Bug#27894: mysqlbinlog formats timestamp wrong in comment

Date stamps lack zero padding, and so are meaningless.

Implement minimum-width zero padding for the percent-escape sequences
used in the logging code.


mysql-test/r/mysqlbinlog2.result:
  Include the long form, so we notice that it may be broken.  Replace
  away specific elements that vary from run to run.
mysql-test/t/mysqlbinlog2.test:
  Include the long form, so we notice that it may be broken.  Replace
  away specific elements that vary from run to run.
mysys/mf_iocache2.c:
  Implement minimum-width padding for printf-style percent-expressions.

into  zippy.cornsilk.net:/home/cmiller/work/mysql/mysql-5.1-new-maint


mysys/my_thr_init.c:
  Auto merged

	errno

Vasil Dimov (at Oracle) noted that pthread_key_create() does not
set errno, so if it fails then we return the wrong error code.

Instead, capture the return value, which is the real error value, 
and instead report that.


mysys/my_thr_init.c:
  pthread functions don't set the global errno, which should be
  obvious.

into  quadxeon.mysql.com:/benchmarks/ext3/TOSAVE/tsmith/bk/maint/51


client/mysql.cc:
  Auto merged
mysql-test/r/strict.result:
  Auto merged
mysql-test/t/strict.test:
  Auto merged
sql/item_cmpfunc.cc:
  Auto merged
mysql-test/include/mix1.inc:
  Manual merge
mysql-test/r/innodb_mysql.result:
  Manual merge

  Add test case for Bug #27210: INNODB ON DUPLICATE KEY UPDATE


mysql-test/r/innodb_mysql.result:
  Add test case for Bug #27210: INNODB ON DUPLICATE KEY UPDATE
mysql-test/t/innodb_mysql.test:
  Add test case for Bug #27210: INNODB ON DUPLICATE KEY UPDATE

mysql-test/r/innodb.result:
  Move test result after incorrect merge

into  pilot.blaudden:/home/msvensson/mysql/mysql-5.1-maint


BitKeeper/deleted/.del-CMakeLists.txt~7:
  Auto merged

into  pilot.blaudden:/home/msvensson/mysql/mysql-5.1-maint


client/mysql_upgrade.c:
  Auto merged

into  pilot.blaudden:/home/msvensson/mysql/mysql-5.1-maint


client/mysql_upgrade.c:
  Auto merged
BitKeeper/deleted/.del-CMakeLists.txt~1:
  Auto merged

client/mysql_upgrade.c:
  Windows fix
scripts/CMakeLists.txt:
  Fix problems with CMakeList causing build to fail
scripts/comp_sql.c:
  Improve comp_sql.c to generate output file with shorter strings

into  mysql.com:/home/ram/work/b26851/b26851.5.0

into  mysql.com:/home/ram/work/b26851/b26851.5.1

into  mysql.com:/home/ram/work/b22824/my50-b22824


sql/item_cmpfunc.cc:
  Auto merged
mysql-test/r/strict.result:
  merging
mysql-test/t/strict.test:
  merging

into  mysql.com:/home/ram/work/b26851/b26851.5.0


client/mysql.cc:
  Auto merged

into  mysql.com:/home/ram/work/b26851/b26851.5.1


client/mysql.cc:
  Auto merged
mysql-test/r/mysql.result:
  Auto merged
mysql-test/t/mysql.test:
  Auto merged

into  mysql.com:/home/ram/work/b22824/my51-b22824


mysql-test/r/strict.result:
  merging
mysql-test/t/strict.test:
  merging
sql/item_cmpfunc.cc:
  merging

client/mysqltest.c:
  Actually include "client_priv.h", as intended in previous changeset.

include client_priv.h in mysqltest.c.  Portability fix.


client/client_priv.h:
  Include client_priv.h, for WEXITSTATUS definition.
  
  Remove code which is already in client_priv.h.
  
  Rearrange OPT_* enum to co-exist with options defined in client_priv.h.
client/mysqltest.c:
  Move WEXITSTATUS #define from mysqltest.c to client_priv.h, as it is
  useful in other clients (in particular, current mysql_upgrade.c won't
  compile without it on some platforms).
  
