[Numpy-discussion] datetime64

Fernando Perez fperez.net@gmail....
Fri Nov 20 02:47:23 CST 2009


Hi Travis,

[...]

On Thu, Nov 19, 2009 at 10:03 PM, Travis Oliphant
<oliphant@enthought.com> wrote:
> Again, the NEP has not been fully implemented yet.   What is implemented
> works as far as I can tell, but could use more tests.   I would like to
> finish the core functionality before 1.4.0 and will try to do that, but
> there is still quite a bit of work to be done.
> If it is not finished, then we can keep the code there, and just document
> what works and what remains to be completed.

thanks for your detailed response.  It seems that you're on the ball
for this (though I'm sure you can use any help available), which is
great to hear.  We don't need this *now*, I just wanted to get a
better understanding of the situation before 1.4 was frozen, as there
hadn't been much response.

But the current plan sounds great, and from what I can see you have
all  the points we've seen so far already in mind.   I hope the
coercion work doesn't prove too difficult or time consuming, as right
now the pure datetime.datetime() API feels rather unwieldy.

Regarding naming:  how about

np.timestamp -> for what today is np.datetime
np.timedelta   -> stays the same

I actually *don't* like the  'date' in datetime, it feels redundant
and pointless.  It's pretty clear to me that a date has to do with
time in a library.  The above names, at least to me, express the
intent of representing absolute and relative time information fairly
clearly, there's a certain nice symmetry to them, and they prevent
stdlib name collisions.

Just an idea...

Thanks again for your detailed reply!

Cheers,

f


More information about the NumPy-Discussion mailing list