[ruby-talk:11410] Re: Time without seconds (updated/fixed)

On Friday 23 February 2001 13:15, David Alan Black wrote:
> On Sat, 24 Feb 2001, W. Kent Starr wrote:
> > On Friday 23 February 2001 10:32, David Alan Black wrote:
> > > Here's a followup question, based on playing around with this:
> > > Why do the two ways of doing this (which I've stashed into
> > > MyTime.floor) result in objects of different types?
> > >
> > >   class MyTime < Time
> > >     def MyTime.floor
> > >
> > >   # Way 1
> > >       t = Time.new
> > >       mt = MyTime.at t - t.sec
> > >
> > >   # Way 2
> > >       tt = MyTime.at (tm = Time.new) - tm.sec
> > >
> > >       [mt,tt].each do |ti|
> > >         puts "#{ti} is of type #{ti.type}"
> > >       end
> > >     end
> > >   end
> > >
> > >   MyTime.floor
> > >
> > >   Output:
> > >   Fri Feb 23 10:31:00 EST 2001 is of type MyTime
> > >   Fri Feb 23 10:31:00 EST 2001 is of type Time
> >
> > FWIW
> >
> > Time.type #-> Class
> > class MyTime < Time
> > end
> > MyTime.type #-> Class
> > a = Time.new
> > b = MyTime.new
> > a.type #-> Time
> > b.type #-> MyTime
> >
> > It appears that a class is of type Class and instances of that class are
> > of type #{Name_Of_That_Class}
> >
> > Interesting (and useful) behavior :-)
>
> Indispensable, even :-)  But I don't see the connection with my question. 
> But see earlier answers from Dave T. and Guy D.
>

Er, no it doesn't :-) (That's what I get for reading the output and not how 
it is derived!)

> (Is there *anyone* on this list without at least one name twin?  Slight
> exaggeration... but still!)
>

Are there more than one Kent? :-)

Regards,

Kent Starr
[e-mail]