|
View previous topic :: View next topic
|
| Author |
Message |
surya4ug
New User
Joined: 08 Jul 2008
Posts: 62
Location: chennai
|
|
|
|
Hello,
I have an input file of LRECL=175,RECFM=FB, this way:
| Code: |
----+----1----+----2----+----3----+----4----+----5----+----6----+----7----+
VERSION: 20080403
PRODUCT - ABC ENTITY 1234
ACCOUNT
00000000000000123456789
00000000000000111334556
00000000000000101334556
00000000000000117657889
PRODUCT - DEF ENTITY 4568
ACCOUNT
00000000000000198876789
00000000000000111332424
00000000000000101324242
00000000000000117687676
PRODUCT - GHI ENTITY 9012
ACCOUNT
00000000000000182648789
00000000000000122472848
00000000000000129823647
00000000000000113497326
|
The records will continue like this and we don't really know how many such series of ACCOUNT, PRODUCT, ENTITY combination records exist in the file. The format, like i mentioned above is uniform though.
I need the output file to be:
| Code: |
----+----1----+----2----+----3----+----4----+----5----+----6----+----7
ACCOUNT PRODUCT ENTITY DATE
00000000000000123456789 ABC 1234 20080403
00000000000000111334556 ABC 1234 20080403
00000000000000101334556 ABC 1234 20080403
00000000000000117657889 ABC 1234 20080403
00000000000000198876789 DEF 4568 20080403
00000000000000111332424 DEF 4568 20080403
00000000000000101324242 DEF 4568 20080403
00000000000000117687676 DEF 4568 20080403
00000000000000182648789 GHI 9012 20080403
00000000000000122472848 GHI 9012 20080403
00000000000000129823647 GHI 9012 20080403
00000000000000113497326 GHI 9012 20080403
|
Please help. I'm trying parallely in this way
-> If we find the word PRODUCT, set a SEQNUM in some position in the file. Similarly for the word ENTITY set another SEQNUM in another position
-> the tricky part is, to fetch the value that is right beside the words PRODUCT, ENTITY and write them by the side of the account numbers as shown in the output file, for that particular PRODUCT,ENTITY group.
When i say group, it is like this:
| Code: |
PRODUCT - ABC ENTITY 1234
ACCOUNT
00000000000000123456789
00000000000000111334556
00000000000000101334556
00000000000000117657889 |
-> For the next group,the corresponding PRODUCT, ENTITY values need to be picked up. |
|
| Back to top |
|
 |
Arun Raj
Moderator
Joined: 17 Oct 2006
Posts: 2482
Location: @my desk
|
|
|
|
Hello,
Here's one way of achieving this in SyncSort. I assumed an input LRECL of 80.
| Code: |
//STEP0100 EXEC PGM=SORT
//SYSOUT DD SYSOUT=*
//T1 DD DSN=&&T1,DISP=(NEW,PASS)
//T2 DD DSN=&&T2,DISP=(NEW,PASS)
//S1 DD DSN=&&S1,DISP=(NEW,PASS)
//SORTIN DD DSN= Input file ---> FB/80
//SYSIN DD *
OMIT COND=(1,1,CH,EQ,C'A')
INREC IFTHEN=(WHEN=INIT,OVERLAY=(81:SEQNUM,8,ZD,START=0)),
IFTHEN=(WHEN=(45,1,CH,EQ,C'P'),OVERLAY=(81:SEQNUM,8,ZD)),
IFTHEN=(WHEN=(45,1,CH,EQ,C' '),OVERLAY=(89:SEQNUM,8,ZD,
81:81,8,ZD,SUB,89,8,ZD,M11,LENGTH=8))
SORT FIELDS=COPY
OUTFIL FNAMES=T1,INCLUDE=(45,1,CH,EQ,C' '),BUILD=(1,88)
OUTFIL FNAMES=T2,INCLUDE=(45,1,CH,EQ,C'P'),BUILD=(1,88)
OUTFIL FNAMES=S1,SAVE,BUILD=(30,3,C',''',39,8,C'''',80:X)
/*
//STEP0200 EXEC PGM=SORT
//SYSOUT DD SYSOUT=*
//SORTOUT DD DSN= Output file ---> FB/80
//SORTJNF1 DD DSN=&&T1,DISP=(OLD,PASS)
//SORTJNF2 DD DSN=&&T2,DISP=(OLD,PASS)
//SYMNAMES DD DSN=&&S1,DISP=(OLD,PASS)
//SYSIN DD *
JOINKEYS FILE=F1,FIELDS=(81,8,A),SORTED
JOINKEYS FILE=F2,FIELDS=(81,8,A),SORTED
REFORMAT FIELDS=(F1:1,23,F2:55,3,F2:70,4)
INREC BUILD=(1,23,29:24,3,47:27,4,61:VER)
SORT FIELDS=COPY
OUTFIL REMOVECC,
