question-mark
Stuck on an issue?

Lightrun Answers was designed to reduce the constant googling that comes with debugging 3rd party libraries. It collects links to all the places you might be looking at while hunting down a tough bug.

And, if you’re still stuck at the end, we’re happy to hop on a call to see how we can help out.

Jupyter Lab build fails after installing extension: path names too long for Windows

See original GitHub issue

I’m not sure where this issue belongs. I tried running these commands (from here https://github.com/matplotlib/jupyter-matplotlib/issues/9#issuecomment-340257700):

pip install jupyterlab==0.28.11 ipympl==0.0.8
jupyter labextension install @jupyter-widgets/jupyterlab-manager@^0.28 jupyter-matplotlib@^0.1
jupyter lab

The step jupyter labextension install @jupyter-widgets/jupyterlab-manager@^0.28 jupyter-matplotlib@^0.1 would fail on the check step. When I listed the installed extensions, they appeared to be installed, but when I would start up jupyter lab, it would recommend that I rebuild. When I would try to build, it would fail, saying that the “The system cannot find the path specified.”

The path in question was

C:\Users\Jeremy\Anaconda3\share\jupyter\lab\staging\node_modules\jupyter-matplotlib\node_modules\@jupyter-widgets\base\node_modules\@jupyterlab\services\node_modules\@jupyterlab\coreutils\node_modules\ajv\node_modules\json-schema-traverse\spec\fixtures\schema.js

I figured out that the problem was that problem was that the path name was too long. See here: https://msdn.microsoft.com/en-us/library/aa365247.aspx#maxpat . I fixed the problem by changing the option in my system registry to allow longer path names. After doing this and performing a clean install, everything worked as expected. I figured I should report this potential problem.

Issue Analytics

  • State:closed
  • Created 6 years ago
  • Comments:6 (4 by maintainers)

github_iconTop GitHub Comments

1reaction
blink1073commented, Nov 7, 2017

Ah, I recently added a check (not in that version of JLab) that verifies node 5+ before trying to install an extension, because it installs maximally flat, which allows use to dedupe packages and avoids the long path problem on Windows.

Let’s call this one closed, thanks!

0reactions
jmbejaracommented, Nov 7, 2017

Hmm. I can’t replicate it myself. I reset LongPathsEnabled and rebooted, uninstalled nodejs and jupyterlab et al, and then started from scratch. However, everything seems to be working fine this time. A possible reason is that when I was first trying to get it to work, I realized I had two copies of nodejs installed—one that I installed a year and a half ago (4.4.7 with npm v. 2.15.8) and the other freshly from conda. I removed the old version while I was tinkering and proceeded to try things to get it to work. Maybe that had something to do with it? While I was trying to replicate the error, I tried with this older version of npm and everything still seemed to work fine. So, I don’t really know what happened. I don’t see the directory @jupyter-widgets being created (or at least persisting) after the build this time.

So, anyway, sorry to bother with this one. I guess it’s probably not worth looking into further. Thanks for checking it though!

Read more comments on GitHub >

github_iconTop Results From Across the Web

Can't add extensions in JupyterLab due to error in build
This is most likely due to another extension you have installed in jupyterlab (for me it was jupyterlab-latex ).
Read more >
What to do when things go wrong - The Jupyter Notebook
Jupyter fails to start​​ If Jupyter gives an error that it can't find notebook , check with pip or conda that the notebook...
Read more >
Git checkouts fail on Windows with "Filename too long error
Go to the registry key Computer\HKEY_LOCAL_MACHINE\SYSTEM\CurrentControlSet\Control\FileSystem\LongPathsEnabled (Type: REG_DWORD) must exist and ...
Read more >
JupyterLab - PyPI
When using a version of Jupyter Notebook earlier than 5.3, the following command must be run after installation to enable the JupyterLab server...
Read more >
Troubleshooting Vertex AI Workbench - Google Cloud
A 502 error might mean that your managed notebooks instance is not ready yet. ... Attempting to install a third party JupyterLab extension...
Read more >

github_iconTop Related Medium Post

No results found

github_iconTop Related StackOverflow Question

No results found

github_iconTroubleshoot Live Code

Lightrun enables developers to add logs, metrics and snapshots to live code - no restarts or redeploys required.
Start Free

github_iconTop Related Reddit Thread

No results found

github_iconTop Related Hackernoon Post

No results found

github_iconTop Related Tweet

No results found

github_iconTop Related Dev.to Post

No results found

github_iconTop Related Hashnode Post

No results found