Backport from 4.1: make sure change_active_index is only called with a properly initialized InnoDB table handle


sql/ha_innodb.cc:
  Backport from 4.1: make sure change_active_index is only called with a properly initialized InnoDB table handle

  Make SHOW INNODB STATUS aware when we are estimating index range size etc.


sql/ha_innodb.cc:
  Make SHOW INNODB STATUS aware when we are estimating index range size etc.

  Fix a typo


sql/ha_innodb.cc:
  Fix a typo

  Fix a possible hang at the adaptive hash index latch if MySQL does query estimations also in the middle of a SELECT statement processing


sql/ha_innodb.cc:
  Fix a possible hang at the adaptive hash index latch if MySQL does query estimations also in the middle of a SELECT statement processing

  Make SHOW INNODB STATUS aware which trx's are waiting because of innodb_thread_concurrency exceeded


innobase/srv/srv0srv.c:
  Make SHOW INNODB STATUS aware which trx's are waiting because of innodb_thread_concurrency exceeded

  Print also the thread ids in SHOW INNODB STATUS also in Linux


innobase/trx/trx0trx.c:
  Print also the thread ids in SHOW INNODB STATUS also in Linux

Allocate bigger default thread stack because of problems with glibc
Fixed bug in UPDATE ... not_null_field=expression_that_returns_null
Fixed bug in replication when using auto_increment and LOAD DATA INFILE


include/my_global.h:
  Guard against compiling without -fno-exceptions
include/my_pthread.h:
  Allocate bigger default thread stack because of problems with glibc
mysql-test/r/null.result:
  Updated result
mysql-test/t/null.test:
  Test of using UPDATE/INSERT with NULL on not null fields.
sql/field_conv.cc:
  Fixed bug in UPDATE ... not_null_field=expression_that_returns_null
sql/sql_load.cc:
  Fixed bug in replication when using auto_increment and LOAD DATA INFILE

  Backport the InnoDB HANDLER bug fix from 4.1


sql/sql_handler.cc:
  Backport the InnoDB HANDLER bug fix from 4.1

  Fix a hang associated with an index tree of height 1 and purging of BLOB fields from it


innobase/row/row0purge.c:
  Fix a hang associated with an index tree of height 1 and purging of BLOB fields from it
innobase/btr/btr0cur.c:
  Fix a hang associated with an index tree of height 1 and purging of BLOB fields from it
innobase/btr/btr0btr.c:
  Fix a hang associated with an index tree of height 1 and purging of BLOB fields from it
innobase/include/btr0btr.h:
  Fix a hang associated with an index tree of height 1 and purging of BLOB fields from it
innobase/include/btr0cur.h:
  Fix a hang associated with an index tree of height 1 and purging of BLOB fields from it

  Add another diagnostic print to determine why the 'queries inside InnoDB' counter might drift upwards


innobase/srv/srv0srv.c:
  Add another diagnostic print to determine why the 'queries inside InnoDB' counter might drift upwards

  Add diagnostic prints to determine why the 'queries inside InnoDB' might drift upwards
ha_innodb.cc:
  Add more print space to SHOW INNODB STATUS; remove possible memory leak in case there is an error


sql/ha_innodb.cc:
  Add more print space to SHOW INNODB STATUS; remove possible memory leak in case there is an error
innobase/trx/trx0trx.c:
  Add diagnostic prints to determine why the 'queries inside InnoDB' might drift upwards

Fixed problem with replication LOAD DATA INFILE when using  --old-rpl-compat.
When executing on master LOAD DATA and InnoDB failed with 'table full' error the binary log was corrupted.


sql/item_cmpfunc.cc:
  Fixed core dump bug in str LIKE "%other_str%" where strings contained characters >= 128.
sql/log_event.cc:
  Fixed problem with replication LOAD DATA INFILE when using  --old-rpl-compat
sql/sql_load.cc:
  When executing on master LOAD DATA and InnoDB failed with 'table full' error the binary log was corrupted.

