Re: Format Level ID

Hi Kurt,

Yes, that's been my experience as well: SQL sometimes results in a different format level ID, for the same (as far as I can tell, anyway) record/field layout.

I would _not_ work around the issue with LVLCHK(*NO), except maybe as a temporary solution -- it throws away all of your safeguards.

Instead, I'd create both files with DDS, so they have the same level ID. Then, as a longer-term project, I'd eliminate the requirement for multi-member files (which are frowned on these days) and once that requirement is gone, use SQL to create both files.



On 1/31/2012 4:49 PM, Kurt Anderson wrote:

It's kind of embarrassing that I've been working with the machine
(Power, iSeries, etc) for over 10 years, but apparently missed
mastering file/format level IDs and when they change, etc.

I'm sure there's some reading I can do, but I have one situation that
I wanted to get input on.

We have a SQL file, TAXSTMTP. We also have a DDS file, CURTAXP.
CURTAXP is like a history/save file of TAXSTMTP. Our "cur" files are
multi-member, so they can't be created via SQL, which is why they are
created using DDS source.

These two files, compiled at 'the same time', one after the other,
results in the following format IDs: TAXSTMTP: 43856FE31EAC4
CURTAXP: 43856FE31EAB4

The only difference being the 2nd to last digit. Is the difference
due to one being SQL and the other DDS, or is that a red herring?

It's coming up b/c we have a program that runs with either CURTAXP or
TAXSTMTP, but it complains (level check) when trying to run over the
file it wasn't compiled over.

Our "cur" files are mostly history files, but there are some used by
programs, which is why I'm hesitant to change the file to
lvlchk(*no).