When only one row was present, the subtraction of nearly the same number
resulted in catastropic cancellation, introducing an error in the
VARIANCE calculation near 1e-15. That was sqrt()ed to get STDDEV, the
error was escallated to near 1e-8.
The simple fix of testing for a row count of 1 and forcing that to yield
0.0 is insufficient, as two rows of the same value should also have a
variance of 0.0, yet the error would be about the same.
So, this patch changes the formula that computes the VARIANCE to be one
that is not subject to catastrophic cancellation.
In addition, it now uses only (faster-than-decimal) floating point numbers
to calculate, and renders that to other types on demand.
mysql-test/r/func_group.result:
Test that the bug is fixed, and that no unexpected behavior arises from the
changes.
mysql-test/t/func_group.test:
Test that the bug is fixed, and that no unexpected behavior arises from the
changes.
sql/item_sum.cc:
Serg's suggestion: Force all VARIANCE calculations to be done with floating-
point types. It's faster, and the SQL standard says we may implement these
functions any way we want.
Additionally, use a form of variance calculation that is not subject to
catastrophic cancellation.
http://static.flickr.com/108/311308512_5c4e1c0c3d_b.jpg
sql/item_sum.h:
Remove unused members and add a comment describing the recurrence relation.
on large length
Problem: Most (all) of the numeric inputs were being coerced into
int (32 bit) sized variables. Works OK for sane inputs; any input
larger than 2^32 (or 2^31 for signed vars) exihibited predictable
wrapping behavior (up to about 10^18) and then started having really
strange behaviour past that point (since the conversion to 64 bit int
from the DECIMAL type can do weird things on out of range numbers).
Solution: 1) Add many tests. 2) Convert input from (u)long type to
(u)longlong. 3) Do (sometimes multiple) sanity checks on input,
keeping in mind that sometimes a negative longlong is not a negative
longlong (if the unsigned_flag is set). 4) Emulate existing behavior
w/rt negative and "small" out-of-bounds values.
mysql-test/r/func_str.result:
Additional test results for #10963
mysql-test/t/func_str.test:
Additional test results for #10963
sql/item_func.cc:
Used larger type for counting, to avoid truncation.
sql/item_strfunc.cc:
Fix for #10963, including comments and cleaned up logic
The Item_func_mod objects never had maybe_null set, so users had no reason
to expect that they can be NULL, and may therefore deduce wrong results.
Now, set maybe_null.
mysql-test/r/func_test.result:
Verify that the predictions are true.
mysql-test/t/func_test.test:
Verify that the predictions are true.
sql/item_func.cc:
MOD functions may be NULL.
mysql-test/r/mysql.result:
Update test result after commenting out the test
mysql-test/t/mysql.test:
Rewrite the test commands so they don't depend on "echo" behaviour
Comment out one as it's a duplicate
into zippy.cornsilk.net:/home/cmiller/work/mysql/mysql-5.0-maint
mysql-test/r/innodb_mysql.result:
Auto merged
mysql-test/r/view.result:
Auto merged
mysql-test/t/ctype_utf8.test:
Auto merged
mysql-test/t/im_daemon_life_cycle.imtest:
Auto merged
mysql-test/t/innodb_mysql.test:
Auto merged
mysql-test/t/sp-error.test:
Auto merged
mysql-test/t/sp.test:
Auto merged
mysql-test/t/view.test:
Auto merged
sql/ha_innodb.cc:
Auto merged
sql/handler.cc:
Auto merged
sql/item_func.cc:
Auto merged
sql/mysqld.cc:
Auto merged
sql/set_var.cc:
Auto merged
sql/sp.cc:
Auto merged
sql/sql_class.h:
Auto merged
sql/sql_delete.cc:
Auto merged
sql/sql_lex.cc:
Auto merged
sql/sql_lex.h:
Auto merged
sql/sql_parse.cc:
Auto merged
sql/sql_prepare.cc:
Auto merged
sql/sql_select.cc:
Auto merged
sql/sql_table.cc:
Auto merged
sql/sql_trigger.cc:
Auto merged
sql/sql_union.cc:
Auto merged
sql/sql_view.cc:
Auto merged
sql/sql_yacc.yy:
Auto merged
sql/table.cc:
Auto merged
tests/mysql_client_test.c:
Auto merged
mysql-test/mysql-test-run.pl:
Manual merge.
