This cut No 7 should finish the part of fixing the parsing of the events :
- Event_timed is no more used during parsing. Less problems because it has
a mutex. Event_parse_data class is used during parsing. It is suited only
for this purpose. It's pretty lightweight
- Late checking of data from parsing is being performed. This should solve
the problems of nested events in SP or other events (for the situation
of no nested bodies). Before if an ALTER EVENT was in a SP, then when the
SP was compiled, and not executed, the actual init_xxx methods of Event_timed
were called, which is wrong.
- It could be a side effect of using a specialized class, but test events_stress is
now 25% quicker.
Cut No8 will start splitting Event_scheduler into 2 parts, the QUEUE will be moved
to Event_queue.
mysql-test/r/events.result:
update result
mysql-test/t/events.test:
disabled is actually wrong, should be disable, but because of the early
checking it was never parsed.
sql/event_data_objects.cc:
move add init_xxx methods from Event_timed to Event_parse_data
Event_parse data does not need definer_user and definer_host
in Event_timed::compile() do not use lex.et, well there is no more lex.et :)
sql/event_data_objects.h:
move parsing responsibilities from Event_timed to Event_parse_data
sql/event_db_repository.cc:
No more Event_timed comes from parsing but Event_parse_data
The initialization of Item*-s from parsing is done late, and not
during the actual parsing. This is the right way to go because
if an ALTER EVENT is inside a SP or CREATE EVENT it should not be
executed (initialized) during parsing, as it was done.
sql/event_db_repository.h:
No more Event_timed comes from parsing but Event_parse_data
The initialization of Item*-s from parsing is done late, and not
during the actual parsing. This is the right way to go because
if an ALTER EVENT is inside a SP or CREATE EVENT it should not be
executed (initialized) during parsing, as it was done.
sql/event_scheduler.cc:
No more Event_timed comes from parsing but Event_parse_data
The initialization of Item*-s from parsing is done late, and not
during the actual parsing. This is the right way to go because
if an ALTER EVENT is inside a SP or CREATE EVENT it should not be
executed (initialized) during parsing, as it was done.
sql/event_scheduler.h:
No more Event_timed comes from parsing but Event_parse_data
The initialization of Item*-s from parsing is done late, and not
during the actual parsing. This is the right way to go because
if an ALTER EVENT is inside a SP or CREATE EVENT it should not be
executed (initialized) during parsing, as it was done.
sql/events.cc:
No more Event_timed comes from parsing but Event_parse_data
The initialization of Item*-s from parsing is done late, and not
during the actual parsing. This is the right way to go because
if an ALTER EVENT is inside a SP or CREATE EVENT it should not be
executed (initialized) during parsing, as it was done.
sql/events.h:
No more Event_timed comes from parsing but Event_parse_data
The initialization of Item*-s from parsing is done late, and not
during the actual parsing. This is the right way to go because
if an ALTER EVENT is inside a SP or CREATE EVENT it should not be
executed (initialized) during parsing, as it was done.
sql/sql_lex.cc:
lex->et_compile_phase and lex->et are no more.
Use lex->event_parse_data
sql/sql_lex.h:
lex->et_compile_phase and lex->et are no more.
Use lex->event_parse_data
sql/sql_parse.cc:
lex->et_compile_phase and lex->et are no more.
Use lex->event_parse_data
ACL checks were moved inside the Events subsystem.
Also ending of the transaction is performed only just
before doing disk operation. Therefore only when needed.
sql/sql_yacc.yy:
lex->et and lex->et_parse_phase are no more
Use the specialized for parsing Event_parse_data
Cut number 6. Move code from sql_show.cc to event_db_repository.cc
that more belongs to the latter.
sql/event_db_repository.cc:
move code that works with mysql.event from sql_show.cc to
event_db_repository.cc . Route through Event_db_repository's interface
which is proxied by class Events. The code relies on a function from
sql_show.cc which does the actual storage in the schema table. I think
it's better to leave the function there because the structure of
I_S.EVENTS is defined in sql_show.cc
sql/event_db_repository.h:
I_S / SHOW EVENTS handling hooks
sql/event_scheduler.cc:
use the pointer to db_repository which Event_scheduler already has
sql/events.cc:
Put a comment to get_instance
sql/events.h:
callback for I_S (sql_show.cc)
sql/sql_show.cc:
move code that belongs more to Event_db_repository than to here.
