Re: Ada, calendar, and daylight savings

[Corey Minyard <minyard@acm.org>]

Marin David Condic wrote:

> Not necessarily. They might be using a time base with launch time being the
> start of the epoc. It depends on what the on-board computer has to do. If
> all it needs to know is that after N seconds of burn it should shut off the
> engines and coast, it really doesn't care if that is 10:30am EST or 23:40
> GMT. An Ada.Calendar package that required UTC for its implementation might
> just be the proverbial Brick S*** House in this application, which if this
> application required a validated compiler might preclude Ada from playing in
> that game.


Sorry, I meant a UTC-like time, which Ada.Real_Time sort of provides as 
long as you don't try to print it out.  But if you are guiding the 
missile with GPS, for instance, then you need to coordinate with GPS 
time from the satellites, at least for positional measurement.

I'm not saying that it needs to use UTC for implementation, but I think 
it at least needs an annex to be able to coordinate time in some way. 
And if it uses daylight savings and timezones, the program needs some 
way to get that information and do translations into different timezones.


> 
> Better that a more sophisticated time package be included as some kind of
> annex or optional package. That's why I suggested that it might make a good
> candidate for a standard component library.


I'm already working on it.  I was just wondering how it should interact 
with Ada.Calendar, and I think I know now.


-Corey