MariaDB Connector/ODBC 1.0.0 Changelog

The most recent Stable (GA) release of MariaDB Connector/ODBC is:
MariaDB Connector/ODBC 3.1.20

Download Release Notes Changelog Overview of MariaDB Connector/ODBC

Release date: 29 Jan 2015

For the highlights of this release, see the release notes.

The revision number links will take you to the revision's page on Launchpad. On Launchpad you can view more details of the revision and view diffs of the code modified in that revision.

  • Revision #55 Wed 2015-01-28 13:41:50 +0100
    • Fixed outdated dependencies of msi installer
  • Revision #54 Tue 2015-01-27 16:49:10 +0100
    • Ignore Visual Studio files in source package
  • Revision #53 Tue 2015-01-27 13:09:17 +0100
    • Added readne and license information Added support for creating source package: soure packages can be created by cpack --target package_source
  • Revision #52 Thu 2015-01-15 15:02:08 +0100
    • Fixed package name
  • Revision #51 Thu 2015-01-15 08:57:24 +0100
    • Removed beta in msi package name Bumped version number to 1.0.0 (GA)
  • Revision #50 Wed 2015-01-14 16:32:16 +0100
    • Partial fix for ODBC-17 Now odbcte32 can use the connector. Added support for SQL_ODBC_API_CONFORMANCE and SQL_ODBC_SQL_CONFORMANCE info types.
  • Revision #49 Wed 2015-01-07 13:37:45 +0100
    • Fixed bug in FindMariaDB.cmake Bumped version number
  • Revision #48 Tue 2015-01-06 00:42:05 +0100
    • Some amendments to the previous patch fixing memoryleaks
  • Revision #47 Mon 2015-01-05 12:13:51 +0100
    • Fix of memory leaks in the connector
  • Revision #46 Fri 2014-12-19 17:45:05 +0100
    • Fixes for posix build.
  • Revision #45 Wed 2014-11-19 13:22:48 +0100
    • Version bump(patch 5->6)
  • Revision #44 Tue 2014-11-11 22:19:50 +0100
    • Change for the ODBC-16 patch: If previously prepared query was not a batch of statements, Stmt->stmt is closed only if new query is the batch. Much more optiomal, and fixes regressions caused by original patch.
  • Revision #43 Mon 2014-11-10 16:27:17 +0100
    • 1) Removed memory leak in MADB_KeyTypeCount - alocated statement was not properly freed 2) Fixed possible buffer overrun in MADB_StmtGetData 3) Changed in all testcase assert to IS - using not overloaded assert was a bad idea for automated tests. 4) Changed testing framework output to be more in accordance with TAP specs - for failed tests now printed "not ok" instead of "fail" 5) Extended framework with meand to set/get server variables/status. Helper function setting variables. remembers original value, and that value is restored after end of test
  • Revision #42 Thu 2014-10-23 11:04:44 +0200
    • bump patch version number
  • Revision #41 Wed 2014-10-22 20:00:44 +0200
    • The fix for ODBC-16. It's adding missing mysql_stmt_close calls. The most important one is in MADB_StmtPrepare(called by both SQLPrepare and SQLExecDirect) closing previosly opened by this STMT object prepared statements
  • Revision #40 Tue 2014-10-21 14:19:00 +0200
    • Fix and the testcase for the bug ODBC-15 - SQLGetInfo did not support one deprecated info type(SQL_ODBC_API_CONFORMANCE)
  • Revision #39 Fri 2014-09-19 17:36:50 +0200
    • Fix and the testcase for ODBC-14 - fixed processing of IPD SQL_DESC_UNNAMED field update by appliction. SQL_UNNAMED value is now allowed, SQL_NAMED causes HY092 error. The error prevented use of queries with parameters in ADO
  • Revision #38 Fri 2014-09-19 16:47:42 +0200
    • bumped minor version
  • Revision #37 Thu 2014-08-14 17:46:37 +0300
    • Something that builds on *nix. DM libs/headers are neither looked for, nor reported they are not found
  • Revision #36 Tue 2014-08-12 17:26:47 +0300
    • Changed all occurrences of SUCCEEDED macro definition use, to more appropriate and more portable, as for a ODBC driver, SQL_SUCCEEDED macro
  • Revision #35 Sun 2014-08-10 02:28:09 +0300
    • Small change for better portability - u_int64 types use changed to int64_t
  • Revision #34 Fri 2014-08-08 21:41:11 +0300
    • Fix for the bug ODBC-13 - everything to make prompting work.
  • Revision #33 Sun 2014-07-13 17:17:53 +0300
