> Would allocating and using SYSTSIN/SYSTSPRT substitute for > The terminal input and output
pretty sure I routinely did this with IKJEFT01 back in the day (not TSR - used to write internal tools for devt folks as an avocation). I did a lot more Rexx than clist. Is there any reason you're using TSR per se? Memory is hazy here, but I thought TSR was intended for invoking clists/rexx from compiled code or vice versa when you want access to the variable pool. I would debug in foreground, then just run the same TMP (01) in a batch job using the contents of proclib(foreground session logon proc name) as a starting point for my JCL. Start with something innocuous as a test, and expand outward to test all the parts of the interface you expect to rely on before spending huge chunks of time on it. Some third party programs intercept clists in various ways, testing is like nuking it from orbit (the only way to be sure). E.g. invoke a clist that just reads from the terminal (here: systsin) and then writes the value back out, perhaps pre/appending its own string. If you're going to depend on accessing a variable pool via TSR, same idea. Best Regards, John Voice US 845-435-9470 BluePages z/VM OpenStack Enablement and KVM ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [email protected] with the message: INFO IBM-MAIN
