GitLab

(exactly, for the bug in 4.1 reported in this bug report). We just make
Load_log_event work like Query_log_event for temp tables.


sql/log_event.cc:
  For LOAD DATA INFILE to be preceded by SET PSEUDO_THREAD_ID if needed, in the mysqlbinlog output,
  we need to use the LOG_EVENT_THREAD_SPECIFIC_F flag in Load_log_event exactly like we already do  
  in Query_log_event.

  CREATE DATABASE statement used the current database instead of the
  database created when checking conditions for replication.
  CREATE/DROP/ALTER DATABASE statements are now replicated based on
  the manipulated database.


mysql-test/t/rpl_until.test:
  Longer sleep to allow slave to stop.
mysql-test/t/rpl_charset.test:
  Position change in binary file.
mysql-test/r/drop_temp_table.result:
  Position change in binlog.
mysql-test/r/rpl_loaddata_rule_m.result:
  Position change in binlog.
mysql-test/r/rpl_charset.result:
  Position change in binlog.
sql/log_event.h:
  Added new flag and parameter to suppress generation of
  USE statements.
sql/log_event.cc:
  Added parameter and code to suppress generation of
  USE statements.
sql/sql_db.cc:
  Suppress generation of USE before CREATE/ALTER/DROP DATABASE
  statements.
sql/log.cc:
  Query_log_event have new extra parameter.
sql/sql_table.cc:
  Query_log_event have new extra parameter.
sql/sql_base.cc:
  Query_log_event have new extra parameter.
sql/sql_update.cc:
  Query_log_event have new extra parameter.
sql/sql_insert.cc:
  Query_log_event have new extra parameter.
sql/sql_rename.cc:
  Query_log_event have new extra parameter.
sql/sql_delete.cc:
  Query_log_event have new extra parameter.
sql/sql_acl.cc:
  Query_log_event have new extra parameter.
sql/handler.cc:
  Query_log_event have new extra parameter.
sql/item_func.cc:
  Query_log_event have new extra parameter.
sql/sql_parse.cc:
  Query_log_event have new extra parameter.

  post-conflict-merge fix (duplicate comment)


sql/log_event.cc:
  post-conflict-merge fix (duplicate comment)

When mysqlbinlog prints LOAD DATA INFILE, let it print the thread id. Some customer would have benefited
much from this in his recovery. All this change does is adding one commented (#) line before the LOAD DATA
command, so it is quite innocuous.

Replication using replicate-rewrite-db did not work for LOAD DATA INFILE.
Now is does.  There was one place in the code that used current database 
instead of the rewrite database.


mysql-test/r/rpl_rewrite_db.result:
  New tests
mysql-test/t/rpl_rewrite_db-slave.opt:
  New tests
mysql-test/t/rpl_rewrite_db.test:
  New tests
sql/log_event.cc:
  Added db to set_fields function so that current db is used.
sql/log_event.h:
  Added db to set_fields function so that current db is used.

Now thd->mem_root is a pointer to thd->main_mem_root and THR_MALLOC is a pointer to thd->mem_root.
This gives us the following benefits:
- Allow us to easily detect if arena has already been swapped before (this fixes a bug in setup_conds() where arena was swaped twice in some cases)
- Faster swaps of arenas (as we don't have to copy the whole MEM_ROOT)
- We don't anymore have to call my_pthread_setspecific_ptr(THR_MALLOC,...) to change where memory is alloced. Now it's enough to set thd->mem_root



