On 11 September 2012 00:40, Paul Gilmartin <paulgboul...@aim.com> wrote: > On Mon, 10 Sep 2012 22:04:53 -0400, Tony Harminc wrote: >> >>If you know that your target program uses only the TIME macro, and >>that it's the default (SVC) form, which is commonly true of >>traditional application programs, then you could set up SVC Screening >>to catch those TIMEs, and edit the results appropriately. >> > I hadn't thought of SVC screening. It's something I don't do; > don't know how to do; am not entitled to do. But I can imagine, > staying close to existing conventions: > > A USERMOD to the initiator (might this even be done in an exit?)
Far easier just to have a program to invoke the application program, as Charles original posited. In these OCO days, there are very few shops indeed that would accept a mod (aka ZAP) to something as central as the initiator. > which: > > If the job step contains DD STDPARM, that data set will be opened and read. Hey - as a bonus the program could support parm strings > 100. :-) > If that data set contains TZ="some string" and that string is a > valid POSIX time zone indication, the SVC screening will be > installed to use that time zone for LOCALTIME. Or just get the needed info from the appropriate UNIX place... > Some tricks necessary in case Daylight Time offset changes during > the job step. > SVC screening does not affect other jobs in the same LPAR, right? Right. It's TCB-related, with an option to progagate it when an ATTACH is done. > What are the obstacles? could the USERMOD/exit code go on cbttape.org? I think the two obstacles are the fact of a mod, and that it will catch only SVC versions of TIME. I would guess that things like LE, and perhaps other common runtime environments, may increasingly use other forms of TIME that wouldn't be caught. Tony H. ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
