Jupyter not respecting pre-set kernel in notebook metadata

Environment data

• VS Code version: 1.63.2
• Jupyter Extension version (available under the Extensions sidebar): v2021.11.1001550889
• Python Extension version (available under the Extensions sidebar): v2021.12.1559732655
• OS (Windows | Mac | Linux distro) and version: Windows 10 19042.1348
• Python and/or Anaconda version: Python 3.6.8
• Type of virtual environment used (N/A | venv | virtualenv | conda | …): N/A
• Jupyter server running: Local | Remote | N/A : Both Local and Remote

Expected behaviour

Jupyter Extension respects defined kernel in the *.ipynb file metadata

Actual behaviour

For each folder I open, I must re-select the kernel before executing.

Steps to reproduce:

[NOTE: Self-contained, minimal reproducing code samples are extremely helpful and will expedite addressing your issue]

1. Create new notebook, and select kernel.
2. Save notebook with kernel selected
3. Move notebook to another folder
4. Open new folder in VSCode and open the notebook.
5. See if kernel is still selected

Example of notebook file in txt form after saving kernel

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.10.0"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}

I tried to find another issue discussing this, but didn’t find an exact match.

Thanks!

[greazer]

After a bunch of discussion here on the team, it was decided that the current shipping behavior should remain for now. The main reasons are:

1. When opening a new notebook that VS Code has never seen, there’s really no 100% way to know that the kernel specified in the metadata actually maps to a kernel on the user’s machine.
2. In the above case, even though the kernel isn’t selected, it is moved to the top of the kernel picker list as the “Suggested” kernel.
3. After a notebook has been executed within VS Code, then the kernel specified in metadata is much more likely to be the “right” one to select by default. So on subsequent re-opens of that notebook it IS set as the kernel by default.

[rchiodo]

Thanks for the bug.

VS code doesn’t actually allow us to specify a specific kernel to set the first time it sees a notebook, but we can probably trick it by forcing a selection.

I’m rather surprised nobody else has logged this yet. I can’t find a dupe though.