In a recent note, "Shmuel Metz (Seymour J.)" said:
> Date: Thu, 15 Dec 2005 07:20:46 -0500
>
> ... IAC, I believe that you can
> do an LMINIT and LMADD without an LMPUT if you want an empty member.
>
Alas, no. I just tried it, with result:
ISPF Dialog Error
**********************************************************************
* ISRLS088 *
* *
* No record is written *
* LMMADD and LMMREP requires previous successful LMPUT service. *
**********************************************************************
M&C expands on this. So, I'll echo Johnny Luo's perplexity concerning
this requirement, which the three of us apparently perceive as frivolous.
I'm attributing it to a pervasive cognitive deficiency among IBM designers:
they apparently can't grasp the concept of "zero", as in "a zero-length
sequence of LMPUTs".
Would it work to LMCOPY from an empty data set? Will LMINIT accept
DATASET('NULLFILE')? ... testing:
>>> "LMINIT dataid(INPVAR) enq(SHR) dataset('NULLFILE')"
+++ RC(8) +++
>>> "ISRZ002 "
>>> "Data set is open "
>>> "Data set is currently being used by ISPF.
??? I hadn't expected this one.
As for:
In a recent note Shmuel said:
> Date: Thu, 15 Dec 2005 07:14:25 -0500
>
> What WRITE? If IEBGENER has not input on SYSUT1 then why would it
> issue a PUT or WRITE to SYSUT2?
>
The question was provoked by Barry Schwarz's remark in:
Linkname: Re: Why does LMPUT(ISPF Library service)?
URL: http://bama.ua.edu/cgi-bin/wa?A2=ind0512&L=ibm-main&D=1&O=D&P=68111
Thinki about how you do it in assembler. You WRITE, then you STOW.
... are you saying Barry is mistaken?
-- gil
--
StorageTek
INFORMATION made POWERFUL
----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO
Search the archives at http://bama.ua.edu/archives/ibm-main.html
