Last part of fix for BUG#7998 "Replication should be more clever about when to replicate RELEASE_LOCK()" + fixes after merge


mysql-test/r/drop_temp_table.result:
  result update
mysql-test/r/mix_innodb_myisam_binlog.result:
  result update
mysql-test/r/rpl000001.result:
  result update
mysql-test/r/rpl_change_master.result:
  result update
mysql-test/r/rpl_deadlock.result:
  result update (merge)
mysql-test/t/rpl000001.test:
  can't rely on GET_LOCK() to do slave synchro (use table lock instead)
mysql-test/t/rpl_change_master.test:
  changing the test as we can't use GET_LOCK() for slave synchro
mysql-test/t/rpl_deadlock.test:
  update (merge) binlog positions
mysql-test/t/rpl_get_lock.test:
  comment

into mysql.com:/home/mysql_src/mysql-5.0-xa


sql/item_func.cc:
  Auto merged
sql/log_event.cc:
  Auto merged

First part of fix for BUG#7998 "Replication should be more clever about when to replicate RELEASE_LOCK()"
(next part is updating test results)


sql/item_func.cc:
  no need for binlogging DO RELEASE_LOCK() anymore. Make GET_LOCK() do nothing in slave thread.
sql/log_event.cc:
  comment fixes
sql/mysqld.cc:
  warning removed
sql/slave.cc:
  fix after merge

into mysql.com:/M50/mysql-5.0

into serg.mylan:/usr/home/serg/Abk/mysql-5.0


sql/ha_innodb.cc:
  Auto merged

mysys/my_mmap.c:
  cleanup
sql/ha_innodb.cc:
  don's start a transaction in prepare().
  restore disabled code for ibbackup

sql/mysql_priv.h:
  Auto merged
sql/mysqld.cc:
  Auto merged
sql/set_var.cc:
  Auto merged
sql/slave.cc:
  SCCS merged
sql/slave.h:
  SCCS merged

innobase/row/row0sel.c:
  Manual merge (Bug #8771)

After merge fix.
The bugfix results in a different error message in this situation.

use function-wide declaration also for loop body.


ndb/src/kernel/blocks/dbdict/Dbdict.cpp:
  Get rid of a duplicate declaration ('Uint32 k;' at function start) that caused 
  compiler failures on 'octane2' (both 32 and 64 bit) and a compiler crash on 'ds20'.
  It would be better not to use those variable declarations with a loop scope,
  they have caused compile problems again and again.

into mysql.com:/home/mysql_src/mysql-4.1-clean


sql/mysqld.cc:
  Auto merged

in slave SQL thread: if a transaction fails because of InnoDB deadlock or innodb_lock_wait_timeout exceeded,
optionally retry the transaction a certain number of times (new variable --slave_transaction_retries).


sql/mysql_priv.h:
  new var slave_transaction_retries
sql/mysqld.cc:
  new variable slave_transaction_retries. Plus fixing a typo.
sql/set_var.cc:
  new global variable slave_transaction_retries (will be one per subslave, when we have multimaster).
sql/slave.cc:
  Slave SQL thread: if a transaction fails because of InnoDB deadlock or innodb_lock_wait_timeout exceeded,
  optionally retry the transaction a certain number of times (--slave_transaction_retries).
sql/slave.h:
  new RELAY_LOG_INFO::trans_retries.

innobase/include/srv0srv.h:
  Auto merged
innobase/srv/srv0srv.c:
  Auto merged
myisam/mi_create.c:
  Auto merged
myisam/mi_open.c:
  Auto merged
myisam/myisamdef.h:
  Auto merged
sql/ha_innodb.cc:
  Auto merged
mysql-test/r/myisam.result:
  SCCS merged
mysql-test/t/myisam.test:
  SCCS merged
sql/ha_innodb.h:
  SCCS merged

into serg.mylan:/usr/home/serg/Abk/mysql-5.0


sql/ha_innodb.cc:
  Auto merged
sql/item_cmpfunc.cc:
  Auto merged
sql/sql_repl.cc:
  Auto merged

