Hi, My main objection about om.* classes was that it does not look very natural. What I understand about these classes is that the main purpose is to abstract the primary key type. Even if the primary key is a number we can get/set the string value and vice-versa. But to achieve this a developer has to learn a new API. When I know the data-type why should I bother about getting a wrapper? For example java.util.List has a method called get(i). But for a linked list, it is not a natural process to access elements randomly. Then why is it in the base class? "what can be done" is different from "what is natural".
My another objection was IdGenerator id and instance. As far as I understand about DB class that it wraps database specific stuffs - like date string, escape character etc. So instead of returning the IdGenerator id, it can return an implementation of IdGenerator. That is instead of String getIDMethodType(), we can have a method like IdGenerator getIdGenerator(). I apologize if these are some trivial issues. Thanks, Rana Bhattacharyya. __________________________________________________ Do You Yahoo!? Yahoo! Finance - Get real-time stock quotes http://finance.yahoo.com -- To unsubscribe, e-mail: <mailto:[EMAIL PROTECTED]> For additional commands, e-mail: <mailto:[EMAIL PROTECTED]>
