Overcoming CRTDUPOBJ limitation

This is just a wild guess, and I'm hopeful that someone else will have a
better idea...

How about using CRTLF for file Z using the DTAMBRS parameter?  (Is that what
you meant by "recreating from the source"? vs. qualifying the library on the
PFILE keywords in the DDS?)

What if you first CDO'd......... nope, skip that

Hmmm, a toughie at best!  Grasping at straws here, but,  Have you tried
OVRDBF?  Do you have a window of time where you could rename the Library1/A
file to something else while the CDO for file Z runs?  If it takes a long time
to CDO, you could rename it back to A right after the CDO starts.

Contrived or not, it does seem like creating the temporary LF source using
file APIs may be your only choice here.  I'd be interested in how you solve
this.

BTW, CDO is my lazy shorthand for CrtDupObj.

Dan Bale
IT - AS/400
Handleman Company
248-362-4400  Ext. 4952

-------------------------- Original Message --------------------------
Hi all

I'm trying to get round a limitation in CRTDUPOBJ as hopefully this
scenario illustrates:

Library1 has physical files A and B, and logical files X and Y.
Library2 has physical file C and logical Z.
File X is a join logical over both physicals A and B.
File Y is a logical over physical B.
File Z is a join logical over both physicals A and C.

I want to create a test environment (in Library3) with all these files.

1) CRTDUPOBJ OBJ(A) FROMLIB(LIBRARY1) OBJTYPE(*FILE) TOLIB(LIBRARY3)
2) CRTDUPOBJ OBJ(B) FROMLIB(LIBRARY1) OBJTYPE(*FILE) TOLIB(LIBRARY3)
3) CRTDUPOBJ OBJ(C) FROMLIB(LIBRARY2) OBJTYPE(*FILE) TOLIB(LIBRARY3)
4) CRTDUPOBJ OBJ(X) FROMLIB(LIBRARY1) OBJTYPE(*FILE) TOLIB(LIBRARY3)
5) CRTDUPOBJ OBJ(Y) FROMLIB(LIBRARY1) OBJTYPE(*FILE) TOLIB(LIBRARY3)
6) CRTDUPOBJ OBJ(Z) FROMLIB(LIBRARY2) OBJTYPE(*FILE) TOLIB(LIBRARY3)

At V4R3M0 1) to 5) completed and 6) failed. At V4R4M0 6) now completes
but logical Library3/Z points to Library1/A, not Library3/A. This is in
keeping with the documentation, but I want Library3/Z to point to A & C
in Library 3 (not much of a test environment otherwise ;-) ).

I've tried deleting and recreating (ADDLFM) the logical file member to
point to the correct file, but it fails. A save/restore doesn't do the
trick either. The only option I know that works is to recreate from the
source. On our development box that isn't a problem, as we have the code
for virtually all our software but it's a bit cumbersome. In a setting
where the source member isn't available I can only think of creating one
(from the the file APIs) and compiling against that. That seems even more
contrived.

Can anyone point out a solution I've missed out (other than a better DB
structure - I'm not able to do anything about that as yet :-( ). I'm
after a generic solution I can include in my DBG/400 software for other
people in the same situation - I guess I'll go with the create source
method if I can't avoid it.

Regards, Martin
--
martin@dbg400.net / jamaro@firstlinux.net
http://www.dbg400.net DBG/400 - DataBase Generation utilities
Free test environment tools for the AS/400 and misc other stuff
+---
| This is the Midrange System Mailing List!
| To submit a new message, send your mail to MIDRANGE-L@midrange.com.
| To subscribe to this list send email to MIDRANGE-L-SUB@midrange.com.
| To unsubscribe from this list send email to MIDRANGE-L-UNSUB@midrange.com.
| Questions should be directed to the list owner/operator: david@midrange.com
+---