Updating a result set

What could pose a problem is max_allowed_packet value. Connected to 5.1.31-log version : 1.5.0_17-b04 vendor : Sun Microsystems Inc.

Attached is my code that works as expected (I created the BLOB file using classes from testsuite. The full Eclipse URL is: -Xmx1024M -XX: Use Parallel GC -Dcom.java6.javac=C:\jvms\jdk1.6.0\bin\-Dcom.java6.rtjar=C:\jvms\jdk1.6.0\jre\lib\-Dcom.default=jdbc:mysql://remote_server:xxxx/test? java.runtime.version : 1.5.0_17-b04 : Windows Server 2008 os.version : null sun.management.compiler : Hot Spot Client Compiler com.

NOT_NULL_FLAG, the "1" value in field-level metadata).

The server doesn't enforce the "ill-advised" updates, that's why the check is in the driver in the first place. There isn't currently a way to set this check on a case-by-case basis in the JDBC driver.

Is it possible to enable that setting for a single query, or must be be a connection-wide setting?

updating a result set-39

I will include the failure-inducing data first, and then the data that does not cause a failure. If I remove the DISTINCT, the query executes as expected (and the PK is properly identified). My guess is the "DISTINCT" causes the optimizer to use a temporary table to resolve a portion of the query, and it's changing some type information that causes the JDBC driver to not "see" the primary key.

Give the database a row entry: INSERT INTO `test_me`.`test_table` VALUES() 3.

Run the code (set BINARY_FILE_NAME with an actual binary file name): import *; import *; public class Blob Update Test ***************************** IN ADDITION: This same problem exists when attempting to update a LONGTEXT field with the Result Set.update Character Stream() method. I also added the extra parameter as you suggested (I added file.size()). None of the routines within the testsuite.regression.

I have boiled-down the problem into something must easier to deal with.

If DISTINCT is removed from the query, it /does not/ fail.

Leave a Reply