deadlock in MYSQL_LOG::new_file()
style fixes


mysql-test/r/innodb.result:
  one more test for truncate
mysql-test/t/innodb.test:
  one more test for truncate
sql/ha_innodb.cc:
  bug#8151 - truncate leaves a transaction open
sql/handler.cc:
  bug#8151 - truncate leaves a transaction open
sql/item.h:
  style fix
sql/item_cmpfunc.cc:
  style fix
sql/item_cmpfunc.h:
  style fix
sql/item_func.cc:
  style fix
sql/item_func.h:
  style fix
sql/log.cc:
  let new_file() to lock LOCK_index,
  don't check for prepared_xids in rotate_and_purge()
  increase thread_safe_increment when LOCK_log is taken
sql/log_event.cc:
  mysqlbinlog now prints a warning if binlog was not closed properly
sql/sql_class.h:
  comments
sql/sql_repl.cc:
  DBUG_ENTER tag corrected

Added a check, if the table, which we are going to create, is open.
This can happen if a MERGE mapped table is TRUNCATEd.


myisam/mi_open.c:
  Bug#8306 - TRUNCATE leads to index corruption
  Made test_if_reopen() globally available.
myisam/myisamdef.h:
  Bug#8306 - TRUNCATE leads to index corruption
  Declared test_if_reopen() as globally available.
mysql-test/r/myisam.result:
  Bug#8306 - TRUNCATE leads to index corruption
  The test result.
mysql-test/t/myisam.test:
  Bug#8306 - TRUNCATE leads to index corruption
  The test case.

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


sql/ha_innodb.cc:
  Auto merged

  Correct indentation + some small style corrections


sql/ha_innodb.cc:
  Correct indentation + some small style corrections

  updated ndb test makefile to make src dist complete


ndb/test/ndbapi/Makefile.am:
  updated ndb test makefile to make src dist complete

into ua141d10.elisa.omakaista.fi:/home/my/bk/mysql-5.0

mysql-test/r/func_group.result:
  Added some tests and fix for Bug#5615.
mysql-test/r/show_check.result:
  Fix for Bug#5615.
mysql-test/t/func_group.test:
  Added some tests.
sql/item_sum.cc:
  Fix for Bug#5615.
sql/item_sum.h:
  Fix for Bug#5615.
sql/sql_select.cc:
  Fix for Bug#5615.
sql/sql_select.h:
  Fix for Bug#5615.

  delete_all_rows(): Revert previous change.
  The transaction has been marked started already
  in external_lock().


sql/ha_innodb.cc:
  delete_all_rows(): Revert previous change.
  The transaction has been marked started already
  in external_lock().

  row_sel_store_mysql_rec(): Do not try to space-pad BLOB fields. (Bug #8771)


innobase/row/row0sel.c:
  row_sel_store_mysql_rec(): Do not try to space-pad BLOB fields. (Bug #8771)

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


sql/ha_innodb.cc:
  Auto merged

  In TRUNCATE TABLE, MySQL may call innobase_xa_prepare() even though there is no active transaction because it was committed in ha_innobase::delete_all_rows(); in innobase_xa_prepare() InnoDB starts then a new transaction; let us adjust the trx->active_trans flag accordingly


sql/ha_innodb.cc:
  In TRUNCATE TABLE, MySQL may call innobase_xa_prepare() even though there is no active transaction because it was committed in ha_innobase::delete_all_rows(); in innobase_xa_prepare() InnoDB starts then a new transaction; let us adjust the trx->active_trans flag accordingly

innobase/dict/dict0crea.c:
  dict_truncate_index_tree(): Commit the mtr after deleting the
  index tree. Add diagnostics for error cases. Let the caller update
  SYS_INDEXES.PAGE_NO to the new root page number. Return the new
  root page number, or FIL_NULL on error.
innobase/include/dict0crea.h:
  dict_truncate_index_tree(): Commit the mtr after deleting the
  index tree. Add diagnostics for error cases. Let the caller update
  SYS_INDEXES.PAGE_NO to the new root page number. Return the new
  root page number, or FIL_NULL on error.
innobase/include/page0page.ic:
  page_mem_free(): Disable the memset() call, to make it possible to
  recover some data if someone accidentally deletes a large number
  of records from a table.
innobase/log/log0recv.c:
  Do not disable InnoDB Hot Backup specific code in MySQL builds.
innobase/row/row0mysql.c:
  row_truncate_table_for_mysql(): Remove an infinite loop in case
  a SYS_INDEXES record has been deleted. Avoid deadlocks by committing
  and restarting the mini-transaction.
sql/ha_innodb.cc:
  ha_innobase::delete_all_rows(): set trx->active_trans = 1