mysql-test/r/ps.result:
Manual merge.
mysql-test/t/ps.test:
Manual merge.
into zippy.cornsilk.net:/home/cmiller/work/mysql/mysql-4.1-maint
configure.in:
Auto merged
mysql-test/t/ps.test:
Auto merged
sql/handler.cc:
Auto merged
sql/sql_delete.cc:
Auto merged
sql/sql_select.cc:
Auto merged
sql/table.cc:
Auto merged
tests/mysql_client_test.c:
Auto merged
myisam/sort.c:
Manual merge.
mysql-test/r/innodb_mysql.result:
Manual merge.
mysql-test/t/innodb_mysql.test:
Manual merge.
mysys/mf_iocache.c:
Manual merge.
into bodhi.local:/opt/local/work/mysql-5.0-runtime-merge
sql/item_sum.cc:
Auto merged
sql/sql_base.cc:
Auto merged
sql/sql_delete.cc:
Auto merged
sql/sql_lex.h:
Auto merged
sql/sql_select.cc:
Auto merged
sql/sql_view.cc:
Auto merged
into bodhi.local:/opt/local/work/mysql-4.1-runtime
mysql-test/r/ps.result:
Auto merged
mysql-test/t/func_gconcat.test:
Auto merged
sql/item_func.cc:
Auto merged
sql/item_func.h:
Auto merged
sql/item_sum.cc:
Auto merged
sql/log_event.cc:
Auto merged
sql/mysql_priv.h:
Auto merged
sql/mysqld.cc:
Auto merged
sql/set_var.cc:
Auto merged
sql/sql_class.h:
Auto merged
sql/sql_delete.cc:
Auto merged
sql/sql_select.cc:
Auto merged
sql/sql_update.cc:
Auto merged
remove timeing dependant part of test
mysql-test/r/ndb_multi.result:
remove timeing dependant part of test
mysql-test/t/ndb_multi.test:
remove timeing dependant part of test
It's not possible to flush the global status variables in 5.0
Update test case so it works by recording the value of handle_rollback
before and compare it to the value after
mysql-test/r/innodb_mysql.result:
Update result file
mysql-test/t/innodb_mysql.test:
It's not possible to reset the global status variables in 5.0 so intead its value
is recorded and compared to the after value.
It should not have changed.
sql/mysqld.cc:
Improve comments
sql/set_var.cc:
Improve comments
sql/sql_class.cc:
Improve comments
sql/sql_class.h:
Improve comments
into perch.ndb.mysql.com:/home/jonas/src/mysql-5.0-ndb
mysql-test/r/ctype_utf8.result:
Auto merged
mysql-test/t/ctype_utf8.test:
Auto merged
sql/sql_base.cc:
Auto merged
sql/sql_delete.cc:
Auto merged
sql/sql_lex.h:
Auto merged
sql/sql_select.cc:
Auto merged
sql/table.cc:
Auto merged
Backport of the fix for bug #8143: A date with value 0 is treated as a NULL value
mysql-test/r/delete.result:
Fix for bug #23412: delete rows with null date field
- test result
mysql-test/t/delete.test:
Fix for bug #23412: delete rows with null date field
- test case
sql/sql_delete.cc:
Fix for bug #23412: delete rows with null date field
- during SELECT queries processing we convert 'date[time]_field is null'
conditions into 'date[time]_field = 0000-00-00[ 00:00:00]' for not null
DATE and DATETIME fields. To be consistent, we have to do the same for DELETE
queries. So we should call remove_eq_conds() in the mysql_delete() as well.
Also it may simplify and speed up DELETE queries execution.
a updatable view.
When there's a VIEW on a base table that have AUTO_INCREMENT column, and
this VIEW doesn't provide an access such column, after INSERT to such
VIEW LAST_INSERT_ID() did not return the value just generated.
This behaviour is intended and correct, because if the VIEW doesn't list
some columns then these columns are effectively hidden from the user,
and so any side effects of inserting default values to them.
However, there was a bug that such statement inserting into a view would
reset LAST_INSERT_ID() instead of leaving it unchanged.
This patch restores the original value of LAST_INSERT_ID() instead of
resetting it to zero.
mysql-test/r/view.result:
Add result for bug#22584: last_insert_id not updated after inserting
a record through a updatable view.
mysql-test/t/view.test:
Add test case for bug#22584: last_insert_id not updated after inserting
a record through a updatable view.
sql/sql_parse.cc:
When we have inserted into a view, and AUTO_INCREMENT column is not
accessed from this view, instead of setting LAST_INSERT_ID to zero set
it to the value it had before this statement was executed.
