On Sat, 18 Jul 2026 20:41:28 -0500, Paul Gilmartin <[email protected]> wrote:
>I see. There's a similar hazard between the JCL system symbols, >DATE and TIME, but no such way to resolve it. Wow! What a rabbit hole! I just tried to look up the answer to your plaint. 1. I would guess (and a guess is of course not a spec, and perhaps I am a cockeyed optimist) that &TIME is fixed for the duration of a job, and consistent with &DATE. I am too lazy to test. 2. I went to confirm that. Wow! JCL Reference refers me to Initialization and Tuning, where system symbols appear to be defined under "Sharing PARMLIB definitions." (Uh, gee, I'm an application programmer trying to use &TIME in my JCL -- what does sharing PARMLIB definitions have to do with me?) I find no answer to my question there, but I do find &TIME defined rather bizarrely as "The time that the dump was requested." Uh, what dump? Charles ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [email protected] with the message: INFO IBM-MAIN
