The most recent release of MariaDB 10.2 is:MariaDB 10.2.44 Stable (GA) Download Now
DownloadRelease NotesChangelogOverview of 10.2
Release date: 18 Apr 2016
For the highlights of this release, see the release notes.
The revision number links will take you to the revision's page on GitHub. On GitHub you can view more details of the revision and view diffs of the code modified in that revision.
Revision #9664240 2016-04-16 20:43:54 +0300 - Merge ../10.2-window-funcs-r10 into 10.2
2016-04-16 20:41:06 +0300
: Assertion `!join->only_const_tables() && fsort' failed
2016-04-15 20:40:01 +0300
: Wrong result with aggregate function as a window function
2016-04-15 17:48:40 +0300
Make test result deterministic.
2016-04-14 01:08:54 -0700 - Merge branch 'bb-10.2-mdev9543' of github.com:MariaDB/server into bb-10.2-mdev9543
2016-04-13 11:06:05 +0200
Make ntile use args[0] for it's argument.
2016-04-13 10:39:06 +0200
Fix another bug in dense_rank.
2016-04-14 00:47:28 -0700
Fixed bug .
2016-04-12 00:00:53 +0200 - Merge ../10.2-window-funcs-r12 into 10.2
2016-04-11 23:59:51 +0200
Fix compiler warning
2016-04-10 07:51:42 +0300
Fix dense_rank returning minimum rank of 2 when using null columns.
2016-04-11 10:44:43 +0200
Use --sorted_result to make test result predictable
2016-04-10 17:22:24 +0200
: Window functions: interplay between window function and other constructs
2016-04-10 16:24:04 +0200
: Assertion `n_rows > 0' failed in Frame_cursor* get_frame_cursor
2016-04-10 11:28:33 +0200
Window functions: return error if aggregate is not supported as window functions
2016-04-10 10:13:55 +0200
Window functions: handle window functions as arguments to other functions
2016-04-09 17:01:01 +0200
Update to previous cset, which added ORDER BY into the wrong clause
2016-04-08 12:02:43 +0200
Make the testcase deterministic
2016-04-08 03:21:25 +0300
Window functions: make "ORDER BY window_func" work
2016-04-07 00:54:39 +0300 - Merge branch '10.2' into bb-10.2-mdev9543
2016-04-06 23:02:31 +0300
: Window functions: wrong sort criteria is used
2016-04-06 20:34:23 +0300
: Window functions: crash with big_tables=1
2016-04-06 19:58:02 +0300
Code cleanup after merge of into 10.2-window-functions
2016-04-06 18:24:11 +0300
Window functions: Better class names
2016-04-05 16:58:50 +0300
Fix ntile to work with --ps-protocol
2016-04-05 19:10:44 +0300
: Window functions: reuse sorting and/or scanning
2016-04-04 15:24:27 -0700
Fixed a problem with setting wrong flags for ordering in the code of compare_window_funcs_by_window_specs().
2016-04-04 22:03:50 +0300
Convert ntile to work with expressions as parameters.
2016-04-04 17:06:12 +0300
Implement ntile window function.
2016-04-04 15:09:31 +0300
Remove no longer used Window_context class.
2016-04-04 14:16:15 +0300
Fix post review comments regarding the usage of List<>.
2016-04-01 15:29:45 -0700
Fix a problem in code for percent_rank(). The problem popped up when running win_percent_cume.test with --ps-protocol.
2016-04-01 14:05:51 -0700
Fixed a problem with --ps-protocol. Window names has to be resolved only once.
2016-04-01 12:00:54 -0700
The implementation of the template bubble_sort assumed that the call-back comparison function returns a positive number when arg1 < arg2, and a negative number when arg1 > arg2. This is not in line with other implementation of sorting algorithm. Changed bubble_sort: now a negative result from the comparison function means that arg1 < arg2, and positive result means that arg1 > arg2. Changed accordingly all call-back functions that are used as parameters in the call of bubble_sort.
2016-03-30 12:43:57 -0700
Fixed a crash in compare_window_frames(). The function did not take into account the case when only one of the pointers to the compared frames is NULL.
2016-03-30 18:39:10 +0300
Make EXPLAIN FORMAT=JSON be able to show the key that's used for sorting.
