into polly.(none):/home/kaa/src/maint/mysql-5.1-maint
mysql-test/r/insert.result:
Auto merged
mysql-test/t/insert.test:
Auto merged
mysql-test/t/type_float.test:
Auto merged
mysql-test/t/variables.test:
Auto merged
sql/field.cc:
Auto merged
into polly.(none):/home/kaa/src/maint/mysql-5.1-maint
mysql-test/t/type_float.test:
Auto merged
mysql-test/t/variables.test:
Auto merged
sql/field.cc:
Auto merged
mysql-test/r/insert.result:
SCCS merged
mysql-test/t/insert.test:
SCCS merged
into polly.(none):/home/kaa/src/maint/mysql-5.0-maint
mysql-test/t/type_float.test:
Auto merged
mysql-test/t/variables.test:
Auto merged
sql/field.cc:
Auto merged
mysql-test/r/insert.result:
SCCS merged
mysql-test/t/insert.test:
SCCS merged
into polly.(none):/home/kaa/src/maint/bug26788/my51-bug26788
mysql-test/r/insert.result:
Auto merged
mysql-test/t/cast.test:
Auto merged
mysql-test/t/insert.test:
Auto merged
mysql-test/t/type_float.test:
Auto merged
mysql-test/t/variables.test:
Auto merged
sql/field.cc:
SCCS merged
numbers into char fields" and bug #12860 "Difference in zero padding of
exponent between Unix and Windows"
Rewrote the code that determines what 'precision' argument should be
passed to sprintf() to fit the string representation of the input number
into the field.
We get finer control over conversion by pre-calculating the exponent, so
we are able to determine which conversion format, 'e' or 'f', will be
used by sprintf().
We also remove the leading zero from the exponent on Windows to make it
compatible with the sprintf() output on other platforms.
mysql-test/r/insert.result:
Added test cases for bug #26788 and bug #31152.
mysql-test/t/cast.test:
Removed --replace_result, since the result is now correct on Windows.
mysql-test/t/insert.test:
Added test cases for bug #26788 and bug #31152.
mysql-test/t/type_float.test:
Removed --replace_result, since the result is now correct on Windows.
mysql-test/t/variables.test:
Removed --replace_result, since the result is now correct on Windows.
sql/field.cc:
Rewrote the code that determines what 'precision' argument should be
passed to sprintf() to fit the string representation of the input number
into the field.
We get finer control over conversion by pre-calculating the exponent, so
we are able to determine which conversion format, 'e' or 'f', will be
used by sprintf().
into mysql.com:/home/hf/work/mrg/my51-mrg
sql/mysqld.cc:
Auto merged
sql/protocol.cc:
Auto merged
storage/federated/ha_federated.cc:
Auto merged
tests/mysql_client_test.c:
Auto merged
include/mysql_com.h:
merging
libmysqld/lib_sql.cc:
merging
mysql-test/r/federated.result:
merging
mysql-test/t/federated.test:
merging
storage/federated/ha_federated.h:
mergin
PS-protocol data is stored in different format - the MYSQL_RECORDS->data
contains the link to the record content, not to array of the links to
the field's contents. So we have to handle it separately for
embedded-server query cache.
libmysqld/emb_qcache.cc:
Bug #30430 crash:./mtr --embedded-server --ps-protocol cache_innodb func_misc...
ps_protocol data now stored in it's particular way.
libmysqld/emb_qcache.h:
Bug #30430 crash:./mtr --embedded-server --ps-protocol cache_innodb func_misc...
Querycache_stream::load_char/store_char ->
load_uchar/store_uchar
mysql-test/r/func_misc.result:
Bug #30430 crash:./mtr --embedded-server --ps-protocol cache_innodb func_misc...
test result
mysql-test/t/func_misc.test:
Bug #30430 crash:./mtr --embedded-server --ps-protocol cache_innodb func_misc...
the test uses t2 table, so let's make sure it doesn't exist befor the start
into four.local.lan:/work/merge/mysql-5.1-dev
mysql-test/suite/rpl/r/rpl_000015.result:
Auto merged
mysql-test/suite/rpl/t/rpl_000015.test:
Auto merged
into four.local.lan:/work/merge/mysql-5.0-dev
BitKeeper/deleted/.del-disabled.def:
SCCS merged
mysql-test/r/rpl000015.result:
Bug does not apply to 5.0 and up
mysql-test/t/rpl000015.test:
Bug does not apply to 5.0 and up.
Bug#31030 rpl000015.test fails if $MYSQL_TCP_PORT != 3306
Note:
This bug does not occur in MySQL 5.0 and up, because
ChangeSet 1.2328.2.1 2006/11/27 for MySQL 5.0 prevents this.
The 5.0 fix uses the environment variable DEFAULT_MASTER_PORT
which is set by mysql-test-run.pl.
mysql-test-run.pl in 4.1 does not set this variable.
There are two alternatives:
1) Backport the 5.0 fix for this test including modifications
to mysql-test-run.pl and mysql-test-run-shell.
This is a not acceptable impact on an old MySQL version.
2) Fix the problem different than in 5.0 like in the current
ChangeSet + do not apply these changes when upmerging to 5.0
mysql-test/r/rpl000015.result:
Updated result
mysql-test/t/disabled.def:
Enable rpl000015
mysql-test/t/rpl000015.test:
Unify the MASTER_PORT number
client library only sets mysql->insert_id when query returned
no recordset. So the embedded library should behave the same way
libmysqld/lib_sql.cc:
Bug #26921 Problem in mysql_insert_id() Embedded C API function.
only set 'affected_rows' and 'insert_id' fields when query
didn't return a recordset
tests/mysql_client_test.c:
Bug #26921 Problem in mysql_insert_id() Embedded C API function.
testcase added
removed unnecessary privilege checks for I_S schema
mysql-test/r/information_schema.result:
test result
mysql-test/t/information_schema.test:
test case
sql/events.cc:
There is no events in I_S so we don't need to execute check_access here.
sql/sql_parse.cc:
removed unnecessary check
1. A bad error message was given when a MERGE table with an
InnoDB child table was tried to use.
2. After selecting from a correct MERGE table and then altering
one of the children to InnoDB, incorrect results were returned.
These bugs have been fixed with the patch for bug 26379 (Combination
of FLUSH TABLE and REPAIR TABLE corrupts a MERGE table).
For verification, I added the test case from the bug report.
mysql-test/r/merge_innodb.result:
Bug#30491 - MERGE doesn't report error when one table is Innodb
Added test result.
mysql-test/t/merge_innodb.test:
Bug#30491 - MERGE doesn't report error when one table is Innodb
Added test case.
The problem was that THD::killed was reset after a command was
read from the socket, but before it was actually handled. That lead
to a race: if another KILL statement was issued for this connection
in the middle of reading from the socket and processing a command,
THD::killed state would be cleaned.
The fix is to move this cleanup into net_send_error() function.
A sample test case exists in binlog_killed.test:
- connection 1: start a new transaction on table t1;
- connection 2: send query to the server (w/o waiting for the
result) to update data in table t1 -- this query will be blocked
since there is unfinished transaction;
- connection 1: kill query in connection 2 and finish the transaction;
- connection 2: get result of the previous query -- it should be
the "query-killed" error.
This test however contains race condition, which can not be fixed
with the current protocol: there is no way to guarantee, that the
server will receive and start processing the query in connection 2
(which is intended to get blocked) before the KILL command (sent in
the connection 1) will arrive. In other words, there is no way to
ensure that the following sequence will not happen:
- connection 1: start a new transaction on table t1;
- connection 1: kill query in connection 2 and finish the transaction;
- connection 2: send query to the server (w/o waiting for the
result) to update data in table t1 -- this query will be blocked
since there is unfinished transaction;
- connection 2: get result of the previous query -- the query will
succeed.
So, there is no test case for this bug, since it's impossible
to write a reliable test case under the current circumstances.
sql/protocol.cc:
Move thd->killed cleanup from dispatch_command() to net_send_error().
sql/sql_parse.cc:
Move thd->killed cleanup from dispatch_command() to net_send_error().
filesort() uses file->estimate_rows_upper_bound() call to allocate
internal buffers. If this function returns a value smaller than
a number of row that will be returned later in find_all_keys(),
that can cause server crash.
Fixed by implementing ha_federated::estimate_rows_upper_bound() to
return maximum possible number of rows.
Present estimation for FEDERATED always returns 0 if the linked to the VIEW.
mysql-test/r/federated.result:
Bug #32374 crash with filesort when selecting from federated table and view.
test result
mysql-test/t/federated.test:
Bug #32374 crash with filesort when selecting from federated table and view.
test case
sql/ha_federated.cc:
Bug #32374 crash with filesort when selecting from federated table and view.
ha_federated::estimate_rows_upper_bound() implemented
sql/ha_federated.h:
Bug #32374 crash with filesort when selecting from federated table and view.
ha_federated::estimate_rows_upper_bound() interface
Parser rejects valid INTERVAL() expressions when associated with
arithmetic operators. The problem is the way in which the expression
and interval grammar rules were organized caused shift/reduce conflicts.
The solution is to tweak the interval rules to avoid shift/reduce
conflicts by removing the broken interval_expr rule and explicitly
specify it's content where necessary.
Original fix by Davi Arnaut, revised and improved rules by Marc Alff
mysql-test/r/parser.result:
Add test case result for Bug#22312
mysql-test/t/parser.test:
Add test case for Bug#22312
sql/sql_yacc.yy:
Resolve shift/reduce conflicts by reorganizing the interval
expression rules.
