Re: [jira] Commented: (DERBY-217) issue with BLOBs and batch updates in 10.1.0.0

[Satheesh Bandaram]

I have briefly looked at the large test... Would it be ok to modify the test next week, after the release? The patch corrects the problem being reported and I will run the regular tests. I will keep the bug open until a test is added here.. Here is the diff. Let me know if anyone has any comments. I have also attached the patch to Jira entry. Satheesh Index: SQLBlob.java =================================================================== --- SQLBlob.java        (revision 180460) +++ SQLBlob.java        (working copy) @@ -187,8 +187,15 @@                         return TypeId.BLOB_PRECEDENCE; // not really used                 } -    public void setInto(PreparedStatement ps, int position) throws SQLExcepti , StandardException { -        ps.setBlob(position,null); +    public void setInto(PreparedStatement ps, int position) +               throws SQLException, StandardException +       { +               if (isNull()) { +           ps.setBlob(position, null); +                       return; +               } + +               ps.setBytes(position, getBytes());      }  } Daniel John Debrunner wrote: Satheesh Bandaram (JIRA) wrote: [ http://issues.apache.org/jira/browse/DERBY-217?page=comments#action_12312921 ] Satheesh Bandaram commented on DERBY-217: ----------------------------------------- I think this bug is a regression caused by the fix for: Derby-174 & Derby-175: Make setNull work with Timestamp and Blob datatypes. Submitted by Shreyas Kaushik ([EMAIL PROTECTED]) If I rollback the fix in my setup, the passes as expected. I believe the setInto() methods added for SQLBlob and SQLTimestamp are incorrect. Fixing this should include adding batch testing of valid values to jdbcapi/parameterMapping. I can help with this testing if needed. Dan.