- Add two null bytes in "buff" variable allowing us to call
get_arg two times also for strings longer than sizeof(buff)
client/mysql.cc:
Make sure there are _two_ zero bytes at the end of buff, to allow 'get_arg' to be
called twice on long strings. The first time it will mark end of string with a
zero and the second time it will "skip ahead" to the first zero, and the find the
second one indicating end of buff.
mysql-test/r/mysql.result:
Update test results
mysql-test/t/mysql.test:
Add tests for "com_connect" function in mysql
Add test reported in bug
into moonlight.intranet:/home/tomash/src/mysql_ab/mysql-5.0-bug18819
mysql-test/r/innodb_mysql.result:
Auto merged
mysql-test/t/innodb_mysql.test:
Auto merged
sql/sql_delete.cc:
Auto merged
If the error happens during DELETE IGNORE, nothing could be send to the
client, thus leaving it frozen expecting the reply.
The problem was that if some error occurred, it wouldn't be reported to
the client because of IGNORE, but neither success would be reported.
MySQL 4.1 would not freeze the client, but will report
ERROR 1105 (HY000): Unknown error
instead, which is also a bug.
The solution is to report success if we are in DELETE IGNORE and some
non-fatal error has happened.
mysql-test/r/innodb_mysql.result:
Add result for bug#18819: DELETE IGNORE hangs on foreign key parent
delete.
mysql-test/t/innodb_mysql.test:
Add test case for bug#18819: DELETE IGNORE hangs on foreign key parent
delete.
sql/sql_delete.cc:
Report success if we have got an error, but we are in DELETE IGNORE, and
the error is not fatal (if it is, it would be reported to the client).
into siva.hindu.god:/usr/home/tim/m/bk/50
configure.in:
Auto merged
mysql-test/r/rename.result:
Auto merged
mysql-test/r/subselect.result:
Auto merged
mysql-test/r/view.result:
Auto merged
mysql-test/t/subselect.test:
Auto merged
mysql-test/t/view.test:
Auto merged
mysys/mf_iocache.c:
Auto merged
sql/item_func.cc:
Auto merged
sql/opt_range.cc:
Auto merged
sql/opt_range.h:
Auto merged
sql/sql_lex.cc:
Auto merged
sql/sql_lex.h:
Auto merged
sql/sql_select.cc:
Auto merged
sql/sql_table.cc:
Auto merged
sql/table.cc:
Auto merged
myisam/sort.c:
Manual merge
mysql-test/r/innodb_mysql.result:
Manual merge
mysql-test/t/innodb_mysql.test:
Manual merge
mysql-test/r/mysqldump-max.result:
Update test result after fix of spelling error
mysql-test/r/rpl_trigger.result:
Update test result after fix of spelling error
mysql-test/r/type_newdecimal.result:
Update test result after fix of spelling error
mysql-test/t/mysqldump-max.test:
Fix spelling error "disable_warnings"
Write shorter "drop table..."
mysql-test/t/rpl_trigger.test:
Fix spelling error, there is no --ignore_warnings
mysql-test/t/type_newdecimal.test:
Fix spelling error
mysql-test/r/ndb_multi.result:
update result file for new behaviour of correcting the 241 error on ha_open
mysql-test/t/ndb_multi.test:
no longer errors on ha_open with 241
To the documentor: The DEFAULT value in SHOW COLUMNS/SELECT FROM I_S.COLUMNS
was truncated to NAME_LEN (which effectively is 64) characters.
mysql-test/r/information_schema.result:
Update result
Because the length of I_S.COLUMNS.COLUMN_DEFAULT was increased to
65535 the column changed it's type from varchar(64) to longtext
The type is longtext if the length is > 65532 (the max varchar len)
mysql-test/t/information_schema.test:
test case for bug#23037: Bug in field "Default" of
query "SHOW COLUMNS FROM table"
Both SHOW COLUMNS and I_S.COLUMNS is TESTED as well the
value returned by both. Using --replace_regex is unfeasible
as the repetition value cannot be > 255.