BUILD/compile-alpha-cxx:
  Don't do depend files with Compaq compiler
configure.in:
  Don't do depend files with Compaq compiler
extra/mysql_waitpid.c:
  Portability fixes
mysql-test/mysql-test-run.sh:
  Better timeout for valgrind

Free row buffer cache after each query for MyISAM tables.
Added table join option FORCE INDEX
Fixed core dump bug when connecting with hostname that could not be resolved.


include/my_base.h:
  Don't count NULL values in cardinalty
myisam/mi_check.c:
  Don't count NULL values in cardinalty
myisam/mi_extra.c:
  Free row buffer cache after each query
myisam/mi_open.c:
  Avoid realloc if cache size doesn't change
myisam/mi_search.c:
  Don't count NULL values in cardinalty
myisam/myisamdef.h:
  Change buffer length from uint to uint32 to make it more portable/predictable
mysql-test/r/myisam.result:
  Test case for cardinality with NULL keys and FORCE INDEX
mysql-test/t/myisam.test:
  Test case for cardinality with NULL keys and FORCE INDEX
sql/lex.h:
  Added table join option FORCE INDEX
sql/mysql_priv.h:
  Added table join option FORCE INDEX
sql/opt_range.cc:
  Added table join option FORCE INDEX
sql/sql_base.cc:
  Added table join option FORCE INDEX
sql/sql_lex.h:
  Added table join option FORCE INDEX
sql/sql_parse.cc:
  Added table join option FORCE INDEX
  Don't use strlen() on hostname without first checking if it's not NULL
sql/sql_select.cc:
  Added table join option FORCE INDEX
sql/sql_yacc.yy:
  Added table join option FORCE INDEX
sql/table.h:
  Added table join option FORCE INDEX

into sergbook.mysql.com:/usr/home/serg/Abk/mysql-4.0

  no _mi_search_next() bug fixed


myisam/ft_nlq_search.c:
  no _mi_search_next() bug fixed

Fix for SHOW VARIABLES in embedded server


Docs/internals.texi:
  Added documentation for join_buffer_size
configure.in:
  Changed version number
sql/log_event.cc:
  Fix for bug in LOAD DATA INFILE
sql/log_event.h:
  Fix for bug in LOAD DATA INFILE
sql/slave.cc:
  Fix for bug in LOAD DATA INFILE
sql/sql_show.cc:
  Fix for SHOW VARIABLES in embedded server

sql/sql_show.cc:
  fix embedded server crash

  Remove the warning message that mem allocation spills from the additional mem pool to the OS; this does not hit performance with modern malloc libraries


innobase/mem/mem0pool.c:
  Remove the warning message that mem allocation spills from the additional mem pool to the OS; this does not hit performance with modern malloc libraries

Fixed test suite for HPUX 10.20 and MacOSX


Build-tools/Do-compile:
  Added timeout to mysqladmin shutdown commands
  Kill old running mysqld started by earlier runs
  Removed run time warning from LD_LIBRARY_PATH
client/mysqladmin.c:
  Return 1 if pid file isn't deleted on shutdown.
  Fix error message if pid file is not deleted
client/mysqltest.c:
  Always allow --debug flag
  (Makes it easier to run mysql-test-run)
mysql-test/mysql-test-run.sh:
  A lot of safety fixes. 
  This fixes some problems with test suite for HPUX 10.20 and MacOSX
sql-bench/bench-init.pl.sh:
  Allow tests to change time limit.
sql-bench/crash-me.sh:
  Indentation cleanups
  Added DROP for a created table
sql-bench/test-alter-table.sh:
  Added default time limit
  Changed test to be estimated to get down run time.
  Fixed that add_multi_col is detected
sql-bench/test-insert.sh:
  Comment cleanup
sql/mysql_priv.h:
  Removed not needed prototype.
sql/mysqld.cc:
  Removed DBUG warnings
  Removed default argument for clean_up() and made it static.
  More comments.
  Ignore SIGHUP during shutdown
sql/net_pkg.cc:
  More comments
sql/slave.cc:
  Added DBUG_PRINT messages

