> I believe not. If that were the case, it would have the final value, XYZZY, > in all steps
Ah. I was too hasty. Charles -----Original Message----- From: IBM Mainframe Discussion List [mailto:[email protected]] On Behalf Of Paul Gilmartin Sent: Friday, November 06, 2015 2:43 PM To: [email protected] Subject: Re: Premature SYMBOL substitution in SYSIN? On Fri, 6 Nov 2015 14:31:21 -0800, Charles Mills wrote: >Hmmm. JCL processing versus run-time? When does the SET V2= happen? When JES >reads the job in? I suspect so. > I believe not. If that were the case, it would have the final value, XYZZY, in all steps. >Or when it runs? > In fact, this example was distilled from a larger job in which symbols were substituted in both a SYSIN and a DD PATH='.../&V2/...' In the DD statements the substitution was as if the substitution were performed logically, sequentially. In the SYSIN the premature substitution occurred. Where I noticed the behavior, the DD PATH= preceded the SYSIN. I haven't tried with the SYSIN preceding the DD PATH=. >//* >// EXPORT SYMLIST=* >//P PROC >//GEN EXEC PGM=IEBGENER >//SYSPRINT DD SYSOUT=(,) >//SYSIN DD DUMMY >//SYSUT2 DD SYSOUT=(,) >//SYSUT1 DD *,SYMBOLS=JCLONLY >GENER step; > with V2=&V2.. >//P PEND >//* >// SET V2=WOMBAT ************************************* >//WOMBAT1 EXEC P >//WOMBAT2 EXEC P >// SET V2=XYZZY ************************************* >//XYZZY1 EXEC P >//XYZZY2 EXEC P >//* -- gil ---------------------------------------------------------------------- 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
