Must not use Item_direct_ref in HAVING because it points to
the new value (witch is not yet calculated for the first row).
mysql-test/r/subselect.result:
Bug #16255: Subquery in where
- test case
mysql-test/t/subselect.test:
Bug #16255: Subquery in where
- test case
sql/item_subselect.cc:
Bug #16255: Subquery in where
Must not use Item_direct_ref in HAVING because it points to
the new value (witch is not yet calculated for the first row).
The crash was caused by invalid sequence of handler::** calls:
ha_smth->index_init();
ha_smth->index_next_same(); (2)
(2) is an invalid call as it was not preceeded by any 'scan setup' call
like index_first() or index_read(). The cause was that QUICK_SELECT::reset()
didn't "fully reset" the quick select- current QUICK_RANGE wasn't forgotten,
and quick select might attempt to continue reading the range, which would
result in the above mentioned invalid sequence of handler calls.
5.x versions are not affected by the bug - they already have the missing
"range=NULL" clause.
mysql-test/r/innodb_mysql.result:
Testcase for BUG#21077
mysql-test/t/innodb_mysql.test:
Testcase for BUG#21077
sql/opt_range.h:
BUG#21077: Possible crash caused by invalid sequence of handler::* calls:
- Make QUICK_SELECT::reset() really reset the quick select
myisam/mi_uniue.c:mi_check_unique() should skip trailing spaces comparing
TEXT and VARTTEXT key segments.
myisam/mi_unique.c:
Fix for bug #20709: Collation not used in group by on 4.1.
myisam/mi_uniue.c:mi_check_unique() should skip trailing spaces comparing
TEXT and VARTTEXT key segments.
Example: assume, we have a 'char(200) collate utf8_unicode_ci' field,
there are two records with _utf8"0x65" and _utf8"0xC3A9" characters;
these values are equal according
to the utf8_unicode_ci collation, but two 600 byte length corresponding keys:
"0x65<0x20 repeats 599 times>" and "0xC3A9<0x20 repeats 598 times>" are not
equal if we count trailing spaces and it may cause inconsequent behavior.
So, let's pass 1 as the skip_end_space parameter value to the mi_compare_text()
function for proper TEXT and VARTTEXT key segments comparison.
mysql-test/r/ctype_utf8.result:
Fix for bug #20709: Collation not used in group by on 4.1.
- test results.
mysql-test/t/ctype_utf8.test:
Fix for bug #20709: Collation not used in group by on 4.1.
- test case.
Problem described in this bug report affects MyISAM tables only.
Running mysqld --flush instructs mysqld to sync all changes to disk
after each SQL statement. It worked well for INSERT and DELETE
statements, but it did sync for UPDATE only in case if there was
index change (change of colum that has an index). If no updated column
has an index, data wasn't synced to disk.
This fix makes UPDATE statement to sync data to disk even if there is
no index change (that is only data change) and mysqld is run with
--flush option.
myisam/mi_update.c:
Every myisam function that updates myisam table must end with
call to _mi_writeinfo(). If operation (second param of
_mi_writeinfo()) is not 0 it sets share->changed to 1, that is
flags that data has changed. If operation is 0, this function
equals to no-op in this case.
mi_update() must always pass !0 value as operation, since even if
there is no index change there could be data change.
privileges
This problem is 4.1 specific. It doesn't affect 4.0 and was fixed
in 5.x before.
Having any mysql user who is allowed to issue multi table update
statement and any column/table grants, allows this user to update
any table on a server (mysql grant tables are not exception).
check_grant() accepts number of tables (in table list) to be checked
in 5-th param. While checking grants for multi table update, number
of tables must be 1. It must never be 0 (actually we have
DBUG_ASSERT(number > 0) in 5.x in grant_check() function).
mysql-test/r/grant.result:
Addition to test case for bug#7391:
- Added grant statement to trigger this problem in 4.1.
- Fixed error messages.
mysql-test/t/grant.test:
Addition to test case for bug#7391:
- Added grant statement to trigger this problem in 4.1.