2016-03-30 16:00:46 +0300
Fix the merge error in (Merge branch '10.2' into bb-10.2-mdev9543)
2016-03-30 02:07:48 -0700
Added sorting window function to minimize the number of scans of the temporary table needed to compute them.
2016-03-28 22:52:18 +0300 - Merge branch 'cume_dist' into bb-10.2-mdev9543
2016-03-28 22:48:32 +0300
Implemented cume_dist function.
2016-03-28 22:11:00 +0300
Convert percent_rank to work with cursors
2016-03-28 16:57:41 +0300
Lay the groundwork for variable number of cursors.
2016-03-28 22:19:55 +0300 - Merge branch 'bb-10.2-mdev9543' of github.com:MariaDB/server into bb-10.2-mdev9543
2016-03-28 20:53:09 +0300
: Window functions: HAVING and GROUP BY - Hook window function computation into the right location. - Add a testcase which shows that HAVING is now checked before the window function computation step.
2016-03-28 18:38:42 +0300
Make window functions computation step show up in EXPLAIN FORMAT=JSON output
2016-03-28 22:18:38 +0300 - Merge branch '10.2' into bb-10.2-mdev9543
2016-03-28 12:02:56 +0300
: Re-engineer the code for post-join operations
2016-03-28 00:55:57 +0300
Remove JOIN_TAB::used_window_func, it is not used anymore
2016-03-27 16:50:04 +0300
Get rid of Window_func_runner::first_run
2016-03-27 12:15:11 +0300
Fix a typo
2016-03-27 12:07:02 +0300
Remove out-of date code
2016-03-27 11:47:19 +0300
Make window function computation a part of the query plan
2016-03-24 03:23:14 +0300
Encapsulate use of List<Cached_item> in a Group_bound_tracker.
2016-03-24 03:08:43 +0300
Code cleanup
2016-03-24 02:57:03 +0300
Encapsulate the switching between different return values in Item_window_func
2016-03-24 02:42:38 +0300
Move the deprecated comments out of the way
2016-03-23 16:10:47 -0700 - Merge branch 'bb-10.2-mdev9543' of github.com:MariaDB/server into bb-10.2-mdev9543
2016-03-24 02:09:17 +0300
Fix a PS re-execution problem and code cleanup
2016-03-21 20:00:05 -0700
Changed the base class for Item_window_func from Item_result_field to Item_func_or_sum. Implemented method update_used_tables for class Item_findow_func. Added the flag Item::with_window_func. Made sure that window functions could be used only in SELECT list and ORDER BY clause. Added test cases that checked different illegal placements of window functions.
2016-03-23 16:09:58 -0700
Changed the base class for Item_window_func from Item_result_field to Item_func_or_sum. Implemented method update_used_tables for class Item_findow_func. Added the flag Item::with_window_func. Made sure that window functions could be used only in SELECT list and ORDER BY clause. Added test cases that checked different illegal placements of window functions.
2016-03-18 10:52:02 -0700
For some window functions an order list must be present.
2016-03-18 01:15:05 +0300
: Window functions: Frame Exclusion support
2016-03-18 00:33:53 -0700
Prohibit using window functions of some types with window frames in full accordance with the SQL standard.
2016-03-17 14:13:54 -0700 - Merge branch 'bb-10.2-mdev9543' of github.com:MariaDB/server into bb-10.2-mdev9543
2016-03-17 22:17:09 +0300
: Buildbot shows a crash in JOIN::make_aggr_tables_info()
2016-03-17 14:13:38 -0700
Fixed bug . Each window name has to be resolved only once.
2016-03-17 21:51:00 +0300
Window functions + ORDER BY : first testcase
2016-03-17 21:50:07 +0300
Remove the wrong check, win_spec->order_list may be empty but it is not NULL.
2016-03-17 16:52:23 +0300
: Window functions: catch invalid window frame specs
2016-03-16 23:35:28 -0700
Fixed the bug concerning execution of prepared statements with window functions. Added the test case for it. Also allowed to use aliases for set functions in partition and order lists that are specified in window functions.
2016-03-16 12:03:43 +0300
Add testcases for frames with bound1 > bound2.
