IBM Mainframe Forum

Hi,

I have below requirement. My file is:
EMP-ID;JULIANDATE;XXXXXX;

Now I want to sum up the records where emp-id and julian date matches and to keep only the first record in output with the count.

Example:
EB111111111;2011068;xxxxxx,yyyyyy,zzzzz,dddddd ------>FIRST RECORD
EB111111111;2011068;xxxxxx,yyyyyy,zzzzz,dddddd ------>SECOND RECORD
EB111111111;2011069;xxxxxx,yyyyyy,zzzzz,dddddd ------->THIRD RECORD
EB122222222;2011068;xxxxxx,yyyyyy,zzzzz,dddddd ------->FOURTH RECORD

THE OUTPUT FILE SHOULD BE:

EB111111111;2011068;xxxxxx,yyyyyy,zzzzz,dddddd,02 ------>FIRST RECORD
EB111111111;2011069;xxxxxx,yyyyyy,zzzzz,dddddd,01 ------->THIRD RECORD
EB122222222;2011068;xxxxxx,yyyyyy,zzzzz,dddddd,01 ------->FOURTH RECORD

I TRIED THE BELOW SORT:

SORT FIELDS=(1,11,CH,A,13,7,CH,A),EQUALS
SUM FIELDS=(70,2,PD)

BUT I AM GETTING SOC7 I ALSO TRIED SUM FIELDS=(70,2,BI) BUT GETTING RC=04 AND IMPROPER RESULTS WITH DATA OVERFLOW ISSUE.

CAN ANYONE PLEASE SUGGEST HOW TO RESOLVE AND GET THE OUTPUT AS NEEDED.

Hi,

In order to get the desired result I did the following. I am creating my input file by using OUTREC so when I am doing OUTREC I did the following:

SORT FIELDS=COPY
OUTREC FIELDS=(1:1,12,13:24,56,69:1C';',70:1C'1',71:10X)
END

I am purposefully putting 1 at position 70.

Now once the file got created I used the below as second step:

SORT FIELDS=(1,11,CH,A,13,7,CH,A),EQUALS
SUM FIELDS=(70,1,ZD)
END

I got a RC=04 with the message ICE152I 0 OVERFLOW DURING SUMMATION - RC=4. I got the desired output and created my file.

The above method works but looks improper to get RC=04 and its temporary fix. Can anyone help me in resolving how to get it done in proper way. I belive I can do all in one single step and multiple steps are not needed.

Please help..I need this urgent.

Hello kpargao.

You posted in the SYNCSORT forum, but based on the ICE152I message, you are running DFSORT. I am going to move your question to the appropriate forum.

Regards,

kpargao,

You need to use INREC instead of OUTREC. Use the following DFSORT JCL which will give you the desired results



The output from this job is

THANKS KOLUSU IT WORKED AND I GOT THE DESIRED OUTPUT WITHOUT ANY OVERFLOW.

Hi,

When you remove the EQUAL from the below code then also the output is same.

Then please can you explain me what is the use of EQUAL in JCL while sorting records.

//SYSIN DD *
INREC BUILD=(1,68,C';01',80:X)
SORT FIELDS=(1,19,CH,A),EQUALS
SUM FIELDS=(70,2,ZD)
//*

Thanks,
Vivek.

There is no EQUAL in JCL. If you are referring to sort control cards then refer to sort control cards - not JCL. Also, this topic concluded over 2 years ago. Please start a new topic in future when you are not replying to a current topic. You can always use a reference to the topic (cut and paste the URL into your post and use the URL tags. Also, use the code tags (use the search facility to search for info about them) when posting "screen-type stuff".

Refer to the sort manual for info about EQUALS. You could also search this section of the forum - again using the search function. Did you search the manual/internet/forum before posting? If not - you should have - read the rules.

Finally - Welcome to the forum!!!! (may not seem like it after that lot but now you know for the future).

vvkswm wrote:Hi,

When you remove the EQUAL from the below code then also the output is same.

Then please can you explain me what is the use of EQUAL in JCL while sorting records.

vvkswm,

Just because you removed the EQUALS from the JCL doesn't mean that you are actually running with NOEQUALS. Your shop might have set it as a DEFAULT. Look at your sysout to see if EQUALS is set as DEFAULT. You can find that on ICE128I. Also remember that the sample input shown above just has 2 duplicates. If you really want to see the difference between EQUALS and NOEQUALS you need to play with input that has more duplicates.

Hi,

I tried the Below JCL with NOEQUALS.Then also the output is same as EQUALS.

Please can you explain me the use of EQUALS in jcl while sorting records.

CODE:

//STEP0100 EXEC PGM=SORT
//SYSOUT DD SYSOUT=*
//SORTIN DD *
----+----1----+----2----+----3----+----4----+----5-
EB111111111;2011068;WWWWWW,YYYYYY,ZZZZZ,DDDDDD
EB111111111;2011068;XXXXXX,YYYYYY,ZZZZZ,DDDDDD
EB111111111;2011069;XXXXXX,YYYYYY,ZZZZZ,DDDDDD
EB111111111;2011070;XXXXXX,YYYYYY,ZZZZZ,DDDDDD
EB122222222;2011068;XXXXXX,YYYYYY,ZZZZZ,DDDDDD
EB122222222;2011068;XXXXXX,YYYYYY,ZZZZZ,DDDDDD
EB122222222;2011069;XXXXXX,YYYYYY,ZZZZZ,DDDDDD
//SORTOUT DD SYSOUT=*
//SYSIN DD *
INREC BUILD=(1,68,C';01',80:X)
SORT FIELDS=(1,19,CH,A),NOEQUALS
SUM FIELDS=(70,2,ZD)
/*

OUTPUT:

----+----1----+----2----+----3----+----4----+----5- ;01
EB111111111;2011068;wwwww,YYYYYY,ZZZZZ,DDDDDD ;02
EB111111111;2011069;XXXXXX,YYYYYY,ZZZZZ,DDDDDD ;01
EB111111111;2011070;XXXXXX,YYYYYY,ZZZZZ,DDDDDD ;01
EB122222222;2011068;XXXXXX,YYYYYY,ZZZZZ,DDDDDD ;02
EB122222222;2011069;XXXXXX,YYYYYY,ZZZZZ,DDDDDD ;01

If your records are identical, how are you going to tell there is any difference between EQUALS and NOEQUALS?