    • 1) Made tests to read values from environment as well. Command line parameters override values from the environment variables. Names of variables are: TEST_DSN TEST_DRIVER TEST_SERVER TEST_UID TEST_PASSWORD TEST_SCHEMA TEST_PORT 2) Added some more name templates to .bzrignore
  • Revision #32 Tue 2014-07-01 01:24:51 +0300
    • Mostly fixes of warnings on Windows. 1) rc files could not be compiled with express version of VS because of "afxres.h" inclusion. That header belongs to MFC that does not come with express version. 2) Big number of changes is making data types casts explicit or changing type of variables(to avoid casts), removing of not referenced variables, changed ++/-- from postfix to prefix form. 3) Changed one helper function(MADB_SetString) parameter type to eliminate signed/unsigned values comparing warnings. 4) Also there are some fixes of existing casts. Some of them look bad enough to had been able to cause problems. 5) Now fixes in tests. Some macros were duplicated - removed copies. In many places positions of parameters of macros were switched. Renamed IS_NUM to is_num because it clashed with other macro in some windows header. 6) Tests did not respect some of command line parameters. Also I added processing of the case when connection to the server could not be established - some tests were crashing. 7) Introduced .bzrignore file
  • Revision #31 Mon 2014-06-23 16:22:18 +0300
    • Recoded rc files from utf16(le) to ansi to let bzr see diff in it
  • Revision #30 Thu 2014-06-26 06:15:57 +0200
    • Fix length calculation for SQL_C_CHAR
  • Revision #29 Wed 2014-06-25 14:53:11 +0200
    • - Fixed length calculation if both OctetLength and IndicatorPtr (== SQL_NTS) was specified. - Bump patch version ------------- Ths lin and the following will be ignored --------------
  • Revision #28 Tue 2014-06-24 19:15:21 +0200
    • Fix for WCHAR fetch: We need to convert the terminating zero charater too.
  • Revision #27 Fri 2014-06-20 16:15:50 +0200
    • Fixed length calculation for array_binding if SQL_NT was specified.
  • Revision #26 Thu 2014-06-12 06:57:41 +0200
    • Fix for ODBC-11: Due to server bug MDEV-5273 we can't get result set metadata after prepare for each statement. Statements like SHOW or CALL procedure name will return metadata after execute. In latter case the number of columns after execute will change and we need to set the ird metadata.
  • Revision #25 Mon 2014-06-02 08:29:21 +0200
    • Fixed string lengths for unicode conversion
  • Revision #24 Thu 2014-05-29 15:22:32 +0200
    • Fix for ODBC-10: Fixed length calculation for SQLWCHAR for Fetch and Execute
  • Revision #23 Wed 2014-05-28 08:35:57 +0200
    • Fix for ODBC-9: buffer overflow during conversion from char* to SQL_NUMERIC_TYPE
  • Revision #22 Mon 2014-05-26 12:58:05 +0200
    • Fix for ODBC-8: Map FieldIdentifier in SQLColAttributes to ODBC 3.0 compliant values
  • Revision #21 Mon 2014-05-19 11:46:21 +0200
    • °Fix for ODBC-6 and ODBC-7 Also includes fix for ADO crash (wrong binding)
  • Revision #20 Thu 2014-05-15 16:28:35 +0200
    • Fix for CONO7: Return SQL_VARCHAR instead of SQL_UNKNOWN_TYPE for NULL fields
  • Revision #19 Wed 2014-05-14 08:29:42 +0200
    • Fix for ODBC-6: Fixed length attribute for SQLWCHAR parameters if the string isn't null terminated (SqlLen_or_IndPtr != SQL_NTS)
  • Revision #18 Wed 2014-04-23 09:32:53 +0200
    • Fixed typecasts for CURSOR_BEHAVIOR constants in SQLGetInfo
  • Revision #17 Thu 2014-04-10 19:00:45 +0200
    • Fix for SQLCursor: If ODBC version is < 3.0 SQLCloseCursor should not return an error in no cursor is open.
  • Revision #16 Thu 2014-04-10 11:44:31 +0200
    • Fixed bug in the msi-installer: The GUID for the driver components was undefined. Uninstallation of prior versions require manual deletion of the MariaDB ODBC driver directory

Comments

Comments loading...
Content reproduced on this site is the property of its respective owners, and this content is not reviewed in advance by MariaDB. The views, information and opinions expressed by this content do not necessarily represent those of MariaDB or any other party.