client/mysqltest.c:
  Remove some not needed defines
  (Things like this should be done in config-win.h)
include/config-win.h:
  Added popen() and pclose() compatibility macros
mysql-test/t/rpl_failed_optimize-master.opt:
  Portability fix
sql/ha_berkeley.cc:
  New thd->memroot handling
sql/item_cmpfunc.cc:
  Simpler arena swapping code
sql/item_func.cc:
  Simpler arena swapping code
sql/item_subselect.cc:
  Simpler arena swapping code
  New thd->mem_root handling
sql/item_sum.cc:
  New thd->mem_root handling
sql/item_timefunc.cc:
  Fixed not-initalized usage errors found by valgrind
sql/log_event.cc:
  New thd->mem_root handling
sql/mysql_priv.h:
  New thd->mem_root handling
sql/mysqld.cc:
  New thd->mem_root handling
sql/opt_range.cc:
  New thd->mem_root handling
sql/repl_failsafe.cc:
  New thd->mem_root handling
sql/set_var.cc:
  New thd->mem_root handling
sql/sql_acl.cc:
  New thd->mem_root handling
sql/sql_base.cc:
  Simpler arena swapping code
  New thd->mem_root handling
sql/sql_class.cc:
  New thd->mem_root handling
sql/sql_class.h:
  Simpler arena swapping code
  New thd->mem_root handling
sql/sql_db.cc:
  New thd->mem_root handling
sql/sql_error.cc:
  New thd->mem_root handling
sql/sql_help.cc:
  New thd->mem_root handling
sql/sql_insert.cc:
  New thd->mem_root handling
sql/sql_parse.cc:
  New thd->mem_root handling
  Added some extra checking of return value of new
sql/sql_prepare.cc:
  New thd->mem_root handling
sql/sql_select.cc:
  New thd->mem_root handling
sql/sql_select.h:
  New thd->mem_root handling
sql/sql_union.cc:
  Simpler arena swapping code
sql/sql_yacc.yy:
  New thd->mem_root handling
sql/table.cc:
  New thd->mem_root handling
sql/thr_malloc.cc:
  New thd->mem_root handling
tests/client_test.c:
  Added drop table to some tests
  Changed some table names to 't1'

as we already have db_len in Log_event. Only if rewrite_db() changed the db we need a strlen
(so we now do the strlen() in rewrite_db). Plus a test (we had none for --replicate-rewrite-db :( ).


sql/log_event.cc:
  The goal is to get of rid of one strlen() per replication event in slave:
  we don't need to compute strlen(thd->db) as we already have db_len in the event;
  only case where we need to do a strlen() is if rewrite_db() changed the db.
  Note that db_len is always a meaningful value. It's 0 if event's db is 0.
sql/slave.cc:
  rewrite_db now returns the len of the returned db.
  print_slave_db_safe() needn't call rewrite_db() as rewrite_db() is already called by caller.
sql/slave.h:
  declaration updates for slave.cc

we do not increment rli->group_master_log_pos if we are just after a SET ONE_SHOT (it's not a standalone event)


mysql-test/r/rpl_charset.result:
  testing interruption of slave SQL thread between SET CHARACTER_SET_SERVER... and the companion INSERT.
mysql-test/t/rpl_charset.test:
  testing interruption of slave SQL thread between SET CHARACTER_SET_SERVER... and the companion INSERT.
sql/log_event.cc:
  we do not increment rli->group_master_log_pos if we are just after a SET
  ONE_SHOT, because SET ONE_SHOT should not be separated from its following
  updating query.

This does not fix any known bug, but is still a good idea.


sql/log_event.cc:
  when we update thd->db in replication, it's safer to update thd->db_length too.
sql/slave.cc:
  when we update thd->db in replication, it's safer to update thd->db_length too.
sql/sql_db.cc:
  comment

We must not reset the charset in slave after each statement, otherwise the SET CHARACTER SET is cancelled immediately.
Instead, we write a SET CHARACTER SET DEFAULT to the master's binlog when needed (like we already do for SET FOREIGN_KEY_CHECKS);
such writing is not necessary in 4.1 (in 4.1 the bug does not exist, as the SET ONE_SHOT syntax is used).
I have written a test and it works, but I'm not pushing the test as it requires building with all charsets.
I have noticed differences between what is inserted in the master's table in 4.0 and 4.1, and alerted Bar.


sql/log.cc:
  When SET CHARACTER SET has been used, we must reset the charset after the writing the statement, in the binlog.
  In 4.1, this resetting is already achieved by the SET ONE_SHOT syntax.
sql/log_event.cc:
  In slave, we must not simply reset the charset after each statement: if we do this, the charset gets
  immediately after executing the SET CHARACTER SET! (BUG#4500).

client code and replication slave code, as far as LOAD DATA INFILE and
other queries' execution is concerned. Duplication of code leads to
replication bugs, because the replication duplicate lags much behind.
Fix for 2 Valgrind errors on slave replicating LOAD DATA INFILE
  - one serious (causing a random test failure in rpl_loaddata in 5.0)
  - one not serious (theoretically a bug but not dangerous): uninited thd->row_count


sql/log_event.cc:
  Fix for 2 Valgrind errors:
  - one serious (causing a random test failure in rpl_loaddata in 5.0): uninited lex in replic of LOAD DATA INFILE on slave
  - one not serious (theoretically a bug but not dangerous): uninited thd->row_count in replication of LOAD DATA INFILE on slave.
  Sanja is likely to rework the fix to the 1st problem tomorrow.

