extract records based on Binary field

chockalingam_rsp · New User Joined: 13 Aug 2009 Posts: 43 Location: chennai

Hi ,

i want to extract records from a file from a particular position which is in binary format .

I want to check for the values for 1 ,2, 3 ,4, 5 .

I know the below syntax is wrong . I am not sure on how to give the values in binary format .

SORT FIELDS=COPY
INCLUDE COND=((2,1,BI,EQ,C'1'),OR,
(2,1,BI,EQ,C'2'),OR,
(2,1,BI,EQ,C'3'),OR,
(2,1,BI,EQ,C'4'),OR,
(2,1,BI,EQ,C'5'))

Can you please help in this .

Bill Woodger · Posted: Fri May 01, 2015 2:26 am

BI is a numeric data-type. You are using a character literal. You need to use a numeric literal.

chockalingam_rsp · New User Joined: 13 Aug 2009 Posts: 43 Location: chennai

So you mean to say if i use in the below format , this will work ?

SORT FIELDS=COPY
INCLUDE COND=((4,1,BI,EQ,1),OR,
(4,1,BI,EQ,2),OR,
(4,1,BI,EQ,3),OR,
(4,1,BI,EQ,4),OR,
(4,1,BI,EQ,5))

The input file has the data in COMP format . This is a 9 byte field that has been declared as S9(9) COMP . So it would occupy 4 bytes . I want to extract records that end with 1, 2,3,4,5 in the 9th position

So how do i need to change my INCLUDE condition ?

Bill Woodger · Posted: Fri May 01, 2015 4:58 am

Syntactically it is correct, but you haven't given enough information to know if it will work.

S9(9) isn't nine bytes, it is nine digits. The S means it has been defined as signed. Can it contain a negative value? If yes, then do you want those negative values to be included as well?

chockalingam_rsp · New User Joined: 13 Aug 2009 Posts: 43 Location: chennai

Hi Bill ,

After putting a HEX ON on the file s. i am seeing records in the below format .

09BA
CE9E
----
LL
3DD9
933A
----

09B6
CEA3
----
LR1
3DDF
9391
----

09B7
CEA7

chockalingam_rsp · New User Joined: 13 Aug 2009 Posts: 43 Location: chennai

when i use copybook the data is in the below format .

4/BI 211771397
4/BI 205025540
4/BI 206023172
4/BI 212709892

My requirement is to extracts records that are ending with 1,2,3,4,5 in the position that i have highlighted

Garry Carroll · Posted: Fri May 01, 2015 8:09 pm

If you are looking to extract records with binary fields using decimal numbers, I would think you need to get DFSORT to convert the number from BI to ZD before you extract based on the ZD field in position 9. Do this as an extension of the input record and drop the extension on output.

Garry.

chockalingam_rsp · New User Joined: 13 Aug 2009 Posts: 43 Location: chennai

Can you please let me know how to go about this ? I am not sure about the syntax that needs to be used for the conversion .

Garry Carroll · Posted: Fri May 01, 2015 8:48 pm

You might try something like...

Garry Carroll · Posted: Fri May 01, 2015 8:50 pm

... typo, sorry

Then drop bytes from 81 in the output phase based on the value I the 89th position...

Garry

chockalingam_rsp · New User Joined: 13 Aug 2009 Posts: 43 Location: chennai

Hi ,

My file is a VB file of length 10500 .

So my statement should be like this .

INREC BUILD(4,10504,15:1,4,BI,TO=ZD,LENGTH=9,10505:X)
SORT FIELDS=COPY
INCLUDE COND=((10513,1,CH,EQ,C'1'),OR,
(10513,1,CH,EQ,C'2'),OR,
(10513,1,CH,EQ,C'3'),OR,
(10513,1,CH,EQ,C'4'),OR,
(10513,1,CH,EQ,C'5'))

Correct me if i am wrong .

Bill Woodger · Posted: Sat May 02, 2015 12:39 am

For variable-length records you should do the extension at the beginning of the record, else all your variable-length records will become fixed-length records which happen to have an RDW.

It doesn't matter what order you put the commands in, DFSORT will always process them in its own order. INCLUDE/OMIT COND= comes before INREC.

There is no need to store all nine digits. If you have a sequence of numbers, test for the sequence, not individually.

chockalingam_rsp · New User Joined: 13 Aug 2009 Posts: 43 Location: chennai

hi Bill ,

I tried this , but something is wrong in my coding , . I am getting the below error .

chockalingam_rsp · New User Joined: 13 Aug 2009 Posts: 43 Location: chennai

Posting my JCL also .

