Re: SYSCOLUMNS & SYSCOLUMN2

On Tue, Nov 26, 2019 at 10:18 AM Jon Paris <jon.paris@xxxxxxxxxxxxxx> wrote:

I'm trying to build utility routines that can work through a trigger buffer - or journal image and identify the individual fields that have been modified.

You should have led with that. How you are looking at the data matters.

I used the term table because even if created with DDS it is still a table that can be accessed with SQL. It's origin is irrelevant to my purpose.

You may not think the origin is relevant to your purpose (which you
didn't reveal until now), but it still may be relevant nonetheless. If
the only tables you were working with were created via SQL, then you
wouldn't be talking to us about the possibility of *MDY dates.

As it stands, what you really care about is how the data is *rendered*
in a trigger buffer or journal image. I personally don't know what
those look like, but if they were built to match what comes out of
DSPFFD, then that seems like the obvious thing to use.

There is a lot of overlap between what you can do with SQL, and what
you can do with DDS and the "traditional" midrange facilities. But
there are areas where they're different, and you're bound to bump up
against them, all the more so when you are working at a fairly low
level (as with the utilities you're writing). And that is what I meant
when I said IBM *could* expose more DDS-specific features through SQL,
but they've chosen not to.

John Y.