On Wed, 18 Aug 2021, at 12:52, Tony Thigpen wrote: > The problem is that he approached his objective from the wrong direction. > > It appears that is objective was to 'avoid changing the JCL for > different control cards'.
I thought that none of what was written addressed the distinction between JCL doing set-up for a batch process versus (I presume) the python running in the foreground. If the python solution was in fact also run in the background it too would have had some JCL. Would the python recover gracefully if necessary datasets were not available when required? I thought it ironic that he somehow thought that the (I thought) convoluted syntax for passing parms (eg dataset allocation space values) to the python interface was somehow less likely to suffer a problem with a typo than the supposedly complicated JCL he wanted to do away with. And did you notice this paragraph ...? "Additionally, the code checks input so the system doesn’t have to. Finally, it keeps people from the typos that happen when one modifies JCL prior to submission. This not only ensures that execution is less error free, but it also allows..." /Less error-free/ ?? -- Jeremy Nicoll - my opinions are my own. ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [email protected] with the message: INFO IBM-MAIN
