Gerhard and all,
I have been kicking the sample around. I have been working with these
control statements using IEBGENER:
V1 = " GENERATE MAXNAME=2,MAXGPS=1"
V2 = " MEMBER NAME=DRSSKEY"
/* V3 = "GROUP1 RECORD IDENT=(8,'FIRSTMEM',1)" */
V3 = "GROUP1 RECORD IDENT"
V4 = " MEMBER NAME=KEYLRS"
My input file is a sequential file. My output is a PDS. Both are
predefined. Generate is required for a PDS. MAXNAME is set to 2 because I
want to create two members: DRSSKEY and KEYLRS with two MEMBER statements.
MAXGRP is set to 1 because I only have 1 GROUP statement. Not sure what GROUP1
does for me, it looks like it's just a label.
Member name DRSSKEY works great. I can "iebgener" a flat file to a PDS
member.
However, I 'm lost on the IDENT. It talks about editing. If I just key
"GROUP1 RECORD IDENT" it wants more. If I give it the 'FIRSTMEM' it appears to
not match and stops with only DRSSKEY created in the PDS from the flat file.
IDENT identifies the last record of a collection of records in the input data
set.
"The first RECORD statement (GROUP1) identifies the last record to be placed
in the first member. The name of this record (FIRSTMEM) appears in the first
eight positions of the input record." This looks to be evaluating what's in my
flat file.
The remaining MEMBER and RECORD statements define the second and third
members. There is no RECORD statement associated with the third MEMBER
statement, the remainder of the input file will be loaded as the third member.
Do I need to know what's in my input file? I just want the whole flat file
as a member (or two) in a PDS. If this does "editing" I don't know what to do
with that.
Q). How can I code the IDENT so it "matches" and creates my second member
in the PDS? Yes, I could run this process twice (or do a copy to get my second
member).
Thank you, Dave
-----Original Message-----
From: IBM Mainframe Discussion List [mailto:[email protected]] On Behalf
Of Gerhard Adam
Sent: Wednesday, March 19, 2014 12:02 PM
To: [email protected]
Subject: Re: Help with IEBCOPY?
From the DFSMSdfp Utilities manual.
In this example, a partitioned data set (consisting of three members) is
created from sequential input.
//TAPEDISK JOB ...
//STEP1 EXEC PGM=IEBGENER
//SYSPRINT DD SYSOUT=A
//SYSUT1 DD DSNAME=INSET,UNIT=tape,LABEL=(,SL),
// DISP=(OLD,KEEP),VOLUME=SER=001234
//SYSUT2 DD DSNAME=NEWSET,UNIT=disk,DISP=(,KEEP),
// VOLUME=SER=111112,SPACE=(TRK,(10,5,5)),
// DCB=(RECFM=FB,LRECL=80,BLKSIZE=2000)
//SYSIN DD *
GENERATE MAXNAME=3,MAXGPS=2
MEMBER NAME=MEMBER1
GROUP1 RECORD IDENT=(8,'FIRSTMEM',1)
MEMBER NAME=MEMBER2
GROUP2 RECORD IDENT=(8,'SECNDMEM',1)
MEMBER NAME=MEMBER3
/*
-----Original Message-----
From: IBM Mainframe Discussion List [mailto:[email protected]] On Behalf
Of Paul Gilmartin
Sent: Wednesday, March 19, 2014 9:55 AM
To: [email protected]
Subject: Re: Help with IEBCOPY?
On Wed, 19 Mar 2014 09:45:26 -0700, Gerhard Adam wrote:
>You're using the wrong utility. You want IEBGENER.
>
Maybe.
>-----Original Message-----
>Behalf Of Hansen, Dave L - Eagan, MN
>Sent: Wednesday, March 19, 2014 9:41 AM
>
>
> I tried "C I=INDD1,O=OUTDD1" and "S M=(DRSSKEY,KEYLRS)". I would
>like the flat file to be copied into two PDS members in the same
>dataset. If I can get one member created that would be nice.
>
What's the format of your flat file? How are the two members delimited.
> The first message FCO710A says there is more information in FCO700A.
>
> The second message FCO701A says it's trying to read from the "unloaded"
>dataset.
>
Anything that's not a PDS is presumed to have been unloaded by IEBCOPY.
> Q). Can I use IEBCOPY to copy a flat file to a PDS?
>
Only if it has been unloaded by IEBCOPY.
> Q). I am running a REXX exec using IEBCOPY. Would you recommend a
>different approach for my REXX exec?
>
IEBGENER? I don't know the full capabilities of IEBGENER. Can it stop some
known delimiter?
At worst, read records with EXECIO and write with EXECIO.
Someone will surely suggest DFSORT or ICEUTIL. Not in my skill set.
-- gil
----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions, send email to
[email protected] with the message: INFO IBM-MAIN
----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions, send email to
[email protected] with the message: INFO IBM-MAIN
----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to [email protected] with the message: INFO IBM-MAIN
