-
Notifications
You must be signed in to change notification settings - Fork 307
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
is_hidden in jupyter_server==1.4.1 import breaks with jupyter_core==4.7.1 #496
Comments
Thank you for opening your first issue in this project! Engagement like this is essential for open source projects! 🤗 |
hi @astrophysaxist - thanks for opening this issue. It's not clear to me when a pip-installed module would take precedence over a conda-installed module within an active conda environment, but it seems clear that the pip-installed version of
One thing worth exploring is that you're using Of course, these experiments only provide data points (and possible workarounds) but don't address the underlying issue that you're picking up packages from older locations. I'm hoping someone with deeper Python experience than myself can help explain when Pip trumps Conda. I'd love to better understand this. |
Hi @kevin-bates, I think you're right. I did each one of the experiments and I do see the same error.
I would've thought the jupyter_core grabbed would be from the conda environment. I'm looking around to see if there's an environmental variable I could change, but so far no luck. |
You might try updating jupyter_core via the active env: |
Looks like we should bump our dependency on jupyter_core to >= 4.6.0 - which is when |
Description
Hi, I've been chasing some errors when I go to run open a jupyter notebook in a virtual environment with jupyter_server==1.4.1 and jupyter_core==4.7.1. I has traced it to what looks like a deprecated import in the base/handler.py file. Here's a snapshot of the errors on the start of a
jupyter notebook --no-browser --port=8889
call:Reproduce
A minimal environment to stand up and test can be made with:
conda create -n test -c rapidsai -c nvidia -c conda-forge jupyter jupyter_server=1.4.1 jupyter-server-proxy=3.0.2 jupyter_core=4.7.1 python=3.7
Activate the environment with
conda activate test
. Then execute ajupyter notebook
command.Context
I'm running this on a CentOS7 machine in a virtual environment not too different from the one above. I've been tracking down why the wrong libraries seem to be grabbed in the jupyter notebook, and discovered this along the way.
Troubleshoot Output
The text was updated successfully, but these errors were encountered: