I'm a little new to this discussion, but Byron Foster (who has been immersed in this a while) is a colleague of mine.
I agree that save() methods be finalized. It seems bad that user code should have the ability to weasel itself in during the almighty save operation that will confuse it. In general, what sort of work/thoughts has gone into handling graphs (with the possibility of cycles) in torque? russ ----- Original Message ----- From: "John McNally" <[EMAIL PROTECTED]> To: "Turbine Developers List" <[EMAIL PROTECTED]> Sent: Wednesday, November 21, 2001 8:43 AM Subject: Re: Torque preSave revisited > Sorry I did not get back to you earlier on this. It seems like most of > the operations you describe should be implemented in the Base* classes. > I am a bit unclear on how adding a preSave method, which I assume would > need to have some generated code in it to handle the logic you suggest, > would work better than just putting the code into the save method. Can > you give an example of how it would solve the infinite recursion > problem? > > The preSave method was added by me briefly into cvs as I was > experimenting with adding security to the bo's. My idea did not pan out > and the method was never used by me and hopefully anyone else. I backed > it out as soon as jason started down the road to move turbine2.2 onto > standalone torque. > > john mcnally > > -- To unsubscribe, e-mail: <mailto:[EMAIL PROTECTED]> For additional commands, e-mail: <mailto:[EMAIL PROTECTED]>
