[Numpy-discussion] datetime dtype possible regression

Mark Wiebe mwwiebe@gmail....
Mon Apr 30 17:20:46 CDT 2012


On Mon, Apr 30, 2012 at 4:11 PM, Charles R Harris <charlesr.harris@gmail.com
> wrote:

>
>
> On Mon, Apr 30, 2012 at 1:19 PM, Mark Wiebe <mwwiebe@gmail.com> wrote:
>
>> I've done some comparisons of 1.6.1 and 1.7 (master), and written up some
>> key differences in a pull request here:
>>
>> https://github.com/numpy/numpy/pull/264/files#diff-0
>>
>> What you've discovered here looks like an interaction between the
>> automatic unit detection and struct dtypes, it's a bug to do with how
>> flexible types work. To match how the struct dtypes work with flexible
>> string dtypes, it should raise an error when trying to create a dtype with
>> just 'M8' and no unit specified, at In [4].
>>
>>
> Hi Mark, is there anything relevant that should go in the 1.6.2 release?
>

I don't think so, I can't think of anything in the datetime fixes which is
small and isolated enough to be worth integrating in the 1.6 series.

-Mark


>
> <snip>
>
> Chuck
>
> _______________________________________________
> NumPy-Discussion mailing list
> NumPy-Discussion@scipy.org
> http://mail.scipy.org/mailman/listinfo/numpy-discussion
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mail.scipy.org/pipermail/numpy-discussion/attachments/20120430/259c9a31/attachment.html 


More information about the NumPy-Discussion mailing list