Fix bug #12779 : never give a row count estimate of 0 to the MySQL query optimizer, as then left join optimizer may beleive it KNOWS that the table is empty; note that this fix may change query optimization of many other queries where one table is empty; note that the proper fix would be to make the query optimizer to know that the row count estimates it receives really are just estimates, it cannot assume they are certain
Fixed portability problem with bool in C programs
Moved close_thread_tables out from LOCK_thread_count mutex (safety fix)
my_sleep() -> pthread_cond_timedwait()
1.) Added a new option to mysql_lock_tables() for ignoring FLUSH TABLES.
Used the new option in create_table_from_items().
It is necessary to prevent the SELECT table from being reopend.
It would get new storage assigned for its fields, while the
SELECT part of the command would still use the old (freed) storage.
2.) Protected the CREATE TABLE and CREATE TABLE ... SELECT commands
against a global read lock. This prevents a deadlock in
CREATE TABLE ... SELECT in conjunction with FLUSH TABLES WITH READ LOCK
and avoids the creation of new tables during a global read lock.
3.) Replaced set_protect_against_global_read_lock() and
unset_protect_against_global_read_lock() by
wait_if_global_read_lock() and start_waiting_global_read_lock()
in the INSERT DELAYED handling.
The problem: base_list::remove didn't modify base_list::last when removing
the last list element.
The fix: If we remove the last element, find the element before it (by walking
from the beginning of the list) and set base_list::last accordingly.
The list gets corrupted in both 4.0 and 4.1. There are no visible problems in
current 4.1 because current 4.1 doesn't call where_cond->fix_fields() after
constant propagation step.
Count null_bits separately from field offsets and adjust them in case of primary key parts.
(Previously a CREATE TABLE with a lot of null fields that was part of a primary key caused MySQL to wrongly count the number of bytes needed to store null bits)
This is a more complete bug fix for #6236
when merging just use "ul"] In configure.in, don't remove $AVAILABLE_LANGUAGES_ERRORS_RULES at end
because config.status may later need this file (if it does not find it
it won't incorporate dependencies of errmsg.sys in sql/share/Makefile).
In sql/share/Makefile.am using "all:" leads to double-"all:" in Makefile.
Added protection against global read lock while creating and
initializing a delayed insert handler.
Allowed to ignore a global read lock when locking the table
inside the delayed insert handler.
Added some minor improvements.
Fixed Field_timestamp::val_int() so now it works correctly in --new mode
(or for TIMESTAMP(19) columns).
Also removed unused Field_timestamp::fill_and_store() method.