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

CREATE INDEX and DROP INDEX every 10,000 rows,
similar to ALTER TABLE.


sql/ha_innodb.cc:
  Commit inserts to the temporary tables created by
  CREATE INDEX and DROP INDEX every 10,000 rows, similar
  to ALTER TABLE.
sql/ha_innodb.h:
  Remove duplicate declaration of srv_max_purge_lag.

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

mysql-test/r/func_group.result:
  Auto merged
mysql-test/r/ps.result:
  Auto merged
mysql-test/r/select_found.result:
  Auto merged
mysql-test/r/show_check.result:
  Auto merged
mysql-test/t/func_group.test:
  Auto merged
mysql-test/t/show_check.test:
  Auto merged
sql/mysqld.cc:
  Auto merged
sql/sql_class.cc:
  Auto merged
sql/sql_show.cc:
  Auto merged
strings/ctype-bin.c:
  Auto merged
sql/sql_select.cc:
  SCCS merged

into hundin.mysql.fi:/home/marko/mysql-4.1

  Changed #warning pragma at the bottom to only occur when being used on non-Windows OS


mysys/my_mmap.c:
  Changed #warning pragma at the bottom to only occur when being used on non-Windows OS

remove old junk and to improve compression ratio.
InnoDB: Make implicit type conversions explicit. (Bug #8826)


innobase/btr/btr0cur.c:
  Add "offsets" parameter to page_cur_delete_rec() calls
  btr_cur_optimistic_delete(): Simplify the logic with a flag variable
  btr_cur_pessimistic_delete(): Compute "offsets" earlier
innobase/include/page0cur.h:
  page_cur_delete_rec(): Add parameter "offsets"
innobase/include/page0page.h:
  page_mem_free(): Replace parameter "index" with "offsets"
innobase/include/page0page.ic:
  page_mem_free(): Replace parameter "index" with "offsets".
  Clear the data bytes of the freed record.
  (The "extra" bytes will be needed by free space management.)
innobase/include/rem0rec.h:
  Remove unnecessary function rec_get_size()
innobase/log/log0recv.c:
  Remove function rec_apply_log_recs_for_backup()
  unless #ifdef UNIV_HOTBACKUP.
innobase/page/page0cur.c:
  Add parameter "offsets" to page_cur_delete_rec().
innobase/page/page0page.c:
  page_create(): Zero fill the data area.
  page_delete_rec_list_start(): Add parameter "offsets"
  to page_cur_delete_rec().
innobase/rem/rem0rec.c:
  Remove unnecessary function rec_get_size().
  Fix compiler warnings about implicit type conversions.  (Bug #8826)
innobase/srv/srv0srv.c:
  Fix compiler warnings about implicit type conversions. (Bug #8826)
innobase/sync/sync0sync.c:
  Fix compiler warnings about implicit type conversions. (Bug #8826)

with view-merge algorithm and prepared statements: in case when some
Item_ref pointing to a view column was substituted with a reference 
pointing to the view expression for that column
Item_ref::ref member of the original Item_ref was left pointing to 
not_found_item (0x1).
As we currently perform expression substition part of the view-merge 
algorithm per each execution of a prepared statement or stored procedure, 
we need to preserve original Item_ref objects usable.


sql/item.cc:
  Set member Item_ref::ref to null whenever the item itself is substituted 
  with another item.
  This is necessary if we want to re-execute a prepared statement next time.
  
  Additionally Item_ref::fix_fields() implementation was cleaned up
  (by Monty and myself) to reduce the number of if branches. This
  doesn't change the logic of this function.

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

into mysql.com:/usr/home/ram/work/4.1.b6089


sql/sql_select.cc:
  Auto merged

into mysql.com:/usr/home/ram/work/4.1.b6089


sql/sql_select.cc:
  Auto merged

into mysql.com:/usr/home/ram/work/4.1.b5615


mysql-test/r/show_check.result:
  Auto merged
sql/sql_select.cc:
  Auto merged