HEADER1=('ACCOUNT',29:'PRODUCT',47:'ENTITY',61:'DATE') |
|
|
| Back to top |
|
|
surya4ug
New User
Joined: 08 Jul 2008
Posts: 62
Location: chennai
|
|
|
|
Hello Arun,
I didnot understand the purpose of the following piece of code, in this context. Could you pls explain me:
| Code: |
| OUTFIL FNAMES=S1,SAVE,BUILD=(30,3,C',''',39,8,C'''',80:X) |
|
|
| Back to top |
|
|
surya4ug
New User
Joined: 08 Jul 2008
Posts: 62
Location: chennai
|
|
|
|
Never mind, Arun. Thats for the version. I got it  |
|
| Back to top |
|
|
surya4ug
New User
Joined: 08 Jul 2008
Posts: 62
Location: chennai
|
|
|
|
Hmm..i didnot imagine that this will be a hinderance, but this is how my input file looks like actually.....
| Code: |
----+----1----+----2----+----3----+----4----+----5----+----6----+----7----+
**********
VERSION: 20080403
REPORT ID: ABCDEF
PRODUCT - ABC ENTITY 1234
ACCOUNT PREV CURR
00000000000000123456789 05/08/00 05/03/01 ----12345--
00000000000000142435556 01/03/06 03/06/02 ***!34552--
00000000003131101334556 06/09/08 01/03/09 27488***
00000000003243536334556
.................................................
04/03/08 ***ABCDEFGHIJK****
22:50:39
PRODUCT - DEF ENTITY 4568
ACCOUNT PREV CURR
12456789 01/08/00 05/03/01 ----1122D--
17127224 09/08/00 06/03/01 ----QEFSF--
10138742 01/08/09 05/08/01 ----AFSFSF--
11134416 05/08/03 02/03/01 ----DFDCASR--
.................................................
04/03/08 ***ABCDEFGHIJK****
22:50:39 |
I tried executing the code you posted and got S013 abend code ( UNSUCCESSFUL SORT ) for the 2nd step. May be because i didnot mention the input file properly, based on which you
wrote the sort card, it would have failed. Noteably,
a) There is data below the PRODUCT, ENTITY tags and beside the account numbers. It is not space, like i mentioned before
b) a dotted line followed by timestamp appears after each of the so called 'group'
| Code: |
.................................................
04/03/08 ***ABCDEFGHIJK****
22:50:39 |
c) Account numbers differ in length from group to group and the records can have spaces between them, like in the input file mentioned above.
I'm trying to change the code accordingly. if you want me to make any adjustments to the previously posted code, please let me know.
Thanks! |
|
| Back to top |
|
|
enrico-sorichetti
Superior Member
Joined: 14 Mar 2007
Posts: 10902
Location: italy
|
|
|
|
| Quote: |
| Hmm..i didnot imagine that this will be a hinderance, but this is how my input file looks like actually..... |
if You knew from the beginning,
can You explain why in h**l You asked for something,
and after somebody has replied with a solution,
only then You come up with the final requirement  
it just shows no respect for people spending their own time testing and providing solutions for Your needs
do You behave in the same way when dealing with Your manager??? |
|
| Back to top |
|
|
surya4ug
New User
Joined: 08 Jul 2008
Posts: 62
Location: chennai
|
|
|
|
It isn't that way Enrico.
i tried changing the fileformat to what was 'desired', but did i ever say that the code that was posted here willnot help? I guess you are under a wrong assumption that i treat my manager and others in a different way....... I have a lot of respect for people out here and even if some one lashes out at me because i didnot make something clear, im fine, because you guys are a great bunch of people and are helping us out time and again. So apologies for assuming something and confusing you.
