Re: RCVF in CL

On 10-Feb-2015 09:20 -0600, John Yeung wrote:

<<SNIP>>
I agree with Buck's sentiment: CL is not a good tool for this task.

Which makes me wonder, and I don't pose these questions to be
snarky:

1. Are there folks who are really, really skilled with CL, but can't
code a lick of RPG? (I guess by a Carl-Sagan-style argument, there
probably are some.)

Many; often operators and system administrators will learn only various scripting [languages] and CL programming. Before free-form RPG for example, I was an ardent anti-RPG person. I would attempt to rewrite all of my customer-provided RPG re-creates in CL [albeit using some tooling that mimics the various database requests of the RPG] so as to minimize reading of the fixed-format code, to just the creation of the ODP and the I/O actions being performed. RPG was a language that I considered cryptic and outdated with little value to learn, and I never got the sense that my feelings were anywhere near approaching unique.

2. Is there such a thing as an IBM midrange system with a CL
compiler but no RPG compiler? Like, can you even buy the CL compiler
without the RPG compiler? (If the CL compiler is always included with
the system no matter what, and the RPG compiler is an extra-cost
option, then I consider the answer to be "yes".)

The "yes" answer would be the correct assessment, because the CL and the compiler are part of the OS. Many systems have just COBOL or just C, because the languages are Licensed Program Product (LPP) that are at-cost options beyond what the OS provides. Also there are many production systems that have no compilers [other than what the OS provides; e.g. CL] which assist in preventing developers from creating [debug capable and debugging] code there without the proper controls; the developers are on their own separate system or partition to allow for separation of activity.