add header record to pds members

mike beckmann · Posted: Sat Feb 06, 2010 12:55 am

I have the following file and would like to for each row take positions 1-13 and add it to the pds file in positions 15-31.

superk · Posted: Sat Feb 06, 2010 2:52 am

I'm not quite sure what approach you're looking for here. But, here's a thought. Use IEBPTPCH to copy the entire PDS to a sequential dataset. Use sort to read your control file and generate a full set of sort IFTHEN statements to match the data. Then, use sort again to read the punched sequential dataset, apply the IFTHEN logic to the header "MEMBER NAME Mnnnnnn" and insert an IEBUPDTE ./ ADD NAME record followed by your header content. End the process by using the IEBUPDTE program to re-build the PDS members that now contains the header record.

Arun Raj · Posted: Sat Feb 06, 2010 8:59 am

mike beckmann,

Are you trying to attach only one header row to each of the PDS members? Or say for e.g, you want to attach the text 'RV_AC_RG_COMP' to all the rows present in M000001 ?

mike beckmann · Posted: Sat Feb 06, 2010 9:58 am

Arun Raj,

I am trying to attach one header to each of the PDS members, the header will be on the same line as the member; so in the example above,
RV_AC_RG_COMP will be the header row for the M000001 pds and
RV_ACC_RATES for M000002, RV_ACC_RG for M000003 and so on...

I could have hundreds of these to match up, I just gave three as an example. I am going to research and attempt SUPERK's suggestion but any others are welcome as well. Thank you.

mike beckmann · Posted: Wed Feb 17, 2010 12:18 am

I have attempted SUPERK's suggestion and ran into a roadblock at "apply the IFTHEN logic to the header "MEMBER NAME Mnnnnnn" "

YA.U.RV.TEST1 is the sequential dataset created from the pds and i have another file YA.U.QMFJOIN with all the IFTHEN statements. i do not know how to apply all the IFTHEN statements to the sequential dataset...

YA.U.RV.TEST1 example:

BROWSE YA.U.RV.TEST1
Command ===>
----+----1----+----2----+----3----+----4---

MEMBER NAME M000080
SELECT DISTINCT MCINET_DSNO.&RACF.GWAYIDS.*
FROM MCINET_DSNO.&RACF.GWAYIDS
WITH UR
MEMBER NAME M000081

YA.U.QMFJOIN:
********************************* Top of Data ****************************
IFTHEN=(WHEN=(9,7,CH,EQ,C'M000080'),OVERLAY=('RV_GWAY1 '))
IFTHEN=(WHEN=(9,7,CH,EQ,C'M000058'),OVERLAY=('RV_GET_IDS '))

Thanks for any help you can provide.
mbeck

gcicchet · Posted: Wed Feb 17, 2010 3:36 am

Hi Mike,

mike beckmann · Posted: Wed Feb 17, 2010 3:57 am

Sorry i will try to explain it better - below does what i want however it only replaces one 'member' - in this case it replaces M000080 with RV_GWAY1. I have a file YA.U.QMFJOIN that has 40 different IFTHEN/WHEN/OVERLAY statements that i want to incorporate into this - is there a way to substitute this file into the INREC statement somehow?

//RUNSORT4 EXEC PGM=SORT
//SYSOUT DD SYSOUT=*
//SORTIN DD DSN=YA.U.RV.TEST1,DISP=SHR
//SORTOUT DD DSN=YA.U.RV.TEST2,
// DISP=(NEW,CATLG,CATLG),
// DCB=(RECFM=FBA,LRECL=130,BLKSIZE=0),
// SPACE=(CYL,(1,1),RLSE)
//SYSIN DD *
OPTION COPY
INREC IFTHEN=(WHEN=(15,7,CH,EQ,C'M000080'),OVERLAY=(15:C'RV_GWAY1 '))
/*

dick scherrer · Posted: Wed Feb 17, 2010 4:02 am

Hello,

Possibly i am missing something, but can the entire set of SYSIN statements not simply be put into the YA.U.QMFJOIN file and then name the file in the JCL rather than DD *?

That or concatenate multiple "pieces". . .

gcicchet · Posted: Wed Feb 17, 2010 4:09 am

Hi,

if I understand you correctly, this is an example of what you need

mike beckmann · Posted: Wed Feb 17, 2010 6:19 am

dick scherrer/superk - Thank you very much!! That is exactly what i was looking to do.