Split a file with only selected rows

GANAPATH · New User Joined: 06 Mar 2006 Posts: 35

Hi,

I have a dataset whose record length is 5000 and this keeps changing everytime we get the file. i would like to create another file from this big single row file that has only 80 record lenth and more rows instead of one big row. How can i do this.

and also is there a way that i can specify when ever u see a particular character like "table" write it in the next line. I hope iam clear

Thanks,
ganapath

amitrath · Posted: Fri Mar 31, 2006 2:45 am

ISPF EDITOR's TEXT FLOW line command provides exactly what you want.

TF?Text Flow
The TF (text flow) line command restructures paragraphs. This is sometimes
necessary after deletions, insertions, or splitting.
Syntax
TF[n]
n The column number to which the text should be flowed. The default is the
panel width when default boundaries are in effect. If you are using
nondefault bounds, the right boundary is used. This is different from the
TFLOW macro command, which always defaults to the right boundary.
If a number greater than the right boundary is specified, the right
boundary is used.
Description
To flow text:
1. Type TF in the line command area of the line at which you want the text to
begin flowing. If you want to specify the rightmost column position for the
restructured text, type a number greater than 1 immediately after the TF
command.
2. Press Enter. The text is flowed from the beginning of that line to the end of the
paragraph.

GANAPATH · New User Joined: 06 Mar 2006 Posts: 35

Hi,

This is very useful, i fixed it just like that,Thanks a lot, i have one more question, can we automate this somehow,
like i keep this in a jcl as a step and that will do the TF command instead of me opening the file and typing TF.Is this possible?

Aprreciate it.
Thanks,
Ganapath

amitrath · Posted: Wed Apr 05, 2006 1:47 am

Not sure about JCL. You cannot access ISPF editor line commands so easily through JCL.

But you can surely automate this through REXX.

superk · Posted: Wed Apr 05, 2006 2:31 am

I think a SORT would work well for you: