On Fri, 17 Oct 2008 10:55:04 -0500, Chase, John wrote: >> >> > 3 //DD01 DD SYSOUT= <== nothing there >> > >> >IGD101I SMS ALLOCATED TO DDNAME (DD01 ) >> >> That's surprising. Paul's SYSOUT=() does the same thing. His SMS > >That's just a "null override". We use that technique in CICS JCL, to >"extend" the //DFHRPL concatenation with region-specific libraries (the >"common" libraries are coded in the startup PROC). Of course, at CTS >3.2 the "dynamic libraries" function was introduced, so "someday" we >might pare the //DFHRPL concatenation down to only those libraries that >cannot be defined "dynamically" to CICS. > Of course. But why does supplying one null positional subparameter, "SYSOUT=()", still constitute a null override, while supplying two null positional subparameters, "SYSOUT=(,)", asserts the SYSOUT parameter? Are there any other JCL keywords with similar behavior?
And this provokes a further question. Suppose I'm overriding a DD SYSOUT statement in a PROC which specifies both class and writer. I might variously want to do each of the following. How can I: o Nullify class and leave writer unchanged? o Nullify writer and leave class unchanged? o Nullify both class and writer? (It's plausible that a PROC might have "//SYSUT2 DD SYSOUT=(B,INTRDR)", but for testing I'd want not to submit a job, but keep the SYSOUT in the spool and browse with SDSF, for example. How do I nullify INTRDR?) Or is it simply impossible to nullify positional subparameters? -- gil ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