Use a callback of class Events. Only 1 function is left here, because
it copies data into the actual rows of I_S.EVENTS and belongs to this file.
sql/sql_show.h:
export this function will be called from event_db_repository.cc
5th cut, moved DB related code to Event_db_repository and
updated accordingly the remanining code.
Moved change/restore_security_context() to class THD
Removed events_priv.h
Next step is to reorganize create/update_event() and parsing for them.
But probably some other refactoring could be done in the meanwhile.
The changes so far pass the test suite.
BitKeeper/deleted/.del-events_priv.h~2e8bce2cf35997df:
Delete: sql/events_priv.h
sql/Makefile.am:
events_priv.h is no more
sql/event_data_objects.cc:
reorganize events code
sql/event_data_objects.h:
reorganize events code
sql/event_db_repository.cc:
reorganize events code
sql/event_db_repository.h:
reorganize events code
sql/event_scheduler.cc:
reorganize events code
sql/event_scheduler.h:
reorganize events code
sql/events.cc:
reorganize events code
sql/events.h:
reorganize events code
sql/mysqld.cc:
reorganize events code
sql/set_var.cc:
reorganize events code
sql/sql_class.cc:
add ::change_security_context() and restore_security_context()
sql/sql_class.h:
add ::change_security_context() and restore_security_context()
sql/sql_db.cc:
reorganize Events code
sql/sql_parse.cc:
reorganize Events code
sql/sql_show.cc:
reorganize Events code
Third cut to simplify parsing phase. Now DROP EVENT works.
Overloaded few functions to be able to use either sp_name or pass two LEX_STRINGs
instead of a Event_timed pointer. This is transitional and eventually the old
functions will be removed. For now DROP EVENT also works, does not need anymore
a parsing object (Event_timed) and definer initialization because everyone who
has EVENT_ACL can drop events, and this is checked on execution time in sql_parse.cc
from the security context, as it should be.
sql/event_data_objects.cc:
overload few functions
sql/event_scheduler.cc:
Event_scheduler::drop_event() actually does not need Event_timed object
but just an identifier, hence pass only sp_name.
Overloaded Event_scheduler::find_event() to work with sp_name object. Eventually
the old version will be removed. This is being done as transitional step to
be able to test frequently code.
sql/event_scheduler.h:
Event_scheduler::drop_event() actually does not need Event_timed object
but just an identifier, hence pass only sp_name.
Overloaded Event_scheduler::find_event() to work with sp_name object. Eventually
the old version will be removed. This is being done as transitional step to
be able to test frequently code.
sql/events.cc:
Change db_drop_event() not to use Event_timed, either coming from parsing
or from Event_timed::drop, but use LEX_STRINGs. sp_name is not convinient
because in Event_timed::drop a temporary object has to be created. Hence,
dereference the sp_name in Events::drop_event() and pass the LEX_STRINGs.
sql/events.h:
Change db_drop_event() not to use Event_timed, either coming from parsing
or from Event_timed::drop, but use LEX_STRINGs. sp_name is not convinient
because in Event_timed::drop a temporary object has to be created. Hence,
dereference the sp_name in Events::drop_event() and pass the LEX_STRINGs.
sql/events_priv.h:
Change db_drop_event() not to use Event_timed, either coming from parsing
or from Event_timed::drop, but use LEX_STRINGs. sp_name is not convinient
because in Event_timed::drop a temporary object has to be created. Hence,
dereference the sp_name in Events::drop_event() and pass the LEX_STRINGs.
sql/sql_parse.cc:
SQLCOM_DROP_EVENT does not need lex->event_parse_data object and
is more like SQLCOM_SHOW_CREATE_EVENT. Therefore, move it to the block that
handles the latter.
sql/sql_yacc.yy:
DROP EVENT does not need a parsing object, just a name.
