I suspect you've found a genuine bug -- my guess is that Eugene references
the variable once during startup code and stores it away [avoiding the need
to re-evaluate it].  Try this: now that the second run of MPEX recognizes
the cutoff, change the cutoff to, say, 1960 and calc a date of 01/01/55.

Also, I suspect this initialization code would execute BEFORE evaluating the
contents of the various MPEXMGR files

Tom Emerson
Sr. Systems Analyst
NDC | e COMMERCE
[log in to unmask]
626-258-4309
626-350-3832 FAX



> -----Original Message-----
> From: Steve Hammond [mailto:[log in to unmask]]
> Sent: Tuesday, September 21, 1999 7:56 AM
> To: [log in to unmask]
> Subject: MPEX puzzlement
>
>
> Okay someone explain this one to me - and Vladimir is
> visiting Thursday, so he might be able to answer it.
>
> In my MPEXMGR file, I set the jcw VESOFTYEARBASE=1950 as
> detailed in the memo from VESOFT I received last week. This
> sets a 'roll-over' year for dates.
>
> I then do the follow
> >>>>>>>>>>>>>>>>>>
> :showtime
> TUE, SEP 21, 1999,  9:50 AM
> :showvar v@
>          ^
> No match found for this variable set. (CIWARN 8116)
> :mpex
>
> MPEX/3000  27N81128  (c) VESOFT Inc, 1980  5.5
>
> %showvar v@
> VESOFTYEARBASE = 1950
> VESOFTDEFAULTPRINT = ;keepamdates
> %calc cdate(01/01/01,"MM/DD/YYYY")
> 01/01/1901
> %exit
>
> END OF PROGRAM
> :showvar v@
> VESOFTYEARBASE = 1950
> VESOFTDEFAULTPRINT = ;keepamdates
> :mpex
>
> MPEX/3000  27N81128  (c) VESOFT Inc, 1980  5.5
>
> %showvar v@
> VESOFTYEARBASE = 1950
> VESOFTDEFAULTPRINT = ;keepamdates
> %calc cdate(01/01/01,"MM/DD/YYYY")
> 01/01/2001
> %exit
>
> END OF PROGRAM
> :
> >>>>>>>>>
> Okay - the jcw is set as the first instance of MPEX starts,
> but it seems to be ignored by it. But when you run MPEX a
> second time, it works.
>
> Interesting.
>
> steve hammond
> [log in to unmask]
>