  Add a final OPT_MAX_CLIENT_OPTION to enum options_client, which can
  be used when a maximum shared client option value is needed.

into  siva.hindu.god:/home/tsmith/m/inno/apr18/51

into  siva.hindu.god:/home/tsmith/m/inno/apr18/51

into  siva.hindu.god:/home/tsmith/m/inno/apr18/50

into  siva.hindu.god:/home/tsmith/m/inno/apr18/51


BitKeeper/deleted/.del-configure.in:
  Use local
mysql-test/r/innodb.result:
  Use local
mysql-test/t/innodb.test:
  Use local
mysys/my_alloc.c:
  SCCS merged
sql/ha_ndbcluster.cc:
  SCCS merged
sql/item.cc:
  SCCS merged
sql/item_cmpfunc.cc:
  SCCS merged
sql/mysqld.cc:
  SCCS merged
sql/sql_delete.cc:
  SCCS merged
sql/sql_select.cc:
  SCCS merged
sql/sql_table.cc:
  SCCS merged
storage/innobase/handler/ha_innodb.cc:
  Use local
storage/innobase/handler/ha_innodb.h:
  Use local
storage/innobase/os/os0file.c:
  Use local

Fixes:

- Bug #26662: mysqld assertion when creating temporary (InnoDB) table on a tmpfs filesystem
  Fix by not open(2)ing with O_DIRECT but rather calling fcntl(2) to set
  this flag immediately after open(2)ing. This way an error caused by
  O_DIRECT not being supported can easily be ignored.
- Bug #23313: AUTO_INCREMENT=# not reported back for InnoDB tables
- Bug #21404: AUTO_INCREMENT value reset when Adding FKEY (or ALTER?)
  Report the current value of the AUTO_INCREMENT counter to MySQL.


mysql-test/r/innodb.result:
  Applied innodb-5.1-ss1404 snapshot
  
  Revision r1404:
  Report the current value of the AUTO_INCREMENT counter to MySQL.
  (Bug #23313, Bug #21404)
  
  ha_innobase::update_create_info(): New function, to report
  the auto_increment_value.
mysql-test/t/innodb.test:
  Applied innodb-5.1-ss1404 snapshot
  
  Revision r1404:
  Report the current value of the AUTO_INCREMENT counter to MySQL.
  (Bug #23313, Bug #21404)
  
  ha_innobase::update_create_info(): New function, to report
  the auto_increment_value.
storage/innobase/handler/ha_innodb.cc:
  Applied innodb-5.1-ss1404 snapshot
  
  Revision r1404:
  Report the current value of the AUTO_INCREMENT counter to MySQL.
  (Bug #23313, Bug #21404)
  
  ha_innobase::update_create_info(): New function, to report
  the auto_increment_value.
storage/innobase/handler/ha_innodb.h:
  Applied innodb-5.1-ss1404 snapshot
  
  Revision r1404:
  Report the current value of the AUTO_INCREMENT counter to MySQL.
  (Bug #23313, Bug #21404)
  
  ha_innobase::update_create_info(): New function, to report
  the auto_increment_value.
storage/innobase/os/os0file.c:
  Applied innodb-5.1-ss1404 snapshot
  
  Revision r1395:
  * Fix Bug#26662 by not open(2)ing with O_DIRECT but rather calling fcntl(2)
  to set this flag immediately after open(2)ing. This way an error caused by
  O_DIRECT not being supported can easily be ignored.
  
  * Add support for skipping the OS caching on Solaris by calling directio()
  instead of fcntl().
  
  Approved by:	Heikki
  
  
  
  Revision r1391:
  Merge the bodies of os_file_handle_error() and os_file_handle_error_no_exit()
  into a generic function which is called from both os_file_handle_error() and
  os_file_handle_error_no_exit()
  
  Approved by:	Marko
storage/innobase/plug.in:
  Applied innodb-5.1-ss1404 snapshot
  
  Revision r1395:
  * Fix Bug#26662 by not open(2)ing with O_DIRECT but rather calling fcntl(2)
  to set this flag immediately after open(2)ing. This way an error caused by
  O_DIRECT not being supported can easily be ignored.
  
  * Add support for skipping the OS caching on Solaris by calling directio()
  instead of fcntl().
  
  Approved by:	Heikki

NULL MERGE: this ChangeSet will be null merged into mysql-5.1

Fixes:

- Bug #26662: mysqld assertion when creating temporary (InnoDB) table on a tmpfs filesystem
  Fix by not open(2)ing with O_DIRECT but rather calling fcntl(2) to set
  this flag immediately after open(2)ing. This way an error caused by
  O_DIRECT not being supported can easily be ignored.
- Bug #23313: AUTO_INCREMENT=# not reported back for InnoDB tables
- Bug #21404: AUTO_INCREMENT value reset when Adding FKEY (or ALTER?)
  Report the current value of the AUTO_INCREMENT counter to MySQL.


innobase/configure.in:
  Apply innodb-5.0-ss1405 snapshot
  
  Revision r1396:
  branches/5.0: Merge r1395 from trunk:
  
  * Fix Bug#26662 by not open(2)ing with O_DIRECT but rather calling fcntl(2)
  to set this flag immediately after open(2)ing. This way an error caused by
  O_DIRECT not being supported can easily be ignored.
  
