MVSFORUMS.com

aniscorp · Beginner Joined: 24 Apr 2006 Posts: 14 Topics: 3

1. The READ PREV syntax is not supported by the compiler and I have to read the previous records in a VSAM file. Can anyone suggest an option.

2. Secondly, the START key does not support "<=" combination. Is there a way out to direectly point to less than or equal to the value of the key.

kolusu · Posted: Mon Apr 24, 2006 8:31 am Post subject:

aniscorp · Beginner Joined: 24 Apr 2006 Posts: 14 Topics: 3

Kolusu,

The version of cobol is:
1PP 5668-958 IBM VS COBOL II Release 4.0 09/15/92

And compilers are:
PROC GROUP...... CIIG1B01
PROCESSOR....... GBNSCBB1

The errors in compile listing is as follows:
1. For Start <= key:
519 IGYPS2106-S "<=" was found in the "START" statement. The statement was dhe "START" statement. The statement was discarded.

2. For READ PREV key
526 IGYPS2072-S "PREV" was invalid. Skipped to the next verb, period or proc Skipped to the next verb, period or procedure-name definition.

And the Code is:

000519 000821 START INDEX-FILE KEY NOT <= INDX-KEY
000520 000822 INVALID KEY
000521 000823 MOVE INDEX-NOT-FOUND-MSG-01
000522 000824 TO INDX-COMM-INDX-ERROR-MSG
000523 000825 MOVE 9 TO INDX-COMM-PROCESS-IND
000524 000826 GO TO 10000-START-INDEX-EXIT.
000525 000827
000526 000828 READ INDEX-FILE PREV RECORD
000527 000829 AT END
000528 000830 MOVE INDEX-NOT-FOUND-MSG-04
000529 000831 TO INDX-COMM-INDX-ERROR-MSG
000530 000832 MOVE 9 TO INDX-COMM-PROCESS-IND
000531 000833 MOVE 'Y' TO INDX-FOUND-SW.

Mervyn · Posted: Tue Apr 25, 2006 4:38 am Post subject:

aniscorp,

Your compiler does indeed support <=, but it doesn't support "NOT <=".
You should use ">" instead. Check here:

http://publibz.boulder.ibm.com/cgi-bin/bookmgr_OS390/BOOKS/IGYL1101/3.33?DT=19930312093006

I think you do have a problem with PREVIOUS.
_________________
The day you stop learning the dinosaur becomes extinct

aniscorp · Beginner Joined: 24 Apr 2006 Posts: 14 Topics: 3

sorry seems that there was a typo in last line while posting the syntax, it is:
>>--START--file-name-1--------------------------------------------------

>-----------------------------------------------------------------------
+-KEY------------EQUAL----------------------------------data-name-1-+
+-IS-+

Mervyn · Posted: Tue Apr 25, 2006 8:24 am Post subject:

You've lost me. Confused

Are you saying the code you posted was not the actual code that caused the problem?
_________________
The day you stop learning the dinosaur becomes extinct

aniscorp · Beginner Joined: 24 Apr 2006 Posts: 14 Topics: 3

Mervyn,

Please ignore my last post, as I might have confused you. But here is the complete summary of the problem:

The code which is in my cobol program is as follows:
START INDEX-FILE KEY NOT <INDX>=, or key NOT <. But there is no mention of Key <=.

Because of this constrain of the START key unable to use the <= condition, the job in the production takes around 10 hours to run and we are in the process of optimizing the same. I had tried all possible methods, i.e increasing the BUFSP, BUFNI, Region etc, but nothing works.
Since this file is searched 3 times each by 9 million records and as there is no direct read possible due to the START key constrain, the program reads one by one record until the <= condition is met.

I hope this explanation would have helped in giving a clear picture of the actual problem.

Regards,
Aniscorp

aniscorp · Beginner Joined: 24 Apr 2006 Posts: 14 Topics: 3

I guess that after posting, the text has again got rearranged and before it is going to create another confusion, Iam re-sending it thru Qucik reply as below:

Mervyn,

Please ignore my last post, as I might have confused you. But here is the complete summary of the problem:

The code which is in my cobol program is as follows:
START INDEX-FILE KEY NOT <INDX>=, or key NOT <. But there is no mention of Key <=.

Because of this constrain of the START key unable to use the <= condition, the job in the production takes around 10 hours to run and we are in the process of optimizing the same. I had tried all possible methods, i.e increasing the BUFSP, BUFNI, Region etc, but nothing works.
Since this file is searched 3 times each by 9 million records and as there is no direct read possible due to the START key constrain, the program reads one by one record until the <= condition is met.

I hope this explanation would have helped in giving a clear picture of the actual problem.

Regards,
Aniscorp

aniscorp · Beginner Joined: 24 Apr 2006 Posts: 14 Topics: 3

slade · Posted: Sat Apr 29, 2006 6:39 pm Post subject:

Hi Aniscorp,

You say the prod version takes 10 hrs+ because of the START cmd restrictions. Why don't you show us the code it uses to get around them?
_________________
Regards, Jack.

"A problem well stated is a problem half solved" -- Charles F. Kettering

aniscorp · Beginner Joined: 24 Apr 2006 Posts: 14 Topics: 3