--replace_regex /A{256}/VALUE/ doesn't work, not to talk about
/A{65532}/VALUE/
/(A{255}){255}/VALUE/ is awfully slow. Thus, simple function with
cursor is used to extract the value.
sql/sql_show.cc:
Default value can have TIMESTAMP, CHAR, VARCHAR, ENUM
VARCHAR can have the longest value, up to 65535. However, because
of table handler limitations the actual limit is 65532 characters,
latin1 charset. However, here is used MAX_FIELD_VARCHARLENGTH macro,
because there could be a storage engine without such limitation.
into bodhi.local:/opt/local/work/mysql-5.0-runtime-merge
mysql-test/t/func_gconcat.test:
Auto merged
sql/item_func.cc:
Auto merged
sql/sql_base.cc:
Auto merged
sql/sql_insert.cc:
Auto merged
sql/sql_lex.cc:
Auto merged
sql/sql_lex.h:
Auto merged
sql/sql_select.cc:
Auto merged
sql/sql_update.cc:
Auto merged
mysql-test/r/view.result:
Manual merge.
mysql-test/t/view.test:
Manual merge.
into dl145s.mysql.com:/data/bk/team_tree_merge/MERGE/mysql-5.0-opt
mysql-test/r/myisam.result:
Auto merged
sql/mysql_priv.h:
Auto merged
sql/sql_base.cc:
Auto merged
sql/sql_select.cc:
Auto merged
This patch reverts a change introduced by Bug 6951, which incorrectly
set thd->abort_on_warning for stored procedures.
As per internal discussions about the SQL_MODE=TRADITIONAL,
the correct behavior is to *not* abort on warnings even inside an INSERT/UPDATE
trigger.
Tests for Stored Procedures, Stored Functions, Triggers involving SQL_MODE
have been included or revised, to reflect the intended behavior.
(reposting approved patch, to work around source control issues, no review needed)
mysql-test/include/sp-vars.inc:
Tests for SQL_MODE='TRADITIONAL'
mysql-test/r/sp-vars.result:
Tests for SQL_MODE='TRADITIONAL'
mysql-test/r/sp.result:
Tests for SQL_MODE='TRADITIONAL'
mysql-test/r/trigger.result:
Tests for SQL_MODE='TRADITIONAL'
mysql-test/t/sp-vars.test:
Tests for SQL_MODE='TRADITIONAL'
mysql-test/t/sp.test:
Tests for SQL_MODE='TRADITIONAL'
mysql-test/t/trigger.test:
Tests for SQL_MODE='TRADITIONAL'
sql/sp_head.cc:
For SQL_MODE='TRADITIONAL',
thd->abort_on_warning should be set only when assigning a *column*
into neptunus.(none):/home/msvensson/mysql/mysql-5.0-maint
client/mysqltest.c:
Auto merged
mysql-test/r/mysqltest.result:
Auto merged
mysql-test/t/mysqltest.test:
Auto merged
into mysql.com:/home/svoj/devel/mysql/engines/mysql-5.0-engines
myisam/sort.c:
Auto merged
mysql-test/r/repair.result:
Auto merged
mysql-test/t/repair.test:
Auto merged
sql/sql_base.cc:
Use local.
into dl145s.mysql.com:/data/bk/team_tree_merge/MERGE/mysql-5.0-opt
mysql-test/r/merge.result:
Auto merged
mysql-test/r/myisam.result:
Auto merged
mysql-test/r/select.result:
Auto merged
mysql-test/r/view.result:
Auto merged
sql/item_cmpfunc.cc:
Auto merged
sql/item_cmpfunc.h:
Auto merged
sql/mysql_priv.h:
Auto merged
sql/opt_range.cc:
Auto merged
sql/opt_range.h:
Auto merged
sql/sql_select.cc:
Auto merged
sql/sql_table.cc:
Auto merged
sql/sql_update.cc:
Auto merged
sql/table.cc:
Auto merged
commands and go directly to result file processing
client/mysqltest.c:
Add command "exit" to mysqltest
mysql-test/r/mysqltest.result:
Add command "exit" to mysqltest
mysql-test/t/mysqltest.test:
Add command "exit" to mysqltest
into mysql.com:/usr/home/ram/work/bug20732/my50-bug20732
sql/opt_range.cc:
Auto merged
sql/opt_range.h:
Auto merged
sql/sql_select.cc:
Auto merged
sql/table.cc:
Auto merged
mysql-test/r/range.result:
SCCS merged
mysql-test/t/range.test:
SCCS merged
