[opencms-dev] To the developers team

[Alexander Kandzior]

Dear Santander,

regarding your questions

> I heard in the OpenCMS news and this list that for the next 
> version of OpenCMS the old pool will be replaced with another 
> pool based on the pool of apache foundation. Well, my questions are:
> - Why not OpenCms uses the pool of the application server 
> instead of use a own pool? Creating a own pool is like 
> "re-invent" the wheel, isn´t it?

AFAIK there is no common standard for DB pooling on the application server
level. This would mean that the DB configuration during setup would be even
more difficult (and different for every servlet engine). 

What we actually did is that we used the Apache Jakarta commons-dbcp package
and the commons-pool package and replaced the old proprietary pool
implementation with these. Actually the same packaes are used by Tomcat for
it's datasources. That's not exactly the same as re-inventing the wheel. 

> - To get (in a easy way) that OpenCMS work in any database, 
> Why not to implement the database access of OpenCMS in a 
> persistence engine like hibernate, wich can use its own pool 
> or the application server pool?

We thought about that and did dome testing with Apache Torque, but the
effort would have been to great. 

Best Regards,
Alex.

Alexander Kandzior
Alkacon Software - The OpenCms Experts
http://www.alkacon.com