Fix a theoretical hang over the adaptive hash latch in InnoDB if one runs INSERT ... SELECT ... (binlog not enabled), or a multi-table UPDATE or DELETE, and only the read tables are InnoDB type, the rest are MyISAM; this also fixes bug #7879 for InnoDB type tables
sql/ha_innodb.cc:
Fix a theoretical hang over the adaptive hash latch in InnoDB if one runs INSERT ... SELECT ... (binlog not enabled), or a multi-table UPDATE or DELETE, and only the read tables are InnoDB type, the rest are MyISAM; this also fixes bug #7879 for InnoDB type tables
case will be added to client_test in 4.1
sql/sql_parse.cc:
A fix for bug #6761: "mysql_list_fields() does not work": init
thd->query_length, which is used next line. Remove a stale comment.
open interface for setting the "delete-on-close" flag.
innobase/os/os0file.c:
Use system-supplied tmpfile() on Netware
sql/ha_innodb.cc:
Remove innobase_mysql_tmpfile() on Netware.
mysql_admin_table() attempted to write to a vio which was 0. I could have fixed mysql_admin_table()
but fixing my_net_write() looked more future-proof.
sql/net_serv.cc:
If no VIO, no write.
This happened only if my_write() couldn't write even one byte.
I cannot easily add a .test for this, but I tested by hand before and after the change.
mysys/my_write.c:
Monty and I could not find a reason why a write should not wait for free disk space
(if disk is full) because it could not write at least one byte; doing so certainly
corrupts tables. my_pwrite() and my_fwrite() don't test for -1, so no problem.
build (to make the test suite pass)
BUILD/compile-dist:
- enable "with-extra-charsets=complex" for the distribution build
(to make the test suite pass)
official binary builds for Linux that are built against a static glibc with
a 128k thread stack size limit can be compiled with a default that doesn't
result in a harmless (but oft-misunderstood) warning message. (Bug #6226)
include/my_pthread.h:
Allow DEFAULT_THREAD_STACK to be set via -DDEFAULT_THREAD_STACK=...
ChangeLog, too (to be more exact when tagging a release that's based on a
merge ChangeSet)
Build-tools/Bootstrap:
- include merge ChangeSets into the ChangeLog, too (to be more exact when
tagging a release that's based on a merge ChangeSet)
the Epoch". (With after review fixes).
mysql-test/r/func_time.result:
Added test for bug #7515 "from_unixtime(0) now returns NULL instead of
the Epoch".
mysql-test/t/func_time.test:
Added test for bug #7515 "from_unixtime(0) now returns NULL instead of
the Epoch".
sql/item_timefunc.cc:
Item_func_from_unixtime:
from_unixtime(0) should return Epoch instead of NULL.
sql/item_timefunc.h:
Item_func_from_unixtime:
- Removed unused method definition.
- fix_length_and_dec() should set maybe_null to true since now
from_unixtime() can return NULL even in case when none of its
arguments is NULL.
Fix the previous bug fix: dropping a table with FOREIGN KEY checks running on it caused a cascade of failed drops while the foreign key check was waiting for a lock
innobase/row/row0mysql.c:
Fix the previous bug fix: dropping a table with FOREIGN KEY checks running on it caused a cascade of failed drops while the foreign key check was waiting for a lock
Fix bug: if we dropped a table where an INSERT was waiting for a lock to check a FOREIGN KEY constraint, then an assertion would fail in lock_reset_all_on_table(), since that operation assumes no waiting locks on the table or its records
row0mysql.c:
Fix bug: InnoDB failed to drop a table in the background drop queue if the table was referenced by a foreign key constraint
innobase/row/row0mysql.c:
Fix bug: InnoDB failed to drop a table in the background drop queue if the table was referenced by a foreign key constraint
innobase/row/row0ins.c:
Fix bug: if we dropped a table where an INSERT was waiting for a lock to check a FOREIGN KEY constraint, then an assertion would fail in lock_reset_all_on_table(), since that operation assumes no waiting locks on the table or its records
sql/mysqld.cc:
A fix (bug #5652: [patch] tcpwrapper support is broken on systems using an unmodified tcpwrapper).
Wrapper for fromhost, hosts_access, eval_client has been removed.
mysql-test/r/func_str.result:
result for test case for a bug in QUOTE() (Bug #7495)
mysql-test/t/func_str.test:
test case for a bug in QUOTE() (Bug #7495)
sql/item_strfunc.cc:
a better fix for a QUOTE() bug (Bug #7495)
to behave well on 5.0 tables (well now you can't use tables from 4.1
and 5.0 with 4.0 because former use utf8, but still it is nice to have
similar code in acl_init() and replace_user_table()).
This also will make such GRANTs working in 5.0 (they are broken now).
mysql-test/r/grant.result:
Added test for GRANT which manipulates user limits.
mysql-test/t/grant.test:
Added test for GRANT which manipulates user limits.
mysql-test/r/func_str.result:
A result for test case for the bug #7495 involving either LTRIM()
or TRIM() within QUOTE() function.
mysql-test/t/func_str.test:
A test case for the bug #7495 involving either LTRIM() or TRIM()
within QUOTE() function.
sql/item_strfunc.cc:
Changes for LTRIM() and TRIM() functions that aleviate the bug
entirely.
