Yeah, if the datasource itself is capable of doing it, then I recommend going with that approach. Beware of underlying drivers that don't support it though.
Cheers, Clinton On Sun, Aug 23, 2009 at 2:46 PM, Rick <ric...@gmail.com> wrote: > So just for my own understanding... If I configured a datasource that > uses pstmt caching (which for example c3p0 can do on a per-connection > basis, which isn't a standard implementation I know), can I just not > even bother opening a Session in iBATIS with ExecutorType.REUSE and > I'd still get caching implemented by the underlying datasource? Sounds > like I'm best not even worrying about it and just use openSession(). > > > On Sun, Aug 23, 2009 at 4:32 PM, Clinton Begin<clinton.be...@gmail.com> > wrote: > > iBATIS only caches at the connection level anyway. The broad support for > > this is lacking, so I just avoided the problem. > > > > 99% of the gain with psmt caching is in loops in a single thread > anyway.... > > > > Clinton > > > > On Sun, Aug 23, 2009 at 2:23 PM, Rick <ric...@gmail.com> wrote: > >> > >> I thought that most datasources allow you to cache preparedStatements > >> at the datasource level, so I'm confused what affect using > >> sqlSessionFactory.openSession(ExecutorType.REUSE) would be if you > >> have, or don't have, your datasource set to cache pstmts and you call > >> (and not call) openSession with ExecutorType.REUSE ? > >> > >> -- > >> Rick R > >> > >> --------------------------------------------------------------------- > >> To unsubscribe, e-mail: user-java-unsubscr...@ibatis.apache.org > >> For additional commands, e-mail: user-java-h...@ibatis.apache.org > >> > > > > > > > > -- > Rick R > > --------------------------------------------------------------------- > To unsubscribe, e-mail: user-java-unsubscr...@ibatis.apache.org > For additional commands, e-mail: user-java-h...@ibatis.apache.org > >
