Make jupyter.runStartupCommands show in the workspace settings

[dbirks]

Fixes #15626

Just a small change to make jupyter.runStartupCommands not only show in the User settings.

Built the extension and confirmed that it now auto-completes in a .vscode/settings.json file (which it wasn't doing before):

And that it shows in the Workspace tab in the Settings:

Thanks for all the effort in maintaining this!

[DonJayamanne]

Thank you for the pull request, however we cannot accept this at this stage as we have concerns over making this setting a workspace scoped setting.
Basically we haven't decided whether to go ahead and enable this setting for workspaces.
@rebornix /cc

[dbirks]

Ah ok, I'm not sure what those concerns are, but maybe I can make my best case briefly and we'll see if we're talking past each other. 🙂

I've found that Azure Synapse and Microsoft Fabric give you a spark variable pre-initialized with the SparkSession, and without this jupyter.runStartupCommands, I've not found a way to duplicate what the live environment is doing. Without this there's that little bit of friction each time of having to add some cells just to your local notebook to get the same behavior.

Having jupyter.runStartupCommands in the User settings is amazing, as I can now mimic the behavior locally, but I can't easily share that setting with the other people on my team, as I can't add it to the .vscode/settings.json in the repo.

Quite a big part of why extensions are helpful for me is their ability to be shared, both by defining the extensions themselves in the extensions.json as you know, and the config for them in the settings.json. Without that I'm limited to just making notes in a readme, and can't source control all of my config directly.

Either way, thanks for all your work on this extension. It's a very helpful thing to have in the toolbox.

[DonJayamanne]

@dbirks Please can you file an issue with the details you have mentioned above
Along with some samples and a bit more details of the issue you are running into.

, I've not found a way to duplicate what the live environment is doing. Without this there's that little bit of friction each time of having to add some cells just to your local notebook to get the same behavior.

This is what I do not understand.
Perhaps some more details of the Environment (Python version, kind of kernel, etc) and issues you are running into.
Please do file a new issue and lets continue the conversation there.