Cannot run Cadabra2 kernel in vscode #6622
Comments
|
@dualer It looks like we haven't even picked up your base conda environment. Ordinarily you need to start |
Ok, I have added the output information above. In fact, when I use aconda to install cadabra2-kernel following the document, I still can't run cadabra2-kernel, so I finally delete miniaconda, and install cadabra2 from source directly. As the document said, I have mentioned it before, I should be able to get cadabra2-kernel automatically after it. |
|
From the log it looks like we did actually find your kernelspec. Not sure why it's not showing up in the kernel list. We might be incorrectly filtering it out. |
|
Yes, the strange thing is that I can run cadabra2-kernel in browser, but not in vscode. So I think there might be something wrong in your extension. |
I think I should give more information: when I unlink python2 using and link them again using I cannot get python2 kernel in vscode, while I can get it in vscode before. Maybe I have changed some variables in some document. This should be related to my question above. |
I have known what's wrong. I had deleted the link |
|
@rchiodo hi rchiodo, I don't know why you closed this issue, and this issue is still unsolved. I said I know what's wrong just means I know why vscode can't find python2 kernel instead of cadabra2 kernel. As @joyceerhl said, your extension might be incorrectly filtering it out. If you think this issue is trivial or something, please give your comment instead just close it like this. |
|
@dualer, sorry, we assumed that you had gotten your setup working. |
Ok, never mind. Do you have any further ideas for this issue? I have spent much much time on it, but I still stop at this point and don't have any progress. |
|
Can you do a We check this by looking at the arguments in the kernelspec. If it uses |
Ok, the result is |
|
@dualer Please could you install the latest version of VS Code Insiders & then install the latest version of the Jupyter extension (I believe this may be fixed in that version). FYI - You need VS Code Insiders 1.59 & the latest Jupyter extension from the marketplace. |
|
@DonJayamanne Are you sure? I have updated vscode insiders & jupyter extension to the latest version, but it still doesn't work, and everything‘s still the same. |
|
Please could you upload the logs, from Jupyter output panel |
I have uploaded it before, you can see it in the first post. |
|
No worries, I believe I've identified the problem. |
Ok, thanks. I look forward to it. |
|
How to validate:
{
"argv": [
"<fully qualified path to python interpreter>",
"-m",
"cadabra2_jupyter",
"-f",
"{connection_file}"
],
"env": {},
"display_name": "Cadabra2",
"language": "python",
"interrupt_mode": "signal",
"metadata": {}
}
All we need to ensure is this kernelspec is listed in the list of kernels (previously we wouldn't display this in the list). |
|
I have modified |
|
@dualer apologies, those comments were for our internal engineers to test this out. |
|
@DonJayamanne That's all right. I have updated insider to the latest version, but I still cannot get the right result. Look forward to your further information or progress. |
|
@dualer I'm sorry to hear that, thanks a lot for helping with the validation. |
|
@DonJayamanne I have updated the output of Maybe this is the key point. |
|
That shouldn't matter |
|
I have uploaded the logs of jupyter again. |
|
I have noticed the folllowing info: This should be the point where it went wrong. |
|
Thanks for the information. |
|
The result is |
|
Hmmm, ok the problem is its |
greazer
added
the
notebook-kernel
|
ok wait for your good news |
|
@dualer does this now work in VS Code insiders? |
|
Ok, It works. Thanks for your work. |
Environment data
Expected behaviour
I want to run Jupyter's cadabra2 kernel in vscode locally.
Actual behaviour
I can run it in browser, but it can't be run in vscode locally. See pictures:
Steps to reproduce:
[NOTE: Self-contained, minimal reproducing code samples are extremely helpful and will expedite addressing your issue]
I work on win10, but
cadabra2just supportslinuxandmac, so I choose vscode's extensionwls-remoteto linkwindowswithlinux. I think this is not the key point, and you can think I work on ubuntu 18.04. The important point is how to installcadabra2-kernel, and you can see 1, 2 for more details. In fact, If we installCadabra2from source, we will can getcadabra2-kernelautomatically as the document said.Logs
Output for
Jupyterin theOutputpanel (View→Output, change the drop-down the upper-right of theOutputpanel toJupyter)The text was updated successfully, but these errors were encountered: