5d8a853331
Merge pull request #1665 from sougou/bugs
...
sqltypes bug: incorrect vitess type returned
2016-04-29 11:10:24 -07:00
9eb47e2de2
bug fix: code improvement
...
Now that we don't mix mysql flags with MySQL types, there's
no need to bit-shift them. Makes the code more readable.
2016-04-28 16:31:02 -07:00
0eecb3c753
sqltypes bug: safer fix for type resolution
...
The previous fix was still vulnerable to unexpected MySQL
behavior. The new fix checks the flags based on the type.
So, it's more robust.
2016-04-28 16:22:33 -07:00
939ec958b2
sqltypes bug: incorrect vitess type returned
...
Bug: b/28409793
MySQL sometimes sets stray flags that have to be ignored.
Specifically, for this query:
select max(uint) from t
MySQL sets the binary flag along with the unsigned flag.
The bug fix changes the behavior where vitess only looks at
the relevant flags and ignores all others.
2016-04-28 14:47:40 -07:00
97b183d622
Merge pull request #1663 from youtube/jdbc
...
Return AutoGeneratedKey as ResultSet
2016-04-28 12:29:50 -07:00
48405d9c8b
Merge pull request #1664 from youtube/ResultSet_Changes
...
Fix from bloating the byte[] values in GetByte()/GetBytes()
2016-04-28 11:30:33 -07:00
0f10c842ff
Merge pull request #1662 from enisoc/java-null-bindvar
...
java: Fix for null bind variable value.
2016-04-28 11:27:42 -07:00
a2d8f65096
Fix from bloating the byte[] values in GetByte()/GetBytes()
2016-04-28 16:14:09 +05:30
7f28127b49
code formating
2016-04-28 15:34:33 +05:30
448724ace5
code formating
2016-04-28 15:26:39 +05:30
dfe4e78855
resolved merge conflict of VitessPreparedStatement class
2016-04-28 15:12:40 +05:30
12db645a4d
AutoGeneratedKeys
2016-04-28 15:05:25 +05:30
bc3bc34c1c
java: Fix for null bind variable value.
...
When Java `null` is passed as a bind variable value, we should translate
it to a non-null, empty `ByteString` at the protobuf level.
2016-04-28 00:01:24 -07:00
4aa8402b50
BinlogPlayer: Do not use unhealthy REPLICA tablets as source.
2016-04-27 21:15:49 -07:00
0434ae24a2
discovery: Provide filter helper functions to reduce code duplication.
...
Note: The GetCurrentMaster() function which will be used by vtworker when we switch to the discovery package.
2016-04-27 21:15:39 -07:00
b44ed6c0ac
Merge pull request #1660 from sougou/v3
...
v3: support sqltypes.Value for string vindexes
2016-04-27 16:47:01 -07:00
35c5aa2018
v3: support sqltypes.Value for string vindexes
...
All vindexes must accept sqltypes.Value because that's
how values are supplied for cross-shard joins.
I've also updated the example and the v3 e2e test.
2016-04-27 15:17:19 -07:00
55b6665b35
Merge pull request #1654 from sougou/v3
...
v3: revised vindexes
2016-04-27 14:52:45 -07:00
da0b99b597
Merge pull request #1658 from alainjobart/v3error
...
V3error
2016-04-27 14:13:18 -07:00
0beecb01da
Merge pull request #1659 from michael-berlin/worker_refactor
...
vtworker: Do not reuse the mutex from StatusWorker in implementations.
2016-04-27 14:04:44 -07:00
b7655d2278
Better version of the keyspace watcher in v3.
...
We now wait synchronously for the first version, when we get them all we
build the initial vschema. We then process changes incrementally as we
did before.
2016-04-27 13:06:34 -07:00
1f8f722528
Only log v3 schema errors after all keyspaces.
...
Otherwise at startup, some keyspace vschemas may depend on others.
We will only log:
- an error when loading vschema once we have all keyspaces.
- any refresh error.
2016-04-27 13:06:34 -07:00
6ace21037a
Merge pull request #1656 from alainjobart/combo
...
Add BeginExecute{,Batch} to gRPC, removing tabletconn.ERR*
2016-04-27 13:05:50 -07:00
49d7fa3333
vtworker: Do not reuse the mutex from StatusWorker in implementations.
...
Removing this dependency makes it easier to reason about which components need synchronization and use the lock.
StatusWorker is now fully isolated and can be accessed only through its exported methods.
Other changes:
- Unexported remaining fields in StatusWorker
- Added State() to worker interface to deduplicate code in unit tests.
- created TabletStatusList which aggregates TableStatus objects and has its own lock now. By resing it, we reduce code duplication in SplitClone and VerticalSplitClone.
- Changed Status* functions to not use the previous global mutex. Instead, StatusWorker and TabletStatusList are called independently now.
2016-04-27 13:01:06 -07:00
ef3f030c6e
v3 vindexes: handle space padding
...
