The most recent stable (GA) release of MariaDB Connector/ODBC is:
MariaDB Connector/ODBC 3.0.5
Release date: 9 Jun 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 #34cf2a3
2016-06-07 17:37:26 +0200
- Fix for the bug ODBC-44 - incorrect binding of TIMESTAMP to TIME type(affected work with MS Access, if table has time and auto_increment fields) Changed existing testcase(t_tstotime) to cover this case as well.
- Revision #5436e8a
2016-06-03 10:32:46 +0200
- Fix of memleak with multistatement - 1 MYSQL_STMT structure had been being lost.
- Revision #e05a6c7
2016-06-02 19:33:33 +0200
- Small testcase to ensure that ';' in string in query does not confuse the connector
- Revision #c301508
2016-06-02 00:02:54 +0200
- Fix for ODBC-41 - basically ensuring that number of columns(counter in descriptor) is reset before issuing new query. Fixed possible memory leak with multistatement
- Revision #23ae3e3
2016-05-26 23:38:08 +0200
- Fix for ODBC-38 - SQLColumns(and some others along with it) returned ODBC3 SQL types, while MS Access is ODBC2. Along with it, catalog test suite was split into 2 suites,and 2 tests there were un-skipped. One happens to pass now, for 2nd one issue is fixed by this revision - connector could crash on "CREATE FUNCTION". But there are some more problems with it.
- Revision #f347cea
2016-05-24 16:30:32 +0200
- Fixed small bug - SQLGetInfo returned a bit wrong value for SQL_DRIVER_NAME info type. And completely wrong on *nix
- Revision #fbd517e
2016-05-20 00:00:15 +0200
- Some changes to make WiX build using last built configuration.
- Revision #65b26b2
2016-05-12 00:34:34 +0200
- Fixed testcase t_sqltables in catalog - it misbehaved and dropped/re-created user's test database. Which is not particularly good as for a test.
- Revision #cf5ce78
2016-05-10 19:09:10 +0200
- Fix of warnings on 64bit. Many of them were clearly bugs in fact, but most of those were harmless on LE machine. Some were real bugs. As part of this, changed some internal functions parameter and/or return types. Also changed types of properties in some structures. e.g. in descriptor header's Count is SQLSMALLINT now(according to specs), InternalLength in records is now "unsigned long" as it is used in MYSQL_BIND for length.
- Revision #4a1a196
2016-05-03 19:54:48 +0200
- (ODBC-37) Variable used for length in bind structure in SQLGetData was bigger than "unsigned long" on 64bit machines. And it wasn't initialized. Application could eventually get garbage as a length. Also fixed tests in multistatement - there were also pointed variable length problems on 64bit.
- Revision #6eefff8
2016-05-03 13:33:20 +0200
- SQLColAttribute wrote to numeric pointer(if it was supplied) even if string attribute had been requested
- Revision #9a00acc
2016-04-27 22:49:19 +0200
- Un-skipped the testcase for bit field. And fixed conversion of bit fields to numeric types(unfortunately for binds only) Structured SQLFetch a bit. Added updating of ird data to SQLMoreResults. We probably had bugs because of that.
- Revision #ac9147e
2016-04-11 17:42:33 +0200
- Version bump -> 2.0.11