ROW('a','b','c) = ROW('A' collate latin1_bin,'b','c') returns 0
When a number is compared to a string, character sets and
collations are not aggregated. e.g. this returned error in 4.1.0:
SELECT 1=_latin2'1';
because character sets was aggregated, and 1 was considered as
a string of latin1 charset during this aggregation.
mysql-test/r/func_gconcat.result:
Update results
mysql-test/t/func_gconcat.test:
Make tests repeatable. The problem is that due to floating point handling / qsort implementation the way rows are read may differ between versions
into sanja.is.com.ua:/home/bell/mysql/bk/work-err-4.1
mysql-test/r/subselect.result:
Auto merged
sql/sql_parse.cc:
Auto merged
sql/sql_select.cc:
Auto merged
acinclude.m4:
Auto merged
mysql-test/r/subselect.result:
Auto merged
scripts/mysql_create_system_tables.sh:
Auto merged
sql/sql_class.h:
Auto merged
sql/sql_yacc.yy:
Auto merged
mysql-test/r/auto_increment.result:
Update after auto_increment fix
mysql-test/t/auto_increment.test:
Another fix for LAST_INSERT_ID()
sql/handler.cc:
Another fix for LAST_INSERT_ID()
sql/sql_class.h:
Another fix for LAST_INSERT_ID()
Fixed bug in ALTER TABLE ... MODIFY integer-column
Added ref_or_null optimization (needed for subqueries)
BUILD/compile-pentium-valgrind-max:
Remove FORCE_INIT_OF_VARS to spot wrong LINT_INIT() options
mysql-test/r/distinct.result:
Update of test results (new optimizer)
mysql-test/r/null_key.result:
Update after ref_or_null optimization
mysql-test/r/subselect.result:
Update after ref_or_null optimization
mysql-test/t/null_key.test:
New tests for ref_or_null optimization
sql/sql_select.cc:
Added ref_or_null optimization
Optimized find_best_combinations() and read-functions
sql/sql_select.h:
Added ref_or_null optimization
sql/sql_yacc.yy:
Fixed bug in ALTER TABLE ... MODIFY integer-column
sql/table.cc:
Safety fix for ALTER TABLE .. MODIFY
sql/unireg.cc:
Safety fix for ALTER TABLE .. MODIFY
(resolving fields of reduced subselect)
mysql-test/r/subselect.result:
test of bug #745
mysql-test/t/subselect.test:
test of bug #745
sql/sql_union.cc:
to get chance of outer resolved_fields resolve fields of result table as usual fileds
Added command 'replace_column' to mysqltest
client/mysqltest.c:
Added user command replace_column
mysql-test/r/merge.result:
Update of tests
mysql-test/t/innodb.test:
User replace_column to make tests repeatable
mysql-test/t/merge.test:
New tests for testing MERGE tables with multiple databases
scripts/mysql_create_system_tables.sh:
Fixed syntax error
sql/ha_myisammrg.cc:
Allow one to use MERGE tables with tables from different databases
sql/mysql_priv.h:
Added function
sql/sql_parse.cc:
Allow one to use MERGE tables with tables from different databases
sql/sql_show.cc:
Extended append_identifier()
and comparison was done binary. Now, a binary string wins a character sting
taking in account their derivation. That means a character field wins a
binary literal and comparison is done according to the character field
collation, not binary.
fixed BUG #726
mysql-test/r/subselect.result:
test of bug #726
mysql-test/t/subselect.test:
test of bug #726
sql/item_cmpfunc.cc:
cond_count moved to SELECT_LEX_NODE
sql/item_subselect.cc:
fixed current_select pointer
sql/sql_base.cc:
cond_count moved to SELECT_LEX_NODE
sql/sql_class.cc:
cond_count moved to SELECT_LEX_NODE
sql/sql_class.h:
cond_count moved to SELECT_LEX_NODE
sql/sql_lex.cc:
cond_count moved to SELECT_LEX_NODE
sql/sql_lex.h:
cond_count moved to SELECT_LEX_NODE
sql/sql_select.cc:
cond_count moved to SELECT_LEX_NODE
fixed bug of current select pointer in subselect execution
fixed layuot
mysql-test/r/subselect.result:
test of update with IGNORE option and error in subselect
mysql-test/t/subselect.test:
test of update with IGNORE option and error in subselect
sql/item_subselect.cc:
fixed bug of current select pointer
sql/mysqld.cc:
error converting to warnings if hd->lex.current_select->no_error id true
layout fixed
sql/sql_lex.cc:
initialization
sql/sql_lex.h:
flag which force converting errors to warnings
sql/sql_select.cc:
switch on error convering to warnings in select executing if IGNORE option present
outer resolved fields now do not marked as dependent if they is not really belong to outer SELECT
one-letter variables name fixed
mysql-test/r/subselect.result:
test of bug 679
mysql-test/t/subselect.test:
test of bug 679
sql/item.cc:
outer resolved fields now do not marked as dependent if they is not really belong to outer SELECT
one-letter variables name fixed
into sanja.is.com.ua:/home/bell/mysql/bk/work-item-4.1
sql/item.cc:
Auto merged
sql/item_subselect.cc:
Auto merged
sql/item_sum.cc:
Auto merged
sql/sql_lex.h:
Auto merged
sql/sql_select.cc:
Auto merged
sql/sql_union.cc:
Auto merged
sql/sql_yacc.yy:
Auto merged
Removed non-generic info (key used blocks)
preload.result, preload.test, lex.h, sql_yacc.yy:
Syntax change for preload statement
sql/sql_yacc.yy:
Syntax change for preload statement
sql/lex.h:
Syntax change for preload statement
mysql-test/t/preload.test:
Removed non-generic info (key used blocks)
mysql-test/r/preload.result:
Removed non-generic info (key used blocks)
BitKeeper/etc/logging_ok:
auto-union
sql/mysql_priv.h:
Auto merged
sql/mysqld.cc:
Auto merged
sql/set_var.cc:
Auto merged
sql/sql_parse.cc:
Auto merged
sql/sql_yacc.yy:
Auto merged
mysql-test/t/rpl_temporary.test:
user change to avoid conflicts with user 'test'
(to which test connect.test gives password "gambling", so depending
on the order of the tests, 'test' has password '' or 'gambling':
better to use an unknown user which will be directed to ''@localhost
with password always '').
into rurik.mysql.com:/home/igor/dev/mysql-4.1-0
include/my_base.h:
Auto merged
include/my_sys.h:
Auto merged
myisam/myisamdef.h:
Auto merged
sql/ha_myisam.cc:
Auto merged
sql/handler.cc:
Auto merged
sql/handler.h:
Auto merged
sql/lex.h:
Auto merged
sql/mysql_priv.h:
Auto merged
sql/mysqld.cc:
Auto merged
sql/set_var.cc:
Auto merged
sql/sql_base.cc:
Auto merged
sql/sql_class.h:
Auto merged
sql/sql_lex.h:
Auto merged
sql/sql_parse.cc:
Auto merged
sql/sql_table.cc:
Auto merged
sql/sql_yacc.yy:
Auto merged