Ref: http://dev.mysql.com/doc/refman/5.7/en/char.html
MySQL ignores trailing spaces. So, vindexes should reflect this.
2016-04-27 12:48:41 -07:00
8c015b0b10
Removing Err* from tabletserver.
...
We now use canonical errors. Note over the RPC channel we were always
sending the canonical errors already, and not really using the old style
errors anyway.
2016-04-27 11:35:16 -07:00
d0bfecf3d7
v3: fix third_party dependency
2016-04-27 11:27:22 -07:00
be79afda7b
Merge pull request #1657 from erzel/fixing_equal_splits_value_to_big_rat
...
Added a workaround for b/28409793
2016-04-27 11:22:09 -07:00
22d0eb4b50
Merge pull request #1655 from michael-berlin/worker_fixes
...
vtworker: Abort when filtered replication is already enabled.
2016-04-27 11:12:42 -07:00
915a3ef9de
Added a workaround for b/28409793
2016-04-27 10:37:57 -07:00
7d38d2a036
Removing tabletconn error codes.
...
They are now unused, we just depend on canonical errors.
2016-04-27 09:37:49 -07:00
c1189537bb
vtgate doesn't use tabletconn.ServerError.Code.
...
Instead, it just relies on tabletconn.ServerError.ServerCode, which is
the vtrpcpb version of the code.
2016-04-27 09:30:15 -07:00
6fbfebc2a4
resolver.isConnError -> resolver.isRetryableError.
...
It's all it's used for anyway, so it's just clearer.
2016-04-27 09:08:32 -07:00
894a34a53e
Removing Code from ScatterConnError.
...
This is part of the next wave of clean-up, to remove tabletconn.ERR_*
values and replace them with vtrpcpb.ErrorCode.
2016-04-27 08:57:33 -07:00
af6a18b817
Adding BeginExecute{,Batch} rpc to gRPC code.
...
It is behind a flag for now, until it's been around for long enough, and
then I'll just enable it by default.
2016-04-27 08:11:32 -07:00
faa3f2f2c2
vtworker: Log head of stacktrace in case of a panic.
2016-04-26 23:52:06 -07:00
dd05dec73c
vtworker: Abort when filtered replication is already enabled.
2016-04-26 23:50:19 -07:00
16174fc5e2
v3: fix []byte vs string bug
...
The v3 router used to convert all []byte to string to
check for uniqueness (a futile exercise). The uniqueness
check was removed, but the conversion remained. Now, I've
removed it, and fixed tests and other dependencies accordingly.
2016-04-26 22:53:06 -07:00
34b6790b16
Merge pull request #1653 from michael-berlin/v3_demo_message
...
examples: Log port of v3 demo on stdout.
2016-04-26 21:57:10 -07:00
169207ebc0
v3: rename binary & char vindexes
...
The new vindexes now have more precise names. This is because
their behavior is not supposed to change over time.
2016-04-26 21:48:57 -07:00
fec6433e28
Merge pull request #1651 from erzel/fix_equal_splits_default
...
Made equal_splits use the first split_column if more than one is give rather than complain
2016-04-26 20:36:26 -07:00
79423e9f6e
v3: fix binary & char vindexes
...
Fix 1: Use proper unicode methods to generate
reliable unique keys for unicode strings. The reference
is here: http://www.unicode.org/reports/tr10/#Multi_Level_Comparison .
We're using level 1 comparison, which is the most
permissive.
Fix 2: Use MD5 hash instead of the weaker 3DES algorithm.
2016-04-26 18:23:27 -07:00
57812424fd
Made equal_splits use the first split_column if more than one is given, rather than complain.
2016-04-26 16:27:34 -07:00
734d20a78e
Merge pull request #1652 from erzel/adding_support_for_random_population_of_vttest_to_go_client
...
Allow initialization of vttest DB with random data from go.
2016-04-26 13:52:03 -07:00
87cf663fd8
examples: Log port of v3 demo on stdout.
2016-04-26 13:21:24 -07:00
5c5b269b79
Allow initialization of vttest DB with random data from go.
2016-04-26 11:46:14 -07:00
e00aed5743
Merge pull request #1650 from alainjobart/beginexecute
...
Adding combo methods Begin+Execute{Batch}
2016-04-26 10:52:30 -07:00
39980a6184
Fixing a few tests that were using GetSessionId.
...
It was all implicit, and lazy, really.
2016-04-26 10:36:16 -07:00
dbc71f23e5
Removing sessionID in one more place in tests.
2016-04-26 10:15:41 -07:00
7eda94d924
Clean up of tabletconntest for no more Session.
2016-04-26 10:02:39 -07:00
sougou
Anthony Yeh
Ashudeep Sharma
Harshit Gangal
Michael Berlin
Alain Jobart
Alain Jobart
Erez Louidor
Erez Louidor