[phobos] std.datetime and tick types

[Andrei Alexandrescu]

Guys, let's settle for HNSeconds and call it a day. There's no real good 
name for it. hectAny name we use, people will have to look it up at 
least once when using it and that's that.

Andrei

On 9/28/10 0:39 PDT, Jonathan M Davis wrote:
> On Monday 27 September 2010 22:08:12 Robert Jacques wrote:
>> On Mon, 27 Sep 2010 11:51:00 -0400, Jonathan M Davis<jmdavisProg at gmx.com>
>> Well, if you make up a unit, then you can guarantee no one will have heard
>> of it before. And really, learning the hecto SI prefix means someone will
>> always understand what hectonanosecond means, as opposed to StdTick or
>> Duration, etc, which would have a tendency to be quickly forgotten or
>> confused with another library type from another language. As for typing, I
>> would expect the 100ns unit to be mainly an internal thing. Most of the
>> time I deal in microseconds/1000.0 (i.e. floating point milliseconds),
>> when dealing with std.perf today.
>
> Well, it's not entirely made up. 100ns is the tick that Windows uses. So, I
> believe that it's used in C# and other Windows-centric code. So, there is a
> precedent. And while ticks/hectonanoseconds are mostly internal, there are
> certain operations where you're going to want to use them, albeit hopefully
> fairly rarely. The documentation does use the term quite a bit though,
> particularly since std ticks pretty much replace time_t - though that can be a
> bit confusing because time_t is a type which holds the number of milliseconds
> since midnight January 1st, 1970 while std ticks are unit of time which are 100
> ns and are used to hold the time from midnight January 1st, 1 AD.
>
> Maybe, I should replace some of the mentios of ticks with something like StdTime
> (would be an alias for long and hold units of 100 ns with midnight January 1st,
> 1 AD as its epoch) and use hectonanosecond where its an issues of units rather
> than type.
>
> Bleh. I don't entirely like any of the options. Having StdTime does seem like a
> good idea, since it does help properly separate the idea of the unit and a type
> with those units from the epoch, but I really don't like hectonanoseconds. It's
> just way too long. milliseconds and microseconds are already seriously pushing
> it. Hectonanoseconds do have the virtue of being explicit however. Some people
> will know what they mean (though I wouldn't expect very many people to), and
> those who don't probably wouldn't have known what ticks indicated anyway, though
> the few people that I've talked to about it haven't liked the idea of having
> hectonanoseconds, since it's really ugly.
>
> - Jonathan M Davis
> _______________________________________________
> phobos mailing list
> phobos at puremagic.com
> http://lists.puremagic.com/mailman/listinfo/phobos