> -----Original Message-----
> From: IBM Mainframe Discussion List [mailto:IBM-MAIN@LISTSERV.UA.EDU]
> On Behalf Of Elardus Engelbrecht
> Sent: Wednesday, October 22, 2014 3:52 PM
> To: IBM-MAIN@LISTSERV.UA.EDU
> Subject: Re: DFSORT/ICETOOL pondering
>
> Thomas Berg wrote:
>
> >Sometimes I have a need to do a selective copy of a dataset where the
> selection is both record sequence and record position/part dependent.
> >And by that I mean the record part/position is varying depending on which
> record it's about.
> >An example (yes, it looks silly):
>
> Who ever gave you homework is silly... ;-D

...it's me... :)

> >I want to copy  position 31 to 40 of all records but from record 8 and 9 I 
> >want
> to copy also (concatenate) position 71 to 80 and from record 6 to 8 I also
> (concatenate in "front") want to copy position 1 to 3.
>
> Hmmm. Hard requirement if you want all of this in ONE pass. Could you be kind
> to post an example of your input and output?

An hypotethical example (monospace font recommended):

Input:
----+----1----+----2----+----3----+----4----+----5----+----6----+----7----+----8
001aaaaaaaaaaaaaaaaaaaaaaaaaaabbbbbbbb01ccccccccccccccccccccccccccccccdddddddd01
002aaaaaaaaaaaaaaaaaaaaaaaaaaabbbbbbbb02ccccccccccccccccccccccccccccccdddddddd02
003aaaaaaaaaaaaaaaaaaaaaaaaaaabbbbbbbb03ccccccccccccccccccccccccccccccdddddddd03
004aaaaaaaaaaaaaaaaaaaaaaaaaaabbbbbbbb04ccccccccccccccccccccccccccccccdddddddd04
005aaaaaaaaaaaaaaaaaaaaaaaaaaabbbbbbbb05ccccccccccccccccccccccccccccccdddddddd05
006aaaaaaaaaaaaaaaaaaaaaaaaaaabbbbbbbb06ccccccccccccccccccccccccccccccdddddddd06
007aaaaaaaaaaaaaaaaaaaaaaaaaaabbbbbbbb07ccccccccccccccccccccccccccccccdddddddd07
008aaaaaaaaaaaaaaaaaaaaaaaaaaabbbbbbbb08ccccccccccccccccccccccccccccccdddddddd08
009aaaaaaaaaaaaaaaaaaaaaaaaaaabbbbbbbb09ccccccccccccccccccccccccccccccdddddddd09
010aaaaaaaaaaaaaaaaaaaaaaaaaaabbbbbbbb10ccccccccccccccccccccccccccccccdddddddd10


Output:
----+----1----+----2----+----3
bbbbbbbb01
bbbbbbbb02
bbbbbbbb03
bbbbbbbb04
bbbbbbbb05
006bbbbbbbb06
007bbbbbbbb07
008bbbbbbbb08dddddddd08
bbbbbbbb09dddddddd09
bbbbbbbb10


> What about the rest of input? Should they stay the same during copy or left 
> out
> of the output?

The rest is discarded.

> >I can see the use of the SUBSET (ICETOOL) for record selection but how do you
> user DFSORT/ICETOOL to solve the combining record and varying position needs
> ?
>
> What about SPLICE and JOIN?

I haven't looked more closely at these. Thought they required a matching file 
etc.

> >I now use a homegrown program for this but it's lacking efficiency to say the
> least and I'm looking at DFSORTs efficiency and speed.
>
> How so? Could you post that program or some snippets? I believe I could do 
> that
> with REXX.

That program IS coded in REXX.  It works but it's a problem with (very) large 
files.
(This program is sort of a Swiss Army Knife with lot of other functions, this 
copy problem is a minor subfunction in it.
I use it e g as a line command at ISPF 3.4 etc.)



Best Regards,
Thomas Berg
___________________________________________________________________
Thomas Berg   Specialist   zOS/RQM/IT Delivery   Swedbank AB (Publ)


----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN

Reply via email to