"You haven't stated whether your pgmA is using the API or the RCVJRNE command. "
Neither one of these. I am using the DSPJRN to *outfile command.
Thanks for ideas.
-----Original Message-----
From: midrange-l-bounces@xxxxxxxxxxxx [mailto:midrange-l-bounces@xxxxxxxxxxxx] On Behalf Of Monnier, Gary
Sent: Monday, October 01, 2012 1:54 PM
To: Midrange Systems Technical Discussion
Subject: RE: make cust file look like a journal record - format
Depends upon how often you need to transfer the entire files. If time is of the essence maybe not. You could put together the read/update program in under an hour while maintaining an audit trail of what was transferred.
You haven't stated whether your pgmA is using the API or the RCVJRNE command. If pgmA is using the API you don't really have any choice but to generate journal entries. If pgmA is using the RCVJRNE command you can spoof it but I don't know what your program or ETL process will do with duplicate sequence numbers.
But here's a way you could use the RCVJRNE command's exit program to do what you want. You will still have to write a new program though.
1. Create an F-Spec for each table being transferred.
2. Define external data structures for the tables in question.
3. Define the prototype for pgmA to match the existing pgmA interface.
4. Define a data structure to match the journal data portion of pgmA's interface.
a. Journal header stuff.
b. Journal entry specific data (this will hold your table's record contents).
4. Read each table but rather than update each record call pgmA.
a. Set the mock journal header information. You know what pgmA is looking for so I don't have any suggestions.
b. Set the JOESD to contain the corresponding external data structure's contents (JOESD = CUSTDS for example).
c. Call pgmA.
If ZC entry types are part of your system's auditing set up you're set. If not, think about how you will explain the transfer to your auditors.
This what you were looking for?
Gary
-----Original Message-----
From: midrange-l-bounces@xxxxxxxxxxxx [mailto:midrange-l-bounces@xxxxxxxxxxxx] On Behalf Of Stone, Joel
Sent: Monday, October 01, 2012 11:26 AM
To: 'Midrange Systems Technical Discussion'
Subject: RE: make cust file look like a journal record - format
Yes but -
Many of these files have millions of records in them - that is a lot of wasted disk space I would think - even if only for the several days we keep journals around.
-----Original Message-----
From: midrange-l-bounces@xxxxxxxxxxxx [mailto:midrange-l-bounces@xxxxxxxxxxxx] On Behalf Of Monnier, Gary
Sent: Monday, October 01, 2012 1:12 PM
To: Midrange Systems Technical Discussion
Subject: RE: make cust file look like a journal record - format
Joel,
All you have to do is write a program that reads then updates the file in question. An RPG program that just reads and updates the file might be faster than an SQL script.
dow (not %EOF(CUST));
read(E) CUST;
write(E) CUST;
enddo;
You don't have to change anything but by updating the record you force an entry to the journal where, you state, your program reads the journal changes. No change to your ETL are required.
Gary
> On 10/1/2012 11:19 AM, Stone, Joel wrote:
I have pgmA which reads cust journal changes and sends them off for > an ETL app.
I would also like to be able to send the entire cust file using the > same pgmA with no changes.
How can I make the cust record format look just like a journal > record (the format that DSPJRN creates).
For example:
Cust file format:
CustName 30
CustAddr 30
CustCity 20
-----
Total 80
I want this to look like a DSPJRN record so I can use the same pgmA > to load up all CUST records for an initial load.
Desired format:
Field Field
Text Name Type Length
Length of entry JOENTL S 5
---------------------|
Sequence number JOSEQN S 10
|
Journal Code JOCODE A 1
|
Entry Type JOENTT A 2
|
Date of entry: Job date f JODATE A 6
|
Time of entry: hour/minut JOTIME S 6
|---------- journal header fields
Name of Job JOJOB A 10
|
Name of User JOUSER A 10
|
Number of Job JONBR S 6
|
Name of Program JOPGM A 10
|
Name of Object JOOBJ A 10
----------------------|
.
.
.
Entry Specific Data - Var JOESD A 80
<--------------------------------- I want my CUST data here!!
Most of the journal fields can empty - but I want my CUST data to be > in JOESD. Can SQL or OPNQRYF do this?
I think I need to take fields JOENTL thru JOOBJ and initialize those.
Then take all CUST fields CUST.* and string them together (alpha,
packed, > zoned) into the one JOESD field.
Is this possible without using an HLL?
Thanks!
--
This is the Midrange Systems Technical Discussion (MIDRANGE-L) mailing list To post a message email: MIDRANGE-L@xxxxxxxxxxxx To subscribe, unsubscribe, or change list options,
visit:
http://lists.midrange.com/mailman/listinfo/midrange-l
or email: MIDRANGE-L-request@xxxxxxxxxxxx Before posting, please take a moment to review the archives at
http://archive.midrange.com/midrange-l.
________________________________________________________________________
This inbound email has been scanned for all viruses by the MessageLabs SkyScan
service.
________________________________________________________________________
______________________________________________________________________
This outbound email has been scanned for all viruses by the MessageLabs Skyscan service.
For more information please visit
http://www.symanteccloud.com
______________________________________________________________________
midrange.com
