At 13:52 -0700 on 06/08/2012, Scott Ford wrote about Re: Getmain question:
Very true, Robert, btw the Getmain worked..i will change the getmain for sure...
I saw that you had spotted the problem as a field in a ACEE Control Block. I was just pointing out that the wrong version of the GETMAIN was being used since you wanted to handle a no-space condition not just ABEND. Often when I open code to fix an error, I take the time to fix coding type errors like this since it is a ticking time bomb just waiting to happen. Some time if the code is not fixed you will get a no-space condition and that ABEND if not fixed. I think it is much better to get an error message than an ABEND when you know it is possible to trap the error occurance.
Scott J Ford Software Engineer http://www.identityforge.com ________________________________ From: Robert A. Rosenberg <[email protected]> To: [email protected] Sent: Friday, June 8, 2012 3:52 PM Subject: Re: Getmain question At 07:36 -0400 on 06/07/2012, Peter Relson wrote about Re: Getmain question: >The only time GETMAIN RU can have a non-0 RC is when CheckZeroRC=YES is >specified. >Thus, there is a *useful* return code for GETMAIN RU only when >CheckZeroRC=YES. >So GETMAIN RU has a "defined return code" of 0 but, as Binyamin wrote, >that is meaningless in this case since >for this invocation the R15 is always 0 upon return (as it abends if the >obtain fails). Since the code checks R15 and issues an error message for Non-Zero (ie: A failure) I wonder why not just go with GETMAIN RC (not RU) to avoid the ABEND since there is a failure trap test in the program. I know that is query is not about what went wrong but IMO it is something that should be addressed when the real reason is found and any needed coding changes are made.
