[Python-Dev] Include datetime.py in stdlib or not?

Alexander Belopolsky alexander.belopolsky at gmail.com
Wed Jul 7 20:37:33 CEST 2010


On Tue, Jul 6, 2010 at 11:54 PM, Terry Reedy <tjreedy at udel.edu> wrote:
> On 7/6/2010 3:59 PM, Alexander Belopolsky wrote:
[.. skipping more general stdlib discussion see "Python equivalents in
stdlib" thread ..]

>> 2. There are other areas of stdlib that can benefit more from pure
>> python equivalents.
>
> Possibly true, but developers do what they do, and this seems mostly done.
>

The reason I want to do datetime module is that there are some
long-standing bugs/RFEs that would require some experimentation to get
it right.  Such experimentation is unfeasible in C where more effort
goes into thinking about integer overflow and reference counting than
into actual design.  Here are some of those issues:

http://bugs.python.org/issue5516 = equality not symmetric for
subclasses of datetime.date and datetime.datetime
http://bugs.python.org/issue2736 = datetime needs an "epoch" method
http://bugs.python.org/issue7584 = datetime.rfcformat() for Date and
Time on the Internet
http://bugs.python.org/issue1100942 = Add datetime.time.strptime and
datetime.date.strptime
http://bugs.python.org/issue8860 = Rounding in timedelta constructor
is inconsistent with that in timedelta arithmetics
http://bugs.python.org/issue1647654 = No obvious and correct way to
get the time zone offset
http://bugs.python.org/issue5288 = tzinfo objects with sub-minute
offsets are not supported (e.g. UTC+05:53:28)
http://bugs.python.org/issue1982 = Feature: extend strftime to accept
milliseconds

>> 3. Reference implementations should be written by a senior CPython
>> developer and not scraped from external projects like PyPy.
>
> I did not see that in my reading of the thread.
>
This POV was brought up in the #python-dev IRC channel.

>  In any case, what matters is quality, not authorship.

I completely agree and I think the sooner the python code gets into
the main tree the more reviews it will get before the next release.


More information about the Python-Dev mailing list