Re: Save library and restore with no data in files

An earlier suggestion on this thread was completely ignored, yet, I believe
it is the best answer, Why not just code up a DDL that will recreate the
objects, including all the dependencies etc, and run that from SQL
scripts?

It A) documents the database structure as well as can be done, better if
you include commentary, and B) it's fast and easy. Another benefit is as
changes are made if you include the SQL statements in your build script,
you've now documented the changes. You can even get the system to generate
the SQL for you. Some simple editing and you're done.

I guess I'm just confused about why this was not considered more
significantly. We all talk about modernization then when an opportunity
like this comes along to actually do it, no one seems interested.

--
Jim Oberholtzer
Chief Technical Architect
Agile Technology Architects


On Thu, May 12, 2022 at 12:13 PM Sam_L <lennon_s_j@xxxxxxxxxxx> wrote:

FWIW: I suspect a CLFPFM is fairly efficient. What is probably more
inefficient is the time to load the data and build the indexes, which
you then immediately dispose of.

Might it be more effective to do CRTDUPOBJ without data to a new library
on the source system, save that, then restore to the target system.

Sam

On 5/11/2022 2:35 PM, Steve McKay wrote:

Yes. It appears that I will have to CLRPFM after I restore. I thought I
remembered a parm to prevent data restore but I guess it was CRTDUPOBJ.

7