RE: RE: CRTDUPOBJ of file with trigger renders trigger inoperative

midrange-l-request@xxxxxxxxxxxx wrote:

>   9. RE: CRTDUPOBJ of file with trigger renders trigger
>      inoperative (Wilt, Charles)
>
>>From the infocenter and various online articles.  It seems CRTDUPOBJ
>works on files with triggers.  But IBM is saying it doesn't work with an
>SQL defined file that has an SQL defined trigger.  So does it work on a
>DDS defined file with an SQL trigger, or a SQL defined file with a RPG
>trigger?
>
>Why the difference?
>
>Not only does it not make sense, it seems to violate the principles that
>and SQL defined table is the same object as a DDS defined physical file.
>
>Heck, this is supposed to be an object oriented OS.  Shouldn't the
>object no how to duplicate itself?


Perhaps the problem is that it _does_ (attempt to) duplicate itself.

That is, what it really needs to do is duplicate a kind of template of itself 
and then make all of the adjustments needed for that template to operate within 
its new context as if the whole context had been duplicated with it.

Sometimes duplication isn't sufficient.

I need to dig up the details, but a general example that is similar went 
something like --

I had a program that made a bunch of API calls back in V4R3. When V5R1 came 
around and the program was first restored in that new environment, it blew 
pretty badly. Why? Because IBM had moved a couple APIs out of one service 
program into another. When the system tried to locate the procedures in the 
same service program after restore as in Version 4, well, "symbol not found" 
errors showed up. The procedures were moved either into or out of one of the 
QP0* service programs, can't recall the details now.

The APIs themselves were the same as they had been from when they were first 
available. The procedure definitions were unchanged. But the same binding 
couldn't happen when my program was restored to a new system. (Pretty 
unpleasant surprise.)

I wouldn't be surprised if the same kind of 'context' problem might happen when 
duplicating combinations of SQL objects.

Just musing.

Tom Liotta