This relates to the calculation of the week number from a given datetime, when calling the strftime method. If you call isocalendar() on the datetime.datetime object for the date ‘2018-12-31’, the week number returned is 1, which is correct. This is the same when checking the week attribute for the pandas timestamp equivalent. However, when you call strftime on this object (either datetime or timestamp), passing the ‘%W’ offset string, it returns 53, and then returns 00 for the remainder of the week. It seems that the rest of the weeks in 2019 are out by 1 when returned using this function. This issue seems to be present with the strptime function also.

Possibly related to 35535.

I think this is not a bug. bpo-35535 is probably also intended behavior, but that is less certain.

The misunderstanding here is that %W does not give you the ISO week number, from the documentation:

    %W: Week number of the year (Monday as the first day of the week) as a decimal number.
        All days in a new year preceding the first Monday are considered to be in week 0.


If you want the ISO week number, I think you need %V, which *is* the ISO week:

    >>> datetime(2018, 12, 31).strftime("%V %W")                                                                                            
    '01 53'

I believe this ticket can be closed.

Hi Paul,

Thank you for the clarification. I can see that %V does indeed return the
correct week number. It seems that when calling strftime, it is possible to
use this in conjunction with %y, but when calling strptime, it is not. Is
this also intended behaviour?

Best regards,

Tommy

On Mon, Jan 28, 2019 at 4:44 PM Paul Ganssle <report@bugs.python.org> wrote:

>
> Paul Ganssle <p.ganssle@gmail.com> added the comment:
>
> I think this is not a bug. bpo-35535 is probably also intended behavior,
> but that is less certain.
>
> The misunderstanding here is that %W does not give you the ISO week
> number, from the documentation:
>
>     %W: Week number of the year (Monday as the first day of the week) as a
> decimal number.
>         All days in a new year preceding the first Monday are considered
> to be in week 0.
>
>
> If you want the ISO week number, I think you need %V, which *is* the ISO
> week:
>
>     >>> datetime(2018, 12, 31).strftime("%V %W")
>
>     '01 53'
>
> I believe this ticket can be closed.
>
> ----------
>
> _______________________________________
> Python tracker <report@bugs.python.org>
> <https://bugs.python.org/issue35841>
> _______________________________________
>

> I believe this ticket can be closed.

Yes, I think so. How you say, on 35535 there is a detailed explanation about the behavior.