Re: CPYF Command Error? -- MIDRANGE-L

Hi, Jerry:

Well, my well-aged (if not well-used) degree is in Psychology:  the 
scientific study of the behavior of flatworms, white mice, pigeons, and 
college sophomores.

I think this is not so much a case of logic as of expectations. I am 
accustomed to working with overrides and library lists, because the ERP II 
package on which our system is based has different libraries for different 
accounting entities, with duplicate files in each library as required. In 
addition, we have a set of files in yet more separate libraries, used by 
the programming staff for development. Control is mostly through job 
description library lists and OVRDBFs. In our environment, hardcoded 
library specifications are often explicitly deprecated; but even when they 
would not create problems, we are in the habit of avoiding them. So for 
me, the precedence taken by an override seems perfectly natural -- the 
"only" way to do it. It is puzzling why there are *exceptions* to this 
rule.

However, for someone who, for whatever reason, is accustomed to working 
with programs that specify libraries explicitly, expecting the hardcoded 
library specification to rule the roost would seem just as natural as the 
other way seems to me.

As I know you understand, it boils down to knowing how CPYF actually 
works. The "why" is peripheral.

Darrell

Darrell A. Martin  -  754-2187
Manager, Computer Operations
dmartin@xxxxxxxxxxxxx



Jerry Adams <jerry@xxxxxxxxxxxxxxx> 
Sent by: midrange-l-bounces@xxxxxxxxxxxx
07/05/2006 03:35 PM
Please respond to
Midrange Systems Technical Discussion <midrange-l@xxxxxxxxxxxx>


To
Midrange Systems Technical Discussion <midrange-l@xxxxxxxxxxxx>
cc

Subject
Re: CPYF Command Error?






Thanks, guys.  I usually do as you suggest, Darrell.  But this was a 
"simple" short shelf life CL so, I guess, I subconsciously (or 
unconsciously) decided to take a short cut.  Just never gave it much 
thought before and never [to my knowledge .-)] encountered it before. 


I did try to find a reference to this behavior in the CL manual but 
didn't find anything.


I'll defer to the consensus that this is normal behavior.  But it still 
doesn't seem logical.  However, we'll just chalk that one up to being a 
history major. 


                 * Jerry C. Adams
*IBM System i Programmer/Analyst
B&W Wholesale Distributors, Inc.* *
voice
                 615.995.7024
fax
                 615.995.1201
email
                 jerry@xxxxxxxxxxxxxxx <mailto:jerry@xxxxxxxxxxxxxxx>



Darrell A Martin wrote:

Jerry:

This is expected behavior. The override takes precedence over the library

specification.

I usually set each override immediately before the point of its intended 
use. I then delete it immediately after that point. At least, I do so

when

there is any chance that I will be processing a different instance of the

file at a different point in the program, and if I suspect that it could 
ever become an ease of maintenance problem later on.

Darrell

Darrell A. Martin  -  754-2187
Manager, Computer Operations
dmartin@xxxxxxxxxxxxx



Jerry Adams <jerry@xxxxxxxxxxxxxxx> 
Sent by: midrange-l-bounces@xxxxxxxxxxxx
07/05/2006 02:41 PM
Please respond to
Midrange Systems Technical Discussion <midrange-l@xxxxxxxxxxxx>


To
Midrange-L <midrange-l@xxxxxxxxxxxx>
cc

Subject
CPYF Command Error?






I encountered something new this afternoon.  Seems like a bug to me but, 
before I report it, I wanted to know if it is, indeed, a bug or anomaly.


I wrote a CL that sets up a test file.  The CL contains:

   OVRDBF   FILE(FILEA)  TOFILE(LIBA/FILEA)

A conversion program runs.  Then I have a CPYF command:

CPYF       FROMFILE(LIBB/FILEB)                         +
            TOFILE(LIBC/FILEA) MBROPT(*REPLACE)               +
            FMTOPT(*MAP *DROP) 


The anomaly (to me, anyway) is that, even though the CPYF's TOFILE 
explicitly names the library as LIBC, it was the LIBA file that got the 
copy.  Once I put a DLTOVR FILEA between the conversion program and the 
CPYF command things went as I expected (i.e., LIBC/FILEA received the 
data).


Am I in error to expect an explicit definition (TOFILE(LIBC/FILEA) in 
this case) to take precedence over an OVRDBF?  I mean, if I had simply 
said TOFILE(FILEA) I could understand what happened; i.e., the override 
being used instead of necessarily searching the library list.


Thanks.