-
Notifications
You must be signed in to change notification settings - Fork 181
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
test_jupyter_path_prefer_env fails #208
Comments
Hi @mgorny. Do you happen to run with @jasongrout - would it make sense to have some remove/restore logic for |
No, I don't have it set in the environment. Also, the mismatched path changes if I change |
Sorry about this Michał. I'm hoping Jason can take a look at this. In the meantime, could you please post the results of:
|
Hmm - I suspect your ENV and SYSTEM paths are the same in that your python is installed is |
|
I am using a regular system IPython install (i.e. |
Uninstalling it doesn't help but also makes two more tests fail. |
What does this python snippet produce? import sys
sys.prefix If either |
Those lines were duplicating the logic from before: jupyter_core/jupyter_core/paths.py Lines 152 to 154 in 4c072dc
I think it is the test that is wrong. What do we expect to happen if we set to prefer env paths, but really there are no env paths because they are system paths? I expect that the user will be the first, but the test is trying to insist on sysprefix (even though it is a system path). So I think the test should also check to see if there are env paths that are not system paths, and if there are not any, it should see if the user path is first. Also, I agree with you that the test should temporarily reset the JUPYTER_PATH variable to cut out that side effect. |
|
This answers it as well:
There are no non-system environment paths... |
Also:
|
I can reproduce this test failure both with 4.7.0 and git master (which is practically the same thing ;-)):
The text was updated successfully, but these errors were encountered: