b277b613f7
model: fix ddl data race ( #2708 )
2017-02-22 18:18:06 +08:00
bb2aace0d6
ddl: Fix the bug that showing status row count is 0 ( #2696 )
2017-02-21 21:01:14 +08:00
417122af42
ddl: save original schema ID for allocator when rename table. ( #2671 )
...
Fixes a bug that renamed table to another database failed to insert.
2017-02-17 22:19:00 +08:00
87f51736ee
*: support the statement of alter table alter column set default ( #2608 )
2017-02-08 15:03:27 +08:00
76bf890f0b
ddl/model: fix ddl log format ( #2599 )
2017-02-07 09:12:31 +08:00
217c26bd0b
*: Support the statement of rename table ( #2444 )
2017-01-15 00:58:10 +08:00
99b3465c1e
ddl: allocate column ID and index ID in TableInfo. ( #2205 )
2016-12-09 12:31:27 +08:00
f5792b5e80
*: the first setp of adding the field of history information in job structure ( #2170 )
2016-12-07 18:29:37 +08:00
1056c56df1
ddl: support modify column ( #1930 )
...
* ddl: support modify column
2016-11-08 15:20:41 +08:00
1c6c145f94
*: Set custom verbose flag to true ( #1956 )
...
* *: set custom verbose flag to true
* *: update .gitignore
2016-11-05 18:00:13 +08:00
1f1a60c02a
*: improve log ( #1944 )
...
Add logging for set system variable.
Add connection ID in log.
Adjust some log level.
Improve readability.
2016-11-04 17:11:10 +08:00
b4e3e58693
*: Clean up ( #1928 )
...
* ddl: remove useless code, add and update comments
2016-11-03 11:40:14 +08:00
d04dea2c57
*: use schema diff to reduce memory allocation ( #1858 )
...
When schema version has changed, we don't need to reload all schemas, only
modified table need to be reloaded, this will reduce the pressure on Go garbage collector.
Also removed used map in *inforSchema.
2016-10-24 15:17:44 +08:00
037dc11f43
ddl: Cancel DDL job when the key is exist ( #1844 )
2016-10-22 20:52:45 +08:00
662cb41a1f
ddl: write DDL binlog when DDL is done. ( #1850 )
...
Previous implementation write DDL binlog as early as when job is added to the queue,
but DDL may be canceled, we should not sync the ddl at that time, there would be a long
time waiting for the DDL to be done.
After this commit, DDL binlog is written right before the DDL job is done, so we don't need to wait.
Also expose a 'GetAllHistoryJobs' method in meta which will be used later.
2016-10-20 14:03:33 +08:00
b399060118
ddl: truncate table by DDL operation. ( #1766 )
...
Currently, truncate table deletes rows in table in one large transaction,
too large transaction is very likely to fail.
This commit use DDL to do the work, it returns instantly, old data will be
deleted by background worker.
2016-09-29 14:22:02 +08:00
6e87d52cfc
*: use defer ( #1781 )
2016-09-29 13:54:07 +08:00
0cbbec594c
model: show RowCount value after the operation of admin show ddl ( #1780 )
2016-09-29 13:24:47 +08:00
5df5085e33
*: Add row count field in a job, fix a drop schema bug and pass race ( #1759 )
2016-09-28 18:11:40 +08:00
dde36a141f
*: Convert string to terror in DDL Job ( #1699 )
...
* *: convert string to terror
2016-09-09 09:58:40 +08:00
e687f908d3
*: tiny clean up ( #1404 )
2016-07-07 09:50:45 +08:00
0450ba0b47
*: unify "See url" format ( #1391 )
...
* *: unify "See url" format
2016-07-05 11:21:10 +08:00
d04b0b997c
*: support index length ( #1350 )
...
For string columns, indexes can be created that use only the
leading part of column values, using col_name(length) syntax
to specify an index prefix length
2016-07-02 13:26:40 +08:00
e742def024
add foreign key on update and on delete support ( #1152 )
...
* add foreign key on update and on delete support
2016-04-27 19:44:00 +08:00
b51ab763df
plan: support limit pushdown ( #1105 )
...
* plan: support limit pushdown
2016-04-20 13:32:20 +08:00
0537b67d2b
*: Add more info into TableInfo and fix bug in show create table ( #1072 )
...
Fix: https://github.com/pingcap/tidb/issues/1061
2016-04-09 15:59:51 +08:00
ce96978182
*: Add index option in IndexInfo
...
1. Add index option in IndexInfo.
2. Show index option content when running show index statement.
2016-03-16 22:09:11 +08:00
be2039cf44
Merge branch 'master' into zimuxia/inspectkv-sql
2015-12-29 20:41:24 +08:00
8993f0e148
*: add test
2015-12-29 20:40:36 +08:00
c3bdfb1fed
stmt/stmts: insert use primary key value as record ID
2015-12-25 21:02:52 +08:00
56ff3d3000
model, ddl: add PKIsHandle field in TableInfo.
2015-12-25 20:29:21 +08:00
1a28ff83a0
Merge remote-tracking branch 'origin/master' into shenli/index-id
...
Conflicts:
ddl/ddl.go
kv/index_iter.go
2015-11-27 14:28:59 +08:00
b9135eadb4
*: Add ID to IndexInfo and use IndexID in prefix instead of IndexName
2015-11-24 21:58:26 +08:00
3d6378092d
model: add helper job function and update test.
2015-11-20 09:30:12 +08:00
9c6b27877d
*: must wait 2 * lease schema lease time when running job.
2015-11-18 09:53:14 +08:00
b230d4b3dc
*: remove no need DDL action.
2015-11-12 10:46:09 +08:00
4eb34f69ce
*: distinguish write/delete reorganization state.
...
we must distinguish different reorganization state after write only/
delete only state.
2015-11-11 13:29:06 +08:00
d0caf9d166
*: save reorganization processed handle in meta.
2015-11-10 11:12:03 +08:00
fbed93ca52
*: support simple statistics
...
we will use show status to show DDL statistics later.
2015-11-05 22:52:49 +08:00
01213eb729
model: rename meas to means
2015-11-04 15:47:28 +08:00
87e7c0a2a6
*: rename reorgnization to reorganization
2015-11-04 14:40:36 +08:00
ce74a843bd
ddl: rename reOrg
2015-11-04 13:03:08 +08:00
aeb57cd7bd
*: add ReOrgHandle to support resume reorg job.
2015-11-04 11:09:35 +08:00
cfb8d0a635
*: Update job's last update time stamp
2015-10-30 19:47:17 +08:00
32fe0db38b
*: add create/drop index
2015-10-28 22:00:30 +08:00
b3fc6db760
*: Job adds SchemaState to log more information
2015-10-28 09:35:05 +08:00
c8bbf7879d
*: support create table and drop table.
2015-10-27 15:05:33 +08:00
a815babeaa
ddl: typo and clean up a bit
2015-10-27 10:56:17 +08:00
d0c80f6a0d
*: add drop schema
2015-10-27 10:09:57 +08:00
2cb3e2add9
*: first commit and support create schema.
2015-10-26 21:54:54 +08:00
