db-derby-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Daniel John Debrunner <...@apache.org>
Subject Re: Question about checkDataSource test and DERBY-1158
Date Wed, 05 Apr 2006 00:37:04 GMT
Kathey Marsden wrote:

> After  updating my client with the change for DERBY-1158 I see the
> following diff  in the client checkDataSource test.
> I think this diff is ok but wanted to check before updating the master
> because I wasn't sure about the lock output.
> It is different than embedded.  I was wondering if someone could confirm
> that these changes are ok for updating the client checkDataSource.out
> master.
> 
> Note: checkDataSource is not yet in the client suite because of  DERBY-1173.
> 
> 
> < sru1-global-2: java.sql.SQLException: Cannot set holdability
> ResultSet.HOLD_CURSORS_OVER_COMMIT for a global transaction.
> ---
> 
>>sru1-global-2: ru(SQL_CURLH000C4) contents {1} {2} {3}
> 
> 361c361
> < sru1-global-4: java.sql.SQLException: Cannot set holdability
> ResultSet.HOLD_CURSORS_OVER_COMMIT for a global transaction.

I think it's ok. Basically now a execution of a Statement that was
created with hold cursors will succeed in a local transaction, but the
returned ResultSet will be close cursors at comment. That would lead to
the exceptions not being thrown and the change in lock counts. This is
partially implemented at the moment, only in some situations from the
client will this be the case. I'm working on making it consistent.

Dan.


Mime
View raw message