Martin Kal�n wrote:
Greetings,
I would appreciate if someone with commit karma could review the following:

[1] http://www.mail-archive.com/[email protected]/msg14543.html

Applied by Unico. Thanks!

[2] http://www.mail-archive.com/[email protected]/msg14542.html

Also applied by Unico (thanks again!) :). However, I had time to regtest with
PostgreSQL today and found out that the user-submitted SQL was not correct
for the case when revisionNumber==null. Since I mainly use Oracle9i, I
previously figured that Oracle and PostgreSQL would behave differently
wrt setNull for statement parameters.

However I verified today that PostgreSQL is behaving like Oracle and since
the CommonRDBMSAdapter has already been fixed for this case, I attached
a new patch for PostgreSQL that will unify the revokePermission code
to look the same in both adapters
http://www.mail-archive.com/[email protected]/msg14710.html

[3] http://www.mail-archive.com/[email protected]/msg14580.html

3. Improves/extends configuration of Commons DBCP. See link to mail archive
   for full details.

Any feedback on this? Would you like me to create a bugzilla issue?

This one is especially important for long-running applications with
some inactivity times (eg overnight inactive webapps in a corporate
environment) where the server drops the DBCP pooled connections.

To remedy this the DBCP configuration mechanism in Slide must be able
to specify a validation query. Above patch fixes this (and permits
any DBCP configurable option to be set from Slide's Domain.xml).


Thanks to all devs for you continous efforts!

Cheers,
 Martin

---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]

Reply via email to