if its possible for you to help me out, i'll be more than happy. if you are not able to also, it is fine because you tried your level best to help me out already. |
|
| Back to top |
|
|
surya4ug
New User
Joined: 08 Jul 2008
Posts: 62
Location: chennai
|
|
|
|
Hi Arun,
I wrote a syncsort card to arrive 'almost' at the first input file, from the second input file i posted previously. Here it is:
| Code: |
INREC IFTHEN=(WHEN=INIT,BUILD=(1:2,23,45:61,13,63:118,15)),
IFTHEN=(WHEN=(1,6,ZD,EQ,NUM),OVERLAY=(1,25,56X)),
IFTHEN=(WHEN=(1,7,CH,EQ,C'ACCOUNT'),OVERLAY=(9:72X'40')),
IFTHEN=(WHEN=(45,7,CH,EQ,C'PRODUCT'),OVERLAY=(44X'40',45,36)),
IFTHEN=(WHEN=(45,7,CH,NE,C'PRODUCT',AND,
1,23,CH,NE,C'ACCOUNT ',
AND,1,6,ZD,NE,NUM),OVERLAY=(1:80X'40')),
IFTHEN=(WHEN=(1,7,CH,EQ,C'-------'),OVERLAY=(1:80X'40'))
SORT FIELDS=COPY
OUTFIL OMIT=(1,1,CH,EQ,C' ',AND,45,7,CH,NE,C'PRODUCT'),
BUILD=(1,80)
|
But the problem i face is that there is lot of other data in the second input file from position 1,25. So, even though i considered all records with numeric data in position 1,6 and also those with the tag ACCOUNT in the above sort card and eliminated those that didnot meet these conditions, iam seeing some alphanumeric data in the output that has first 6 digits as numeric and the rest being alphabetic.
Here is a sample of what i got as output
| Code: |
570877068 ABCDEG
570877068 BCUDEH
570877068 JHISINU
PRODUCT - ABC ENTITY 1234
ACCOUNT
00000000059101000000001
00000000059101000000002
PRODUCT - DEF ENTITY 4568
ACCOUNT
00000000059101000000003
00000000059101000000004 |
It should only be:
| Code: |
PRODUCT - ABC ENTITY 1234
ACCOUNT
00000000059101000000001
00000000059101000000002
PRODUCT - DEF ENTITY 4568
ACCOUNT
00000000059101000000003
00000000059101000000004 |
So, is there a way to check for consecutive alphabets ( any number of ) in a field using SYNCSORT? That will help me get rid of the unwanted data in the output. Upon overcoming this problem, i will be able to use the sort card you posted.
Please guide me. |
|
| Back to top |
|
|
surya4ug
New User
Joined: 08 Jul 2008
Posts: 62
Location: chennai
|
|
|
|
Hi Arun,
Please let me know if there is a way to accomplish this task. Your inputs would be very valuable. |
|
| Back to top |
|
|
surya4ug
New User
Joined: 08 Jul 2008
Posts: 62
Location: chennai
|
|
|
|
| Just to make it clear..The consecutive alphabetic characters could start anywhere with in the field and end up anywhere ,They need not necessarily be towards the end or beginning of the field. |
|
| Back to top |
|
|
dick scherrer
Moderator Emeritus
Joined: 23 Nov 2006
Posts: 19243
Location: Inside the Matrix
|
|
| Back to top |
|
|
surya4ug
New User
Joined: 08 Jul 2008
Posts: 62
Location: chennai
|
|
|
|
Excellent, it worked!! Thank you Dick and Arun for your help in accomplishing an important task. As always, i've reveived very timely help from you guys.
I'll ensure from now that the final file creation criteria is mentioned clearly right from the beginning of the topic. |
|
| Back to top |
|
|
Arun Raj
Moderator
Joined: 17 Oct 2006
Posts: 2482
Location: @my desk
|
|
|
|
| You're welcome. Would you mind posting the final solution here? |
|
| Back to top |
|
|
surya4ug
New User
Joined: 08 Jul 2008
Posts: 62
Location: chennai
|
|
|
|
Sure Arun, here we go....Initially when you posted the code, we picked VERSION from the top of the file to fetch the date value adjacent to it. Now, i tweaked the logic bit to fetch the date from some other place of the input file.
