Bug#8367 "low log doesn't gives complete information about prepared
statements"
Implement status variables for prepared statements commands (a port of
the patch by Andrey Hristov).
See details in comments to the changed files.
No test case as there is no way to test slow log/general log in
mysqltest.
mysql-test/r/ps_grant.result:
Now execute is logged with tag 'Execute' (changed result file).
sql/mysql_priv.h:
- remove obsolete macro.
- add declarations for new status variables.
- export function log_slow_statement, which now is used in sql_prepare.cc
sql/mysqld.cc:
Add status variables for prepared statements API: now we record
mysql_stmt_close, mysql_stmt_reset, mysql_stmt_prepare, mysql_stmt_execute
mysql_stmt_send_long_data, PREPARE, EXECUTE, DEALLOCATE.
sql/sql_parse.cc:
- account DEALLOCATE prepare as a Com_stmt_close command (close of a
prepared statement).
sql/sql_prepare.cc:
- fix a bug in SQL syntax for prepared statements + logging:
if we use --log and EXECUTE stmt USING @no_such_variable;, the
server crashed because the old code assumed that the variable
returned by get_var_with_binlog is never NULL.
- account statistics for
mysql_stmt_{prepare,execute,close,reset,send_long_data} in
Com_stmt_{prepare,execute,close,reset,send_long_data} correspondingly.
- log slow statements into the slow log early, when thd->query
points to a valid (with expanded placeholder values) query.
The previous version was logging it in sql_parse, when thd->query
is empty. Prevent the server from logging the statement twice by
setting thd->enable_slow_log= FALSE.
- now in case of EXECUTE stmt in SQL syntax for prepared statements the
general log gets two queries, e.g.
Query EXECUTE stmt USING @a, @b, @c
Execute INSERT INTO t1 VALUES (1, 2, 3)
This makes the behavior consistent with PREPARE command, which
also logs the statement twice.
Bug#8367 "low log doesn't gives complete information about prepared
statements"
Implement status variables for prepared statements commands (a port of
the patch by Andrey Hristov).
See details in comments to the changed files.
No test case as there is no way to test slow log/general log in
mysqltest.
CURSOR_TYPE_READ_ONLY". The bug was that we (me) don't perform proper
cleanups of the prepared statement when done fetching from a cursor.
Another patch.
sql/mysql_priv.h:
Rename reset_stmt_for_execute to init_stmt_before_use (to correspond to
cleanup_stmt_and_thd_after_use).
sql/sp_head.cc:
Rename.
sql/sql_prepare.cc:
Move common cleanup code to a cleanup function, call it when we close
a cursor.
sql/sql_select.cc:
Cleanup.
sql/sql_select.h:
No need for init_thd, this code has been inlined in Cursor::open.
tests/mysql_client_test.c:
Add a test case for Bug#10729 "mysql_stmt_attr_set CURSOR_TYPE_READ_ONLY"
(problem reusing a prepared statemnt if there was a cursor)
CURSOR_TYPE_READ_ONLY". The bug was that we (me) don't perform proper
cleanups of the prepared statement when done fetching from a cursor.
Another patch.
In order to make multi-delete SP friendly we need to have all table
locks for the elements of main statement table list properly set
at the end of parsing.
Also performed small cleanup: We don't need relink_tables_for_multidelete()
any longer since the only case now when TABLE_LIST::correspondent_table
is non-zero are tables in auxilary table list of multi-delete and these
tables are handled specially in mysql_multi_delete_prepare().
mysql-test/r/sp-threads.result:
Added test case for bug #11158 "Can't perform multi-delete in stored
procedure".
mysql-test/t/sp-threads.test:
Added test case for bug #11158 "Can't perform multi-delete in stored
procedure".
sql/mysql_priv.h:
- Removed third argument from the declaration of multi_delete_precheck()
as nowdays we calculate number of tables in multi-delete from which
we are going to delete rows right at the end of statement parsing.
- Introduced definition of multi_delete_set_locks_and_link_aux_tables()
which is responsible for propagation of proper table locks from
multi-delete's auxilary table list to the main list and binding
corresponding tables in these two lists.
sql/sql_base.cc:
Removed relink_tables_for_multidelete() routine and its invocations.
