how to match record from 2 datasets

murmohk1 · Posted: Mon May 28, 2007 3:39 pm

miosne,

Please explain your requirement clearly. Your post is bit confusing

miosne · New User Joined: 13 Mar 2006 Posts: 63 Location: italy

I'll wanna found different records between in1 and in2.
when I found them I'll wanna keep the rec's from in1.

krisprems · Posted: Mon May 28, 2007 4:13 pm

miosne
do u want all the records in IN1 that are not in IN2?

miosne · New User Joined: 13 Mar 2006 Posts: 63 Location: italy

all the records in IN1 that are not in IN2 and the differents

t.y.

krisprems · Posted: Mon May 28, 2007 6:39 pm

miosne

In this SPLICE example, the input records from data sets FILE1 and FILE2 are separated into the following
output data sets:
? BOTH: records that appear in FILE1 and FILE2
? F1ONLY: records that only appear in FILE1
? F2ONLY: records that only appear in FILE2

Frank Yaeger · Posted: Mon May 28, 2007 10:06 pm

krish,

Would you please point to the complete Smart DFSORT Trick you're regurgitating instead of just posting part of it. In this case, it's the "Create files with matching and non-matching records" Smart DFSORT Trick at:

www.ibm.com/servers/storage/support/software/sort/mvs/tricks/

Please copy that down and use it in the future!

krisprems · Posted: Tue May 29, 2007 12:51 pm

frank
i had considered the same example from sortnew.pdf, but i could not point to any specific topic. So pasted the code here.

miosne · New User Joined: 13 Mar 2006 Posts: 63 Location: italy

My system does not support "inrec overlay"

miosne · New User Joined: 13 Mar 2006 Posts: 63 Location: italy

My English is not good.
If I explain myself badly you have patience.
If I am too much insistent,it is because I never do not succeed to explain to me well.

miosne · New User Joined: 13 Mar 2006 Posts: 63 Location: italy

in short I have 2 files that they contain records that they arrive from vsam with a sort therefore to key parity.
I must obtain all the records that they have been modified, in accidental way inside of all the rows.

infinitely thanks who has helped me to write and you that you have much patience

Frank Yaeger · Posted: Tue May 29, 2007 9:02 pm

Frank Yaeger · Posted: Tue May 29, 2007 9:11 pm

dick scherrer · Posted: Tue May 29, 2007 11:19 pm

Hello miosne,

Please post a few records from in1 and in2 and what you want for output if those few records were the entire input for a run. If you need to "match" records from the 2 datasets, please mention what position the "match key" begins in and how long it is. Also, what is the lrecl of each the files? Are they FB or VB or ?

This

miosne · New User Joined: 13 Mar 2006 Posts: 63 Location: italy

The position of the key is 1,9.
The lrecl is 450 FB.
The part to verify begins to the position 10.
Here some examples of the two files:
IN1
<---key--><----------data------------>
000000001simone
000000002antonio miraglia
000000003monti
000000004roberto rossi
000000005norberto marzagalli
000000006cristiano castelli

IN2
<---key--><----------data------------>
000000001simone lissi
000000002antonio miraglia
000000003aldo monti
000000004roberto rossi

OUT
<---key--><----------data------------>
000000001simone
000000003monti
000000005norberto marzagalli
000000006cristiano castelli

Comparison the first one with the second and I must obtain the variations to key parity (key 000000001 and 000000003)
moreover the eventual ones in more present in the first one (key 000000005 and 000000006)

Many thanks and having a good day from Italy

miosne

krisprems · Posted: Wed May 30, 2007 8:35 pm

miosne
Please check this code

Frank Yaeger · Posted: Wed May 30, 2007 9:39 pm

miosne,

Kriprems is on the right track, but you don't need to create three output data sets given that you only need one. Here's a modified DFSORT/ICETOOL job that will create just the one data set you want:

miosne · New User Joined: 13 Mar 2006 Posts: 63 Location: italy

while I hope you having a good day.
watch the source below
//IN1 DD DSN=&&ENFTR02,DISP=(OLD,DELETE)
//IN2 DD DSN=MIFENDPN.PROD.ENFD00S.OLD,DISP=SHR
//T1 DD DSN=&&ENFTR03,UNIT=SYSDA,DISP=(,PASS),
// SPACE=(CYL,(5,5),RLSE),
// DCB=(RECFM=FB,LRECL=452)
//OUT12 DD DSN=MIFENDDN.SVIL.ENSI012.OUT12,DISP=(,CATLG),
// SPACE=(CYL,(5,1),RLSE),
// DCB=(RECFM=FB,LRECL=450)
//OUT1 DD DSN=MIFENDDN.SVIL.ENSI012.OUT11,DISP=(,CATLG),
// SPACE=(CYL,(5,1),RLSE),
// DCB=(RECFM=FB,LRECL=450)
//OUT2 DD DSN=MIFENDDN.SVIL.ENSI012.OUT22,DISP=(,CATLG),
// SPACE=(CYL,(5,1),RLSE),
// DCB=(RECFM=FB,LRECL=450)
//TOOLIN DD *
COPY FROM(IN1) TO(T1) USING(CP01)
COPY FROM(IN2) TO(T1) USING(CP02)
SPLICE FROM(T1) TO(OUT12) ON(1,9,CH) ON(10,440,CH) WITH(10,440) -
WITH(452,1) KEEPNODUPS USING(CP03)
/*
//CP01CNTL DD *
OUTREC FIELDS=(1,450,451:C'11')
/*
//CP02CNTL DD *
OUTREC FIELDS=(1,450,451:C'22')
/*
//CP03CNTL DD *
OUTFIL FNAMES=OUT12,INCLUDE=(451,2,CH,EQ,C'12'),BUILD=(1,450)
OUTFIL FNAMES=OUT1,INCLUDE=(451,2,CH,EQ,C'11'),BUILD=(1,450)
OUTFIL FNAMES=OUT2,INCLUDE=(451,2,CH,EQ,C'22'),BUILD=(1,450)
//*

the return code is 00 the response i dont understand because I found out1 and out2 empty and all the records in out12.

you have good day and salutes to all

miosne

Frank Yaeger · Posted: Thu May 31, 2007 8:24 pm

miosne,

You don't have MOD for T1 as you should. I don't know if that's the only problem.

You might want to try adapting my job as its more efficient for what you said you wanted to do.

miosne · New User Joined: 13 Mar 2006 Posts: 63 Location: italy

I do not want to adapt your job to mine.
I ask excuse.
I have only wants to make in order to learn .
I will be more careful, and I will try to follow more your indications.

Thank you very much, miosne.