I guess we would have accomplished the task thru EZYTRIEVE with a bit of lesser effort and coding, but certainly, i prefer SYNCSORT and believe me, its a lot faster and just took about 0.8 mins to run all the 4 steps, as opposed to nearly 5-6 mins thru EZYTRIEVE. This is considerable difference because, i need to execute the same piece of code for nearly 360 input files.
| Code: |
//*************************************************************
//**** FORMAT THE INITIAL FILE TO REMOVE UNWANTED TAGS ****
//**** AS WE WE NEED ONLY ACCOUNT INFORMATION HERE ****
//*************************************************************
//STEP1 EXEC PGM=SORT
//SYSOUT DD SYSOUT=*
//SORTIN DD DISP=SHR,DSN=MY INPUT FILE
//SORTOUT DD DSN=&&OUTPUT FILE1,
// DISP=(NEW,CATLG,DELETE)
//SYSIN DD *
INREC IFTHEN=(WHEN=INIT,BUILD=(1:6,23,28:65,13,45:122,15,64:64,8,
73:51,8)),
IFTHEN=(WHEN=(1,6,ZD,EQ,NUM),OVERLAY=(1,25,35X,62:64,8,3X,73:73,8)),
IFTHEN=(WHEN=(1,7,CH,EQ,C'ACCOUNT'),OVERLAY=(9:72X'40')),
IFTHEN=(WHEN=(28,7,CH,EQ,C'PRODUCT'),OVERLAY=(27X'40',28,32,21X)),
IFTHEN=(WHEN=(28,7,CH,NE,C'PRODUCT',AND,
1,23,CH,NE,C'ACCOUNT ',
AND,1,6,ZD,NE,NUM),OVERLAY=(1:80X'40')),
IFTHEN=(WHEN=(1,7,CH,EQ,C'-------'),OVERLAY=(1:80X'40'))
SORT FIELDS=COPY
OUTFIL OMIT=(1,1,CH,EQ,C' ',AND,28,7,CH,NE,C'PRODUCT'),
BUILD=(1,80)
/*
//*************************************************************
//**** INCASE ANY 'CUSTOMER' LEVEL DATA APPEARS IN THE ****
//**** FILE WHEN THE 'FIRST 6 DIGIT NUMERIC' INCLUSION ****
//**** DATA CHECK FAILS, THIS STEP WILL REMOVE ALL THE ****
//**** RECORDS HAVING 4 CONSECUTIVE ALPHABETS STARTING ****
//**** FROM 7TH POSITION THRU 23RD IN THE OUTPUT OF ****
//**** STEP1 ****
//*************************************************************
//STEP2 EXEC PGM=ICEMAN
//SYSOUT DD SYSOUT=*
//SYMNAMES DD *
CHECK,'ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz'
//SORTIN DD DISP=SHR,DSN=&&OUTPUT FILE1
//SORTOUT DD DSN=&&OUTPUT FILE2,
// DISP=(NEW,CATLG,DELETE),
//SYSIN DD *
OPTION COPY
OMIT FORMAT=SS,
COND=(7,1,EQ,CHECK,&,
8,1,EQ,CHECK,&,
9,1,EQ,CHECK,&,
10,1,EQ,CHECK,OR,
*
8,1,EQ,CHECK,&,
9,1,EQ,CHECK,&,
10,1,EQ,CHECK,&,
11,1,EQ,CHECK,OR,
*
9,1,EQ,CHECK,&,
10,1,EQ,CHECK,&,
11,1,EQ,CHECK,&,
12,1,EQ,CHECK,OR,
*
10,1,EQ,CHECK,&,
11,1,EQ,CHECK,&,
12,1,EQ,CHECK,&,
13,1,EQ,CHECK,OR,
*
11,1,EQ,CHECK,&,
12,1,EQ,CHECK,&,
13,1,EQ,CHECK,&,
14,1,EQ,CHECK,OR,
*
12,1,EQ,CHECK,&,
13,1,EQ,CHECK,&,
14,1,EQ,CHECK,&,
15,1,EQ,CHECK,OR,
*
13,1,EQ,CHECK,&,
14,1,EQ,CHECK,&,
15,1,EQ,CHECK,&,
16,1,EQ,CHECK,OR,
*
14,1,EQ,CHECK,&,
15,1,EQ,CHECK,&,
16,1,EQ,CHECK,&,
17,1,EQ,CHECK,OR,
*
15,1,EQ,CHECK,&,
