Okay. Mea culpa.. COND= is documented at least adequately. (And I was speaking only about IARV64, not GETMAIN or STORAGE, except in my response to @Ed's point.)
I think I read only the description of COND=YES: The request is conditional. The request is not abnormally ended for resource unavailability. which certainly does not scream "but will be ABENDed for everything else" and missed the discussion of COND= in general which goes on at some length including In all cases, the request will be abnormally ended for invalid requests. The one valid documentation complaint is that in the Guide, the discussion of COND= is in the wrong place, under the heading about Guard Storage. @Peter, do you want me to put in an RCF, or will you? I suppose the paragraph quoted immediately above might be clearer on what is an invalid request, as opposed to a request that cannot be fulfilled at this time. An IBMer writes me off-list that the IARV64 behavior is identical to conditional GETMAIN in that invalid conditional GETMAIN requests also lead to an ABEND, not a return code. I guess I have not written enough invalid GETMAINs recently to have noticed the distinction. Charles -----Original Message----- From: IBM Mainframe Discussion List [mailto:IBM-MAIN@LISTSERV.UA.EDU] On Behalf Of Peter Relson Sent: Sunday, January 27, 2019 5:42 AM To: IBM-MAIN@LISTSERV.UA.EDU Subject: Re: IARV64 - why ABEND rather than return with reason code? I too was surprised at "no conditional GETMAINs". No conditional FREEMAINs I can understand. As to "expectations", when you're talking about what can be conveyed with 8 characters (4 for COND, 1 for equal, 3 for YES), or for just about anything else, you really should not "expect" much other than that things will behave as documented. You made an assumption. Was it an unreasonable assumption? No. But it was an assumption nevertheless. And it happened not to be a correct one. There's always room for documentation improvement. As I said, COND=YES means for the most part, return code instead of abend when "no storage available". If that is not clear in the doc, it should be made so. No one would choose document all the specific reasons that could be other than "no storage available". We're talking about the STORAGE macro (since GETMAIN does not have a COND keyword, only a COND indicator, such as the "C" in "RC"). Please re-read the documentation and comment on problems with that documentation, not making assumptions that there is no such documentation about the behavior of COND=YES. Peter Relson z/OS Core Technology Design ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