New handler::index_flags() definition to make it easy to check the full used key and a specific key part.
Added key part to optimize_range() to fix problems when using fields in key parts.



sql/examples/ha_archive.h:
  New handler::index_flags() definition
sql/examples/ha_example.h:
  New handler::index_flags() definition
sql/field.cc:
  New optimize_range() definition
sql/field.h:
  New optimize_range() definition
sql/ha_berkeley.cc:
  New handler::index_flags() definition
sql/ha_berkeley.h:
  New handler::index_flags() definition
sql/ha_heap.h:
  New handler::index_flags() definition
sql/ha_innodb.h:
  New handler::index_flags() definition
sql/ha_isam.h:
  New handler::index_flags() definition
sql/ha_isammrg.h:
  New handler::index_flags() definition
sql/ha_myisam.h:
  New handler::index_flags() definition
sql/ha_myisammrg.h:
  New handler::index_flags() definition
sql/ha_ndbcluster.cc:
  New handler::index_flags() definition
sql/ha_ndbcluster.h:
  New handler::index_flags() definition
sql/handler.h:
 ...

processlist on slave":
we now report in SHOW PROCESSLIST that we are writing to the temp
files or loading the table. When we are writing to the tmp file:
| 3  | system user |                 |    | Connect | 6    | Making temp file /tmp/SQL_LOAD-2-1-2.data | 
and when we are actually loading the .data temp file into the table:
| 3  | system user |                 | test | Connect | 2    | | LOAD DATA INFILE '/tmp/SQL_LOAD-2-1-2.data' INTO TABLE `t` <...> |


sql/log_event.cc:
  Replication of LOAD DATA INFILE:
  we now report in SHOW PROCESSLIST that we are creating
  the temp files or loading the table.
  Plus removing a line which had a comment "should not be needed"
  and a guarding assertion which we have never heard fail (and logic
  says it should not fail).

In mysqlbinlog, there was a problem with how we escaped the content of a string user variable.
To be perfect, we should have escaped with character_set_client. But this charset is unknown
to mysqlbinlog. So the simplest is to print the string in hex. This is unreadable but
100% safe with any charset (checked with Bar), no more need to bother with character_set_client.


mysql-test/r/rpl_charset.result:
  hex strings
mysql-test/r/rpl_user_variables.result:
  hex strings
mysql-test/r/user_var.result:
  hex strings
sql/log_event.cc:
  In mysqlbinlog, there was a problem with how we escaped the content of a string user variable.
  To be perfect, we should have escaped with character_set_client. But this charset is unknown
  to mysqlbinlog. So the simplest is to print the string in hex. This is unreadable but
  100% safe with any charset (checked with Bar), no more need to bother with character_set_client.

by binlogging some SET ONE_SHOT CHARACTER_SETetc,
which will be enough until we have it more compact and more complete in 5.0. With the present patch,
replication will work ok between 4.1.3 master and slaves, as long as:
- master and slave have the same GLOBAL.COLLATION_SERVER
- COLLATION_DATABASE and CHARACTER_SET_DATABASE are not used
- application does not use the fact that table is created with charset of the USEd db (BUG#2326).
all of which are not too hard to fulfill. 
ONE_SHOT is reserved for internal use of mysqlbinlog|mysql and works only for charsets,
so we give error if used for non-charset vars.
Fix for BUG#3875 "mysqlbinlog produces wrong ouput  if query uses
 variables containing quotes" and BUG#3943 "Queries with non-ASCII literals are not replicated
 properly after SET NAMES".
Detecting that master and slave have different global charsets or server ids.