2016-03-16 00:50:14 -0700
The class Window_spec now has pointers to the partition and order lists of the type SQL_I_List rather then the objects of this type. It allows to replace easily one instance of such a list for another. Besides it will facilitate to compare two lists if they originate from the same window specification. In fact any direct assignment for objects of the type SQL_I_List was not valid.
2016-03-16 01:57:59 +0200
Implemented avg() window function.
2016-03-16 01:42:45 +0200
Fix 2 more bugs in item_windowfunc.
2016-03-16 01:14:43 +0200
Remove no longer needed TODO.
2016-03-16 01:08:09 +0200
Make sure to return the result value when calling Item_windowfunc::val_real()
2016-03-16 01:26:39 +0300
Continuation of "Implemented a counter within Item_sum_sum" a few commits before
2016-03-15 16:03:51 +0300
: Re-engineer the code for post-join operations
2016-03-14 19:11:03 +0200
Implemented a counter within Item_sum_sum
2016-03-14 17:49:23 +0300
Add a testcase for RANGE-type frames and NULL values
2016-03-14 15:42:00 +0200
Preliminary implementation for the aggregate sum function as a window function
2016-03-14 14:13:59 +0300
: RANGE-type frames for window functions
2016-03-13 03:44:40 +0300
Add a testcase for non-integer range frame.
2016-03-13 03:34:31 +0300
: RANGE-type frames for window functions
2016-03-11 23:29:52 +0300
Better comments
2016-03-11 23:00:15 +0300
Use correct frame bounds when window frame was not specified
2016-03-11 20:23:24 +0300
: Wrong window frame when using RANGE BETWEEN N FOLLOWING AND PRECEDING
2016-03-10 17:46:47 +0300
: Wrong window frame when using RANGE BETWEEN N FOLLOWING AND PRECEDING
2016-03-07 13:37:04 +0200
Add temporary test cases for bitwise window functions.
2016-03-07 12:30:04 +0200
Create a default frame bound when no frame is specified.
2016-03-07 11:58:58 +0200
Implement BIT_(AND|OR|XOR) functions as window functions.
2016-03-06 23:55:34 +0300
: RANGE-type frames for window functions
2016-03-06 23:10:20 +0300
: RANGE-type frames for window functions
2016-03-03 18:45:37 +0200
Implement percent_rank window function
2016-02-28 22:00:00 +0200
Convert if statements to switch case.
2016-03-02 14:01:43 +0300
Remove comments that may be confusing
2016-03-02 13:08:55 +0300
Remove all comments starting with 'psergey'
2016-03-02 13:02:58 +0300
Rename window frame start/end to top/bottom. Bikeshed should be green.
2016-03-01 15:07:57 +0300
: Compute Aggregate functions as window functions
2016-02-28 01:37:18 +0300
Re-factoring in window cursors code split Frame_unbounded
2016-02-26 02:08:45 +0300
: Window function produces incorrect value
2016-02-19 23:20:09 -0800
Fixed a problems in the parser. Resolved window names. Checked some constraints for window frames. Added test cases for window name resolution.
2016-02-18 01:25:26 +0300
: Compute Aggregate functions as window functions
2016-02-16 00:33:53 +0200
Fix review comments.
2016-02-16 00:22:12 +0200
Implement DENSE_RANK function.
2016-02-15 18:46:02 +0300
Testcase fix and code cleanup for window functions
2016-02-15 18:40:04 +0300
Forgot to add sql_select.cc to the previous commit
2016-02-15 17:16:30 +0300
Cleanup: remove JOIN::table_access_tabs, top_table_access_tabs_count.
2016-02-14 21:00:05 +0300
Post-merge fixes. win.test passes but further cleanup is needed.
2016-02-07 01:06:56 +0300
More testcases, fixed comments
2016-02-06 01:53:17 +0300
Initial implementation of RANK() window function
2016-02-05 16:50:50 +0300
More comments
2016-02-05 14:24:35 +0300
More testcases. Added .result file
2016-02-05 14:12:17 +0300
Got sort-and-read single-pass window function computation to work
2016-02-05 00:52:17 +0300
Added comments
2016-02-04 18:41:50 +0300
Window functions: moving ahead
2016-02-04 00:40:48 +0300
Moved window function computation code from JOIN::exec_inner() into a separate function, JOIN::process_window_functions().
