Most changes to Python require a NEWS entry.

Please add it using the blurb_it web app or the blurb command-line tool.

All commit authors signed the Contributor License Agreement.

Hi @calestyo , Would you like to add a test for this?

@orsenthil Well I could at least try, but two issues here:

• I think it would be handy to have a test that does this for any function that is ought to accept the PathLike-interface, but that in turn doesn't seem to be possibly generically.
• The test get a bit ugly, since one cannot create os.DirEntry objects, even if I'd just made one for os.ismount(). Right now these do:
  

  cpython/Lib/test/test_posixpath.py

  Lines 178 to 180 in e34c82a

  	def test_ismount(self):
  	self.assertIs(posixpath.ismount("/"), True)
  	self.assertIs(posixpath.ismount(b"/"), True)

but one cannot yield / as os.DirEntry via scandir().

Without this patch, I would expect this

self.assertIs(posixpath.ismount(b"/"), True)

to fail or behavior differently than with this patch.

What might I be missing?

Without this patch, I would expect this

self.assertIs(posixpath.ismount(b"/"), True)

to fail or behavior differently than with this patch.

That I don't understand… as far as I understand it, it should have worked previously with both str and bytes but not when e.g. using it with a os.DirEntry.

A Python core developer has requested some changes be made to your pull request before we can consider merging it. If you could please address their requests along with any other requests in other reviews from core developers that would be appreciated.

Once you have made the requested changes, please leave a comment on this pull request containing the phrase I have made the requested changes; please review again. I will then notify any core developers who have left a review that you're ready for them to take another look at this pull request.

And if you don't make the requested changes, you will be put in the comfy chair!

Could you add a test that passes a bytes path-like to ismount()?

@orsenthil @JelleZijlstra … I've added a test using Path. Does that seem ok for you?

I have made the requested changes; please review again

Thanks for making the requested changes!

@JelleZijlstra: please review the changes made to this pull request.

The tests are failing.

Uagh... seems Path itsellf claims it would accept a PathLike object, e.g. pathlib.PurePath(*pathsegments) says:

Each element of pathsegments can be either a string representing a path segment, an object implementing the os.PathLike interface which returns a string, or another path object:

but it doesn't really, but only such that use strings. Isn't that yet another bug?

Anyway... I don't know how to do a proper test then. Is there any other PathLike object type in the library?

The only I'd know is os.DirEntry which one cannot directly create. I could use os.scandir(b"/")__next__() but there's the result could be a mointpoint or not.

Any ideas?

Oh and is a test here really worth it? The fix is trivial, should someone ever remove it, the check could also just be removed by accident as well.

As I've already said above, to which noone replied,... it would IMO only make sense, if all these functions could be tested for whether they accept the right types, at once.

Uagh... seems Path itsellf claims it would accept a PathLike object, e.g. pathlib.PurePath(*pathsegments) says:

Each element of pathsegments can be either a string representing a path segment, an object implementing the os.PathLike interface which returns a string, or another path object:

but it doesn't really, but only such that use strings. Isn't that yet another bug?

The problem in your test is that Path only supports str paths, not bytes paths. This is well-established behavior and not a bug.

Anyway... I don't know how to do a proper test then. Is there any other PathLike object type in the library?

You can use test.support.os_helper.FakePath.

Oh and is a test here really worth it?

Yes, every bugfix needs an associated test.

You can use test.support.os_helper.FakePath.

thx... still fails on one platform though... no idea why

Looks spurious, I retried it.

Thanks @calestyo for the PR, and @JelleZijlstra for merging it 🌮🎉.. I'm working now to backport this PR to: 3.10, 3.11.
🐍🍒⛏🤖

GH-99455 is a backport of this pull request to the 3.11 branch.

GH-99456 is a backport of this pull request to the 3.10 branch.