mysql-test/r/rpl_server_id1.result:
  it's normal to not run as I have added a test to compare server ids of master and slave
  at startup and stop if equal (unless --replicate-same-server-id)
mysql-test/r/rpl_user_variables.result:
  result update (as we now print charset of user var).
mysql-test/r/user_var.result:
  result update
mysql-test/t/rpl_server_id1.test:
  no need to select as slave is not running
mysql-test/t/user_var.test:
  testing if the content of user vars is escaped when mysqlbinlog prints them,
  and if the name is backquoted.
sql/lex.h:
  new keyword ONE_SHOT
sql/log.cc:
  when writing to the binlog, before writing the actual statement, write some SET ONE_SHOT CHARACTER_SET_CLIENT etc
  for the slave to know the charset variables (which are important as they affect the inserted data).
sql/log_event.cc:
  print charset and collation of user var in mysqlbinlog and SHOW BINLOG EVENTS.
  escape the content of the var. Backquote its name.
  Will ask Bar to check that using my_charset_bin for escaping is ok.
sql/set_var.cc:
  understand SET CHARACTER_SET_CLIENT=10 (don't require a string, accept a number).
  Refuse changing of GLOBAL CHARACTER_SET_SERVER/COLLATION_SERVER if binlog or slave,
  as it will make the master or slave make wrong assumptions.
  A function to catch SET ONE_SHOT on non-charset variables (which is forbidden)
sql/set_var.h:
  no_support_one_shot to know if the var supports ONE_SHOT (only charset vars do, soon timezones).
  Accept int arg in SET CHARACTER_SET_etc
sql/slave.cc:
  when I/O slave thread starts, verify that master's and slave charsets match.
  And by the way verify that server ids are different.
  Don't fail if UNIX_TIMESTAMP() can't be done on master (very old master), that's
  not fatal.
sql/sql_class.cc:
  one_shot
sql/sql_class.h:
  one_shot
sql/sql_lex.h:
  one_shot
sql/sql_parse.cc:
  when SET ONE_SHOT is used, verify that it's only used for charset/collation vars;
  otherwise refuse.
sql/sql_yacc.yy:
  ONE_SHOT keyword in SET

VC++Files/innobase/innobase.dsp:
  non-existent file removed
client/mysql.cc:
  local opt_max_allowed_packet and opt_net_buffer_length introduced
client/mysqldump.c:
  local opt_max_allowed_packet and opt_net_buffer_length introduced
include/mysql.h:
  mysql_get_parameters() interface added
  #define max_allowed_packet added
include/mysql_com.h:
  these should not be exported
libmysql/libmysql.c:
  mysql_get_parameters implementations
libmysql/libmysql.def:
  interface changed
libmysql_r/Makefile.am:
  MYSQL_CLIENT define added
libmysqld/lib_sql.cc:
  line moved to be above the '#include "mysql.cc"'
libmysqld/libmysqld.c:
  mysql_get_parameters implementation (embedded)
libmysqld/libmysqld.def:
  interface changed
sql/log_event.cc:
  should be like that in this case
tools/mysqlmanager.c:
  compiler warns on this line

mysql-test/r/type_float.result:
  Fix result after merge
sql/log_event.cc:
  Fixed typo

  Fix remaining cases of Bug #3596: fix possible races caused by an obsolete value of thd->query_length in SHOW PROCESSLIST and SHOW INNODB STATUS; this fix depends on the fact that thd->query is always set to NULL before setting it to point to a new query


sql/sql_class.h:
  Fix remaining cases of Bug #3596: fix possible races caused by an obsolete value of thd->query_length in SHOW PROCESSLIST and SHOW INNODB STATUS; this fix depends on the fact that thd->query is always set to NULL before setting it to point to a new query
sql/ha_innodb.cc:
  Fix remaining cases of Bug #3596: fix possible races caused by an obsolete value of thd->query_length in SHOW PROCESSLIST and SHOW INNODB STATUS; this fix depends on the fact that thd->query is always set to NULL before setting it to point to a new query
sql/log_event.cc:
  Fix remaining cases of Bug #3596: fix possible races caused by an obsolete value of thd->query_length in SHOW PROCESSLIST and SHOW INNODB STATUS; this fix depends on the fact that thd->query is always set to NULL before setting it to point to a new query
sql/slave.cc:
  Fix remaining cases of Bug #3596: fix possible races caused by an obsolete value of thd->query_length in SHOW PROCESSLIST and SHOW INNODB STATUS; this fix depends on the fact that thd->query is always set to NULL before setting it to point to a new query
sql/sql_db.cc:
  Fix remaining cases of Bug #3596: fix possible races caused by an obsolete value of thd->query_length in SHOW PROCESSLIST and SHOW INNODB STATUS; this fix depends on the fact that thd->query is always set to NULL before setting it to point to a new query
sql/sql_parse.cc:
  Fix remaining cases of Bug #3596: fix possible races caused by an obsolete value of thd->query_length in SHOW PROCESSLIST and SHOW INNODB STATUS; this fix depends on the fact that thd->query is always set to NULL before setting it to point to a new query
sql/sql_show.cc:
  Fix remaining cases of Bug #3596: fix possible races caused by an obsolete value of thd->query_length in SHOW PROCESSLIST and SHOW INNODB STATUS; this fix depends on the fact that thd->query is always set to NULL before setting it to point to a new query

in hard-coded replication messages, always put small-length info (error codes, explanation of the error) at the beginning,
so that it is not cut by truncation if the query is very long (which happens if the query goes first).


