Fix opening active notebook from kernel - custom drive

[itsmevichu]

Fixes #15931

[jupyterlab-probot]

Thanks for making a pull request to jupyterlab!
To try out this branch on binder, follow this link:

[welcome]

Thanks for submitting your first pull request! You are awesome! 🤗

If you haven't done so already, check out Jupyter's Code of Conduct. Also, please make sure you followed the pull request template, as this will help us review your contribution more quickly.

You can meet the other Jovyans by joining our Discourse forum. There is also a intro thread there where you can stop by and say Hi! 👋

Welcome to the Jupyter community! 🎉

[itsmevichu]

I've noticed that localPath is intentionally used for the session creation. Is there a specific reason for this?

[krassowski]

Thank you! The test failure appears related, if this change is indeed desired we should update the the tests too:

  ● docregistry/context › Context › #constructor() › should set the session path with local path

    expect(received).toEqual(expected) // deep equality

    Expected: "373e1985-29d0-487c-9153-91ae151c002b.txt"
    Received: "TestDrive:373e1985-29d0-487c-9153-91ae151c002b.txt"

      64 |         });
      65 |
    > 66 |         expect(context.sessionContext.path).toEqual(localPath);
         |                                             ^
      67 |       });
      68 |     });
      69 |

      at Object.<anonymous> (test/context.spec.ts:66:45)


Error: expect(received).toEqual(expected) // deep equality

Expected: "373e1985-29d0-487c-9153-91ae151c002b.txt"
Received: "TestDrive:373e1985-29d0-487c-9153-91ae151c002b.txt"

[krassowski]

@jtpio this looks reasonable to me but wanted to check with you as the author of #15931 and someone who recently worked with the drive API - is this the right solution?

[jtpio]

The change does look reasonable indeed.

As originally noticed in jupyterlite/jupyterlite#1227 there were some changes in JupyterLab 4 when it comes to the handling of these paths. Which led to this comment: #14519 (comment)

Looks like it was needed for the collaboration extension, although not sure why exactly. It could be because the collaboration drive replaces the default file browser.

[jtpio]

Trying from the Binder for this PR (which has jupyter-collaboration installed), it does seem to be fixing the UX issue mentioned #14519 (comment) about the drive prefix being missing when hovering on the kernel in the running tab. However "New Console for Notebook" seems to be starting a different kernel, likely because the paths are now different:

[itsmevichu]

@jtpio, I've tried opening the console via 'New Console for Notebook,' where it prompts to select a kernel. Upon choosing the notebook's kernel, the console opens with that selection as expected.

open-console.mp4

[jtpio]

@itsmevichu normally the expected behavior is the same kernel being reused for both the notebook and the console:

jupyterlab-new-console-for-notebook.webm

[Zsailer]

@itsmevichu how's it going here?

It looks like this PR will solve this issue on jupyter-collaboration. 🚀

I was planning to work on PR but saw you had started one. Great work! I'm happy to help test and review going forward.

I also see the regression @jtpio mentioned when opening a console from a previous session when testing your PR. Let us know if you have questions here to push this PR through. Thanks!

[Zsailer]

However "New Console for Notebook" seems to be starting a different kernel, likely because the paths are now different:

@jtpio I think this is may be caused by jupyter-collaboration's replacement of the file browser, because I only see this behavior when jupyter-collaboration is installed (I tested this PR locally with + without JC).

[jtpio]

Thanks @Zsailer for looking into this!

this is may be caused by jupyter-collaboration's replacement of the file browser, because I only see this behavior when jupyter-collaboration is installed (I tested this PR locally with + without JC).

OK so if this PR fixes the default install (without RTC), maybe we can look into fixing this in jupyter-collaboration directly?