- Fixed error messages.
sql/sql_update.cc:
check_grant() accepts number of tables (in table list) to be checked
in 5-th param. For this particular check number of tables must be 1.
It must never be 0 (actually we have DBUG_ASSERT(number > 0) in 5.x
in grant_check() function).
set.
(Ramil's patch, recreated.)
heap/hp_delete.c:
Reset info->lastkey_len for further heap_rnext/heap_rprev calls.
mysql-test/r/heap_btree.result:
Test for bug #9719: DELETE with WHERE on HEAP table just deletes first
row of matched set.
mysql-test/t/heap_btree.test:
Test for bug #9719: DELETE with WHERE on HEAP table just deletes first
row of matched set.
into moonbone.local:/work/tmp_merge-4.1-opt-mysql
sql/item_strfunc.cc:
Auto merged
sql/mysql_priv.h:
Auto merged
sql/sql_parse.cc:
Auto merged
sql/sql_select.cc:
Auto merged
into mysql.com:/Users/kent/mysql/bk/mysql-4.1
configure.in:
Auto merged
scripts/make_binary_distribution.sh:
Auto merged
support-files/mysql.spec.sh:
SCCS merged
Man page for mysqld command move to section 8 (bug#21220)
configure.in:
Man page for mysqld command move to section 8 (bug#21220)
man/Makefile.am:
Man page for mysqld command move to section 8 (bug#21220)
Man page for "mysqld" command move to section 8 (bug#21220)
scripts/make_binary_distribution.sh:
Man page for "mysqld" command move to section 8 (bug#21220)
When processing aggregate functions all tables values are reset
to NULLs at the end of each group.
When doing that if there are no rows found for a group
the const tables must not be reset as they are not recalculated
by do_select()/sub_select() for each group.
mysql-test/r/subselect2.result:
* Bug #20792: Incorrect results from aggregate subquery
- test suite for the bug. This is dependent on InnoDB despite
the fact that the bug and the fix are not InnoDB specific.
This is because of the table flag HA_NOT_EXACT_COUNT.
When this flag is off (as in MyISAM) both t2 and t3 become of
join type 'system' as they are estimated to have 1 record and
and this statistics can be trusted (according to the absence of
HA_NOT_EXACT_COUNT).
mysql-test/t/subselect2.test:
* Bug #20792: Incorrect results from aggregate subquery
- test suite for the bug
sql/sql_select.cc:
* Bug #20792: Incorrect results from aggregate subquery
- when clearing results if there are not rows found for group
the const tables must not be reset as they are not recalculated
for each group.
Too many cursors (more than 1024) could lead to memory corruption.
This affects both, stored routines and C API cursors, and the
threshold is per-server, not per-connection. Similarly, the
corruption could happen when the server was under heavy load
(executing more than 1024 simultaneous complex queries), and this is
the reason why this bug is fixed in 4.1, which doesn't support
cursors.
The corruption was caused by a bug in the temporary tables code, when
an attempt to create a table could lead to a write beyond allocated
space. Note, that only internal tables were affected (the tables
created internally by the server to resolve the query), not tables
created with CREATE TEMPORARY TABLE. Another pre-condition for the
bug is TRUE value of --temp-pool startup option, which, however, is a
default.
The cause of a bug was that random memory was overwritten in
bitmap_set_next() due to out-of-bound memory access.
mysys/my_bitmap.c:
Local 'bitmap_size' is measured in bytes, no need to multiply it by 8.
sql/sql_select.cc:
Clear the temp_pool_slot bit only if we have set it previously.
When optimizing conditions like 'a = <some_val> OR a IS NULL' so that they're
united into a single condition on the key and checked together the server must
check which value is the NULL value in a correct way : not only using ->is_null
but also check if the expression doesn't depend on any tables referenced in the
current statement.
This additional check must be performed because that optimization takes place
before the actual execution of the statement, so if the field was initialized
to NULL from a previous statement the optimization would be applied incorrectly.
mysql-test/r/select.result:
Bug #21019: First result of SELECT COUNT(*) different than consecutive runs
- test case
mysql-test/t/select.test:
Bug #21019: First result of SELECT COUNT(*) different than consecutive runs
- test case.
Note that ALTER TABLE is important here : it happens to
leave the Field instance for t1.b set to NULL, witch is vital for
demonstrating the problem fixed by this changeset.
sql/sql_select.cc:
Bug #21019: First result of SELECT COUNT(*) different than consecutive runs
- check whether a value is null taking into account its table dependency.
- Modify test case to workaround the test tool problem
- (Null merge into 5.0)
mysql-test/t/mysqlbinlog.test:
Send the ouput from "mysqlbinlog" to a file and then read execute it with "mysql"
This is a workaround for a windows bug in the test tools in mysql-4.1, which makes
it impossible to use "|" to send the output from one program directly to the other.
This has been fixed in mysql-5.0.
innobase/ibuf/ibuf0ibuf.c:
Applied innodb-4.1-ss32 snapshot.
ibuf_fixed_addr_page(): Add parameter space. As the insert
buffer B-tree is only located in the system tablespace
(space 0), IBUF_TREE_ROOT_PAGE_NO is only special in space 0.
The problem was in that opt_sum_query() replaced MIN/MAX functions
with the corresponding constant found in a key, but due to imprecise
representation of float numbers, when evaluating the where clause,
this comparison failed.
When MIN/MAX optimization detects that all tables can be removed,
also remove all conjuncts in a where clause that refer to these
tables. As a result of this fix, these conditions are not evaluated
twice, and in the case of float number comparisons we do not discard
result rows due to imprecise float representation.
As a side-effect this fix also corrects an unnoticed problem in
bug 12882.
mysql-test/r/func_group.result:
BUG#20954 - test result adjustment.
Adjusted the test result of bug 12882 which was not preperly fixed.
The current patch corrects the problem that was fully corrected by the
patch for 12882.
The problem was that opt_sum_query() indicated that the optimizer may
remove all tables because all MIN/MAX/COUNT functions are constants,
but this lead to an empty result instead of NULL because the WHERE
clause was still evaluated.
The current fix removes all conjuncts in the where clause that
reference the removed tables, and thus corrects the problem.
mysql-test/r/select.result:
BUG#20954 - added test
mysql-test/r/subselect.result:
BUG#20954 - test result adjustment.
The fix removes those conditions in a where clause that refer to
tables optimized away by MIN/MAX optimization (opt_sum_query()).
mysql-test/t/select.test:
BUG#20954 - added test
sql/sql_select.cc:
Fix for BUG#20954: avg(keyval) retuns 0.38 but max(keyval) returns an empty set
When MIN/MAX optimization detects that all tables can be removed,
also remove all conjuncts in a where clause that refer to these
tables. As a result of this fix, these conditions are not evaluated
twice, and in the case of float number comparisons we do not discard
result rows due to imprecise float representation.
As a side-effect this fix also corrects an unnoticed problem in
bug 12882.
LIKE craashed with a pattern having letters in the range 128..255
(e.g. A WITH ACUTE or C WITH CARON) because of wrong cast from
signed char to unsigned int.
mysql-test/r/ctype_cp1250_ch.result:
Adding test case
mysql-test/t/ctype_cp1250_ch.test:
Adding test case
strings/ctype-win1250ch.c:
Fixing wrong cast from "signed char" -> "uint" to
"signed char" -> "unsigned char" -> uint, to properly
handle bytes 128..255.
- Send confusing output to /dev/null
mysql-test/t/mysqlbinlog.test:
Send confusing error messages to /dev/null so they don't appear in erro log if test case fails
- The mysql-test/std_data/bug15328.cnf file was not copied to install or dist dir.
mysql-test/Makefile.am:
Add std_data/*.cnf files to dist_hook and install-data-local
scripts/make_win_src_distribution.sh:
Copy *.cnf files as well.