sql/log_event.cc:
  in hard-coded replication messages, always put small-length info (error codes, explanation of the error) at the beginning,
  so that it is not cut by truncation if the query is very long (which happens if the query goes first).

if you are printing LOAD DATA INFILE and its USE as comments, don't update 'last_db' (because you
have not actually changed the db in the server).


sql/log_event.cc:
  if you are printing LOAD DATA INFILE and its USE as comments, don't update 'last_db' (because you
  have not actually changed the db in the server).

sql/log_event.cc:
  non_cachable_table flag changed to cacheable_table
sql/repl_failsafe.cc:
  non_cachable_table flag changed to cacheable_table
sql/slave.cc:
  non_cachable_table flag changed to cacheable_table
sql/sql_acl.cc:
  non_cachable_table flag changed to cacheable_table
sql/sql_base.cc:
  non_cachable_table flag changed to cacheable_table
sql/sql_cache.cc:
  non_cachable_table flag changed to cacheable_table
sql/sql_help.cc:
  non_cachable_table flag changed to cacheable_table
sql/sql_insert.cc:
  non_cachable_table flag changed to cacheable_table
sql/sql_parse.cc:
  non_cachable_table flag changed to cacheable_table
sql/sql_prepare.cc:
  renamed DEBUG tag
sql/table.h:
  non_cachable_table flag changed to cacheable_table

too big by 6 bytes. So I add code to substract 6 bytes if the master is 3.23.
This is not perfect (because it won't work if the slave I/O thread has not
noticed yet that the master is 3.23), but as long as the slave I/O thread
starts Exec_master_log_pos will be ok.
It must be merged to 4.1 but not to 5.0 (or it can be, because of #if MYSQL_VERSION_ID),
because 5.0 already works if the master is 3.23 (and in a more natural way:
in 5.0 we store the end_log_pos in the binlog and relay log).
I had to move functions from slave.h to slave.cc to satisfy gcc.


sql/log_event.cc:
  make the event's length 6 bytes shorter if the master is 3.23
sql/slave.cc:
  Moving several st_relay_log_info methods out of the declaration of the struct,
  because gcc complained that 'mi' was not declared.
  Substracting 6 bytes from the event's length in inc_pos() if the master is 3.23.
sql/slave.h:
  moving several methods out of the declaration of st_relay_log_info.
  Removing 'inline', let's have the compiler decide.

Fixed bugs in group_concat with ORDER BY and DISTINCT (Bugs #2695, #3381 and #3319)
Fixed crash when doing rollback in slave and the io thread catched up with the sql thread
Set locked_in_memory properly


include/mysql_com.h:
  Fixed compiler warning
libmysqld/emb_qcache.cc:
  Removed not used variable
libmysqld/lib_sql.cc:
  Removed not used variable
myisam/mi_locking.c:
  Added comment
myisam/mi_rnext.c:
  Fixed bug in concurrent insert
myisam/mi_rprev.c:
  Simple optimization
mysql-test/r/func_gconcat.result:
  New tests
mysql-test/t/func_gconcat.test:
  New tests
mysql-test/t/func_group.test:
  Cleanup
sql-common/client.c:
  Removed compiler warning
sql/derror.cc:
  Better comments
sql/field.cc:
  Removed not used function/variable
sql/field.h:
  Removed not needed variable
sql/ha_innodb.cc:
  Removed not used function
sql/item.cc:
  Fixed compiler warning
sql/item_cmpfunc.cc:
  Fixed compiler warning
sql/item_func.cc:
  Fixed compiler warning
sql/item_geofunc.cc:
  Fixed ...

fixed error code in union test


mysql-test/t/union.test:
  right code worr 4.1 error
sql/log_event.cc:
  assign non-cachable for Item status for all stack tables for safety
sql/repl_failsafe.cc:
  assign non-cachable for Item status for all stack tables for safety
sql/slave.cc:
  assign non-cachable for Item status for all stack tables for safety
sql/sql_acl.cc:
  assign non-cachable for Item status for all stack tables for safety
sql/sql_base.cc:
  assign non-cachable for Item status for all stack tables for safety
  non-cachable status processing
sql/sql_cache.cc:
  assign non-cachable for Item status for all stack tables for safety
sql/sql_class.cc:
  removed ald way to prevent using stack tables for caching Items in PS
sql/sql_class.h:
  removed old way to prevent using stack tables for caching Items in PS
sql/sql_help.cc:
  assign non-cachable for Item status for all stack tables for safety
sql/sql_insert.cc:
  assign non-cachable for Item status for all stack tables for safety
  removed old way to prevent using stack tables for caching Items in PS
sql/table.h:
  non-cachable for Item status for tables

sql/log_event.cc:
  Cast for VC++ compiler error
sql/opt_range.cc:
  Cast for VC++ compiler error

We introduce a new function mysql_test_parse_for_slave().
If the slave sees that the query got a really bad error on master
(killed e.g.), then it calls this function to know if this query
can be ignored because of replicate-*-table rules (do not worry
about replicate-*-db rules: they are checked so early that they have
no bug). If the answer is yes, it skips the query and continues. If
it's no, then it stops and say "fix your slave data manually" (like it
did before this change).


mysql-test/r/rpl_error_ignored_table.result:
  result update
mysql-test/t/rpl_error_ignored_table-slave.opt:
  ignore more tables
mysql-test/t/rpl_error_ignored_table.test:
  we test if a killed query on the master, is ignored on the slave
  if the tables it involves should be excluded because of replicate-*-table
  rules.
sql/log_event.cc:
  If the query got a really bad error on the master (thread killed etc),
  parse it to test the table names: if the replicate-*-do|ignore-table
  rules say "this query must be ignored" then we exit gracefully;
  otherwise we warn about the bad error and tell DBA to check/fix it.
  Before this change, we always warned and stopped.
sql/mysql_priv.h:
  new function
sql/slave.cc:
  don't print error immediately as we need to do one more test to
  be sure.
sql/sql_parse.cc:
  we add a function mysql_test_parse_for_slave(), to be used only
  by the slave if it wants to know if the query should be ignored or not;
  so this function only parses the query, does not execute it.

- the one about BUG#2921
- the one about relay log flushing
Both will be rewritten in a next changeset
(this one will not be pushed before the next changeset).


sql/log.cc:
  undo
sql/log_event.cc:
  undo
sql/repl_failsafe.cc:
  undo
sql/slave.cc:
  undo
sql/slave.h:
  undo
sql/sql_repl.cc:
  undo

BUILD/compile-pentium-debug-max:
  Added --with-raid
configure.in:
  Removed -DFN_NO_CASE_SENCE for Mac OS X as this is not always true
mysql-test/install_test_db.sh:
  Added --skip-warnings
mysql-test/mysql-test-run.sh:
  Fixes to get --gdb and --ddd to work
mysql-test/r/lowercase_table2.result:
  Test for lower_case_table_names=2 and temporary tables
mysql-test/r/lowercase_table3.result:
  Fixed error message
mysql-test/r/multi_update.result:
  Test of behaviour of multi-table-delete and alias
mysql-test/t/lowercase_table2.test:
  Test for lower_case_table_names=2 and temporary tables
mysql-test/t/lowercase_table3.test:
  Fixed error
mysql-test/t/multi_update.test:
  Test of behaviour of multi-table-delete and alias (Bug #2940)
mysys/mf_iocache.c:
  Renamed _flush_io_cache to my_b_flush_io_cache
sql/ha_myisam.cc:
  Added comment
sql/lock.cc:
  Extra debugging
sql/log.cc:
  New parameter to flush_relay_log_info
sql/log_event.cc:
  New...

INSERT DELAYED works only for one-row inserts (in latest 4.0 versions
at least). So killing a delayed_insert thread does not spoil replication:
the rows which actually went into the table are exactly those listed
in the binlog. So when the delayed_insert thread is killed, don't log
it as 'killed', because it causes superfluous stops on the slave.


sql/log_event.cc:
  INSERT DELAYED works only for one-row inserts (in latest 4.0 versions
  at least). So killing a delayed_insert thread does not spoil replication:
  the rows which actually went into the table are exactly those listed
  in the binlog. So when the delayed_insert thread is killed, don't log
  it as 'killed', because it causes superfluous stops on the slave.

