How is ISPF packed data compressed? Is it simple run length encoding? Couldn't 
some clever soul write some sort of generalized un-PACK command or utility 
program that could somehow front-end other commands? Use pipes or a 
behind-the-scenes temporary dataset or ... ?

Yeah, I know, wrong place in the stack. I'm not a DFSMS guy. What does PACK do 
for you that dataset compression does not? Do we really need ISPF-level 
compression when we have system-level compression?

Hmmm. Am I reading this right? Compressed implies extended-format and PDSE does 
not support extended-format? That's ... um, unfortunate.

Charles


-----Original Message-----
From: IBM Mainframe Discussion List [mailto:[email protected]] On Behalf 
Of Paul Gilmartin
Sent: Thursday, February 15, 2018 5:02 PM
To: [email protected]
Subject: Re: J line command was: TSO temp dataset

On Fri, 16 Feb 2018 00:32:53 +0000, Jesse 1 Robinson wrote:

>Nice catch! I never considered the effect of PACK. The reason why ISPF 
>allocates and copies to a separate data set is that the JCL on the screen does 
>not necessarily match the current physical content on disk. The JCL may not 
>have been saved yet (or ever), or !aha! it may be PACKed on disk. TSO SUBMIT 
>and apparently the J command read the disk copy directly to the internal 
>reader, so PACKed records are just garbage. I do not PACK my data, so J worked 
>fine for me. 
> 
How modular is the un-PACK code?  Wouldn't it be reasonable for J to recognize 
PACKed data as other functions do (how?) and un-PACK it on-the-fly while 
passing it to INTRDR?

----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to [email protected] with the message: INFO IBM-MAIN

Reply via email to