  Fix bug: an index cursor can theoretically be restored in a wrong place
log0log.c:
  Fix bug: if combined log file size is >= 2 GB in a 32-bit computer InnoDB can write log to a wrong position


innobase/log/log0log.c:
  Fix bug: if combined log file size is >= 2 GB in a 32-bit computer InnoDB can write log to a wrong position
innobase/btr/btr0pcur.c:
  Fix bug: an index cursor can theoretically be restored in a wrong place

Added rename table for BDB. This fixes a bug in ALTER TABLE with BDB tables.


BUILD/SETUP.sh:
  Enabled uninitalized warnings for debug builds.
Build-tools/Do-compile:
  Added options --no-mysqltest and --no-benchmarks
  Log time before each stage
  Removed warnings on some OS (at least hpux11)
acinclude.m4:
  Fixed call to wrong function
configure.in:
  Don't use -O3 as compiler option with aCC
innobase/include/univ.i:
  Removed compiler warning (on Hpux11)
innobase/os/os0file.c:
  Removed compiler warning (on Hpux11)
innobase/trx/trx0sys.c:
  Removed compiler warning (on Hpux11)
innobase/trx/trx0trx.c:
  Removed compiler warning (on Hpux11)
innobase/ut/ut0ut.c:
  Removed compiler warning (with gcc)
myisam/mi_check.c:
  Added missing 'static'
myisam/mi_test1.c:
  Added missing 'static'
mysys/my_getopt.c:
  Added missing 'static'
sql/ha_berkeley.cc:
  Added rename table. This fixes a bug in ALTER TABLE with BDB tables.
  Fixed compilation problem on OSF1
sql/ha_berkeley.h:
  Added rename table. This fixes a bug in ALTER TABLE with BDB tables.
sql/mini_client.cc:
  Added reinterpret cast to avoid some warnings.
sql/sql_parse.cc:
  Added reinterpret cast to avoid some warnings.
  Removed usage of strnlen() as this gave portability problems.
sql/sql_union.cc:
  Added reinterpret cast to avoid some warnings.
sql/table.cc:
  Added reinterpret cast to avoid some warnings.
tools/mysqlmanager.c:
  Added missing 'static'

Extended max_allowed_packet for clients to 1G
Fixed bug in sending compressed rows >= 16M
Fix bug in skiping too long packets from clients.
Added checking of wrong command number sent by client.


include/mysql.h:
  Added max_allowed_packet as option parameter.
include/mysql_com.h:
  Added COM_END to be able to check for wrong commands.
libmysql/libmysql.c:
  Extended max_allowed_packet for clients to 1G
  Added support for max_allowed_packet in option files read by mysql_option()
mysys/my_compress.c:
  Debugging output
sql/net_pkg.cc:
  Fixed wrong handling of blobs >= 16M
sql/net_serv.cc:
  Changed MAX_THREE_BYTES -> MAX_PACKET_LENGTH
  More DEBUG output and more comments
  Fixed bug in sending compressed rows >= 16M
  Optimized sending of big packets (fewer memcpy and compress data in bigger blocks)
  Fix bug in skiping too long packets from clients.
  (old code didn't always work for big packets)
sql/sql_class.h:
  Changed type of variable
sql/sql_parse.cc:
  Added checking of wrong command number sent by client.
  Changed handling of too big packets to make code safer
tests/big_record.pl:
  E

into mysql.r18.ru:/usr/home/ram/work/mysql-4.0

BitKeeper/etc/logging_ok:
  Logging to logging@openlogging.org accepted

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

into sinisa.nasamreza.org:/mnt/work/mysql-4.0

that connection does not close in the case when client sends a 
buffer larger then max_allowed_packet.

sql/ha_innodb.cc:
  Auto merged