gh-141497: make ipaddress.IP{v4,v6}Network.hosts() return an iterator

[XChaitanyaX]

• Issue: Incomplete return type for ipaddress.IPv4Network.hosts() #141497

[dr-carlos]

A short test should be added in the relevant part(s) of test_ipaddress.py to ensure that the hosts() method always returns an iterator.

[dr-carlos]

The tests looks good, but this should still pass on main. Could you add similar tests for the specific cases affected by this PR?

cpython/Lib/test/test_ipaddress.py

Lines 1507 to 1510 in 17a1a05

	# special case where the network is a /32
	addrs = [ipaddress.IPv4Address('1.2.3.4')]
	str_args = '1.2.3.4/32'
	tpl_args = ('1.2.3.4', 32)

cpython/Lib/test/test_ipaddress.py

Lines 1525 to 1527 in 17a1a05

	addrs = [ipaddress.IPv6Address('2001:658:22a:cafe::1'), ]
	str_args = '2001:658:22a:cafe::1/128'
	tpl_args = ('2001:658:22a:cafe::1', 128)

[XChaitanyaX]

but this should still pass on main

the workflow shows an error in test import, I think it is a known issue I saw it somewhere....

[sharktide]

LGTM

[sharktide]

Although we probably should re-run the failed test. It seems unrelated.

[XChaitanyaX]

Is there a way to rerun the workflow?? I don't see any in checks

[sharktide]

Is there a way to rerun the workflow?? I don't see any in checks

You can't, a core developer or official triager has to. I am not officially either. It is fine for now. They probably will when they see this PR of you haven't pushed any commits in the meantime because losing commits re-runs all tests.

[sharktide]

Core review? @serhiy-storchaka

[bedevere-app]

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.

[serhiy-storchaka]

LGTM.

[miss-islington-app]

Thanks @XChaitanyaX for the PR, and @serhiy-storchaka for merging it 🌮🎉.. I'm working now to backport this PR to: 3.13.
🐍🍒⛏🤖

[miss-islington-app]

Thanks @XChaitanyaX for the PR, and @serhiy-storchaka for merging it 🌮🎉.. I'm working now to backport this PR to: 3.14.
🐍🍒⛏🤖

[miss-islington-app]

Sorry, @XChaitanyaX and @serhiy-storchaka, I could not cleanly backport this to 3.13 due to a conflict.
Please backport using cherry_picker on command line.

cherry_picker 6b1bdf6c7a6c87f12a247a125e25f8e721cc731e 3.13

[bedevere-app]

GH-141673 is a backport of this pull request to the 3.14 branch.

[XChaitanyaX]

Should I make a pr for 3.13?? @serhiy-storchaka

[serhiy-storchaka]

If it's not too difficult for you. Otherwise, I'll do it.

[bedevere-app]

GH-141695 is a backport of this pull request to the 3.13 branch.

[WilliamDEdwards]

For other users coming across this issue: this is a breaking change.

Ran into this while upgrading a 3.11 project to 3.13. One environment uses 3.13.11 and chokes on /128s and /32s:

single_ip_address_with_cidr = '10.40.1.1/32'

    def single_ip_address_with_cidr_to_ip_address(single_ip_address_with_cidr: str) -> str:
        """Transform a single IP address with CIDR (185.233.172.0/32) to a CIDR-less version (e.g. 185.233.172.0)."""
>       return str(ipaddress.ip_network(single_ip_address_with_cidr).hosts()[0])  # type: ignore[index]
                   ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
E       TypeError: 'tuple_iterator' object is not subscriptable

... whereas another environment on 3.13.5 is fine.

[picnixz]

I don't think it's correct to use [0] in this case. The docs (even before this patch) say:

Returns an iterator over the usable hosts in the network

The fact that someone is using type: ignore[index] also seems to indicate that they don't want to rely on this assumption. So I would suggest you to report this to their repository instead.