MariaDB Java Client 1.1.1 Changelog

Download | Release Notes | Changelog | Java Client Overview

Release date: 01 Mar 2013

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 #414 Wed 2013-02-27 23:17:31 +0100
  • CONJ-27 : Implement tcpAbortiveClose option, for "hard" socket closing.
  • If set, client will abort connection in a way that TCP connection does not enter usual TCP_WAIT state upon close, and free local port. This option can be handy in situation where connections are created and destroyed in rapid succession and new connection can't be created because of port shortage.
• Revision #413 Wed 2013-02-27 23:00:07 +0100
  • Removed unused classes.
• Revision #412 Mon 2013-02-18 15:21:03 +0100
  • Fix test warning
• Revision #411 Mon 2013-02-18 15:20:40 +0100
  • Small refactoring : remove yet another unnecessary level of abstraction - MySQLQueryFactory. Use MySQLQuery directory
• Revision #410 Thu 2013-02-14 22:32:33 +0100
  • CONJ-25 : Close open streaming result sets, when closing connection.
  • Prior to this patch a connection would hang during close(), if streaming result set was not fully read.
• Revision #409 Wed 2013-02-13 23:05:29 +0100
  • CONJ-24 - first() is not legal to use with streaming result set, throw exception if it used
• Revision #408 Wed 2013-02-13 17:28:01 +0100
  • Refactor code for getGeneratedKeys() - remove unnecessary classes.
• Revision #407 Wed 2013-02-13 11:05:14 +0100
  • CONJ-23 : JVM does not exit if statement timeout is used.
  • The reason is that timer thread does not exit. The fix is to Use Timer constructor that marks timer thread a daemon.
• Revision #406 Tue 2013-02-12 22:30:04 +0100
  • Do not generate javadoc during the build (no reason for it,as long as we implement standard JDBC interfaces)
• Revision #405 Tue 2013-02-12 22:29:01 +0100
  • Use arrays rather than lists to represent resultset metadata and cached result set row, for array are more efficient (faster element access, less space)
• Revision #404 Tue 2013-02-12 15:56:23 +0100
  • Implement lazy parsing of result set metadata. Avoid calling expensive String constructors unless necessary.
• Revision #403 Mon 2013-02-11 20:11:51 +0100
  • Remove unportable IPV6 test (requires working dual stack and 5.5)
• Revision #402 Thu 2013-02-07 23:13:02 +0100
  • Speed up Connection.getWarnings() - avoid unnicessary "show warnings", if last statement did not produce any warnings.
• Revision #401 Sun 2013-02-03 21:39:59 +0100
  • CONJ-20 Ensure that getObject() returns byte array for CHAR BINARY.
  • Also make sure that getColumnType,getColumnClassName,getColumnTypeName return values indicating binary type for fixed binary type
• Revision #400 Sun 2013-02-03 00:32:30 +0100
  • CONJ-17 : ResultSetMetaData.getColumnName() returned empty string for special cases, where result set column was not a column from any table. For example, column name returned by ResultRSetMetaData.getColumnName(1) for the query " select count(*) from table" would be empty string ""
  • The fix is to return column label (called "virtual column name" here http://dev.mysql.com/doc/internals/en/text-protocol.html#packet-Protocol::ColumnDefinition) instead of "".
• Revision #399 Sat 2013-02-02 22:04:50 +0100
  • CONJ-19 : DatabaseMetaData.getColumns() handled MySQL YEAR datatype as SMALLINT. Fixed to return DATE by default and SMALLINT only n case yearIsDateType=false is set
• Revision #398 Sat 2013-02-02 14:37:34 +0100
  • Fix test case
• Revision #397 Sat 2013-02-02 14:37:14 +0100
  • CONJ-15 - DatabaseMedataData.getColumns() returned incorrect value in "COLUMN_SIZE" column for character data.
  • Fix is to use I_S.COLUMNS.CHARACTER_MAXIMUM_LENGTH instead of I_IS.COLUMNS.CHARACTER_OCTET_LENGTH to calculate COLUMN_SIZE
  • Also, return COLUMN_SIZE for temporal datatypes instead of NULL.
• Revision #396 Sat 2013-02-02 13:35:04 +0100
  • CONJ-16 : Introduced nullCatalogMeansCurrent parameter, compatibly to ConnectorJ.
  • There are applications that depend on ConnectorJ default behavior, even if this behavior deviates from standard.
• Revision #395 Thu 2013-01-17 23:34:26 +0100
  • CONJ14 - Fix MySQLStatement to return -1 from getUpdateCount() and null from getResultSet() in case there are no more results.