  * Add support for skipping the OS caching on Solaris by calling directio()
  instead of fcntl().
innobase/os/os0file.c:
  Apply innodb-5.0-ss1405 snapshot
  
  Revision r1394:
  branches/5.0: Merge r1391 from trunk:
  
  Merge the bodies of os_file_handle_error() and os_file_handle_error_no_exit()
  into a generic function which is called from both os_file_handle_error() and
  os_file_handle_error_no_exit()
  
  
  Revision r1396:
  branches/5.0: Merge r1395 from trunk:
  
  * Fix Bug#26662 by not open(2)ing with O_DIRECT but rather calling fcntl(2)
  to set this flag immediately after open(2)ing. This way an error caused by
  O_DIRECT not being supported can easily be ignored.
  
  * Add support for skipping the OS caching on Solaris by calling directio()
  instead of fcntl().
mysql-test/r/innodb.result:
  Apply innodb-5.0-ss1405 snapshot
  
  Revision r1405:
  branches/5.0: Merge r1404 from trunk:
  
  Report the current value of the AUTO_INCREMENT counter to MySQL.
  (Bug #23313, Bug #21404)
  
  ha_innobase::update_create_info(): New function, to report
  the auto_increment_value.
mysql-test/t/innodb.test:
  Apply innodb-5.0-ss1405 snapshot
  
  Revision r1405:
  branches/5.0: Merge r1404 from trunk:
  
  Report the current value of the AUTO_INCREMENT counter to MySQL.
  (Bug #23313, Bug #21404)
  
  ha_innobase::update_create_info(): New function, to report
  the auto_increment_value.
sql/ha_innodb.cc:
  Apply innodb-5.0-ss1405 snapshot
  
  Revision r1405:
  branches/5.0: Merge r1404 from trunk:
  
  Report the current value of the AUTO_INCREMENT counter to MySQL.
  (Bug #23313, Bug #21404)
  
  ha_innobase::update_create_info(): New function, to report
  the auto_increment_value.
sql/ha_innodb.h:
  Apply innodb-5.0-ss1405 snapshot
  
  Revision r1405:
  branches/5.0: Merge r1404 from trunk:
  
  Report the current value of the AUTO_INCREMENT counter to MySQL.
  (Bug #23313, Bug #21404)
  
  ha_innobase::update_create_info(): New function, to report
  the auto_increment_value.

into  qualinost.(none):/home/mtaylor/src/mysql/mysql-5.1-new-maint


Makefile.am:
  Auto merged
configure.in:
  Auto merged

into  qualinost.(none):/home/mtaylor/src/mysql/mysql-5.0-maint


configure.in:
  Auto merged

Makefile.am:
  Removed debian/control creation from the top level dist-hook
configure.in:
  Create debian/control from configure again
debian/rules:
  Protect debian/control from the debuild distclean run. 
  Prevent a makefile variable loop when settin MYSQL_BASE_VERSION

into  siva.hindu.god:/home/tsmith/m/bk/maint/51


client/mysql_upgrade.c:
  Auto merged
client/mysqltest.c:
  Auto merged
mysql-test/lib/mtr_cases.pl:
  Auto merged
mysql-test/mysql-test-run.pl:
  Auto merged
mysql-test/r/type_datetime.result:
  Auto merged
mysys/my_alloc.c:
  Auto merged
sql/field.h:
  Auto merged
sql/item.h:
  Auto merged
sql/mysql_priv.h:
  Auto merged
sql/mysqld.cc:
  Auto merged
sql/sql_parse.cc:
  Auto merged
sql/sql_select.cc:
  Auto merged

into  siva.hindu.god:/home/tsmith/m/bk/maint/50


mysys/my_alloc.c:
  Auto merged
sql/mysqld.cc:
  Auto merged

into  pilot.blaudden:/home/msvensson/mysql/mysql-5.1-maint


scripts/Makefile.am:
  Auto merged

into  pilot.blaudden:/home/msvensson/mysql/mysql-5.1-maint


client/mysql_upgrade.c:
  Auto merged

into  pilot.blaudden:/home/msvensson/mysql/mysql-5.1-maint


client/mysql_upgrade.c:
  Auto merged
client/mysqltest.c:
  Auto merged

into  pilot.blaudden:/home/msvensson/mysql/mysql-5.0-maint


client/mysqltest.c:
  Auto merged