How about a PGM=CGETENV with a 4K/16K/64K? memory area for the C program to copy the ENV area into?
On Thu, Oct 5, 2023 at 5:43 AM David Cole <[email protected]> wrote: > > Hi Jon, Hi Peter, > > I must say, your insights have been quite helpful. Thank you! > > Jon, You raise a good point. I neglected to say why I wanted to get > at the environment variables. Ok, here's why... > > > > Classic z/XDC uses keyword ddnames as a very easy way to pass a fair > number of processing settings into z/XDC. It's easy for the user to > use, and it's super easy for z/XDC to find. > > We are considering writing a XDCUSS[A] analog to XDCCALL[A], and we'd > like to continue using the keyword ddname approach for passing > settings into z/XDC. > * However, my developer says that's not feasible, so I'm looking > for alternative approaches. > * To my naive mind, it seems to me that environmental variables > could be such an alternative. > * However, my developer says that environmental variables, while > available to C programs, are unavailable to assembler. > * I find that astonishing. > My developer has had little prior experience with USS; however, he is > an extraordinarily quick study, and has assimilated a lot of > knowledge in a very short time. Nevertheless, I thought I'd appeal to > the broader community in the hopes of learning what more experienced > minds have to say. > > > So our initial goal is to use environment variables for z/XDC's > internal control purposes. > But once that is accomplished, the broader purpose of providing > displays to the user will probably be undertaken. > > Dave > > > > At 10/5/2023 02:22 AM, Jon Perryman wrote: > >On Thu, 5 Oct 2023 05:46:48 +0000, Farley, Peter > ><[email protected]> wrote: > > > > >Why does any programmer need to care where the environment > > variables are stored? > > > >Normally, I would agree but XDC is a very special case with very > >broad requirements. As a full z/OS system debugger, Dave Cole has > >many requirements that are atypical. E.g. does he want the ability > >to display environment variables from all processes instead of just > >the process being debugged? He also has restrictions such as > >debugging SRB, SVC, PC routines and more. More than likely, he > >doesn't want to use SRB's & IRB's to gather the data when access > >registers are simpler. > > > >---------------------------------------------------------------------- > >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 -- Mike A Schwab, Springfield IL USA Where do Forest Rangers go to get away from it all? ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [email protected] with the message: INFO IBM-MAIN
