test error with python=3.11 and numpy=1.24

[d-v-b]

on the v3 branch, if I run hatch run test:pytest tests/test_group_v3.py, I get the following test failure only for python=3.11 and numpy=1.24:

──────────────────────────────────────────────────────────── test.py3.11-1.24-optional ─────────────────────────────────────────────────────────────
============================= test session starts ==============================
platform linux -- Python 3.11.5, pytest-8.1.1, pluggy-1.4.0
rootdir: /home/bennettd/dev/zarr-python
configfile: pyproject.toml
plugins: cov-5.0.0, asyncio-0.23.6
asyncio: mode=Mode.STRICT
collected 0 items / 1 error

==================================== ERRORS ====================================
___________________ ERROR collecting tests/test_group_v3.py ____________________
ImportError while importing test module '/home/bennettd/dev/zarr-python/tests/test_group_v3.py'.
Hint: make sure your test modules/packages have valid Python names.
Traceback:
../../.cache/pyapp/distributions/_11600656464259912759/python/lib/python3.11/importlib/__init__.py:126: in import_module
    return _bootstrap._gcd_import(name[level:], package, level)
tests/test_group_v3.py:4: in <module>
    from zarr.v3.group import AsyncGroup, Group, GroupMetadata
E   ModuleNotFoundError: No module named 'zarr'
=========================== short test summary info ============================
ERROR tests/test_group_v3.py
!!!!!!!!!!!!!!!!!!!! Interrupted: 1 error during collection !!!!!!!!!!!!!!!!!!!!
=============================== 1 error in 0.15s ===============================

The error looks like it might be due to improper building of zarr-python in the test environment. I don't have time to track this down, so I'm flagging it in case anyone wants to pick this up.

[maxrjones]

hmm interesting, I would be glad to track this down but am having trouble reproducing the error. The tests/test_group_v3.py all pass locally for the v3 branch and https://github.com/zarr-developers/zarr-python/actions/runs/8452862012/job/23154219080?pr=1726 also seems to look fine for those tests across all environments. I'm planning to look at #1656 (comment) today, so I'll see if it comes up during that work.

[d-v-b]

cool, thanks for checking! hopefully it's just something weird about my deployment

[maxrjones]

@d-v-b, I didn't run into the issue. Perhaps it's worth trying removing all the existing environments with hatch env prune before running the tests again?

[d-v-b]

@d-v-b, I didn't run into the issue. Perhaps it's worth trying removing all the existing environments with hatch env prune before running the tests again?

Thank you, this did the trick! closing as solved.