e7065d355c
BUG#121915: Merge 4.1->5.0
2005-10-19 01:52:06 +04:00
e9f2f9437a
BUG#12915: Added single-table UPDATE/DELTE ... ORDER BY ... LIMIT
...
optimization: now can use index to find records to update/delete
when there is no WHERE clause.
2005-09-30 15:21:37 +04:00
775f17667b
Merge with 4.1
2005-06-07 00:31:53 +03:00
29fd1f2fd9
Move USE_PRAGMA_IMPLEMENTATION to proper place
...
Ensure that 'null_value' is not accessed before val() is called in FIELD() functions
Fixed initialization of key maps. This fixes some problems with keys when you have more than 64 keys
Fixed that ROLLUP don't always create a temporary table. This fix ensures that func_gconcat.test results are now predictable
2005-06-03 23:46:03 +03:00
113a99626d
Merge of 4.1->5.0. This contained the fixes for GCC 4.0
2005-06-01 21:56:30 -07:00
ba9b9f8993
Mainly cleanups for gcc 4.0. Some small pieces from looking at -Wall. Removed a number of dumb things in ha_tina.
2005-06-01 17:34:10 -07:00
34cec09efb
Merge neptunus.(none):/home/msvensson/mysql/bug10241
...
into neptunus.(none):/home/msvensson/mysql/mysql-5.0
2005-05-09 11:26:48 +02:00
5b5565c30a
BUG#10241 cygwin port: invalid pragma interface directives
...
- Introduce ifdefs so we can control when to use #pragma interface on cygwin
2005-05-04 15:05:56 +02:00
a7a1d4dc1e
Fix for BUG#10244:
...
Make get_quick_select_for_ref() accept estimated # records as parameter and
set QUICK_RANGE_SELECT::records, as this value is used to allocate buffers
in Multi-Range Read.
2005-04-29 01:16:32 +04:00
fb0845f848
Post WL#2126 fixes:
...
* remove get_next_init and move its functionality to QUICK_RANGE_SELECT::reset()
* added more comments
* added empty FT_SELECT::reset()
2004-12-29 16:07:27 +03:00
60e35755ef
WL#2126 - Multi_read_range.
...
Added the required structures and functions for
handing over multiple key ranges to the table handler.
2004-12-23 21:45:10 +01:00
e6a7386c12
Fix for BUG#4177:
...
* Make index merge quick selects code allow perform several scans.
* Delay additional handler objects creation till row retrieval is started.
2004-11-21 11:51:19 +03:00
bc5a9111d7
backport Serg's fix of FT interface (BUG#6523)
2004-11-16 22:58:02 +02:00
e2cd3dd1ce
WL#1724 "Min/Max Optimization for Queries with Group By Clause"
...
- after-review changes
- merged with the source tree from 204-08-27
2004-08-27 16:37:13 +03:00
838eaec709
Fix and testcase for BUG#4984
...
The crash is eliminated but still it is weird/inefficent that ROR-intersection is used when performing updates in empty table.
2004-08-11 16:40:46 +04:00
aa35efd292
after merge fixes
2004-07-22 00:26:33 +02:00
f9de714945
misc fixes for compile-time errors
2004-07-21 00:45:08 +02:00
2fd11169ab
failing fulltext_order_by.test fixed
2004-07-15 17:30:55 +02:00
5b3c418b48
After merge fixes
...
Note: The following tests fails
- fulltext (Sergei has promised to fix)
- rpl_charset (Guilhem should fix)
- rpl_timezone (Dimitray has promised to fix)
Sanja needs to check out the calling of close_thread_tables() in sp_head.cc
2004-07-15 04:19:07 +03:00
3f1c4ba745
handler interface cleanups:
...
more logical table/index_flags
return HA_ERR_WRONG_COMMAND instead of abstract methods where appropriate
max_keys and other limits renamed to max_supported_keys/etc
max_keys/etc are now wrappers to max_supported_keys/etc
ha_index_init/ha_rnd_init/ha_index_end/ha_rnd_end are now wrappers to real {index,rnd}_{init,end} to enforce strict pairing
2004-06-23 12:29:05 +02:00
ee0a516ccd
Manual merge
2004-06-14 11:18:56 +04:00
2e93dfaf42
More code cleanup, debug prints removed
...
