Thanks. Mostly idle curiosity on my part.

Yeah, I know submit LRECL is a bugaboo of yours. Can't say I disagree.

Charles

-----Original Message-----
From: IBM Mainframe Discussion List [mailto:[email protected]] On Behalf 
Of Paul Gilmartin
Sent: Wednesday, August 27, 2014 3:40 PM
To: [email protected]
Subject: Re: z/OS FTP JCL LOCSITE

On Wed, 27 Aug 2014 14:23:04 +0200, Charles Mills wrote:

>So my submitted dataset would be (for example) LRECL=1024 with embedded 
>DD * FTP subcommands. JCL would process columns 1-72 and ignore 73 to 
>1023. FTP would process columns 1 to 1023. Is that right? Would I even 
>have to specify DD *,LRECL=1024
> 
IIRC, z/OS 2.1 allows DD *,LRECL=, but not BLKSIZE nor RECFM.  1.13 allows none 
of those.

>or would QSAM pick it up from the fact that the original dataset was 
>LRECL=1024?
> 
QSAM, or perhaps JES2 picks it up, apparently not from the INTRDR allocation 
attributes, but from the longest record (trailing blanks are significant), 
perhaps assuming a minimum of 80.  If all records are of equal length (perhaps 
only one record), RECFM=FB, otherwise RECFM=VB. Note that this biases toward 
RECFM=FB,LRECL=80.  I suspect JES3 is different.  All empirical; I submitted an 
RCF requesting clarification.  (Quite a while ago; I should ping it.)

FTP JESLRECL imposes a maximum of 254.

>And I guess ISPF SUB does not handle this but INTRDR does? What about TSO 
>SUBMIT?
>
ISPF SUBMIT quietly (ugh!) truncates to F80; TSO SUBMIT fails if not F80.

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

Reply via email to