"(binlog, position) stored by InnoDB for a replication slave can be wrong".
This code contains conditional #if to distinguish between versions;
it should be merged into 4.1 and 5.0.


sql/ha_innodb.cc:
  correcting the binlog position stored in InnoDB in a slave.
sql/log_event.cc:
  better code to store the binlog position in InnoDB for a slave.
sql/slave.h:
  Better code for storing the binlog position in InnoDB for a slave.

"wrong Relay_Log_Pos if Rotate is in the middle of a transaction in relay log"
increment 'pending' instead of 'relay_log_pos'.


mysql-test/r/rpl_trunc_binlog.result:
  result update
sql/log_event.cc:
  - corrected error message
  - in Rotate_log_event::exec_event(), if we're in a transaction (which
  can happen if the I/O slave thread reconnected while reading a
  transaction), don't increment relay_log_pos but pending instead.
  Otherwise, relay_log_pos becomes garbage (and so if slave stops at that
  moment, it will never restart).

VC++Files/innobase/innobase.dsp:
  odbc0odbc.c was removed from the tree.
VC++Files/libmysql/libmysql.dsp:
  strxnmov.c is needed in this module.
sql/log_event.cc:
  Cast required to resolve compile error on Windows - byte * is 
  not equivalent to char *.
sql/opt_range.cc:
  Cast to const char * required to resolve compile error on Windows.
BitKeeper/etc/logging_ok:
  Logging to logging@openlogging.org accepted

mysql-test/r/subselect.result:
  test suite for BUG#2838
mysql-test/t/subselect.test:
  test suite for BUG#2838
sql/item.cc:
  revision of fix_fields() calls
  removed unnecessary variabl, and used correct last parameter of fix_fields()
sql/item_cmpfunc.cc:
  revision of fix_fields() calls (BUG#2838)
sql/item_func.cc:
  revision of fix_fields() calls
sql/item_row.cc:
  changed for efficience (and to be similar for Item_func::fix_fields)
sql/item_subselect.cc:
  fixed last arguments for fix_fields call
sql/item_sum.cc:
  layout fixed
  revision of fix_fields() calls
sql/log_event.cc:
  revision of fix_fields() calls
sql/set_var.cc:
  revision of fix_fields() calls
sql/sql_base.cc:
  revision of fix_fields() calls
sql/sql_class.cc:
  revision of fix_fields() calls
sql/sql_handler.cc:
  revision of fix_fields() calls

Added more DBUG statements
Ensure that we are comparing end space with BINARY strings
Use 'any_db' instead of '' to mean any database. (For HANDLER command)
Only strip ' ' when comparing CHAR, not other space-like characters (like \t)


BitKeeper/deleted/.del-ctype_tis620.result-old~3578ceb0b8284685:
  Delete: mysql-test/r/ctype_tis620.result-old
BitKeeper/deleted/.del-ctype_tis620.test-old~ffb1bbd2935d1aba:
  Delete: mysql-test/t/ctype_tis620.test-old
client/mysqlbinlog.cc:
  Added DBUG statements
  Added call of my_end() to free all used memory on exit
heap/hp_info.c:
  After merge fixes
heap/hp_open.c:
  After merge fixes
include/heap.h:
  After merge fixes
include/m_ctype.h:
  Use pchar instead of 'int' for character parameters.
  Added 'my_binary_compare()'
include/m_string.h:
  Fixed wrong define
innobase/ibuf/ibuf0ibuf.c:
  After merge fixes
innobase/srv/srv0start.c:
  After merge fixes
mysql-test/r/alter_table.result:
  Fixed results after merge
mysql-test/r/auto_increment.result:...

