Why did you make this a JDBC property instead of adding it as a
property of the XADataSourceLoader MBean?  I think that would be the
better approach (and it avoids having to remove it before the real driver
sees it).  The StatementFactory seems like a good idea; I'll take a swing
at integrating your patches into the main Minerva source as soon as I get
a chance.

Aaron

On Thu, 5 Apr 2001 [EMAIL PROTECTED] wrote:
> Patches item #414023, was updated on 2001-04-05 08:00
> You can respond by visiting:
> http://sourceforge.net/tracker/?func=detail&atid=376687&aid=414023&group_id=22866
>
> Category: None
> Group: None
> Status: Open
> Priority: 5
> Submitted By: Bill Burke (patriot1burke)
> Assigned to: Nobody/Anonymous (nobody)
> Summary: rewrite minerva statement caching
>
> Initial Comment:
> I rewrote the statement caching in Minerva for a number
> of reasons:( tar attached)
>
> - Make the allowable open statements configurable from
> jboss.jcml
> - Consolidate statement caching logic into one class
> for less code duplication
> - To be able to keep track of all open statements a
> connection has so that we release prepared statements
> when a maximum of allowed open statements(plain
> Statements plus PreparedStatements) is reached.
>
> What did I change?
>
> - You can configure the maximum allowable open
> statements by putting in a property in the Properties
> tag in the XADataSourceLoader xml stuff in jboss.jcml
>
>      <attribute
> name="Properties">MaxOpenStatements=99</attribute>
>
> "-1" value means caching is turned off
> "0" value means caching is unlimited
>
> - org.opentools.minerva.jdbc.StatementFactory contains
> all the logic for statement creation and caching.
> Prepared statements only are cached, but all open
> statements are counted.  If a createStatement will put
> the open statements over MaxOpenStatements, then a
> prepared statement will be closed and released from the
> cache.
>
> - ConnectionInPool and XAClientConnection delegate to a
> StatementFactory for creating, preparing and closing
> statements.
>
> - there is no longer a psCaches HashMap contained in
> ConnectionInPool as a static variable.  This is instead
> renamed to sfCaches and hold a hashmap of connections
> to StatementFactories.
>
> - XADataSourceImpl reads the properties and extracts
> MaxOpenStatements.  I also remove MaxOpenStatements
> from the Properties so that the JDBC driver doesn't get
> screwed up by a property it doesn't understand.
>
> Tell me what you think, and if you need more
> information please let me know.  I have tested the code
> with caching turned on and off and unlimited and it
> seems to work pretty good with the application we're
> writing to JBoss.  Let me know if you think these
> changes are relevant.
>
> Regards,
> Bill
>
>
> ----------------------------------------------------------------------
>
> You can respond by visiting:
> http://sourceforge.net/tracker/?func=detail&atid=376687&aid=414023&group_id=22866
>
> _______________________________________________
> Jboss-development mailing list
> [EMAIL PROTECTED]
> http://lists.sourceforge.net/lists/listinfo/jboss-development
>


_______________________________________________
Jboss-development mailing list
[EMAIL PROTECTED]
http://lists.sourceforge.net/lists/listinfo/jboss-development

Reply via email to