[MDB2] new code

Hi, well after the proxy idea didnt really worked out as I had hoped I started to play with some other concepts. Well while I was messing with the code I did end up persuing the idea of result objects instead of returning resources. While benchmarking I noticed that the various intval($result) are really expensive. So it turns out that returning objects as results is cheaper than having to do all those intval's to internally store metadata about the result sets. This also opens up the possibility of making specialized result set objects for buffered and unbuffered result sets. Anyways in terms of performance we are now really close to ADODB for simple query-fetch scenarios. And that even though MDB has the optional datatype abstraction and the possibility to specify the fetch method (ordered or assoc) in the fetch method. I currently have two different types of result set classes: Buffered and Unbuffered. I am not yet 100% sure if that is the way to go, but it seems a bit more clean and might even improve performance noticeably on RDBMS's that don't have internal buffering mechanisms and where the user doesn't want buffer to happen at all. Regards, Lukas Smith smith@backendmedia.com _______________________________ BackendMedia www.backendmedia.com berlin@backendmedia.com Linn Zwoch Smith GbR Pariser Str. 44 D-10707 Berlin Tel +49 30 83 22 50 00 Fax +49 30 83 22 50 07