Always use all LOAD DATA options in replication.
Fixed problem in mysqlbinlog where LOAD DATA options could be overwritten.


client/mysqlbinlog.cc:
  Ensure that we don't call init_dynamic_array() before my_init() has been done
configure.in:
  Always use libtool with "--preserve-dup-deps" to prevent link problems with g++
mysql-test/r/mysqlbinlog.result:
  new results
mysql-test/r/range.result:
  new results
mysql-test/t/range.test:
  add missing DROP TABLE
sql/log_event.cc:
  Always use all LOAD DATA options. This is needed to handle commands of type "FIELDS TERMINATED BY ''"
  Allocate memory for Create_file_log_event options as these may be overwritten in mysqlbinlog
sql/log_event.h:
  Free buffer

client/mysqlbinlog.cc:
  fixed bug #1378 "mysqlbinlog for remote host is broken"
  by using shared part of code for local and remote variant 
  (function process_event)
mysql-test/r/mysqlbinlog.result:
  fixed results for remote variant 
  (they are the same as for local now)
mysql-test/t/mysqlbinlog.test:
  discarded note 'V. Vagin should ...' 
  because V.Vagin done
sql/log_event.cc:
  added special local_fname flag to Load_log_event for 
  using in Load_log_event::check_fname_outside_temp_buf 
  instead of ugly condition
sql/log_event.h:
  added special local_fname flag to Load_log_event for 
  using in Load_log_event::check_fname_outside_temp_buf 
  instead of ugly condition

reset errors (in thd) before executing the event. Otherwise if an event is ignored
because of replicate-*-table rules (error ER_SLAVE_IGNORED_TABLE) this error code
may remain in thd->net and the next event may pick it. 


sql/log_event.cc:
  Reset errors (in thd) before executing the event. Otherwise if an event is ignored
  because of replicate-*-table rules (error ER_SLAVE_IGNORED_TABLE) this error code
  may remain in thd->net and the next event may pick it.

Fixed bug in ORDER BY on a small column (Bug #2147)
Fixed error from pthread_mutex_destroy() when one had wrong errmsg file



client/mysqltest.c:
  Added handling of error on query send
  (Needed for init_connection.test)
mysql-test/mysql-test-run.sh:
  Added tracing of mysqldump and mysqlbinlog
mysql-test/r/init_connect.result:
  Updated tests
mysql-test/r/order_by.result:
  Added test for bug filesort bug
mysql-test/t/init_connect-master.opt:
  Added proper quoting (for Solaris and OSF)
mysql-test/t/init_connect.test:
  Portability fix
mysql-test/t/order_by.test:
  Added test for bug #2147 (bug in filesort)
sql/filesort.cc:
  Fixed bug in filesort (Bug #2147)
sql/item.h:
  Clear 'fixed' on cleanup (For prepared statements)
sql/item_func.cc:
  Protect mutex destroy.
  (Fixed error from pthread_mutex_destroy() when one had wrong errmsg file)
sql/log_event.cc:
  Portability fix
sql/sql_class.h:
  Fixed compiler warning
sql/sql_prepare.cc:
  Portability fi...

"EE_ error codes (EE_DELETE, EE_WRITE) end up in the binlog, making slave stop".
The problem was that during execution of the command on the master, an error
can occur (for example, not space left on device, then mysqld waits and when
there is space it completes successfully: so finally it worked but the error
EE_WRITE remains in thd->net.last_errno and thd->net.last_error).
To know if finally the command succeeded, we test the 'error' variable in
every place, and if it shows no failure we reset thd->net.last_err* using
the function THD::clear_error() which is backported from 4.1.
A new test to see if now only real errors get to the binlog (note: the test
uses "rm").

Also a bit of memory free/alloc saving in log_event.cc (do not free the whole
mem_root after every query in the slave SQL thread: we can keep the initial
block of it; which will be freed when the thread terminates).


sql/log_event.cc:
  In the slave SQL thread, it's a waste to free the initial block of the mem_root
  after every query. We can instead keep it. It will be freed when the thread
  terminates (in THD::~THD()).
sql/sql_acl.cc:
  clear the error in thd->net.last_errno as there was no error
sql/sql_base.cc:
  clear the error in thd->net.last_errno as there was no error
sql/sql_class.h:
  Backport of THD::clear_error() from 4.1:
  clears the error in thd->net.last_errno
sql/sql_db.cc:
  clear the error in thd->net.last_errno as there was no error
sql/sql_delete.cc:
  clear the error in thd->net.last_errno as there was no error
sql/sql_insert.cc:
  clear the error in thd->net.last_errno as there was no error
sql/sql_parse.cc:
  clear the error in thd->net.last_errno as there was no error
sql/sql_rename.cc:
  clear the error in thd->net.last_errno as there was no error
sql/sql_table.cc:
  clear the error in thd->net.last_errno as there was no error
sql/sql_update.cc:
  clear the error in thd->net.last_errno as there was no error