Discovery runs `conda info` command multiple times during startup when large number of conda interpreters are present

VS Code version

1.63.0

Extension version

v2021.12.1559732655

OS type

macOS

OS version

11.4

Python distribution

Anaconda

Python version

3.8.5

Language server

Default

Expected behaviour

I would expect the extension to just use the already selected Python interpreter on the left and provide indexer features (like the ability to click through functions) on startup.

Actual behaviour

The “Discovering Python Interpreters” dialog spins forever. If I go to “Output” and then select “Python” on the right drop down, the issue is the extension is looping through hundreds of python versions available on the cluster I am ssh’ing into.

For example, I see a constant output of the format, for everyone of hundreds of conda envs. I would just like to skip this discovering step and use the one I already selected.

> /home/XXX/.conda/envs/1e59305f3-855/bin/python ~/.vscode-server/extensions/ms-python.python-2021.12.1559732655/pythonFiles/interpreterInfo.py
> conda info --json

Steps to reproduce

Have many python versions available in your conda env.

Logs

No response

Code of Conduct

• I agree to follow this project’s Code of Conduct

[karrtikr]

This only fixes the issue that conda info command is only run once during startup. So discovery may still take longer and we may run interpreterInfo.py scripts but that is to be expected. Just make sure the same command isn’t run twice.

Thank you, just ran it, seemed to take about 5+ mins to load still

Stable VSIX should bring you to this state you mentioned above.

The fact that language server isn’t started will be fixed by https://github.com/microsoft/vscode-python/issues/18671.

[karrtikr]

The fix is out in the latest stable release, so simply updating to the latest version of VSCode and extension should ideally fix it.