Repository navigation
pathlib.Path.absolute not marked as "new in version 3.11" #99387
Description
New methods are generally marked as "new in version X". For example, pathlib.Path.is_relative_to is marked as "new in version 3.9" : https://docs.python.org/3/library/pathlib.html#pathlib.PurePath.is_relative_to
pathlib.Path.absolute was added in version 3.11, but is not marked as such: https://docs.python.org/3/library/pathlib.html#pathlib.Path.absolute
Activity
sobolevn commentedon Nov 11, 2022
@tudortimi this is a good first issue to contribute. Would you like to do it? If you don't have time, no worries - I can help you.
tudortimi commentedon Nov 11, 2022
@sobolevn I'll give it a try.
AA-Turner commentedon Nov 12, 2022
pathlib.Path.absolutehas been in the standard library sincepathlibwas addded in 3.4, though only documented in GH-26153 -- I am not sure what the precedent is here?https://github.com/python/cpython/blame/v3.4.0/Lib/pathlib.py#L1002-L1018
A
Reddragonemperor58 commentedon Nov 12, 2022
Hi, I am new to open source and this will be my first one so I would like to try it
Reddragonemperor58 commentedon Nov 12, 2022
I have gone through python's helping with documentation, is that enough or is there anything else I need to go through before I start making changes?
brandtbucher commentedon Nov 12, 2022
I'm not sure that documenting when a function was first documented makes much sense. Saying "New in version ..." is really saying when the feature first became available (and is typically used only if that feature was added after its module or class was first made available). Even the directive itself is called
versionadded.I recommend we just close this issue.
tudortimi commentedon Nov 12, 2022
@brandtbucher But was the function widely used if it wasn't documented? I for one always used
resolve()to get the absolute path, because the table mentioned that it was the equivalent ofos.path.abspath():Updated tech docs as per issue python#99387
hauntsaninja commentedon Nov 14, 2022
Agree with Brandt that closing this issue seems appropriate. Although maybe we could think about backporting #26153 (this would e.g. update the table above in 3.10's docs)?
brettcannon commentedon Nov 15, 2022
Without tests I didn't want to backport the doc changes as I don't know what state the code was i. But I also agree that documenting it as new in 3.11 seems weird, too, since the name has previously existed. I say we just leave the docs alone.