On Thu, 26 Mar 2020, at 19:18, Paul Gilmartin wrote: > On Thu, 26 Mar 2020 19:09:40 +0000, Jeremy Nicoll wrote: > > >On Thu, 26 Mar 2020, at 04:23, Bruce Hewson wrote: > > > >> So far I have not seen anyone identify where this is fully documented. > > > >Apart from the documented behaviour of MVSVAR (which I found with Google) > > > Cite?
There's a link in my earlier reply. Maybe we're at cross-purposes. The OP never made it clear precisely what they didn't like about the 'strange' result they got. I thought it was just the unsurprising (to me having skim read the documentation I found) uppercasing of some of the text. But I see, looking back, a leading "&" in the results string ... which I don't understand, and had managed not to see that the substituted string is supposed to resolve to a single var name, not a bunch of words. That is I no longer understand why they (appear to have) expected the "testing &sysname in &sysplex" phrase to have worked, nor do I have any idea why there's ampersands in there. -- 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
