Wirt writes:
...
> The question is: is that day calculation correct? Although when I
...
> However I was very pleased at HPWorld '96 when Vladimir Volokh wanted to
> show me the new calendar features in MPEX that Eugene had recently
> programmed into MPEX (which, BTW, also fails on December 31, 9999).

(Fails after 9999-12-31, I think you meant)

> Thus, after a fit of dueling calendars on side-by-side terminals, we
> demonstrated to each other that our two calendars agreed perfectly. That
> demonstration was quite reassuring. Eugene's date calculations so far
> have been the only completely independent verfication that QueryCalc's
> calendar is correct -- or at least wrong in the same way MPEX is.

From HP-UX & AIX:

   cal 12 9999
          December 9999
   Sun Mon Tue Wed Thu Fri Sat
                1   2   3   4
    5   6   7   8   9  10  11
   12  13  14  15  16  17  18
   19  20  21  22  23  24  25
   26  27  28  29  30  31

BTW, even a "standard" routine like cal varies from Unix to Unix:

   HP-UX:   cal 1 10000
   Bad argument

   AIX:     cal 1 10000
   0702-001 Specify month as number between 1 and 12
           Specify year as number between 1 and 9999.

POSIX on MPE/iX, OTOH, appears to be implemented by a Stanford fan:

   MPE/iX:  cal 1 10000
   cal: not found

:)

--
Stan Sieler                                          [log in to unmask]
                                     http://www.allegro.com/sieler.html