Re: Oops prevention: is there an easy way to detect if the compilers are disabled?

Mark S. Waterbury wrote:

Compile any valid source, such as QGPL/QRPGLESRC member PROOF with CRTBNDRPG to test ILE RPG IV compiler, and compile QGPL/QRPGSRC member PROOF with CRTRPGPGM to test the "legacy" (RPG/400) compilers. If you MONMSG these CRTxxx commands (assuming you run this test from within a CL program), those commands should throw an escape message if the license is expired.

I'd rather I didn't have to depend on any particular source member being there, unless it was one the application actually owned, since some schmuck might have deleted QGPL/QRPGLESRC.PROOF.

So I tried a test program attempting to compile from a display file, rather than a source PF (something obviously guaranteed to fail). Unfortunately, while the joblog messages were different, the PGSDS came back with RNS9310 regardless of whether the failure was the result of the compiler failing to find the member, or the compiler being disabled (correct me if I'm wrong, but I don't appear to have any easy way to test the "complier not even installed" case). And on the box with the disabled compiler, I still got RNS9310 even if I specified OPTION(*NOGEN).

But that led me to another idea: what if I did a *NOGEN, but with a source member guaranteed to exist, but not guaranteed to contain valid RPG source?

It works. Better than expected. Even if I give it a DDS source member, if the compiler is there, and working, if I do OPTION(*NOGEN) OUTPUT(*NONE), it comes back without an exception if the compiler is working, but throws one if it's disabled, and it doesn't litter the user's spool files with pure-bovine-scat source listings.

--
JHHL