Fixed an incorrect optimizer choice for ror-intersect(key, clustered_primary)
Typo bug fixed in multitable update
2004-06-01 15:31:58 +04:00
5843eec521
Manual merge
2004-05-29 20:55:46 +04:00
dff98a8ef7
* New index_merge EXPLAIN output format
...
* Fixed a problem with wrong query results for partially covering keys in ROR-index_merge
* ROR-intersection retrieval plan choice algorithm now uses less disk IO
- and properly processes clustered PK range scans
* Fixed several minor range optimizer problems
* Added more comments
* Code cleanup
2004-05-29 02:04:01 +04:00
71eddc362e
Merging 4.1 to 5.0.
2004-05-26 17:04:45 +02:00
314a8bf114
key_cmp -> key_cmp_if_same
...
New records_in_range() interface (similar to read_range())
Macros for faster bitmap handling
Simplify read_range() code (#WL1786)
New general key_cmp() function to compare keys
2004-05-16 14:48:32 +03:00
ec92106535
Post-merge fixes.
...
Note: One sp.test still fails (prime), and rpl_server_id2.test fails (will be fixed by guilhem ASAP).
2004-05-14 16:00:57 +02:00
3b29cad05c
Manual merge
2004-05-13 01:49:47 +04:00
a46d7542c9
This is first cset for WL#1394 "Optimize index merge when all involved index ranges include only values with equal keys"
...
The main idea is to exploit the fact that key scans for "key=const" return ordered sequences of rowids.
2004-05-13 01:38:40 +04:00
43546e7a82
Moved reading of ranges from opt_range.cc to handler.cc
...
This gives the handler more optimization possiblities and is needed for NDB cluster
Fixed not-initialized memory error detected by valgrind
2004-04-08 13:58:06 +03:00
e94d935221
Cleanup/optimizations of structures and key usage to make it easier to move key-range-search to handler
2004-04-08 01:50:59 +03:00
3d2794ae9a
Fix for Bug#3183
2004-03-15 23:11:58 +03:00
f9e73c7716
Post-merge fixes.
2004-02-11 18:21:55 +01:00
c8585f3283
Merge 4.1 into 5.0.
2004-02-10 18:44:02 +01:00
480b082cf4
merge
2004-02-05 18:14:48 +01:00
6bfede9639
Added SQL_SELECT::cleanup() to make it easier to reuse SQL_SELECT item's for sub selects. Needed for proper fix for bug #2479 .
...
Added 'select_to_file' class to be able to merge identical code for select_export and select_dump
2004-02-05 10:22:08 +01:00
c4728c276a
Code cleanup:
...
we don't need opt_range.h and opt_range.cc files at all.
2004-02-03 14:35:24 +04:00
f231d9826b
just tried to find all 'skipp' and replace it with 'skip'.
2004-02-02 20:25:39 +04:00
67c6d5113c
Precise read time estimates for index_merge/Unique
2003-12-18 06:08:00 +03:00
24c180941f
Two-sweeps read index_merge plus several small index_merge fixes and improvements
2003-12-09 23:40:02 +03:00
3d32afd736
Post-merge fixes for index_merge/Unique/5.0
2003-11-24 20:54:57 +03:00
5f9718adfb
Merging index_merge/Unique into 5.0
2003-11-23 15:17:46 +03:00
4f721196f3
Post-merge fixes
2003-11-22 00:20:48 +03:00
30b3c3bf1b
Merged, most likely needs post-merge fix
2003-11-21 00:27:54 +03:00
4016048d81
logging_ok:
...
Logging to logging@openlogging.org accepted
sql_union.cc, sql_select.cc, opt_range.h, opt_range.cc, opt_ft.h:
Post-merge fixes
Many files:
new file
2003-11-20 23:27:11 +03:00
1c61a92ba7
index_merge implementation using Unique class, to be merged into 5.0
2003-11-19 00:37:17 +03:00
4894565432
merging in index_merge (in progress, not yet working)
2003-11-13 22:14:37 +03:00
ca6c4cc30b
Added index_merge access method
2003-11-13 17:52:02 +03:00
6e17f4f20d
fixes to Bitmap class
2003-10-24 22:44:48 +02:00
f944fd4a65
merged
2003-10-21 11:58:43 +02:00
