Copy content to new location.

ramsri · Active User Joined: 18 Oct 2008 Posts: 380 Location: India

Hi,

File - RECFM=FB, LRECL=90

Input:

Naish · New User Joined: 07 Dec 2006 Posts: 82 Location: UK

Craq Giegerich · Posted: Fri Jul 13, 2012 5:25 pm

You need to explain the rules of getting from input to output a little better since the character in position 30 seems to vary.

ramsri · Active User Joined: 18 Oct 2008 Posts: 380 Location: India

The rule is to copy the word having XXX to 30th position irrespective of its position in input.

Thanks.

Craq Giegerich · Posted: Fri Jul 13, 2012 5:37 pm

Naish · New User Joined: 07 Dec 2006 Posts: 82 Location: UK

dick scherrer · Posted: Fri Jul 13, 2012 6:45 pm

Hello,

You need to start over and post what is really wanted. Your posts are rather inconsistent and do not explaikn what you are trying to accomplish.

You show a "rule" that has 5 Xs, data that has 4 and 5 Xs, data that has some Xs as well as some leading character, and finish with:

ramsri · Active User Joined: 18 Oct 2008 Posts: 380 Location: India

by the time I realized that I wanted add more info to it the EDIT did not allow me because 10 mintues were over..... The word containing XXXX from each input record should be moved to 30th column is what the SORT should achieve. That is a total of 6 characters including the numeral at the last should be moved to new position. That is why I used

Naish · New User Joined: 07 Dec 2006 Posts: 82 Location: UK

By the way you got the right output (pun intended).

Anyway, assuming that you have such an input, suggest that you PARSE into 4 fields and then search in each of the PARSED field and then OVERLAY accordingly.

Hope this helps.

dick scherrer · Posted: Fri Jul 13, 2012 7:46 pm

Hello,

Bill Woodger · Posted: Fri Jul 13, 2012 8:04 pm

Does your value always end with XXXX1?

To put it another way, can you describe the required field accurately. Is the file FB? Length>?

ramsri · Active User Joined: 18 Oct 2008 Posts: 380 Location: India

Bill, its LRECL=90, RECFM=FB.

Its a 6 character field that will have XXXX appearing in the middle. This fields first position may have different alphabet and last position will always be a numeral.

Thanks.

Bill Woodger · Posted: Fri Jul 13, 2012 9:17 pm

Four PARSEd fields, as Naish said, OVERLAY at the end of the existing record. Test the "middle" of each field for "XXXX". Personally, I'd validate the first and last bytes as well, unless feeling 100% secure about the possibility of not getting a false hitt. I wouldn't use SS looking for s substring.

Plonk the one that matches XXXX (if any) into the correct position on the record.

Assuming that this is connected to your other recent ones, are you going to be looking for multiple values?

Skolusu · Posted: Fri Jul 13, 2012 10:16 pm

ramsri,

Use the following DFSORT JCL which will give you the desired results

Bill Woodger · Posted: Sat Jul 14, 2012 6:23 am

ramsri,

Are you "piecing together" a solution with lthe ittle bits of code supplied to you?

If yes, please instead post the full requirement. There is a chance that people are wasting time with your "movable feast" descriptions.

Is X the letter X, or does it stand for "any letter of the alphabet in any position"? Will you be looking for more than one set of characters? What are you doing with the data once it is in position 30? What topics are you thinking of starting next?

ramsri · Active User Joined: 18 Oct 2008 Posts: 380 Location: India

Hi Kolusu, thanks a lot.......I got the results.

Bill, I won't extend this topic anymore

Bill Woodger · Posted: Tue Jul 24, 2012 4:33 am

Good for you, Ramsri. Thanks for the feedback.