into mysql.com:/usr/home/bar/mysql-4.1.b12829
mysql-test/r/func_gconcat.result:
Auto merged
mysql-test/t/func_gconcat.test:
Auto merged
sql/item_sum.cc:
Auto merged
adding test case.
item_cmpfunc.cc:
Bug#12611
ESCAPE + LIKE do not work when the escape char is a multibyte one
Additional fix for 8bit character sets:
escape character must be converted into
operation character set.
sql/item_cmpfunc.cc:
Bug#12611
ESCAPE + LIKE do not work when the escape char is a multibyte one
Additional fix for 8bit character sets:
escape character must be converted into
operation character set.
mysql-test/t/func_like.test:
adding test case.
mysql-test/r/func_like.result:
adding test case.
Bug 12611 : ESCAPE + LIKE do not work when the escape char is a multibyte one
Forgot to commit this file in the previous changeset,
together with other files.
sql/item_cmpfunc.h:
Bug 12611 : ESCAPE + LIKE do not work when the escape char is a multibyte one
Forgot to commit this file in the previous changeset,
together with other files.
into sanja.is.com.ua:/home/bell/mysql/bk/work-4.1
sql/mysql_priv.h:
Auto merged
sql/sql_cache.cc:
Auto merged
sql/sql_cache.h:
Auto merged
configure.in:
merge
mysql-test/r/fulltext_var.result:
merge
sql/mysqld.cc:
merge
sql/set_var.cc:
merge
sql/mysql_priv.h:
initialisation moved to mysqld.cc
sql/mysqld.cc:
initialisation moved to mysqld.cc
sql/sql_cache.cc:
support of concurent query cache resizing:
- resizing made atomic
- check stack size after each quard mutex lock
sql/sql_cache.h:
initialisation moved to mysqld.cc
removed uneed parameter (now it is always under guard mutex protection or called from destruction)
mysql-test/r/mysqltest.result:
Disable test for failing --exec and --system since output can't be piped to /dev/null on all platforms
mysql-test/t/mysqltest.test:
Disable test for failing --exec and --system since output can't be piped to /dev/null on all platforms
Avoid "echo -e" by using a temporary .sql file for tests that require more than one line.
bug#9948
changed client-charset behavior in 4.1.x libmysql, which issue BC prob
after review fixes
sql/mysql_priv.h:
bug#9948
changed client-charset behavior in 4.1.x libmysql, which issue BC prob
after review fixes
sql/mysqld.cc:
bug#9948
changed client-charset behavior in 4.1.x libmysql, which issue BC prob
after review fixes
Test case for bug #12859 group_concat in subquery cause incorrect not null.
mysql-test/r/func_gconcat.result:
Test case for bug #12859 group_concat in subquery cause incorrect not null.
mysql-test/t/func_gconcat.test:
Test case for bug #12859 group_concat in subquery cause incorrect not null.
myisam/mi_check.c:
a fix (bug #4214: Table corruption with myisampack and large BLOB objects).
- pass version to the save_pack_length().
myisam/mi_packrec.c:
a fix (bug #4214: Table corruption with myisampack and large BLOB objects).
- code cleanup: read_pack_length() and calc_pack_length() introduced,
save_pack_length() modified: now the behavior depends on packing version
- save packing version in the share->pack.version
- pass it to the read_pack_length()
myisam/mi_static.c:
a fix (bug #4214: Table corruption with myisampack and large BLOB objects).
- packing version set to 2
myisam/myisamdef.h:
a fix (bug #4214: Table corruption with myisampack and large BLOB objects).
- packing version slot introduced (see MI_PACK)
myisam/myisampack.c:
a fix (bug #4214: Table corruption with myisampack and large BLOB objects).
- code cleanup
- pass version to the calc_pack_length() and save_pack_length()
+ added option to dumpe rest of page after exnd of data
ndb/src/kernel/blocks/dblqh/redoLogReader/records.cpp:
fixed small bug in ndb redolog printer
ndb/src/kernel/blocks/dblqh/redoLogReader/records.hpp:
fixed small bug in ndb redolog printer
Handle systems like default FC4 where 127.0.0.1 doesn't always map to 'localhost' first.
(Bug #11822)
sql/hostname.cc:
Short-circuit ip_to_hostname() lookup for INADDR_LOOPBACK to allways return 'localhost'.
sql/sql_parse.cc:
Push special handling of 127.0.0.1 into ip_to_hostname().
Always abort if node failure occured between startTransaction and commit
(can later be improved by wl#2610)
ndb/src/kernel/blocks/dbtc/DbtcMain.cpp:
Always abort if node failure occured between startTransaction and commit
make sure not to cut log to early
(specificly not use LCP with maxGciStarted that has not yet completed)
ndb/include/kernel/signaldata/SignalData.hpp:
Add signal data printer for START_FRAG_REQ
ndb/include/kernel/signaldata/StartFragReq.hpp:
Add signal data printer for START_FRAG_REQ
ndb/src/common/debugger/signaldata/SignalDataPrint.cpp:
Add signal data printer for START_FRAG_REQ
ndb/src/common/debugger/signaldata/StartRec.cpp:
Add signal data printer for START_FRAG_REQ
ndb/src/kernel/blocks/dbdih/DbdihMain.cpp:
Add maxGciStarted/Completed to event report
bug: dont use LCP for calcKeepGci in maxGciSTarted is not completed
mysql-test/r/mysqltest.result:
Update test results
mysql-test/t/mysqltest.test:
Send output from failed command to /dev/null in order to avoid recording the output from different shells
into mysql.com:/usr/home/bar/mysql-4.1.b12363
mysql-test/t/variables.test:
Auto merged
sql/set_var.cc:
Auto merged
mysql-test/r/variables.result:
SCCS merged
Makefile.am:
Manual merge: For "test", use Kent's new version using the Perl script with port variables,
but also have the "test-force" target (used in the RPM spec!) (still using Shell).