We don't need them in 5.0 since the only case now when
TABLE_LIST::correspondent_table is non-zero are tables in auxilary table
list of multi-delete and these tables are handled specially in
mysql_multi_delete_prepare().
sql/sql_lex.h:
LEX::table_count
Added description of new role of this LEX member for multi-delete.
Now for this statement we store number of tables from which we should
delete records there.
sql/sql_parse.cc:
multi_delete_precheck():
Moved code which is responsible for iterating through auxilary table
list and binding its elements with corresponding elements of main
table list, and properly updating locks in it to separate function -
multi_delete_set_locks_and_link_aux_tables(). This is because in order
to make multi-delete SP friendly we need to have all locks set properly
at the end of statement parsing. So we are introducing new function
which will be called from parser.
We also calculate number of tables from which we are going to perform
deletions there and store this number for later usage in
LEX::table_count.
Also removed some no longer needed code.
sql/sql_prepare.cc:
mysql_test_multidelete():
Now multi_delete_precheck() takes only two arguments, so we don't
need to pass fake third parameter.
sql/sql_yacc.yy:
delete:
In order to make multi-delete SP friendly we need to have all table
locks for the elements of main statement table list properly set
at the end of parsing.
In order to make multi-delete SP friendly we need to have all table
locks for the elements of main statement table list properly set
at the end of parsing.
Also performed small cleanup: We don't need relink_tables_for_multidelete()
any longer since the only case now when TABLE_LIST::correspondent_table
is non-zero are tables in auxilary table list of multi-delete and these
tables are handled specially in mysql_multi_delete_prepare().
sql/sql_prepare.cc:
Cleanup mysql_test_* family of calls: now we don't send error message
from mysql_stmt_prepare, so no need to support -1 return code
(meaning error is not sent to client) in these functions.
Move unit->cleanup() to mysql_stmt_prepare() as it's done in most
of the mysql_test_ functions, and is a no-op for those which don't
call unit->prepare(). This should make fixing of Bug#10729 (cursors)
easier.
don't evaluate subqueries during statement prepare, even if they
are not correlated.
With post-review fixes.
sql/mysql_priv.h:
Add UNCACHEABLE_PREPARE to mark subqueries as non-constant in
mysql_stmt_prepare
sql/sql_lex.cc:
Add a missing assert: noone can call unit::set_limit from
mysql_stmt_prepare.
sql/sql_lex.h:
Comment fixed.
sql/sql_parse.cc:
Mark new SELECT_LEXes as uncacheable if they created during
statement prepare.
sql/sql_prepare.cc:
Switch off the uncacheable flag when prepare is done.
Fixed bug in mysql_stmt_fetch() when retrieving rows to return
mysql-test/r/ps.result:
Fix to not get warnings if mysql_client_test.test fails
mysql-test/t/index_merge_ror.test:
Proper fix for 64 bit intel (which gives uses another, equal good index)
mysql-test/t/ps.test:
Fix to not get warnings if mysql_client_test.test fails
sql-common/client.c:
More debugging
sql/sql_prepare.cc:
Fixed bug in mysql_stmt_fetch() when retrieving rows to return
sql/sql_select.cc:
More debugging
tests/mysql_client_test.c:
More debugging
(crash on attempt to re-execute a statement with an open cursor) +
post-review fixes.
include/errmsg.h:
Add a special error message when we attempt to mysql_stmt_fetch
from a statement which has no result set.
libmysql/errmsg.c:
Error message text for CR_NO_RESULT_SET
libmysql/libmysql.c:
Move the code which frees result sets on client and closes the cursor
on server, resets long data state on client and server.
This makes one function out of two (mysql_stmt_reset and
mysql_stmt_free_result), thus aggregating all related reset work
in one place.
sql-common/client.c:
Fix one place where we flushed the pending result set of a statement,
but didn't set unbuffered_fetch_cancelled flag.
sql/share/errmsg.txt:
Fix format of ER_UNKNOWN_STMT_HANDLER error message (needs to
be fixed separately in 4.1). Add two new error messages
for the case when we fetch from when there is no cursor
and for the case when we attempt to execute a statement while there is
a cursor.
sql/sql_prepare.cc:
Return error when we fetch while there is no open cursor and
when we call execute while there is a pending cursor.
