Front page | perl.datetime |
Postings from January 2003
Re: Picking up the ball
Thread Previous
|
Thread Next
From:
Dave Rolsky
Date:
January 11, 2003 12:36
Subject:
Re: Picking up the ball
Message ID:
Pine.LNX.4.51.0301111422300.25009@urth.org
On Fri, 10 Jan 2003, Brad Hughes wrote:
> > -- It should handle fractional seconds. Based on a brief perusal of
> > Date::ICal, I think this can be added without too much difficulty.
>
> By "fractional", I hope you mean "counted numbers of subsecond chunks"
> instead of a real number representing a fraction of a second. I have no
> strong preference for what subsecond chunk is used. (A VMS timestamp's
> "chunk" is 100 nanosends, or 1/100 of a second, i.e. hh:mm:ss.cc)
I assume you're suggesting this because it's kind of odd to imagine 1/3 of
a second? That seems logical.
Anyone want to speak up and say that they really needs fractions of a
second before it's too late?
Time::HiRes returns microseconds, so it's a given we have to support at
_least_ microseconds, because this should work without loss of precision:
my $dt = DateTime->new( epoch => Time::HiRes::time() );
The TAI64 standard supports attoseconds, but that just seems excessive.
I think that if the base class is well-designed, it'll be possible to swap
in a DateTime::Implementation::TAI64 object wherever a DateTime object
works. Conversion to and from this implementation would be lossy, but
I think that's ok as long as this is well documented.
For now, I'd just say let's support microseconds. But I'm just making
this up. I should stress to people on this list that I am _not_ an expert
on date/time issues. I just have a good idea of what a unified set of
datetime classes should look like API-wise. So I really hope that people
with expertise in various areas speak up and help with implementation.
-dave
/*=======================
House Absolute Consulting
www.houseabsolute.com
=======================*/
Thread Previous
|
Thread Next