Visual C++ dont relate 'byte' and 'char*' that close, added casts
mysql.sln:
Only build mysqlserver.lib if embedded target
VC++Files/mysql.sln:
Only build mysqlserver.lib if embedded target
sql/log_event.cc:
Visual C++ dont relate 'byte' and 'char*' that close, added casts
Enable row based replication for Windows build
replication.m4:
Made help text more clear, default is with row based replication enabled
config/ac-macros/replication.m4:
Made help text more clear, default is with row based replication enabled
include/config-win.h:
Enable row based replication for Windows build
Drop 'mysqlslap'
Added database name 'ndbsynctest'
sql_table.cc, create.test, create.result:
Patch for bug#17530 done by evgen, extracted from 5.0
mysql-test/r/create.result:
Patch for bug#17530 done by evgen, extracted from 5.0
mysql-test/t/create.test:
Patch for bug#17530 done by evgen, extracted from 5.0
sql/sql_table.cc:
Patch for bug#17530 done by evgen, extracted from 5.0
mysql-test/include/drop-on-restart.inc:
Drop 'mysqlslap'
storage/ndb/src/common/transporter/Transporter.cpp:
use sockaddr_in in declarations to guarantee sin_addr alignment
storage/ndb/src/mgmsrv/Services.cpp:
use sockaddr_in in declarations to guarantee sin_addr alignment
BUG# 17432: Partitions: wrong result, SELECT ... WHERE <column> is null
mysql-test/r/partition.result:
result block for tests
mysql-test/t/partition.test:
test cases for bug #17432 and 17430
sql/sql_partition.cc:
improve null handling by returning LONGLONG_MIN for values that are NULL
client/mysqlimport.c:
Remove disabled --use-threads option
mysql-test/r/mysqldump.result:
Update result
mysql-test/t/disabled.def:
Remove disable of mysqldump
mysql-test/t/mysqldump.test:
Remove the disable_parsing on tests for --use-threads to mysqlimport
mysql-test/r/rpl_bit_npk.result:
Updated results
mysql-test/t/disabled.def:
rpl_bit_npk now works
sql/field.h:
Field_bit::cmp_binary_offset wrongly used base class method that does not work for Field_bit
This was discussed with Monty and should be pushed into 5.0 too
sql/log_event.cc:
Added checks for null bits
Swapped use of m_after_image (was m_search_record) and table->record[1] to use record[i]
in the same way as other MySQL code (i.e. use record[1] for scan data).
Removed use of cmp_binary in record_compare (it is currently wrong to use that
without copying the null bits to the compare data record)
sql/log_event.h:
Name change to indicate new semantics
Problem was a null pointer
Fixed by passing correct database name to slave filtering mechanism
Added test.
sql/sql_parse.cc:
Fix slave crash when issuing 'ALTER DATABASE' with specifying db name
Don't pass null pointer to db_ok_with_wild_table()
Conditionally link with libmysqlclient.la or libmysqlclient_r.la
configure.in:
Corrected help text for --with-debug option
Changed default to be with thread safe client
Added THREAD_SAFE_CLIENT to be used in 'if' in make files
configure.in:
Corrected help text for --with-debug option
Changed default to be with thread safe client
Added THREAD_SAFE_CLIENT to be used in 'if' in make files
client/Makefile.am:
Conditionally link with libmysqlclient.la or libmysqlclient_r.la
Moving assignments to table_map_id for thread-safe handling of
table shares.
sql/ha_ndbcluster_binlog.cc:
Assign_new_table_id() now takes table share instead of table.
sql/mysql_priv.h:
New protptype for assign_new_table_id().
sql/sql_base.cc:
Assign_new_table_id() now takes a table share instead of a table.
Moving call to assign_new_table_id() into get_table_share().
sql/table.cc:
Setting default values of table_map_id and table_map_version inside
alloc_table_share() and init_tmp_table_share().
Removing the settings from open_table_from_share().
This is a re-clone after tagging, so the version must be set back to 5.1.7-beta.
configure.in:
This is a re-clone after tagging, so the version must be set back to 5.1.7-beta.
post-review fixes
mysql-test/r/events_bugs.result:
fix result
sql/event.cc:
fix compilation problem on windows
sql/event_executor.cc:
show the right host in show processlist (should be event_scheduler@localhost)
sql/event_timed.cc:
a bit more debug infor
fix a problem introduced with previous push :( . clean everything that's whitespace.
found because of new test cases which were crashing.
sql/sql_show.cc:
fix compile problem on windows
WL#1032
sql_mode setting was disregarded during create/alter event and wasn't set during
event execution.
(post-review small fixes)
sql/event.cc:
store the sql_mode when the body is stored
- always on create event
- during alter event if the user changes the body
sql/event.h:
add sql_mode as member variable
sql/event_timed.cc:
- exchange thd->variables.sql_mode before and after execution
- set the sql_mode of the anonymous SP -> event_timed::sphead
sql/sql_show.cc:
- show real content in I_S.EVENTS.SQL_MODE , disallow NULL
if we've got a tree of type ALWAYS or MAYBE.
sql/opt_range.cc:
BUG#15448 (group_min_max test failure): Don't buypass group-min-max optimizer entry point
if we've got a tree of type ALWAYS or MAYBE.
The effect observed by this bug has been introduced by this cset:
ChangeSet 1.1967.5.2 2005/12/26 08:40:09 sergefp@mysql.com. The cset introduced PARAM::remove_jump_scans
which purpose was to avoid making [always unsuccessful] attempts to construct index_merge for construct
index_merge for conditions like "t.key1partX=c1 OR t.key1partY=c2". The effect of this change was that some
queries that previously had produced a SEL_TREE representing candidate for index_merge(X,Y) now started to
produce SEL_TREE with type=ALWAYS, and group-min-max analysis for these trees has been bypassed.
In this fix we make SEL_TREE with type=ALWAYS to be processed in the same way as (SEL_TREE*)NULL, its equivalent.
Fixes failing sp.test
mysql-test/r/backup.result:
A new error message for deprecated statements.
mysql-test/t/sp.test:
Disable warnings in the test for Bug#13012
sql/share/errmsg.txt:
A new error message for deprecated statements.
sql/sql_yacc.yy:
Use a better worded error message.
storage/ndb/src/ndbapi/NdbBlob.cpp:
DBUG
storage/ndb/include/ndbapi/Ndb.hpp:
method to set db/schema from table name + format check in internalize_table_name
storage/ndb/include/ndbapi/NdbDictionary.hpp:
method to set db/schema from table name + format check in internalize_table_name
storage/ndb/src/ndbapi/Ndb.cpp:
method to set db/schema from table name + format check in internalize_table_name
sql/ha_ndbcluster_binlog.cc:
set injector_ndb db/schema before calling NDB (may be more cases..). only place to get it is table internal name
