> Do to a bug in pgtypeslin I've been experimenting with the day of week
> stuff a little bit and I found that we do not agree with the output of
> the cal command on older dates. I have no idea which one is correct:
>
> mm=# select to_char('1000/01/01'::date,'D');
> to_char
> ---------
> 4
> (1 row)
>
> Thus we believe 1000/1/1 is a Wednesday. cal says:
>
> Januar 1000
> So Mo Di Mi Do Fr Sa
> 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
>
> Thus 1000/1/1 is a Monday. Is this a bug in cal? Or do we produce the
> wrong output? Or do I simply misread the output?
What do we do for September 1752?
bash-2.05a$ cal 9 1752 September 1752
Su Mo Tu We Th Fr Sa 1 2 14 15 16
17 18 19 20 21 22 23
24 25 26 27 28 29 30