On Mon, 21 Mar 2022 at 10:54, Itschak Mugzach
<00000305158ad67d-dmarc-requ...@listserv.ua.edu> wrote:
>
> Thanks all, I will try to simplify and better describe my needs. I have an
> application that stores data. I want to create an interface to this
> application using an address call. for example (just an example, my
> interest is on the myapp code):
> address myapp get key(44)
> address myapp put key(44) data('some data')
> address myapp loc Data('some data')Ah - at last. Pretty clearly for this you don't need a host command environment. Your example doesn't show any interaction with the REXX variable pool, or anything else that couldn't be done using an existing HCE such as TSO. Even if you do want e.g. your "get key(44)" to set some REXX variable, you can do that in an ordinary TSO command, i.e. the ability to interact with some parts of REXX is independent of having your own HCE. But... If you have no useful environment to start with - perhaps you are running REXX in a subtask of some long-running transaction program or the like - then you can indeed write your own HCE that works as you show. It's not very difficult, and is described adequately if not brilliantly in the TSO/E Rexx Reference. (BTW, keep in mind that you can issue address myapp just once, and it will be the default going forward. So you can then just have myapp get key(44) or whatever.) A good many years ago I wrote a REXX interface to HLASM, which is a fairly constrained environment. So you can invoke a REXX program during assembly/macro processing, interact with REXX variables in your macro, and so on. This is the kind of environment where you can't just assume that e.g. TSO or ISPF or UNIX HCEs are available. I was able to do it using only the REXX Reference book and the HLASM book. Mention has been made of the need to use HLASM to interface with REXX. I would argue that the IBM design is for PL/X rather than assembler. It's pretty clear that the mapping macros are a really bad design for assembler, and have PL/Xy approaches to things. But I may be wrong... Tony H. ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