Fix mysql_stmt_reset to close the open cursor if there is any.
sql/sql_select.cc:
free_items and free_root moved to Cursor::close().
sql/sql_select.h:
A comment added.
tests/mysql_client_test.c:
A test case for Bug#9478, test the case of mysql_stmt_reset
called for client-side cached result set and for the case with open cursor.
All strcpy replaced with strmov (review request).
sql/sql_base.cc:
Auto merged
sql/sql_insert.cc:
Auto merged
sql/sql_parse.cc:
Auto merged
sql/sql_prepare.cc:
Auto merged
sql/sql_select.cc:
Auto merged
sql/sql_view.cc:
Auto merged
sql/table.cc:
Auto merged
sql/table.h:
Auto merged
Fixed bug #8528.
Representation for single-table views was made similar to
representation for multi-table views.
view.test:
Added test case for bug #8528.
view.result:
Added test case for bug #8528. Fixed other test cases.
mysql-test/r/view.result:
Added test case for bug #8528. Fixed other test cases.
mysql-test/t/view.test:
Added test case for bug #8528.
sql/sql_base.cc:
Fixed bug #8528.
Representation for single-table views was made similar to
representation for multi-table views.
sql/sql_delete.cc:
Fixed bug #8528.
Representation for single-table views was made similar to
representation for multi-table views.
sql/sql_insert.cc:
Fixed bug #8528.
Representation for single-table views was made similar to
representation for multi-table views.
sql/sql_parse.cc:
Fixed bug #8528.
Representation for single-table views was made similar to
representation for multi-table views.
sql/sql_prepare.cc:
Fixed bug #8528.
Representation for single-table views was made similar to
representation for multi-table views.
sql/sql_select.cc:
Fixed bug #8528.
Representation for single-table views was made similar to
representation for multi-table views.
sql/sql_update.cc:
Fixed bug #8528.
Representation for single-table views was made similar to
representation for multi-table views.
sql/sql_view.cc:
Fixed bug #8528.
Representation for single-table views was made similar to
representation for multi-table views.
sql/table.cc:
Fixed bug #8528.
Representation for single-table views was made similar to
representation for multi-table views.
sql/table.h:
Fixed bug #8528.
Representation for single-table views was made similar to
representation for multi-table views.
Fixed bug #8528.
Representation for single-table views was made similar to
representation for multi-table views.
view.test:
Added test case for bug #8528.
view.result:
Added test case for bug #8528. Fixed other test cases.
with cursor". The patch refactors do_select/sub_select
functions, which implement the nested loop algorithm, and reuses them to
fetch rows for cursors as well.
Pushing with view.test failing (--ps-protocol).
sql/sql_prepare.cc:
Cursor::fetch() now returns void
sql/sql_select.cc:
A fix for Bug#9520 "SELECT DISTINCT crashes server with cursor":
* rename sub_select returns codes to be able to track down what's going
on in which case.
* move record processing and outer join record processing to a separate
function, out of sub_select read-record loop.
* use generalized sub_select() nested loop function for
cursors instead of own loop implementation used in Cursor::fetch() before
sql/sql_select.h:
Replace all return values of sub_select family with enum.
Add JOIN::resume_nested_loop flag to indicate we are restarting the nested loop
for execution of next chunk of cursor's rows.
tests/mysql_client_test.c:
A test case for Bug#9520 "SELECT DISTINCT crashes server with cursor"
with cursor". The patch refactors do_select/sub_select
functions, which implement the nested loop algorithm, and reuses them to
fetch rows for cursors as well.
Pushing with view.test failing (--ps-protocol).
New more SP-locking friendly approach to handling locks in multi-update.
Now we mark all tables of multi-update as needing write lock at parsing
stage and if possible downgrade lock at execution stage (For its work
SP-locking mechanism needs to know all lock types right after parsing
stage).
mysql-test/r/sp-threads.result:
Added test for bug #9486 "Can't perform multi-update in stored procedure".
mysql-test/t/sp-threads.test:
Added test for bug #9486 "Can't perform multi-update in stored procedure".
sql/sp_head.cc:
SP_TABLE, sp_head::merge_table_list()/add_used_tables_to_table_list():
Since some queries during their execution (e.g. multi-update)
may change type of lock for some of their tables and thus change
lock_type member for some of elements of table list, we should
store type of lock in SP_TABLE struct explicitly instead of using
lock_type member of TABLE_LIST object pointed by SP_TABLE::table.
sql/sql_lex.h:
Removed no longer used LEX::multi_lock_option member.
sql/sql_prepare.cc:
mysql_test_update():
We don't need to bother about LEX::multi_lock_option if we convert
multi-update to update anymore. Since nowdays multi-update uses
TABLE_LIST::lock_type for specifying lock level of updated tables
instead of LEX::multi_lock_option.
sql/sql_update.cc:
mysql_update()/mysql_multi_update_prepare():
Now we mark all tables of multi-update as needing write lock at parsing
stage and if possible downgrade lock at execution stage. Old approach
(don't set lock type until execution stage) was not working well with
SP-locking (For its work SP-locking mechanism needs to know all lock
types right after parsing stage).
mysql_multi_update():
We should return FALSE if no error occurs.
sql/sql_yacc.yy:
update:
Now we mark all tables of multi-update as needing write lock at parsing
stage and if possible downgrade lock at execution stage. Old approach
(don't set lock type until execution stage) was not working well with
SP-locking (For its work SP-locking mechanism needs to know all lock
types right after parsing stage).
New more SP-locking friendly approach to handling locks in multi-update.
Now we mark all tables of multi-update as needing write lock at parsing
stage and if possible downgrade lock at execution stage (For its work
SP-locking mechanism needs to know all lock types right after parsing
stage).
mysql-test/r/ps.result:
Test results: adding tests for CALL statement in prepared mode.
mysql-test/t/ps.test:
Adding tests for CALL statement in prepared mode.
into sanja.is.com.ua:/home/bell/mysql/bk/work-multi-5.0
mysql-test/r/view.result:
Auto merged
sql/sql_derived.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
mysql-test/r/group_by.result:
Auto merged
mysql-test/r/metadata.result:
Auto merged
mysql-test/r/union.result:
Auto merged
mysql-test/t/union.test:
Auto merged
sql/item.h:
Auto merged
sql/sql_parse.cc:
Auto merged
sql/sql_select.cc:
Auto merged
support-files/mysql.server.sh:
Auto merged
(BUG##9398, BUG#8703)
fixed wrong join view detection in multi-delete which lead to server crash
mysql-test/r/lowercase_view.result:
added new tests of updation and selection from the same table
mysql-test/r/view.result:
added new tests of updation and selection from the same table
added test of multidelete command over join view which lead to server crash
test suite from bugs #9398 and #8703
mysql-test/t/lowercase_view.test:
added new tests of updation and selection from the same table
mysql-test/t/view.test:
added new tests of updation and selection from the same table
added test of multidelete command over join view which lead to server crash
test suite from bugs #9398 and #8703
sql/sql_base.cc:
changed procedure of finding tables
sql/sql_class.cc:
added derived table procession detection
sql/sql_class.h:
added derived table procession detection
sql/sql_delete.cc:
fixed detection of selection from table which update for multidelete
sql/sql_derived.cc:
added derived table procession detection
sql/sql_lex.cc:
added detection os SELECTs processed inside derived tables
removed old mechanism of multidelete/multiupdate table duplication detection (which can't work with views)
sql/sql_lex.h:
added detection os SELECTs processed inside derived tables
removed old mechanism of multidelete/multiupdate table duplication detection (which can't work with views)
sql/sql_parse.cc:
removed wrong test of join view (for multidelete in can be not only first table)
sql/sql_prepare.cc:
added detection os SELECTs processed inside derived tables (reset it for reusing in PS/SP)
sql/sql_select.cc:
added detection os SELECTs processed inside derived tables
sql/sql_update.cc:
fixed detection of selection from table which update for multiupdate