16,1,EQ,CHECK,&,
17,1,EQ,CHECK,&,
18,1,EQ,CHECK,OR,
*
16,1,EQ,CHECK,&,
17,1,EQ,CHECK,&,
18,1,EQ,CHECK,&,
19,1,EQ,CHECK,OR,
*
17,1,EQ,CHECK,&,
18,1,EQ,CHECK,&,
19,1,EQ,CHECK,&,
20,1,EQ,CHECK,OR,
*
18,1,EQ,CHECK,&,
19,1,EQ,CHECK,&,
20,1,EQ,CHECK,&,
21,1,EQ,CHECK,OR,
*
19,1,EQ,CHECK,&,
20,1,EQ,CHECK,&,
21,1,EQ,CHECK,&,
22,1,EQ,CHECK,OR,
*
20,1,EQ,CHECK,&,
21,1,EQ,CHECK,&,
22,1,EQ,CHECK,&,
23,1,EQ,CHECK)
/*
//*************************************************************
//**** FORMAT THE OUTPUT OF STEP2 BY ADDING SEQUENCE NUM ****
//**** TO RECOGNISABLE TAGS LIKE 'PRODUCT' etc TO ****
//**** SEGREGATE THE FILE TO 'PRODUCT' TAG FILE AND A FILE****
//**** WITH RECS NOT HAVING A 'PRODUCT' TAG ****
//*************************************************************
//STEP3 EXEC PGM=SORT
//SYSOUT DD SYSOUT=*
//T1 DD DSN=&&T1,DISP=(NEW,PASS),
// SPACE=(CYL,(1500,1400),RLSE),BLKSIZE=0,
// VOL=(,,,90)
//T2 DD DSN=&&T2,DISP=(NEW,PASS),
// SPACE=(CYL,(1500,1400),RLSE),BLKSIZE=0,
// VOL=(,,,90)
//SORTIN DD DSN=&&OUTPUT FILE2,DISP=SHR
//SYSIN DD *
OMIT COND=(1,7,CH,EQ,C'ACCOUNT')
INREC IFTHEN=(WHEN=INIT,OVERLAY=(81:SEQNUM,8,ZD,START=0)),
IFTHEN=(WHEN=(28,1,CH,EQ,C'P'),OVERLAY=(81:SEQNUM,8,ZD)),
IFTHEN=(WHEN=(28,1,CH,EQ,C' '),OVERLAY=(89:SEQNUM,8,ZD,
81:81,8,ZD,SUB,89,8,ZD,M11,LENGTH=8))
SORT FIELDS=COPY
OUTFIL FNAMES=T1,INCLUDE=(28,1,CH,EQ,C' '),BUILD=(1,88)
OUTFIL FNAMES=T2,INCLUDE=(28,1,CH,EQ,C'P'),BUILD=(1,88)
/*
//*************************************************************
//**** JOIN THE SEQUENCE NUMS ADDED TO THE OUTPUT FILES IN****
//**** STEP3 AND GENERATE THE FINAL INFOPAC FILE WITH ****
//**** ACCOUNT,PRODUCT,ENTITY,CURRENT MAINTENANCE DATE, ****
//**** PREVIOUS MAINTENANCE DATE ****
//*************************************************************
//STEP4 EXEC PGM=SORT
//SYSOUT DD SYSOUT=*
//SORTOUT DD DSN=&&OUTPUT FILE3,
// DISP=(NEW,CATLG,DELETE)
//SORTJNF1 DD DSN=&&T1,DISP=(OLD,PASS)
//SORTJNF2 DD DSN=&&T2,DISP=(OLD,PASS) \
//SYSIN DD *
JOINKEYS FILE=F1,FIELDS=(81,8,A),SORTED
JOINKEYS FILE=F2,FIELDS=(81,8,A),SORTED
REFORMAT FIELDS=(F1:1,23,F2:38,3,F2:52,8,F1:62,8,F1:73,8)
INREC BUILD=(1,23,29:24,3,40:27,8,50:35,8,70:43,8,30X)
SORT FIELDS=(1,23,CH,A)
SUM FIELDS=NONE
OUTFIL REMOVECC,
HEADER1=('ACCOUNT',29:'PRODUCT',40:'ENTITY',
50:'CURR MAINT DATE',70:'PREV MAINT DATE')
/* |
Thanks again! |
|
| Back to top |
|
|
dick scherrer
Moderator Emeritus
Joined: 23 Nov 2006
Posts: 19243
Location: Inside the Matrix
|
|
|
|
Good to hear it is working - thank you for letting us know and posting your solution
d |
|
| Back to top |
|
|
|
|
  |
All times are GMT + 6 Hours |
|
