On 2013-09-29 23:53, Brian Schenkenberger, VAXman- wrote:
Sampsa Laine <sampsa at mac.com> writes:
This is a little bit unfair for the blue guys :)
=20
ISPF is far more than a set of screens to invoke TSO commands. The =
panels (that's how ISPF screens are called) are just a part of the whole =
thing. ISPF integrates with TSO, so we are talking about:
=20
- The TSO command language (CLIST language) and also the REXX =
language.
- The panel facility.
- The skeleton facility.
=20
It is quite easy to "simulate" the ISPF panels (you just need DCL to =
do it, I did it when I worked with DEC machines in a mostly IBM-centric =
company), but that would be just the user interface. The APIs provided =
with ISPF are way beyond that...
I was planning on developing a simple format for defining the panels, =
mapping the options to DCL and any params / switches they need. I am =
aware that ISPF can be used for way more than the basic IDE / sysop =
functions that it comes with by default, don't some ISVs actually build =
their software using ISPF panels as the interface?
I just thought a IBM mainframe lookalike interface to VMS would be =
amusing if nothing else.
It's not amuzing; nor are the myriad absurd DCL menuing interfaces I've
seen implemented at all too many sites over the years.
If you really want to implement a menu interface on VMS, have a look-see
into the SMG$ library. Roll-your-own ESCAPE sequences called out in DCL
command procedures leads to problems with terminal compatibility, as well
as horrors when trying to debug[*] and maintain it.
Apart from the SMG$ library, he might want to play with FMS. But I agree with you that I
personally don't find it amusing at all.
Johnny
--
Johnny Billquist || "I'm on a bus
|| on a psychedelic
trip
email: bqt at softjar.se || Reading murder books
pdp is alive! || tryin' to stay hip" -
B. Idol