Store it as lex->spname. Pretty similar handling to the one of
SHOW CREATE EVENT.
Infrastructure built. Added the foreseen files and change Makefile.am/CMakeLists.txt
accordingly.
libmysqld/Makefile.am:
add new files. this is first cut of WL3337u
sql/CMakeLists.txt:
add more files to build
sql/Makefile.am:
add new files. this is first cut of WL3337
sql/event_scheduler.cc:
event_timed.h -> event_data_objects.h (WL#3337)
sql/events.cc:
event_timed.h -> event_data_objects.h (WL#3337)
sql/share/errmsg.txt:
new error message
sql/event_data_objects.cc:
event_timed.h -> event_data_objects.h (WL#3337)
sql/event_data_objects.h:
event_timed.h -> event_data_objects.h (WL#3337)
sql/sql_parse.cc:
event_timed.h -> event_data_objects.h (WL#3337)
sql/sql_show.cc:
event_timed.h -> event_data_objects.h (WL#3337)
sql/sql_yacc.yy:
event_timed.h -> event_data_objects.h (WL#3337)
BitKeeper/etc/ignore:
auto-union
libmysqld/Makefile.am:
Auto merged
sql/Makefile.am:
Auto merged
sql/events.cc:
Auto merged
sql/mysql_priv.h:
Auto merged
sql/mysqld.cc:
Auto merged
sql/sql_db.cc:
Auto merged
sql/sql_parse.cc:
Auto merged
sql/sql_show.cc:
Auto merged
sql/sql_yacc.yy:
Auto merged
sql/time.cc:
Auto merged
Unify method naming -> create/update/drop_event
Move class Event_timed to event_timed.h
class Events is in events.h (renamed from event.h)
The implementation is in events.cc (renamed from event.h)
BitKeeper/deleted/.del-event_executor.cc~f4a4645b973838ab:
Delete: sql/event_executor.cc
include/my_time.h:
add a boundary
libmysqld/CMakeLists.txt:
event.cc -> events.cc
libmysqld/Makefile.am:
event.cc -> event.cc
sql/CMakeLists.txt:
event.cc -> events.cc
sql/Makefile.am:
event.cc -> events.cc
event_priv.h -> events_priv.h
sql/event_scheduler.cc:
event.h -> events.h
add_event -> create_event
replace_event -> update_event()
event_timed_compare_q is only used in event_scheduler.cc , so moving it
from event.cc and making it static
sql/event_scheduler.h:
add_event -> create_event
replace_event -> update_event
sql/event_timed.cc:
moved extern interval_type_to_name to mysql_priv.h
sql/mysql_priv.h:
moved my_time_compare to time.cc
sql/mysqld.cc:
event.h -> events.h
sql/sql_db.cc:
event.h -> events.h
sql/sql_parse.cc:
event.h -> events.h
class Event_timed moved to event_timed.h
sql/sql_show.cc:
event.h -> events.h
class Event_timed moved to event_timed.h
sql/sql_yacc.yy:
event.h -> events.h
class Event_timed moved to event_timed.h
sql/events.cc:
add_event -> create_event
replace_event -> update_event
event_timed_compare_q moved to event_scheduler.cc
sql/events.h:
class Event_timed moved to event_timed.h
sql/events_priv.h:
my_time_compare moved to mysql_priv.h
event_timed_compare_q is static in event_scheduler.cc
sql/time.cc:
moved interval_type_to_name from event.cc to here.
