Hi! Regarding the row ID (and similar methods) discussion:
I found the input quite valueable, however, I still miss how it can be of value in the exact case of our identity map. The purpose of the identity map is basically to ensure that a database record (representing a persistent object) is only kept in memory once per request. The problem tried to be solved is, that different parts of an application fetch the same object and receive a copy instead of a reference. In addition, we try to introduce some caching on a higher level, so that different parts of an application do not issue the same query twice per request. The described featzures in this part of the thread might be helpful in some cases for persistent objects. But we a) cannot rely on features that are supported only by a subset of the databases we support and b) cannot change the database schema of people. Anyway, thanks for your input. I learned quite a bit about features of modern RDBMs. :) Regards, Toby -- Mit freundlichen Grüßen / Med vennlig hilsen / With kind regards Tobias Schlitt (GPG: 0xC462BC14) eZ Components Developer [EMAIL PROTECTED] | eZ Systems AS | ez.no -- Components mailing list [email protected] http://lists.ez.no/mailman/listinfo/components
