql/udf_example.def:
  BitKeeper file /home/georg/work/mysql/prod/mysql-5.0-win/sql/udf_example.def

into  trift2.:/MySQL/M50/push-5.0

Docs/manual.chm:
  BitKeeper file /MySQL/M50/push-5.0/Docs/manual.chm

into  chilla.local:/home/mydev/mysql-5.0-toteam

into  trift2.:/MySQL/M50/push-5.0


configure.in:
  Auto merged
sql/mysqld.cc:
  Auto merged
sql/sql_base.cc:
  Auto merged
sql/sql_parse.cc:
  Auto merged
sql/stacktrace.c:
  Auto merged
sql/stacktrace.h:
  Auto merged

into  mysql.com:/users/lthalmann/bk/MERGE/mysql-5.0-merge


mysql-test/r/rpl_insert_id.result:
  Merge 4.1-5.0
mysql-test/r/rpl_max_relay_size.result:
  Merge 4.1-5.0
mysql-test/t/rpl_insert_id.test:
  Merge 4.1-5.0
mysql-test/t/rpl_max_relay_size.test:
  Merge 4.1-5.0

mysql-test/r/rpl_insert_id.result:
  Updated result file

into  chilla.local:/home/mydev/mysql-5.0-toteam


sql/opt_range.cc:
  Auto merged
sql/sql_select.cc:
  Auto merged

into  mysql.com:/users/lthalmann/bk/MERGE/mysql-5.0-merge


sql/sql_yacc.yy:
  Auto merged

mysql-test/r/mysqldump.result:
  Fix of result file (adding the echo's)
mysql-test/r/rpl_view.result:
  Added end of 5.0 echo
mysql-test/t/rpl_view.test:
  Added end of 5.0 echo

into  mysql.com:/users/lthalmann/bk/MERGE/mysql-5.0-merge


ndb/src/kernel/blocks/dbdih/DbdihMain.cpp:
  Auto merged
sql/ha_ndbcluster.cc:
  Auto merged
sql/sql_class.h:
  Auto merged
sql/sql_parse.cc:
  Auto merged
sql/sql_yacc.yy:
  Auto merged
client/mysqldump.c:
  Merge main->rpl
mysql-test/r/mysqldump.result:
  Restore of main mysqldump test files in rpl tree (only added echo and one test case for bug 13926)
  Printouts not yet added to result file
mysql-test/t/mysqldump.test:
  Restore of main mysqldump test files in rpl tree (only added echo and one test case for bug 13926)

into  mysql.com:/users/lthalmann/bk/MERGE/mysql-4.1-merge


sql/sql_yacc.yy:
  Auto merged