2016-02-03 23:58:00 +0300
Make the temp.table have columns for window function values.
2016-02-02 20:44:15 +0200
Force need_tmp to true if we have window functions.
2016-02-14 00:16:01 +0300
Fix buildbot failure
2016-02-13 21:58:48 +0300
Fix maria.maria test by updating test results.
2016-02-12 20:33:56 -0800
Initial patch for the implementation of window functions (): - All parsing problems look like resolved - Stub performing name resolution of window functions in simplest queries has been added.
2016-02-09 12:35:59 -0800
This is the consolidated patch for : "Re-factor the code for post-join operations".
2016-04-08 12:12:26 +0400
Modifying ctype_gbk_export_import.test to help "meld" not to confuse/join individual records when displaying diff.
2016-04-08 11:04:34 +0400
Better coverange in ctype_gbk_export_import. Adding tests for the sequence 0xEE5C5C which should be treated in GBK context as GBK double-byte character 0xEE5C (with 5C in the second byte) followed by 0x5C (as a normal REVERSE SOLIDUS).
2016-04-08 07:45:18 +0400
ctype_gbk_export_import: adding an utf8 column, for better coverage.
2016-04-08 00:39:26 +0400
Fixing a test failure in ctype_gbk_export_import.test
2016-04-07 20:09:57 +0300
Fixed compiler warnings
2016-04-07 19:58:28 +0300
Removed file-key-management from mysql_install_db as someone may have deciced to use it even before bootstrap.
2016-04-07 17:22:16 +0400
Adding more option combinations into ctype_gbk_export_import.
2016-04-07 17:09:53 +0400
Adding a test for "mysqldump -Tdir dbname tabname" followed by "LOAD DATA INFILE" with various command line and CHARACTER SET clause options.
2016-04-07 16:21:29 +0400
Adding a new test ctype_gbk_export_import covering export with SELECT INTO OUTFILE followed by import with LOAD DATA INFILE, with various command line --default-character-set=xxx and "CHARACTER SET xxx" clause combinations.
2016-04-06 18:08:16 +0300
Fix the buildbot: file-key-management is provided by a plugin, so should be loose.
2016-04-06 13:12:49 +0300
Fix a typo: use attribute(...) not attribute(...)
2016-04-06 10:31:38 +0400
LOAD XML INFILE does not handle well broken multi-byte characters
2016-04-06 09:13:49 +0400
LOAD DATA INFILE silently truncates incomplete byte sequences
2016-04-06 01:01:09 +0200
fix compile error on Windows, intrduced by previous monty's push
2016-04-06 01:18:38 +0300
Fixed but when generating .sys files
2016-04-05 17:44:12 +0300
Cleanups:
2016-04-05 16:52:40 +0300
Added new range of MariaDB error messages, starting from 3000
2016-04-05 09:35:27 +0400
Updating conf_to_src.c according to the current CHARSET_INFO structure.
2016-04-04 09:52:24 +0400
Clean-up: Adding a class Term_string to share some LOAD DATA code
2016-04-02 00:30:40 +0400
Clean-up: Changing READ_INFO to use a String object instead of buff, buff_length, end_of_buff. This unifies "LOAD DATA" and "LOAD XML", so they now both use String for a temporary value storage. This change will make further work for easier.
2016-04-02 00:18:58 +0400
LOAD DATA INFILE does not work well with a TEXT column when using sjis
2016-03-31 14:22:25 +0400
LOAD DATA INFILE does not work well with gbk in some cases LOAD DATA does not work with multi-byte strings in LINES TERMINATED BY when IGNORE is specified
2016-03-31 11:04:48 +0400
Clean-up CHARSET_INFO: strnncollsp: diff_if_only_endspace_difference
2016-03-25 20:51:22 +0400
- enable connection log in mysqltest by default
2016-03-29 12:10:44 +0300
Follow-up #2 for switching PSEUDO_THREAD_ID to ulonglong: fix another 32-bit test
2016-03-29 01:03:03 +0300
Follow-up for switching PSEUDO_THREA_ID to ulonglong: fix the 32-bit test
2016-03-28 12:55:15 +0400 - Merge pull request #166 from iangilfillan/10.2
2016-03-22 21:21:49 +0200
man page updates: mysqlbinlog options, mysqlcheck clarification and 10.2 updates
2016-03-25 15:41:10 +0400
Fixing the return data type of my_charlen() from "uint" to "int", as it can return negative values. The typo was introduced in the patch for in 10.2.0.
2016-03-25 10:36:33 +0100 - Merge branch '10.1' into 10.2
2016-03-25 06:42:44 +0400
Removing my_strnncoll_mb_bin() and my_strnncollsp_mb_bin(), as they are not used any more. We now use function templates from strcoll.ic instead.
2016-03-24 09:35:14 +0100
more result updates
2016-03-23 22:45:48 +0100
after merge fixes
2016-03-23 22:36:46 +0100 - Merge branch '10.1' into 10.2
2016-03-23 22:12:00 +0100
thread cache: cleanup DBUG state when caching a thread
2016-03-23 15:10:08 +0200
Fixed failures from changing values of thread_stack and thread_cache_size Added --thread_cache_size=0 to some tests that was depending on not having a thread cache.
2016-03-23 14:36:59 +0200
Fixed failures from changing values of thread_stack and thread_cache_size Added --thread_cache_size=0 to slow_launch_time_func.test as otherwise the new thread would start too fast to be counted as a slow_launch_thread.
2016-03-23 14:26:43 +0400
Removing duplicate code: sharing implementation of "strnxfrm" between gbk_chinese_ci and big5_chinese_ci.
2016-03-23 12:44:31 +0400
Removing unused code in ctype-bin5.c
2016-03-23 12:37:19 +0400
Fixing compilation warnings introduced in:
2016-03-23 09:21:37 +0100
Make step bigger to reflect modern needs.
2016-03-22 23:42:13 +0200
Increase value of thread_cache_size to 32 Added 5 minute timeout before automaticlally removing threads from thread cache.
2016-03-22 21:51:59 +0200
Removed TABLE->sort to make it possible to have multiple active calls to filesort and init_read_record() for the same table. This will simplify code for WINDOW FUNCTIONS ()
2016-03-22 21:51:49 +0100
: protocol: COM_MULTI command
2016-03-22 22:36:40 +0300
: protocol: COM_MULTI command (part 3)
2016-03-22 19:02:05 +0100
Fix valgrind error - memory leak in yassl's SSL_remove_state
2016-03-22 18:44:46 +0100
fix most annoying warnings on Windows
2016-03-22 21:19:57 +0400
Fixed Windows compilation failure
2016-03-22 18:24:09 +0300
Don't select 'digest' column from PERFORMANCE_SCHEMA.events_statements_history_long
2016-03-22 18:01:40 +0400
Cleanup All_share_tables_list definition
2016-03-21 11:46:03 -0700 - Manual Merge of branch 'bb-10.2-mdev8789' into 10.2
2016-02-17 14:30:25 -0800
Addressed the issues raised in the review for the main patch of . Fixed a bug in TABLE_LIST::print. Fixed another bug for the case when the definition of a WITH table contained column list while the join in the main query used two instances of this table.
2015-12-22 08:56:30 -0800
Added sql_cte.cc for embedded.
2015-12-21 12:13:39 -0800
Fixed compile errors of the merge of the patch for with 10.2.
2015-12-17 23:52:14 +0300
Implement non-recursive common table expressions
2016-03-21 11:00:35 +0100
: protocol: COM_MULTI command (post-post-review changes)
2016-03-20 21:00:58 +0200
Fixed compiler failures and warnings
2016-01-07 16:00:02 +0100
: protocol: COM_MULTI command (part 3)
2016-01-05 20:44:45 +0100
: protocol: COM_MULTI command (part 2)
2015-11-26 11:21:56 +0100
: protocol: COM_MULTI command (part 1)
2016-03-16 10:55:12 +0400
Remove cs->cset->ismbchar()
2016-03-11 16:55:57 +0100
: ALTER TABLE does not work from client
2016-03-14 10:22:42 +0400
Unexpected modification of value and warning about out of range value upon ALTER Limitation on DOUBLE or REAL length is ignored with INSERT .. SELECT
2016-03-09 17:01:41 +0100
fix openssl memory leak in main thread
2016-03-09 16:42:45 +0100
Fix leak from missing my_thread_end
2016-03-09 10:19:09 +0100
Fix threadpool memory leak and connect2 test
2016-03-08 11:05:32 +0200
[] Update test cases for sysvars_server_embedded.
2016-03-08 10:50:04 +0200
[] Make show_create_user testcase not run on embedded build
2016-03-08 00:35:03 +0200
[] Post review fixes and cleanups.
2016-01-21 13:20:40 +0200
[] Added show create user implementation.
2016-01-19 14:33:00 +0200
[] Update test cases
2016-01-19 14:30:19 +0200
[] Updated syntax for SHOW CREATE USER
2016-01-19 13:01:28 +0200
[] Update test cases
2016-01-18 02:16:59 +0200
[] Implement alter user and tested create user
2016-01-17 17:00:19 +0200
[] Update grammar for new syntax
2016-01-17 16:55:29 +0200
Fix warnings in sql_acl.cc
2016-03-08 10:28:26 +0100
Fix threadpool after it was broken by
2016-03-01 10:21:42 +0400
- Simplify audit event dispatching
2016-02-28 20:53:07 +0400
Fixed plugins.cracklib_password_check failure
2016-02-28 18:18:29 +0300
: EXPLAIN FORMAT=JSON should show outer_ref_cond
2016-02-28 14:54:56 +0400
Yet more fixes covering thread_id type change
2016-02-27 16:57:14 +0400
Yet more fixes covering thread_id type change
2016-02-27 13:27:42 +0400
Corrected format string for long long thread_id
2016-02-26 14:34:27 +0400
Add continuous binary log backup to mysqlbinlog. --raw, --stop-never and --stop-never-slave-server-id=id options added to the mysqlbinlog tool.
2016-02-26 18:33:53 +0400
Fixed log_tables and mysqlbinlog_row_minimal tests
2015-12-29 19:09:11 +0400
- Table cache cleanups
2015-12-29 14:58:17 +0400
- Table cache cleanups
2015-12-29 14:37:54 +0400
- Table cache cleanups
2016-02-24 14:31:43 +0400
- information_schema.user_variables
2016-02-26 15:32:34 +0400
Fixed compilarion failure on 32bit systems
2016-02-26 14:55:20 +0400
Fixed change_user and func_misc in embedded
2016-02-26 13:37:31 +0400
Fixed plugins.feedback_plugin_send failure
2016-02-26 12:13:41 +0400
Fixed main.null failure in embedded
2015-12-15 23:34:32 +0100
after merge fix debian builds
2016-02-25 18:19:55 +0100 - Merge branch '10.1' into 10.2
2016-02-16 19:26:59 +0200
Cache check_table_binlog_row_based and mark_trx_read_write Benefits: - Speeds up insert,write and delete by avoiding 1-2 function calls per write/update/delete. - Avoiding calling write_locked_table_maps() if not needed. - The inlined code is much smaller than before - Updating of table->s->cached_row_logging_check moved to when table is opened - Moved some bool values together in handler class to get better alignment.
2016-02-10 00:20:23 +0100
Fix compilation
2016-02-08 22:34:41 +0200
Changed my_thread_id to int64 to fix compilation problem with my_atomic_add32_explicit on windows Fixed that server_audit.c also works if one compiles with safemalloc Fixed compiler warnings
2016-02-07 15:23:08 +0200
Corrected freeing of thd when running with wsrep and thread pool This is needed because of the new code where THD creation was moved to the new thread
2016-02-01 12:48:29 +0200
We can't call set_server_version() in signal handler as it's not callable after server has started (assert in sys_var::where get_sys_var_value_origin())
2016-02-01 12:45:39 +0200
Speed up connection speed by moving creation of THD to new thread
2016-01-30 00:11:27 +0200
Optimized intkorr() and intstore(functions) for intel 64 bits. (Benchmarked, about 30 % faster and 50 % smaller than original)
2016-02-06 18:07:54 +0100
cleanup: remove a couple of unused cmake checks and defines
2016-02-06 18:14:54 +0200 - Merge branch '10.1' into 10.2
2016-01-17 21:21:39 +0100
: Client Server capability negotiation for MariaDB specific functionality
2016-01-17 21:23:47 +0100
Fix client unit test to drop all used tables.