//JS210R EXEC PGM=SORT
//SYSOUT DD SYSOUT=X
//SYSPRINT DD SYSOUT=X
//SYSUDUMP DD SYSOUT=X
//SORTIN DD DSN=INPUT.FILE.VB,
// DISP=SHR
//SORTOUT DD DSN=OUTPUT.FILE.EXTRACT,
// DISP=SHR
//SYSIN DD *
INREC BUILD(1,4,15,4,BI,TO=ZD,LENGTH=1,9)
SORT FIELDS=COPY
OUTFIL INCLUDE=((9,1,CH,EQ,C'1'),OR,
(9,1,CH,EQ,C'2'),OR,
(9,1,CH,EQ,C'3'),OR,
(9,1,CH,EQ,C'4'),OR,
(9,1,CH,EQ,C'5'))
BUILD=(1,4,10500)
/*
//

Bill Woodger · Posted: Mon May 04, 2015 10:00 pm

Please use the Code tags to preserve space.

You've not coded a comma before the BUILD.

chockalingam_rsp · New User Joined: 13 Aug 2009 Posts: 43 Location: chennai

Hi Bill ,

i had corrected that comma , and the JCL ran fine . But when i check the output file , i am seeing all the input records in my output file .

chockalingam_rsp · New User Joined: 13 Aug 2009 Posts: 43 Location: chennai

How to over come this issue ?

RahulG31 · Active User Joined: 20 Dec 2014 Posts: 446 Location: USA

Your 15,4 in INREC with length=1 is actually column 5. So, you need to compare column 5 and not 9.

chockalingam_rsp · New User Joined: 13 Aug 2009 Posts: 43 Location: chennai

I am confused ...

the field that i want to convert to ZD starts in the 15th position and is of length 4 bytes. In this i want to check for the 9th position

4/BI 211771397
4/BI 205025540
4/BI 206023172
4/BI 212709892

Please let me know how to correct my syntax ?

Bill Woodger · Posted: Mon May 04, 2015 11:38 pm

chockalingam_rsp · New User Joined: 13 Aug 2009 Posts: 43 Location: chennai

Nope , it not working . i am still seeing all the records .

19th is the relative position (15 + 4 bytes)

SYSIN :
INREC BUILD(1,4,19,4,BI,TO=ZD,LENGTH=1,9)
SORT FIELDS=COPY
OUTFIL INCLUDE=((5,1,CH,EQ,C'1'),OR,
(5,1,CH,EQ,C'2'),OR,
(5,1,CH,EQ,C'3'),OR,
(5,1,CH,EQ,C'4'),OR,
(5,1,CH,EQ,C'5')),
BUILD=(1,4,10500) ----------> 10500 is the total length of the record
WER108I SORTIN : RECFM=V ; LRECL= 10500; CISIZE = 12288
WER073I SORTIN : DSNAME=INPUT.FILE.VB
WER257I INREC RECORD LENGTH = 10501
WER110I SORTOUT : RECFM=VB ; LRECL= 10504; BLKSIZE= 27998
WER074I SORTOUT : DSNAME=OUTPUT.FILE.EXTRACT
WER405I SORTOUT : DATA RECORDS OUT 0; TOTAL RECORDS OUT 0
WER054I RCD IN 1512850, OUT 1512850

RahulG31 · Active User Joined: 20 Dec 2014 Posts: 446 Location: USA

Can you try this:

chockalingam_rsp · New User Joined: 13 Aug 2009 Posts: 43 Location: chennai

Bill ,

Still i am getting all the records in the output file .

SYSIN :
INREC BUILD(1,4,19,4,BI,TO=ZD,LENGTH=9,5)
SORT FIELDS=COPY
OUTFIL INCLUDE=((13,1,CH,GE,C'1'),OR,
(13,1,CH,LE,C'5')),
BUILD=(1,4,14)
WER108I SORTIN : RECFM=V ; LRECL= 10500; CISIZE = 12288
WER073I SORTIN : DSNAME=INPUT.FILE.VB
WER257I INREC RECORD LENGTH = 10513
WER238I POTENTIALLY INEFFICIENT USE OF INREC
WER110I SORTOUT : RECFM=VB ; LRECL= 10504; BLKSIZE= 27998
WER074I SORTOUT : DSNAME=OUTPUT.FILE.EXTRACT
WER405I SORTOUT : DATA RECORDS OUT 1513130; TOTAL RECORDS OUT 1513130
WER054I RCD IN 1513130, OUT 1513130

Bill Woodger · Posted: Tue May 05, 2015 12:35 am

Code tags. Code tags.

chockalingam_rsp · New User Joined: 13 Aug 2009 Posts: 43 Location: chennai

Hi Bill ,

Please find my spool output .