Reverted log tables alteration.
scripts/mysql_system_tables_fix.sql:
Altering log table is wrong, because
- it is not possible to alter log table when it is in use;
- log table may use MyISAM engine, which perfectly supports NULLs.
Reverted log tables alteration.
The following clarification should be made in The Manual:
Standard SQL is quite clear that, if new columns are added
to a table after a view on that table is created with
"select *", the new columns will not become part of the view.
In all cases, the view definition (view structure) is frozen
at CREATE time, so changes to the underlying tables do not
affect the view structure.
mysql-test/r/view.result:
Update result file.
mysql-test/t/view.test:
Add a test case for BUG#26676: VIEW using old table schema in a session.
Default values of variables were not subject to upper/lower bounds
and step, while setting variables was. Bounds and step are also
applied to defaults now; defaults are corrected quietly, values
given by the user are corrected, and a correction-warning is thrown
as needed. Lastly, very large values could wrap around, starting
from 0 again. They are bounded at the maximum value for the
respective data-type now if no lower maximum is specified in the
variable's definition.
client/mysql.cc:
correct maxima in options array
client/mysqltest.c:
adjust minimum for "sleep" option so default value is no longer
out of bounds.
include/m_string.h:
ullstr() - the unsigned brother of llstr()
include/my_getopt.h:
Flag if we bounded the value (that is, correct anything aside from
making value a multiple of block-size)
mysql-test/r/delayed.result:
We throw a warning now when we adjust out of range parameters.
mysql-test/r/index_merge.result:
We throw a warning now when we adjust out of range parameters.
mysql-test/r/innodb.result:
We throw a warning now when we adjust out of range parameters.
mysql-test/r/innodb_mysql.result:
We throw a warning now when we adjust out of range parameters.
mysql-test/r/key_cache.result:
We throw a warning now when we adjust out of range parameters.
mysql-test/r/packet.result:
We throw a warning now when we adjust out of range parameters.
mysql-test/r/ps.result:
We throw a warning now when we adjust out of range parameters.
mysql-test/r/subselect.result:
We throw a warning now when we adjust out of range parameters.
mysql-test/r/type_bit.result:
We throw a warning now when we adjust out of range parameters.
mysql-test/r/type_bit_innodb.result:
We throw a warning now when we adjust out of range parameters.
mysql-test/r/variables.result:
correct results: bounds and step apply to variables' default values, too
mysql-test/t/variables.test:
correct results: bounds and step apply to variables' default values, too
mysys/my_getopt.c:
- apply bounds/step to default values of variables (based on work by serg)
- print complaints about incorrect values for variables (truncation etc.,
by requestion of consulting)
- if no lower maximum is specified in variable definition, bound unsigned
values at their maximum to prevent wrap-around
- some calls to error_reporter had a \n, some didn't. remove \n from calls,
let reporter-function handle it, so the default reporter behaves like that
in mysqld
sql/mysql_priv.h:
correct RANGE_ALLOC_BLOCK_SIZE (cleared with monty)
sql/mysqld.cc:
correct maxima to correct data-type.
correct minima where higher than default.
correct range-alloc-block-size.
correct inno variables so GET_* corresponds to actual variable's type.
sql/set_var.cc:
When the new value for a variable is out of bounds, we'll send the
client a warning (but not if the value was simply not a multiple of
'blocksize'). sys_var_thd_ulong had this, sys_var_long_ptr_global
didn't; broken out and streamlined to avoid duplication of code.
strings/llstr.c:
ullstr() - the unsigned brother of llstr()
Fix is to remove any references to the current hostname when running
mysql_install_db --cross-bootstrap. (The dist-hook make target makes
this call, and the resulting data directory is included in the source
distribution as win/data/*.)
Also, a few other clean-ups to mysql_install_db while there.
Makefile.am:
Adapt to clean-up in mysql_install_db (--windows becomes --cross-bootstrap)
scripts/mysql_install_db.sh:
Filter out references to the current hostname when performing
a cross-bootstrap installation by removing any lines which
contain the string "@current_hostname".
Deprecate the old --windows option; use --cross-bootstrap
instead, since it more accurately reflects the purpose.
Other clean-up: the wrong syntax was being used to test the
exit status of mysqld --bootstrap. It mostly worked, as long
as mysqld succeeded. However, it was not robust.
scripts/mysql_system_tables_data.sql:
Rename local @hostname variable to @current_hostname, which is a more
unique label to search on. mysql_install_db now filters out all
lines which include "@current_hostname" during a --cross-bootstrap
installation.
Added 64 bit Mac OS X hard coded settings, for universal binaries
include/my_global.h:
Added 64 bit Mac OS X hard coded settings, for universal binaries
