I have not looked at IRXEXCOM in years but in these days of relatively abundant storage can't you have a buffer big enough for the largest variable you expect? If the Rexx variables are really larger than -- say, 256K -- then perhaps redesign the Rexx?
Charles -----Original Message----- From: IBM Mainframe Discussion List [mailto:[email protected]] On Behalf Of Willy Jensen Sent: Wednesday, March 17, 2021 4:26 AM To: [email protected] Subject: Re: Read REXX variables directly by an assembler program The reason is that IRXEXCOM GET does a MOVE, meaning that you must already have a buffer big enough to hold the data. For a normal GET you can just redo the request with a bigger buffer and IRXEXCOM will tell you how big it needs to be. For GETNEXT however you must start again from the top as a a specific GET will destroy the information that REXX stored for GETNEXT. At least that is how I read the documentation, and experimentation seems to prove it. So perhaps what I really need is a GET LOCATE. ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [email protected] with the message: INFO IBM-MAIN ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [email protected] with the message: INFO IBM-MAIN