BitKeeper/etc/ignore:
Added libmysqld/events.cc to the ignore list
Remove compiler warnings
mysql-test/extra/binlog_tests/mix_innodb_myisam_binlog.test:
Fixed regex to get rid of xid= in output
mysql-test/r/mysqlcheck.result:
Drop client_test_db if mysql_client_test fails
mysql-test/t/mysqlcheck.test:
Drop client_test_db if mysql_client_test fails
mysql-test/t/mysqldump.test:
Use --defaults-file instead of --defaults-extra-file to avoid reading my.cnf files
sql/event_scheduler.cc:
Remove compiler warnings
sql/ha_partition.cc:
Don't divide with 0 when checking current auto_increment value
sql/handler.cc:
After merge fix
sql/mysqld.cc:
Remove comiler warning
sql/partition_info.cc:
After merge fix
Fixed memory reference overrun for some wrong partition definitions
sql/sql_base.cc:
After merge fix
Always set *leaves in setup_tables_and_check_access() (assume argument is always given).
sql/sql_select.cc:
Simple optimization
sql/sql_show.cc:
Remove compiler warning
sql/sql_table.cc:
After merge fix
storage/csv/ha_tina.cc:
Ensure table maps are restored properly
storage/ndb/include/ndbapi/Ndb.hpp:
Remove compiler warning
mysql-test/lib/init_db.sql:
remove definer from PK
mysql-test/r/events.result:
update results
mysql-test/r/system_mysql_db.result:
update results
mysql-test/t/events.test:
remove I_S.EVENTS test and move it to events_grant.test
scripts/mysql_fix_privilege_tables.sql:
change table definition
sql/event.cc:
make events non-user specific (namespace change)
sql/event.h:
make events non-user specific (namespace change)
sql/event_priv.h:
make events non-user specific (namespace change)
sql/event_scheduler.cc:
make events non-user specific (namespace change)
sql/event_timed.cc:
make events non-user specific (namespace change)
sql/sql_parse.cc:
make events non-user specific (namespace change)
sql/sql_show.cc:
SHOW EVENTS is available to everyone who has EVENT on specific schema.
No additional privileges are needed to see others' events.
- user A has events in db1 and db2
- user B has events in db1 and db3
A will see all his events from db1 and db2 as well as B's events
from db1 but not from db3.
B will see her events from db1 and db3. In addition B will see
only A's events from db1 but not db2.
Rename load_event() to load_named_event(), to avoid name clash on HP-UX
sql/event_scheduler.cc:
Rename load_event() to load_named_event(), to avoid name clash on HP-UX
sql/event_scheduler.h:
Rename load_event() to load_named_event(), to avoid name clash on HP-UX
make us see at least with gcc the function proto. this can be extended
probably for windows too.
Fix build on intelxeon3 (sun compiler and others)
sql/event_scheduler.cc:
__FUNCTION__ is not available everywhere.
make us see at least with gcc the function proto. this can be extended
probably for windows too.
sql/event_scheduler.cc:
remove unused variables
fix win32 build problem - pthread_mutex_lock() is void there
sql/event_scheduler.h:
pthread_mutex_lock() is void on win32. This fixes the win32 build.
sql/event_timed.cc:
remove unused code & variables
- Scheduler is either initialized at server start or never.
Starting & stopping is now suspending & resuming.
- The scheduler has clear OO interface
- Now all calls to the scheduler are synchronous
- GLOBAL event_scheduler uses thd::sys_var_tmp (see set_var.cc)
- External API is encapsulated into class Events
- Includes fixes for all comments of Kostja's review of 19.05.2005
Starting to merge into 5.1-release (5.1.10) and push
BitKeeper/etc/ignore:
Added libmysqld/event_scheduler.cc to the ignore list
libmysqld/Makefile.am:
executor -> scheduler
mysql-test/r/events.result:
update result
mysql-test/r/events_bugs.result:
update result
mysql-test/r/events_logs_tests.result:
update result
mysql-test/r/events_microsec.result:
update result
mysql-test/r/events_scheduling.result:
update result
mysql-test/r/events_stress.result:
update result
mysql-test/t/disabled.def:
enable these tests
mysql-test/t/events.test:
optimize the test a bit for speed, save some seconds runtime
remove FULL from SHOW EVENTS
mostly use I_S.EVENTS
mysql-test/t/events_bugs.test:
Skip irrelevant for the current design tests - all events are loaded
on server startup. Change in mysql.event will be visible on next server start.
