On Saturday, 07/18/2009 at 05:15 EDT, "Frank M. Ramaekers" <framaek...@ailife.com> wrote: > That?s an idea, but I don?t want so many caveats. I don?t want to have to > worry if the IBM supplied (for instance) is in EXEC, EXEC2 or REXX, nor if it > is REXX that it has an ADDRESS COMMAND initially. I just to frontend it with > the same name (to be transparent).
CMS is not very amenable to that sort of manipulation. When you invoke an exec, the exec will seem to "disappear" from the normal command search order. That is, if TEST EXEC contains address CMS "TEST" it will get RC= -3. So if you want to invoke an exec of the same name, you must explicitly use the EXEC command. But if you want to avoid invoking the same exec, you need to EXECLOAD the right one! It also means that you have to know whether you're invoking an EXEC or a MODULE. Well, you don't HAVE to, but that just means adding more code to respond to the EXECLOAD failure. > It more detail, I just want to frontend the IOCP EXEC with an EXEC that will > verify a successful RC. If successful, it will FTP the source and PDF the > listing and FTP it to a DR server. IOCP EXEC: /* This exec front-ends the IOCP EXEC on the S-disk */ arg parms address Command "EXECLOAD IOCP EXEC S" "EXEC IOCP" parms src = rc if src = 0 then do /* whatever you want */ end "EXECDROP IOCP" exit src Alan Altmark z/VM Development IBM Endicott
