You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Describe the bug
I think the importlib_metadata should be refined for Python 3.8
I have the following error from pipenv:
The conflict is caused by:
chaostoolkit-lib 1.28.0 depends on importlib-metadata~=1.2; python_version < "3.8"
flake8 5.0.4 depends on importlib-metadata<4.3 and >=1.1.0; python_version < "3.8"
keyring 23.8.2 depends on importlib-metadata>=3.6; python_version < "3.10"
My understanding is that for Python 3.8, importlib_metadata should be pinned to ~=1.4 but I'm guessing it pulls bigger than this with importlib-metadata>=3.6; python_version < "3.10"
Am I reading it wrong and the problem lies elsewhere?
The text was updated successfully, but these errors were encountered:
The problem is with flake8, which sets upper bound (importlib-metadata<4.3).
@jaraco, the maintainer of keyring, submitted a pull request to flake8 to port to new API of importlib-metadata (using a compatibility shim) but it was closed because the flake8 maintainer did not want an additional depenedency (PyCQA/flake8#1011).
Other people asked flake8 developers to unpin importlib-metadata and use new API, but the issues were also closed (PyCQA/flake8#1438, PyCQA/flake8#1522).
I think a potential way forward may be to make flake8 support both old and new APIs without need for a new dependency (maybe with try/except blocks), and finally remove upper bound on importlib-metadata.
Thanks @mitya57 for the clear context. I'm sorry if I aimed at the wrong project. Though it doesn't seem like opening a new issue on flake8 will help anyway. sigh.
Describe the bug
I think the
importlib_metadata
should be refined for Python 3.8I have the following error from pipenv:
My understanding is that for Python 3.8,
importlib_metadata
should be pinned to~=1.4
but I'm guessing it pulls bigger than this withimportlib-metadata>=3.6; python_version < "3.10"
Am I reading it wrong and the problem lies elsewhere?
The text was updated successfully, but these errors were encountered: