All I need is a Codespace preconfigured for Jupyter Notebooks and Pylance disabled. But after several months of trying, I still can't get it working reliably!

[kostrykin]

Select Topic Area

Question

Body

The title pretty much says it all. For education purposes, I need to preconfigure a Codespace for students, which has the ms-toolsai.jupyter extension installed, and Pylance as well as all other types of linters disabled. To accomplish that, I started experimenting with different configurations of devcontainer.json. However, Codespaces created using that behaved inconsistently, for example, sometimes the Jupyter extension was loaded, sometimes not.

I then found that it is now possible to use prebuilt containers, which I hoped was more reliable. And indeed, Codespaces created using my prebuilt container seem to be more consistent.

Still, there is a not-so-small chance, that when creating a Codespace using this container, some arbitrary errors will pop up inside the Codespace. Here is an example. Below are two screenshots, both show a newly created Codespace from this repo, right after creating a new test.ipynb file.

As you can see, in one case, there is an error about Pylance reported. In the other case it is not. Note that I had the Codespace deleted and created newly between taking the two screenshots. Moreover, in the first case, a Jupyter Kernel seem to have been selected automatically, whereas in the second case it is not selected automatically (cf. top right corner of the screenshots).

The ingredients for building the container are found in this repo. I build the container by using

devcontainer build --workspace-folder ingredients --image-name mobi-devcontainer-python:test

The Dockerfile is pretty flat and only installs some requirements:

FROM mcr.microsoft.com/devcontainers/python:3.11-bullseye

COPY requirements.txt /tmp/pip-tmp/
RUN pip3 --disable-pip-version-check --no-cache-dir install -r /tmp/pip-tmp/requirements.txt
RUN rm -rf /tmp/pip-tmp

And the .devcontainer.json file used for building the container takes care of (i) loading the Jupyter extension, (ii) hiding spurious Python kernels, and (iii) disables linting for all *.ipynb files:

{
  "name": "Docker image for Python course",
  "build": {
    "dockerfile": "Dockerfile",
    "context": "."
  },
  "customizations": {
    "vscode": {
      "extensions": [
        "ms-toolsai.jupyter"
      ],
      "settings": {
        "python.defaultInterpreterPath": "/usr/local/bin/python",
        "notebook.kernelPicker.type": "all",
        "jupyter.kernels.filter": [
          {
            "path": "/usr/bin/python3",
            "type": "pythonEnvironment"
          }
        ],
        "files.exclude": {
          "**/.*": true,
          "**/*.md": true
        },
        "python.analysis.exclude": ["**"],
        "python.analysis.ignore": ["**/*.ipynb"],
        "python.languageServer": "Pylance"
      }
    }
  }
}

At least this is what it is supposed to do, as far as I understand it.

The prebuilt container can be tested simply by loading the corresponding image,

{
  "image": "docker.io/kostrykin/mobi-devcontainer-python:2023.1",
  "customizations": {
    "codespaces": {
      "openFiles": []
    }
  }
}

or simply created a Codespace for the repo: https://github.com/kostrykin/mobi-devcontainer-python/tree/main

Help is very much appreciated!

[github-actions[bot]]

🕒 Discussion Activity Reminder 🕒

This Discussion has been labeled as dormant by an automated system for having no activity in the last 60 days. Please consider one the following actions:

1️⃣ Close as Out of Date: If the topic is no longer relevant, close the Discussion as out of date at the bottom of the page.

2️⃣ Provide More Information: Share additional details or context — or let the community know if you've found a solution on your own.

3️⃣ Mark a Reply as Answer: If your question has been answered by a reply, mark the most helpful reply as the solution.

Note: This dormant notification will only apply to Discussions with the Question label. To learn more, see our recent announcement.

Thank you for helping bring this Discussion to a resolution! 💬

[zwasham]

Hi @kostrykin,

Did you ever figure this out? I am also stumped in my quest to disable linting. I don't understand why Python > Analysis: Diagnostics Mode does not have an "Off" option.

Zach

[zwasham]

I solved my problem by disabling the Pylance extension.

[kostrykin]

@zwasham Disabling Pylance in the .devcontainer.json file turned off linting for me, but not reliably. At times, when firing up a new codespace, it will still be active.