Super C is giving the following error, how to rectify ?

kabyab · Posted: Wed Jun 06, 2007 7:53 pm

I' m using Superc(ISRSUPC) in a JCL and need to compare specific columns. In order to do that, I have coded the following:

//SYSIN DD *
CMPCOLM 1:2,21:01,23:04,27:04,31:04,35:04,39:04,43:05,48:08,56:04,
CMPCOLM 60:03,63:05,68:04,72:02,74:05,79:05
/*

The second line of CMPCOLM is given as it would not fit into first line.

I am getting the below warning in the o/p file

THE FOLLOWING PROCESS STATEMENTS (USING COLUMNS 1:72) WERE PROCESSED:
CMPCOLM 1:2,21:1
ISRS031W AN INVALID START COLUMN VALUE WAS SPECIFIED.

ISRS078I FALSE MATCHES CORRECTED. RESULTS MAY NOT REFLECT ALL MATCHES. SEE "ADDITIONAL MANUAL ON SUPERC"

Please let me know how to correct it

Phrzby Phil · Posted: Wed Jun 06, 2007 8:43 pm

CPMCOLM want <first>:<last>, not <first>:<length>

kabyab · Posted: Thu Jun 07, 2007 12:47 pm

Thank you very much for your reply. It worked fine.
But it can make only 15 comparison. But I have a requirement for 17 comparisons.

CMPCOLM
1:2,21:21,22:22,23:26,27:30,31:34,35:38,39:42,43:47,48:55,56:59
CMPCOLM 60:62,63:67,68:71,72:73

The above works fine

But my requirement is as below
CMPCOLM 1:2,21:21,22:22,23:26,27:30,31:34,35:38,39:42,43:47,48:55,56:59,
60:62,63:67,68:71,72:73,74:78,79:83

Please help.

Phrzby Phil · Posted: Thu Jun 07, 2007 6:28 pm

You can certainly consolidate your contiguous columns, e.g., 21:27.

I see no reason to split these, even if to you they are different fields.

With SORT parms, though, I like to separate contiguous fields as a documentation aid, so maybe that's your thinking.

Nonetheless, with the 15 restriction, you have no choice.

Anuj Dhawan · Posted: Thu Jun 07, 2007 8:44 pm

Hi,

Why don't you try this,

CMPCOLM 1:2 21:83

I hope you can replace your CMPCOLM arguments with the above.

kabyab · Posted: Fri Jun 08, 2007 2:03 pm

Thanks for your reply Phil and Anuj.

I felt both solution( (21:27 by Phil) , (21:83 (Anuj)) suggested were the same , as it is contigous from 21:83 . I tried both and got the same output

But output of SUPERC and COMPAREX were not the same.

There is record count mismatch between COMPAREX output and SUPERC

I need to convert COMPAREX to SUPERC and get the same output

FYI- After SUPERC, I use a SORT step to get rid of I(insert) which SUPERC normally writes to the o/p file. Ignore D(Delete) which is also written to o/p file as insert are my concern. So the final output will have only data without summary, I or D.

Please help.

Anuj Dhawan · Posted: Fri Jun 08, 2007 3:48 pm

kabyab · Posted: Fri Jun 08, 2007 5:18 pm

Hi,

There is no error messages as such. Only there is record count mismatch.

Job of SuperC
//SUPERC EXEC PGM=ISRSUPC,
//* PARM='CHNGL'
// PARM=('DELTAL','LINECMP',
//* 'NOSUMS NOPRTCC',
//* '',
// '')
//NEWDD DD DSN=file1,
// DISP=SHR
//OLDDD DD DSN=file2,
// DISP=SHR
//SYSIN DD *
CMPCOLM 1:2,21:83
/*
//OUTDD DD DISP=(NEW,CATLG),
// DSN=file3,
// RECFM=FB,LRECL=91,
//* MGMTCLAS=SHRTBKUP,
// SPACE=(CYL,(500,50),RLSE),
// UNIT=DISK
//*
//SORT1 EXEC PGM=SORT
//SORTIN DD DSN=file3,
// DISP=SHR
//*
//SORTOUT DD DSN=file4,
// DISP=(NEW,CATLG,DELETE),DCB=(RECFM=FB,LRECL=91),
// SPACE=(CYL,(500,50),RLSE),
// UNIT=DISK
//SORWK01 DD UNIT=SYSDA,
// SPACE=(CYL,(1,1),RLSE)
//SORWK02 DD UNIT=SYSDA,
// SPACE=(CYL,(1,1),RLSE)
//SYSOUT DD SYSOUT=*
//SYSIN DD *
SORT FIELDS=COPY
INCLUDE COND=(2,3,SS,EQ,C'I -')
INREC FIELDS=(6,91)
/*

Corresponding Comparex Job
//step1 EXEC PGM=COMPAREX
//*
//SYSUT1 DD DISP=(OLD,KEEP,KEEP),
// DSN=file1

//SYSUT2 DD DISP=SHR,
// DSN=file2

/
//SYSIN DD DISP=SHR,
// DSN=&CNTLIB(CTL1)
//
//SYSUT3 DD DISP=(NEW,CATLG),
// DSN=file3,
// DCB=( RECFM=FB,LRECL=91),
// MGMTCLAS=
// SPACE=
//
//*
//*
CTL1 expands to
HALT=COND
DATA
FIELD=(01,02,C)
KEY=(03,18,C,A)

others fileds upto 83
COPYDIFF
FORMAT=15
KILLRC=NO
DASH=C'*'

I/p file sample

2367890123244343 xx......................... .... ..................
23789012334356 xx......................... .... ..................
235766767676767 xx......................... .... ..................
234124141234314 xx......................... .... ..................
231234141341241 xx......................... .... ...Ø..øê..........

The output required is difference i.e present in new file but not in old file

The sysout message of SUPERC speaks more about the SORT step
(Let me know incase you need it)

Thanks.

dick scherrer · Posted: Fri Jun 08, 2007 7:50 pm

Hello,

You will get better replies if you post all of the sysout messages fro both runs (not all of the records, just the messages).

kabyab · Posted: Mon Jun 11, 2007 3:58 pm

Hi Everyone,
Thank you very much for all your help, finally I managed to get the comparison right.
I had to include the KEY=(03,18,C,A) parameters as part of the CMPCOLM.
Instead of using CMPCOLM 1:2,21:83,
we used CMPCOLM 1:2,3:83 and got the results right.

I have one more query.
In the SUPERC JCL as already given in my previous post, I am using SORT to remove the prefixed 'I -' character from the Super C output dataset.

//SORT1 EXEC PGM=SORT
//SORTIN DD DSN=file3,
// DISP=SHR
//SORTOUT DD DSN=file4,
// DISP=(NEW,CATLG,DELETE),DCB=(RECFM=FB,LRECL=91),
// SPACE=(CYL,(500,50),RLSE),
// UNIT=DISK
//SORWK01 DD UNIT=SYSDA,
// SPACE=(CYL,(1,1),RLSE)
//SORWK02 DD UNIT=SYSDA,
// SPACE=(CYL,(1,1),RLSE)
//SYSOUT DD SYSOUT=*
//SYSIN DD *
SORT FIELDS=COPY
INCLUDE COND=(2,3,SS,EQ,C'I -')
INREC FIELDS=(6,91)
/*

Is there any SuperC parameters that we can use instead of SORT to achieve the same?
The parameters I am using for SUPERC are:

//SUPERC EXEC PGM=ISRSUPC,
// PARM=('DELTAL','LINECMP',
// '')