Bill McKeeman

Patrick, I considered DST, and implemented entries in case the use knew DST was in effect. Tabulating DST, even for the U.S., required knowing geographical boundaries for the surprisingly varied laws regulating time. Such an effort was well beyond my own needs, and probably more work than payoff except for customers like AMTRAK. I would be interested in someone else trying it, but not me.

I am impressed the scope and presentation of this toolbox. Very nice work. I have a couple of nits to comment upon.
The last example in vpi.m seems to have lost its text.
I recommend that the examples be such that, if cut and pasted to the command line, run. I did this with my World Time Zones FX entry in response to a criticism and was pleased at the result.
I experimented with replacing
INT.digits = double(fliplr(N))-48;
with
INT.digits = int8(fliplr(N))-int8('0');
The package still seems to run and takes 8x less storage for the vpi numbers.

The version of perms has a bug. It sometimes gives the even permutations instead of the odd permutations and vice versa. It has been corrected and reissued as FX32978. This version, FX10237 should soon be deleted.