2016-01-26 15:15:13 +0100
Make --slave-parallel-workers an alias for --slave-parallel-threads
2016-01-16 18:45:26 +0400
Illegal mix of collation when using GROUP_CONCAT in a VIEW CREATE TABLE SELECT MAX(int_column) creates different columns for table vs view
2016-01-14 17:23:23 +0400
Adding "const" qualifier to the argument of decimal_actual_fraction().
2016-01-12 17:03:29 +0400
Split filesort.cc:make_sortkey() and filesort.cc::sortlength() into virtual methods in Type_handler
2016-01-11 17:20:16 +0400
Split Copy_field::get_copy_func() into virtual methods in Field
2015-12-30 19:39:31 +0100
Fix process handle leak in buildbot. GenerateConsoleCtrlEvent sent to non-existing process will add a process handle to this non-existing process to console host process conhost.exe
2015-12-29 18:06:00 +0400
- On shutdown, report the user and the host executed that.
2015-12-29 19:37:11 +0400 - Merge remote-tracking branch 'origin/10.1' into 10.2
2015-12-23 15:52:34 +0400
- mysql_install_db launches mysqld 3x
2015-11-25 18:12:19 +0400
- On shutdown, report the user and the host executed that.
2015-12-28 12:18:41 +0400
Adding "const" qualifier to methods Field::eq_def() and Copy_field::get_copy_func().
2015-12-27 00:39:39 +0400
Split memcpy_field_possible() into virtual methods in Field
2015-12-26 21:25:48 +0400
Split field_conv_incompatible() into methods
2015-12-24 15:46:18 +0400
- Support for faking server version
2015-12-24 11:37:16 +0400
Changing the parameter of Field::eq_def() from "Field *" to "const Field *"
2015-12-23 16:59:48 +0400 - Merge pull request #123 from ottok/ok-debpkg-10.2
2015-11-24 11:57:10 +0200
Debian packaging: extend libcrack hack to create correct control file
2015-12-23 14:15:00 +0400
Add Field::store_hex_hybrid()
2015-12-04 21:51:34 +0400
Moving the "ha_field_option_struct *option_struct" member from Column_definition to Create_field, as it's not needed neither for make_field(), nor for SP variables, SP parameters, SP return values.
2015-12-04 16:38:42 +0400
Wrap create_virtual_tmp_table() into a class, split into different steps
2015-12-03 23:59:47 +0400
Detect cmp_type() and result_type() from field_type() Part8: Derive Item_copy from Type_handler_hybrid_field_type
2015-12-03 23:42:05 +0400
Detect cmp_type() and result_type() from field_type() Part7: Derive Item_cache from Type_handler_hybrid_field_type
2015-12-03 15:20:57 +0400
Fixing DBUG_ASSERT() introduced in e3fed3b9b4f488e9ad1afa57333ae80249e6cb17 (a change adding a helper method Column_definition::make_field()) to take into account the EXECUTE stage of a prepared statement. This DBUG_ASSERT() caused crashes in a few tests when running "mtr --ps".
2015-12-03 10:10:59 +0400
Detect cmp_type() and result_type() from field_type() Part6: Deriving Item_type_holder from Type_handler_hybrid_real_field_type
2015-12-02 19:30:47 +0100
: Disable malloc size callback prior to exit()
2015-12-02 19:23:02 +0100
restore the check for HAVE_CXX_NEW, it is actually used
2015-12-02 16:22:19 +0400
Detect cmp_type() and result_type() from field_type() Part5: Derive Item_param from Type_handler_hybrid_field_type
2015-12-02 12:14:06 +0400
Detect cmp_type() and result_type() from field_type() A cleanup: removing members and functions that become unused: - Item_func_hybrid_field_type::cached_result_type - sp_map_result_type()
2015-12-02 10:49:16 +0400
Detect cmp_type() and result_type() from field_type() Part4: Deriving Item_temporal_hybrid_func from Type_handler_hybrid_field_type
2015-12-02 09:26:34 +0400 *
Detect cmp_type() and result_type() from field_type() Part3: Deriving Item_splocal from Type_handler_hybrid_field_type
2015-12-01 14:20:09 +0400
Detect cmp_type() and result_type() from field_type() Part2: deriving Item_sum_sum from Type_handler_hybrid_field_type, removing "Item_result Item_sum_sum::hybrid_type".
