> - is the performance hit sigificant?

Significant is relative.

> could the problem be solved by the particular application
> in other ways. That is: Why is an unsaved generator a
> problem in this instance?


> (Bearing in mind that the
> transaction that last used it is going to be rolled back.)

NOT TRUE. Read the thread. The record can be inserted and the generator not be updated. The program crashes and the generator is now the wrong value. My customers are experiencing this problem and causing frustration.

> I am not suggesting that the problem isn't real, but I am
> suggesting that the problem is not one that would effect most
> applications.

"Most" would be anything less than 50% would be true for that statement. If your QA program only fixed those problems that affect most of your customers, you are going to have a lot of crappy software that no one will like. (Most of the customers don't use XE2, but should Jason spend time working on that? Most of the customers don't probably don't use TIB_Cursor, so if there is a bug in it, should he fix it? Your logic is poor customer support!).

> So if the problem is truly application specific
> then so should the solution be.

It isn't application specific--Kevin is getting it to. And I can reproduce it in a small application. (I'm not sure how my customers are getting the problem but they are seeing it).

Why do you have the nerve to wait months before jumping into a conversation and suggest, after I have done hours and hours of work, to suggest it isn't a big problem? I would not have spent the time of creating the demo app if the problem wasn't significant.

Ed Dressel