If AUTOCOMMIT=1, do not acquire an InnoDB table lock in LOCK TABLES; this makes porting of old MyISAM applications to InnoDB easier, since in that mode InnoDB table locks caused deadlocks very easily
sql/ha_innodb.cc:
If AUTOCOMMIT=1, do not acquire an InnoDB table lock in LOCK TABLES; this makes porting of old MyISAM applications to InnoDB easier, since in that mode InnoDB table locks caused deadlocks very easily
Fix InnoDB bug: on HP-UX, with a 32-bit binary, InnoDB was only able to read or write <= 2 GB files; the reason was that InnoDB treated the return value of lseek() as a 32-bit integer; lseek was used on HP-UX-11 as a replacement for pread() and pwrite() because HAVE_BROKEN_PREAD was defined on that platform
innobase/os/os0file.c:
Fix InnoDB bug: on HP-UX, with a 32-bit binary, InnoDB was only able to read or write <= 2 GB files; the reason was that InnoDB treated the return value of lseek() as a 32-bit integer; lseek was used on HP-UX-11 as a replacement for pread() and pwrite() because HAVE_BROKEN_PREAD was defined on that platform
mysql-test/mysql-test-run.sh:
Backport an improvement from 4.1: If the tool is run with '--force', give a list of all test cases that failed at the end.
This is essential for automated analysis of the build logs file.
even with zero month and day" aka "Date decoding trouble"
Two digit year should be interpreted correctly as year in 20th or 21st
century even with zero month and day. Only exception should be zero date
'00-00-00' or '00-00-00 00:00:00'.
mysql-test/r/type_datetime.result:
Added test for bug #7297 "Two digit year should be interpreted correctly
even with zero month and day"
mysql-test/t/type_datetime.test:
Added test for bug #7297 "Two digit year should be interpreted correctly
even with zero month and day"
sql/time.cc:
str_to_TIME(): Two digit year should be interpreted correctly as year
in 20th or 21st century even with zero month and day. Only exception
should be zero date '00-00-00' or '00-00-00 00:00:00'.
In Item_ref::Item_ref set maybe_null (and other fields fix_fields sets) to be the
same as in (*ref), because Item_ref::fix_fields() will not be called. Previously
maybe_null was 0 always and this produced a bogus state where
maybe_null==0 && is_null() == true
which broke evaluation for some upper-level Items, like AND and OR.
mysql-test/r/group_by.result:
Test for BUG#6976
mysql-test/t/group_by.test:
Test for BUG#6976
sql/item.cc:
Comment added
sql/item.h:
Fix for BUG#6976:
in Item_ref::Item_ref(Item**...) fix all fields because fix_fields() will not be called.
Call mark_as_null_row in join_read_const and join_read_system.
mysql-test/r/multi_update.result:
Testcase for BUG#5837
mysql-test/t/multi_update.test:
Testcase for BUG#5837
sql/table.h:
Added comments
Fix for the 0xA0 character problem in the InnoDB FOREIGN KEY parser: if my_isspace() treats 0xA0 as space, then let InnoDB do the same; this might break some multi-byte charset id's, though for big5, ujis, sjis this seems not to change the current behavior (I checked the tables in /share/charsets); this fix must NOT be merged to 4.1 because in 4.1 everything is in UTF-8
sql/ha_innodb.cc:
Fix for the 0xA0 character problem in the InnoDB FOREIGN KEY parser: if my_isspace() treats 0xA0 as space, then let InnoDB do the same; this might break some multi-byte charset id's, though for big5, ujis, sjis this seems not to change the current behavior (I checked the tables in /share/charsets); this fix must NOT be merged to 4.1 because in 4.1 everything is in UTF-8
innobase/dict/dict0dict.c:
Fix for the 0xA0 character problem in the InnoDB FOREIGN KEY parser: if my_isspace() treats 0xA0 as space, then let InnoDB do the same; this might break some multi-byte charset id's, though for big5, ujis, sjis this seems not to change the current behavior (I checked the tables in /share/charsets); this fix must NOT be merged to 4.1 because in 4.1 everything is in UTF-8
innobase/include/dict0dict.h:
Fix for the 0xA0 character problem in the InnoDB FOREIGN KEY parser: if my_isspace() treats 0xA0 as space, then let InnoDB do the same; this might break some multi-byte charset id's, though for big5, ujis, sjis this seems not to change the current behavior (I checked the tables in /share/charsets); this fix must NOT be merged to 4.1 because in 4.1 everything is in UTF-8
srv_printf_innodb_monitor(): Removed extraneous rewind() call.
The caller of this function calls rewind() if necessary.
In this way, we avoid rewind()ing stderr in
srv_lock_timeout_and_monitor_thread().
innobase/srv/srv0srv.c:
srv_printf_innodb_monitor(): Removed extraneous rewind() call.
The caller of this function calls rewind() if necessary.
In this way, we avoid rewind()ing stderr in
srv_lock_timeout_and_monitor_thread().
