There seem to be way too many ways to access various subsets of System and
JCL symbols. I have C/C++ code that I want to be able to read its own
config file (a UNIX file, that is) and perform substitution for system
symbols. I do not want the user to have to run an extra job step or the
like.

There is program ASASYMBM that is a standalone load module in linklist. Can
I invoke it from C/C++? Do any of its environmental requirements clash with
those likely to be encountered in a C/C++ program running either as EXEC
PGM=, or invoked from a shell prompt?

There is also the program IEASYMCK in SYS1.SAMPLIB. This calls ASASYMBM,
and provides at least some guidance in using it.

Then there is EZACFSM1, which is a TCP/IP utility that is evidently of more
general applicability. But it is that separate job step that I don't want
to have to run. Can I invoke it from my C/C++ code? Does it accept DDname
overrides or UNIX files (fat chance, he guesses...)?

Finally, there is an LE library function __le_ceegtjs(...) that retrieves
"exported JCL symbols" one at a time, but doesn't do substitution. It's
less than clear if this covers anything other than the target of a JCL
EXPORT statement (what does EXPORT really *do* anyway...?). At least this
is C/C++ native.

Although it's not so important, I would like to be able to see user-defined
symbols from the JCL as well as System Symbols. But I imagine this will be
problematic if my program is run from a shell prompt/BPXBATCH, because it's
likely to be in another address space.

The overall effect I want is to be able to say to end users of the C/C++
code roughly "you can use System [and JCL?] symbols in the config file as
you would in JCL or a PARMLIB member". This would include consistent use of
& and . and .. during substitution.

Suggestions?

Tony H.

----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to [email protected] with the message: INFO IBM-MAIN

Reply via email to