Version 5.1.33 is a maintenance release of the production 5.1 branch. It is suitable for use with MySQL Server versions 5.5, and 5.6.
Many parameters specified with options when making an Ant build of Connector/J from source (for example,
com.mysql.jdbc.testsuite.url.sha256default
) were not set as system properties and thus were not available for use in the JUnit tests in the testsuite. With this fix, all com.mysql.jdbc.test.* and com.mysql.jdbc.testsuite.* parameters are forwarded to the JUnit tests. (Bug #19505524)The 4-byte UTF8 (
utfbmb4
) character encoding could not be used with Connector/J when the server collation was set to anything other than the default value (utf8mb4_general_ci
). This fix makes Connector/J detect and set the proper character mapping for anyutfmb4
collation. (Bug #19479242, Bug #73663)XA connections failed with a class cast exception for a load-balanced configuration with
pinGlobalTxToPhysicalConnection=true
. This was because some XA-related classes usedcom.mysql.jdbc.ConnectionImpl
in method parameters during calls. This fix makes the classes usecom.mysql.jdbc.Connection
instead in those cases. (Bug #19450418, Bug #73594)The
MANIFEST.MF
file for the JAR package of Connector/J included an empty line between the attributesImport-Package
andName
, which was interpreted as a mark for the end of the main section of the manifest file. This fix removes the empty line. (Bug #19365473, Bug #73474)changeUser()
calls might result in a connection exception (“Got packets out of order”) or a freeze if theuseCompression
option was set to true. (Bug #19354014, Bug #19443777, Bug #73577)The tests
testByteStreamInsert
andtestUseCompress
in the Connector/J testsuite failed with the message “Row size too large (> 8126)...” when they were run against the MySQL server version 5.6.20. This is due to a known limitation with the server version, for whichinnodb_log_file_size
has to be set to a value greater than 10 times the largest BLOB data size found in the rows of the tables plus the length of other variable length fields (that is, VARCHAR, VARBINARY, and TEXT type fields) (see the MySQL 5.6.20 changelog for details). This fix adds, for MySQL 5.6.20, a check forinnodb_log_file_size
to assert that its value is 335544320 or larger and throws an exception if it is not so, asking the user to increase the size ofinnodb_log_file_size
. (Bug #19172037)-
Using Java 6 or later, running
LOAD DATA
with a prepared statement resulted in anIndexOutOfBoundsException
. This was a regression introduced by an issue in the patch for Bug #72008, which has now been fixed. (Bug #19171665, Bug #73163)References: This issue is a regression of: Bug #72008.
-
Connector/J failed the test for Bug# 64205 in the testsuite, thus might return a garbled error message for an invalid query when connected to MySQL server 5.5 or later. The failure was a regression introduced by the fix for Bug #18836319, which made the issuing of
SET NAMES
dependent on the character set difference between the server and the connection. This fix corrects the way by which the error messages are interpreted, so that they will come out right at any stage of a connection. (Bug #19145408)References: This issue is a regression of: Bug #18836319.
Errors occur when Connector/J mapped Windows time zones to Olson time zones for “Caucasus Standard Time” and “Georgian Standard Time”. This fix corrected and updated all the time-zone mappings in Connector/J using the data from the IANA Time Database v.2014g and the Unicode Common Locale Data Repository (CLDR) v.25. (Bug #17527948, Bug #70436)
-
The test case
testsuite.simple.XATest.testRecover
failed when run against MySQL servers 5.7.0 to 5.7.4, because Connector/J did not understand the new output format forXA RECVOER
that was introduced since MySQL 5.7.0 for fixing Bug #19505524. That fix has recently been reverted in MySQL 5.7.5, so that XA RECOVER output is now, by default, in the old format again. The test case, which runs well with MySQL 5.7.5, is now skipped for server versions 5.7.0-5.7.4. (Bug #17441747)References: This issue is a regression of: Bug #19505524.