Repository navigation
Many method parameters in the datetime module are positional-only in the C implementation but positional-or-keyword in the Python implementation #89979
Description
| BPO | 45821 |
|---|---|
| Nosy | @abalkin, @pganssle, @AlexWaygood |
Note: these values reflect the state of the issue at the time it was migrated and might not reflect the current state.
Show more details
GitHub fields:
assignee = None
closed_at = None
created_at = <Date 2021-11-16.19:21:08.113>
labels = ['type-bug', '3.9', '3.10', '3.11', 'extension-modules', 'library']
title = 'Many method parameters in the datetime module are positional-only in the C implementation but positional-or-keyword in the Python implementation'
updated_at = <Date 2021-11-21.13:45:40.155>
user = 'https://github.com/AlexWaygood'bugs.python.org fields:
activity = <Date 2021-11-21.13:45:40.155>
actor = 'AlexWaygood'
assignee = 'none'
closed = False
closed_date = None
closer = None
components = ['Extension Modules', 'Library (Lib)']
creation = <Date 2021-11-16.19:21:08.113>
creator = 'AlexWaygood'
dependencies = []
files = []
hgrepos = []
issue_num = 45821
keywords = []
message_count = 3.0
messages = ['406423', '406592', '406715']
nosy_count = 3.0
nosy_names = ['belopolsky', 'p-ganssle', 'AlexWaygood']
pr_nums = []
priority = 'normal'
resolution = None
stage = None
status = 'open'
superseder = None
type = 'behavior'
url = 'https://bugs.python.org/issue45821'
versions = ['Python 3.9', 'Python 3.10', 'Python 3.11']
Activity
AlexWaygood commentedon Nov 16, 2021
The following methods have parameters that are positional-only in the C implementation of the
datetimemodule, but positional-or-keyword in the pure-Python implementation:These inconsistencies make it extremely difficult to provide an accurate stub for these methods in typeshed.
AlexWaygood commentedon Nov 19, 2021
It seems to me that there are two ways of resolving this:
(1) Change the Python implementation to match the C implementation (make these parameters positional-only in the Python implementation).
(2) Change the C implementation to match the Python implementation (make these parameters positional-or-keyword in the C implementation).
If the decision is that solution (1) is the way to go, I'd be happy to submit a PR fixing this. If the decision is that solution (2) would be better, however, I'm not confident enough with writing C code to submit a PR.
The C implementation is most widely used, so I think strategy (1) would not have too many backwards-compatibility concerns. Strategy (2) is still probably slightly more backwards-compatible, however.
AlexWaygood commentedon Nov 21, 2021
Ref to discussion on typeshed: python/typeshed#6343
StanFromIreland commentedon Mar 9, 2025
I would say (1) is fine (with no backporting) as I doubt there are many use cases for using the Python implementation over the much faster C. I can implement one (or two I guess, using clinic for just a few functions is also not the nicest imo)
serhiy-storchaka commentedon Jul 7, 2025
I support solution (1). We can add support for keyword arguments later if they are useful. See also #135312.
@AlexWaygood, do you prefer to create PR yourself or leave it to others?
AlexWaygood commentedon Jul 7, 2025
I'm somewhat stretched for time right now — I probably won't have time to make a PR myself any time soon :-)