2015-12-01 13:13:23 +0400
Detect cmp_type() and result_type() from field_type() (A dependency task for Add a plugin to field types)
2015-11-27 20:50:19 +0400
A patch for Add a plugin to field types (column types) Adding Column_definition::make_field() as a convenience wrapper for ::make_field.
2015-11-24 19:55:52 +0400
- Support for faking server version
2015-11-26 15:04:55 +0400
Removing sp_variable::type, as it was always set to the same value with sp_variable::field_def.type, so there was data redundancy.
2015-11-26 14:40:40 +0400
Removing the unused "field_type" parameter in sp_head::fill_field_definition().
2015-11-19 11:32:37 +0400
- Removal of the tool "mysql_zap"
2015-11-18 20:15:36 +0400
- Remove mysqlbug
2015-11-19 16:54:45 +0400
- remove "fast mutexes"
2015-11-24 14:43:09 +0400
- The "aria_recover" variable should be renamed "aria_recover_options" to match MyISAM
2015-11-19 16:50:09 +0400
- Obsolete sql_alloc() in favor of THD::alloc() and thd_alloc()
2015-11-18 23:54:01 +0400
- Obsolete sql_calloc() in favor of THD::calloc() and thd_calloc()
2015-11-18 22:55:17 +0400
- Obsolete sql_strmake() in favor of THD::strmake() and thd_strmake()
2015-11-18 19:50:53 +0400
- Obsolete sql_strdup() in favor of THD::strdup() and thd_strdup()
2015-11-18 19:21:30 +0400
- Obsolete sql_memdup() in favor of THD::memdup() and thd_memdup()
2015-11-25 11:57:20 +0400
Split Create_field Part2: moving a few other fields from Column_definition to Create_field. sizeof(sp_variable) is now 200 bytes (vs 248 bytes in 10.1)
2015-11-25 11:22:10 +0400
A joint patch for:
2015-11-24 22:20:32 +0100
cleanup: remove Field->stored_in_db, Create_field->stored_in_db
2015-11-24 11:04:10 +0100
cleanup: remove useless internal fied flag
2015-11-21 20:27:49 +0100
followup for optimizer_search_depth change
2015-11-24 12:44:35 +0400
Changing %type of opt_place from to <const_simple_string>. A prerequisite change for: - sql_yacc.yy: add %type create_field for field_spec and column_def - sql_yacc.yy: get rid of the rules "opt_if_not_exists_table_element" and "opt_if_exists_table_element" - Split Create_field
2015-11-22 07:20:15 +0100
cmake: message_once
2015-11-20 13:41:22 +0100
cmake: MYSQL_PARSE_ARGUMENTS -> CMAKE_PARSE_ARGUMENTS
2015-11-20 13:14:33 +0100
cmake: remove unused files
2015-11-05 15:16:37 +0100
cmake: remove unused checks, options, and symbols
2015-11-20 12:57:23 +0100
cmake: rename symbols used internally by check_compiler_flag.cmake
2015-11-20 18:43:22 +0100
fix debian packaging for 10.2
2015-11-23 18:55:01 +0400
Get rid of LEX::length and LEX::dec A preparatory task for: Add a plugin to field types (column types)
2015-11-23 17:52:09 +0400
BINLOG_SYM in keywords_sp causes shift/reduce conflicts
2015-11-14 21:28:15 +0200
Bump debian/changelog for 10.2 series
2015-11-23 14:49:23 +0400
Wrong error message and shift/reduce conflicts in the RETURNS clause
2015-11-20 11:18:16 +0100 - Merge branch '10.1' into 10.2
2015-11-20 11:14:21 +0100
update tokudb to 10.2
2015-11-19 10:56:30 +0100
Removed depricated optimizer_search_depth.
2015-11-11 13:27:20 +0100
bump the VERSION
This page is licensed: CC BY-SA / Gnu FDL