How to enable loading widget from local nbextension folder? #10664
-
|
Hi! I'm developing a custom notebook extension (https://github.com/lab-cosmo/chemiscope, Unfortunately, trying to use it inside vscode shows me this error: After enabling downloads & restarting the kernel, the error becomes: This is somehow expected, since the chemiscope js is not distributed through these CDN (since it is always installed as part of the python package). Looking through the issues, it seems that vscode-jupyter should be able to load the js code from the local Code to reproduce (works in !pip install chemiscope ase
import ase
import chemiscope
frames = [ase.Atoms("CO2", positions=[(0, 0, 0), (-1.2, 0, 0), (1.2, 0, 0)])]
properties = {"foo": [1], "bar": [1]}
chemiscope.show(frames=frames, properties=properties) |
Beta Was this translation helpful? Give feedback.
Replies: 3 comments 6 replies
-
|
please could you install the latest pre-release version of the Jupyter extension, and let us know whether that works. you will need to install vscode insiders as well. would be Ken to know if this works as expected, or if we still need to make a few more fixed. |
Beta Was this translation helpful? Give feedback.
-
|
I still have the same behavior and error messages with vscode-insider (version 1.69.0-insider Commit: 0b3574dcef8f35fec4ee4f83dc958c1f16ef6fce Date: 2022-07-01T05:16:52.086Z) and jupyter pre-release v2022.7.1001841019 The jupyter output looks like this The JS console contains errors like: I'll try adding a version to the JS code (I think I left it empty until now) and see if that helps. EDIT: adding a version to the Python code ( |
Beta Was this translation helpful? Give feedback.
-
|
Please unzip this file and install the VSIX using the command I've tested this at my end and the |
Beta Was this translation helpful? Give feedback.
-
|
Let me know how this goes (also please note, I've added a screenshot of the sample you provided in our release notes as well microsoft/vscode-docs#5445) If you can provide a code sample with a more complex output that would be appreciated, similar to the output found on the home page https://github.com/lab-cosmo/chemiscope/blob/master/docs/src/img/screenshot.png (asking for this as that would look better than the one for CO2) |
Beta Was this translation helpful? Give feedback.
-
|
Thanks a lot! I'll give it a look & send you a more realistic example |
Beta Was this translation helpful? Give feedback.
-
|
It works very well! Here is an example loading the same data as the home page: import gzip
import json
import numpy as np
import ase
import chemiscope
data = json.load(gzip.open("Qm7b.json.gz"))
frames = []
for structure in data["structures"]:
x = np.array(structure["x"])
y = np.array(structure["y"])
z = np.array(structure["z"])
names = structure["names"]
frames.append(ase.Atoms(symbols=names, positions=np.vstack([x, y, z]).T))
properties = data["properties"]
chemiscope.show(frames=frames, properties=properties, settings={"pinned": [0, 6940, 885, 573]})With the data file available here: https://chemiscope.org/examples/Qm7b.json.gz |
Beta Was this translation helpful? Give feedback.
-
|
thank your for confirming the fix and also for providing an awesome output for the release notes |
Beta Was this translation helpful? Give feedback.
-
Beta Was this translation helpful? Give feedback.
making this as resolved as part of #8241 and #10678