Don't use numeric error codes.
mysql-test/t/events_logs_tests.test:
optimize the test a bit for speed
mysql-test/t/events_microsec.test:
Skip irrelevant for the current design tests - all events are loaded
on server startup. Change in mysql.event will be visible on next server start.
Don't use numeric error codes.
mysql-test/t/events_scheduling.test:
broader test
mysql-test/t/events_stress.test:
Rework the test to the new architecture of suspending/resuming.
Use less events, no need for thousands, hundreds is still ok.
sql/Makefile.am:
executor -> scheduler
sql/cmakelists.txt:
executor -> scheduler
sql/event.cc:
- remove todo comments
- remove unneded evex_queue abstraction functions
- move events_init() and events_shutdown() from event_executor.cc to here
- export db_create_event
- remove evex_load_and_compile_event, part of class Event_scheduler
- integrate the public interface found in event.h and used by sql_parse.cc
to use the new class Event_scheduler.
sql/event.h:
- add COND_finished so if one thread kills a running event it waits on this
- export callback event_timed_definer_equal, event_timed_identifier_equal(),
event_timed_name_equal and event_timed_db_equal()
to be used by Event_scheduler::drop_matching_events()
- cleanup event.h
- encapsulated all external interface into class Events
sql/event_executor.cc:
make it empty, will delete after that
sql/event_priv.h:
- more things in the private header
- remove event queue abstraction functions. tightly bind to QUEUE
- export privately db_drop_event, db_find_event, db_create_event()
- made change_security_context() and restore_security_context() free functions
sql/event_timed.cc:
- fix calculation of time when ENDS is set (STARTS is always set)
- during Event_timed::compile() set the right Security_ctx. Prevents a crash
during Event_scheduler::load_events_from_db()
- add Event_timed::kill_thread()
- implement event_timed_*_equal()
- made change_security_context() and restore_security_context() free functions.
- Comments cleanups
sql/lex.h:
new word scheduler for SHOW SCHEDULER STATUS (available only debug builds)
sql/log.cc:
move these from event_scheduler.cc
sql/mysql_priv.h:
refactor kill_one_thread
export sql_print_message_func and sql_print_message_handlers
sql/mysqld.cc:
In close_connections, called by kill_server() skip the main scheduler
thread and use events_shutdown() for shutting down the scheduler, in the same
manner it's done for RPL.
Add a new value to --event-scheduler :
0 <- No scheduler available
1 <- Start with scheduler enabled
2 <- Start with scheduler suspended
sql/repl_failsafe.cc:
refactor thd::system_thread to be an enum
sql/set_var.cc:
move sys_var_event_executor::update() to set_var.cc
executor -> scheduler
use thd::sys_var_tmp
sql/set_var.h:
executor -> scheduler
sql/share/errmsg.txt:
3 new error messages
sql/sql_class.cc:
refactor thd::system_thread to be an enum . more type-safety
sql/sql_class.h:
refactor thd::system_thread to be an enum . more type-safety
sql/sql_db.cc:
get the error from evex_drop_schema_events
sql/sql_error.h:
export warning_level_names
sql/sql_lex.h:
new command SHOW SCHEDULER STATUS, available only in debug build and
for debug purposes.
sql/sql_parse.cc:
refactor kill_one_thread() -> does the *dirty* work, and sql_kill
just the reporting.
add handler for SQLCOM_SHOW_SCHEDULER_STATUS
sql/sql_show.cc:
fix verbosity handling (this will be obsoleted anyway by the fix for 17394).
sql/sql_yacc.yy:
remove FULL from SHOW EVENTS
add SHOW SCHEDULER STATUS in debug builds
sql/table.cc:
Fix valgrind warning.
