This issue tracker has been migrated to GitHub, and is currently read-only.
For more information, see the GitHub FAQs in the Python's Developer Guide.

Title: Undocumented behavior in strptime for ISO week dates
Type: behavior Stage:
Components: Library (Lib) Versions: Python 3.10
Status: open Resolution:
Dependencies: Superseder:
Assigned To: Nosy List: Jonatan Skogsfors, belopolsky, p-ganssle
Priority: normal Keywords:

Created on 2022-01-31 13:06 by Jonatan Skogsfors, last changed 2022-04-11 14:59 by admin.

Messages (1)
msg412200 - (view) Author: Jonatan Skogsfors (Jonatan Skogsfors) Date: 2022-01-31 13:06
Consider the following code :
Python 3.10.2 (main, Jan 31 2022, 12:03:48) [GCC 9.3.0] on linux
Type "help", "copyright", "credits" or "license" for more information.
>>> from datetime import datetime
>>> datetime.strptime("2021-53-1", "%Y-%W-%w").strftime("%c")
'Mon Jan  3 00:00:00 2022'

And also:
>>> datetime.strptime("2021-53-1", "%G-%V-%u").strftime("%c")
'Mon Jan  3 00:00:00 2022'

2021 is a "short year" with 52 weeks (e.g. since December 31 is on a Friday). The 53rd week of 2021 is interpreted by Python as week 1 of 2022. This can be convenient but as I understand it, it is not a part of ISO 8601 and it is not documented as a deviation in the documentation for datetime.

As a comparison, the special behavior of week 0 (also not part of ISO 8601) is documented. In the documentation for format directive "%W" it is stated that "All days in a new year preceding the first Monday are considered to be in week 0."

If the behavior for week 53 is intended (as opposed to raising an error) I think it should be documented.
Date User Action Args
2022-04-11 14:59:55adminsetgithub: 90750
2022-01-31 13:18:52AlexWaygoodsetnosy: + belopolsky, p-ganssle
2022-01-31 13:06:25Jonatan Skogsforscreate