Re: IBM Mainframe JCL Conversion Tools

In article <ohhSb.561$b77.1266@dfw-service2.ext.raytheon.com>,
cmcampb@adelphia.net_remove_this wrote:

>Does anyone know of tools which will convert the horrible JCL construct
>COND= to IF ... THEN ... ELSE ... ENDIF?
>
>We can't forbid the use of COND= in our JCL Standards Checker (which is
>written in COBOL, so that I'm not 100% off topic), because there is so
>much use of COND= in existing jobs.
>
>If I could get all existing uses converted, then we could prohibit the
>keyword entirely.

One can use IF ... THEN statements in JCL?

--
 ========================================
==============================
ISLAM: Winning the hearts and minds of the world, one bomb at a time.

Lawrence,

Nice explanation :-)

I wonder what the guy was thinking when he dreamt that up. And no, I
don't think I'd like to smoke what he's been smoking.

---

Doug

dwscott@ieee.org

Jr.,

> I think it was done by someone comfortable with NAND (not and) and NOR
> (not or) gates.

Yes, I got that far in understanding it. It's obviously a hardware
engineer who invented it. What I can't understand how - even back in the
sixties - how anyone else could accept and adopt it. I know we had some
weird languages around at the time, but Sheesh! It's worse than RPG!

---

Doug

dwscott@ieee.org

Doug Scott wrote:

> Lawrence,
>
> Nice explanation :-)
>
> I wonder what the guy was thinking when he dreamt that up. And no, I
> don't think I'd like to smoke what he's been smoking.
I think it was done by someone comfortable with NAND (not and) and NOR
(not or) gates.  Unix also has its convolutions.  That said, I agree
that COND was an abomination that never should have seen the light of day.
>
> ---
>
> Doug
>
> dwscott@ieee.org
>
>

In article <VA.00000727.00fc6bfd@ieee.org>, Doug Scott <dwscott@ieee.org> writes:
> 
>
> WFL? Anyway ANYTHING is easier to use and more powerful than JCL.

I submit that OS/400 CL is harder to use and less powerful than JCL,
despite being the only IBM-supplied "scripting" mechanism included
with the OS.  At least for OS/400 V3 and earlier.

--
Michael Wojcik                  michael.wojcik@microfocus.com

This record comes with a coupon that wins you a trip around the world.
-- Pizzicato Five

"Chuck Stevens" <charles.stevens@unisys.com> wrote
 
>
> While ALGOL may have *spawned* or even *inspired* Pascal, it didn't *becom
e*
> Pascal.

There was a request for proposals for what was to be Algol-68.
Various different proposals were made including what Wirth called
Algol-W (and also such things as RRE's Alogol-68R).  Wirth later
renamed his language as Pascal.

Of course Algol-W never was 'Algol' or 'Algol-68'.

Chuck,

> it is
> anachronistic to say that ALGOL is "Pascal-like" and that therefore MCP
> honors the tradition of Pascal-type languages.

Sigh. Did I say that? I might have said that the Burroughs stuff was based o
n
Algol-type languages. I know very well that Pascal derives loosely from Algo
l.

> It is as anachronistic as to
> indicate that Unisys MCP systems honor the traditions established by PL/1 
or
> ADA by using an ALGOL-60 dialect, or that the music of J. S. Bach owes muc
h
> to the compositions of Felix Mendelssohn.

What planet are you from? Algol was the beginning. There were many
derivatives, including Pascal and MCP (so I'm told). It would be stretching
things to say that PL/I was derived from Algol - it merely borrowed a few
constructs.

---

Doug

dwscott@ieee.org

Joe,

> I don't think JCL is incomprehensible -- just lame.

:-) OK. Lame it is.

> It lacks obvious
> features that IBM should have added about the time they went from paper
> tape to punch cards.

??? IBM were never into Paper tape. They ALSO supported paper tape, as
80-column images. They had too much invested in punched card machinery -
in fact, that's what held up the development of IBM computers - Thomas J
could foresee that computers could signal the end of punched cards.

> But all IBM ever did was add an IF statement and say 'use Rexx if you
> want something else'...

Wel, that's advancing a lot of years. REXX is a fine language. The best
thing I like about it is that the author threw away the rule book on
syntax/semantics, and implemented intuitive constructs.

---

Doug

dwscott@ieee.org

"Doug Scott" <dwscott@ieee.org> wrote in message
news:VA.0000072d.006297ed@ieee.org...
 
>
> Sigh. Did I say that? I might have said that the Burroughs stuff was based
on
> Algol-type languages.

What you wrote was "Burroughs went heavy for Pascal-like languages, so I'm
not surprised that MCP still honours that tradition."   Unisys MCP is not
"honoring the traditions" established by Pascal-like languages for the same
reason that Bach didn't honor the traditions established by Mendelssohn.

> I know very well that Pascal derives loosely from Algol.

Not all that loosely, by what I read in the historical record!

>    ... Algol was the beginning. There were many
> derivatives, including Pascal and MCP (so I'm told).

MCP is not a *derivative of* ALGOL, it is *written in* a derivative of ALGOL
(NEWP).

> It would be stretching
> things to say that PL/I was derived from Algol - it merely borrowed a few
> constructs.

I think the basic block/procedure structure of PL/1 did more than "borrow a
few constructs" from ALGOL.  From what I can tell, the language can
simplistically be characterized as a synthesis of the block structure and
loop-control mechanisms of ALGOL with the record declaration capabilities of
COBOL, while keeping a side-glance at the FORTRAN application.  All things
to all people.  Such the 1968-vintage language spec from IBM I have at hand
leads me to believe, anyway.

-Chuck Stevens

Actually to be accurate, Turbo Pascal is not the same as Delphi.
While both are elementally "Pascal" in the language sense, Delphi
changed the language constructs to the point that Delphi could be
considered a descendent of Turbo Pascal and therefore technically a
different language with different skill sets.  While someone versed in
Delphi could go and use Turbo Pascal with very little adjustment,
someone versed in Turbo Pascal would be comfortable with the syntax
conventions of Delphi, but would have to re-learn most of the standard
constructs.

On Mon, 23 Aug 2004 21:32:27 +0100, Doug Scott <dwscott@ieee.org>
wrote:

>You've heard of Borland's Turbo Pascal, aka Delphi, I presume? Again, not p
ure
>Pascal, but a stronger (IMHO) derivative.