Upgrade to jupyterlab 4

[HaudinFlorence]

Update jupyterlab-git for jupyterlab 4

Fixes #1245

[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! 🎉

[github-actions]

👈 Launch a binder notebook on branch HaudinFlorence/jupyterlab-git/upgrade_jupyterlab_git

[mahendrapaipuri]

What is the status of this PR? Is there any timeline when it will be finished?

[HaudinFlorence]

What is the status of this PR? Is there any timeline when it will be finished?

Thanks @mahendrapaipuri for your interest. It is still a work in progress.

[krassowski]

@HaudinFlorence it looks like this one accumulated a number of merge conflicts. I or one of my colleagues would be happy to push this forward by resolving the conflicts and fixing up any outstanding issues. Did you plan to work on this PR anytime soon, or should we go ahead and open a new PR (or a PR against your branch - whichever preferred)?

[HaudinFlorence]

@HaudinFlorence it looks like this one accumulated a number of merge conflicts. I or one of my colleagues would be happy to push this forward by resolving the conflicts and fixing up any outstanding issues. Did you plan to work on this PR anytime soon, or should we go ahead and open a new PR (or a PR against your branch - whichever preferred)?

@krassowski Thanks for proposing some help. We are planning to rework on this PR in the future. But the migration process has shown that the update of jupyterLab-git to jupyterLab4 requires the migration of nbdime to CodeMirror6. So we have been working on it in the past few weeks and are still currently on it. We will then go back to this PR. In the meantime, if you want to rebase it, please feel free to do it. Thanks in advance for your help.
cc @fcollonval @JohanMabille

[krassowski]

Sounds like a plan! If you get stuck with some issue on CodeMirror6 transition feel welcome to ping me, I am now working on LSP which has a large surface of contact with CodeMirror6.

[dlqqq]

@HaudinFlorence Hello Florence! I recently upgraded both Jupyter Scheduler and Jupyter AI to JupyterLab 4, and I'd be happy to help with this migration. Our team at AWS has a vested interest in getting jupyterlab-git JL4-compatible before the end of the year, ideally within the next month or two, so we are happy to be of assistance.

Would you be open to meeting virtually sometime to see how I can help? I also fully understand if you're already 99% finished and would like to complete it alone, just let me know. My email is on my GitHub profile.

[HaudinFlorence]

@dlqqq Thanks for your message and for proposing some help. We are still in the process of migrating nbdime to codemirror6 and are close to have a working version with most of the current features jupyter/nbdime#664. Then we will go back to this PR.

[psychemedia]

just a comment on timelines; I'm in the edu sector and like many other educators we're on a sprint over summer updating things for September (the academic year start); not having jupyterlab-git is a blocker for many educators who want to migrate to JupyterLab v4 and expect students to use git in support of coursework (things tend to get locked in for the forthcoming academic year over the summer).

[HaudinFlorence]

@psychemedia Thanks for your message. We are still on nbdime update and then (the sooner we can) we will move back to finalizing the update of jupyterlab-git.

[JasonWeill]

@HaudinFlorence Thanks for your work on the related nbdime PR! Is there anything we can do to help that PR or this PR move forward? Do you have an estimate for when you would expect this PR to be ready for submission?

[HaudinFlorence]

@HaudinFlorence Thanks for your work on the related nbdime PR! Is there anything we can do to help that PR or this PR move forward? Do you have an estimate for when you would expect this PR to be ready for submission?

@JasonWeill Thanks for your message and proposition for help. The PR on updating nbdime to jupyterlab4 and so to CM6 is almost ready for review (it will be tagged to request review today hopefully). Feel free to have a look at it: jupyter/nbdime#664
Then it will depend on the time required for the review of nbdime PR and for having a new version to be used with jupyterlab-git. It is thus a bit difficult to give a precise timeline. But things are progressing and should be rather quick once a new version of nbdime is ready.

[HaudinFlorence]

The PR updating nbdime to jupyterlab4 is ready for review here: jupyter/nbdime#664. Reviews, testing and comments are welcome. Thanks.

[vkaidalov]

Hi @HaudinFlorence!
Given that jupyter/nbdime#673 is almost finished, would you be so kind to tell how much time will be needed to finish this MR? At least a very rough estimation, e.g. somewhere in the end of this month? Thanks

[HaudinFlorence]

Hi @HaudinFlorence! Given that jupyter/nbdime#673 is almost finished, would you be so kind to tell how much time will be needed to finish this MR? At least a very rough estimation, e.g. somewhere in the end of this month? Thanks

Hi @vkaidalov,
Thanks for your message. The estimated time to finish this PR is mid October but there are still blockers making any target date very provisional.

[JasonWeill]

@HaudinFlorence Thank you very much for your hard work on the nbdime PR! What other blockers affect this PR?

[HaudinFlorence]

@HaudinFlorence Thank you very much for your hard work on the nbdime PR! What other blockers affect this PR?

Hi @JasonWeill,
Thanks for your message. The blocker that is the most difficult to evaluate is related to the improvement of nbdime API for diff and merge views in such a way that it can be reused in jupyterlab-git for plain text files as CM6 does not have merge addon replacement.

[krassowski]

Can you elaborate on what is missing in https://github.com/codemirror/merge? Maybe we could contribute to it to enable reuse in jupyterlb-git? Are there upstream issues open that we could help with? For the record there are different ways of contributing and the author was raising funds some time ago: https://discuss.codemirror.net/t/merge-view-implementation/5072

[Khagesh16]

Please, provide jlab 4 support

[HaudinFlorence]

Can you elaborate on what is missing in https://github.com/codemirror/merge? Maybe we could contribute to it to enable reuse in jupyterlb-git? Are there upstream issues open that we could help with? For the record there are different ways of contributing and the author was raising funds some time ago: https://discuss.codemirror.net/t/merge-view-implementation/5072

Hi @krassowski. Thanks for your comments. To be honnest, I don't know how to answer these points and how much effort it would be to make a 3 way-merge view in CM6 if it is the point raised here.

[fcollonval]

For all people willing to push this forward, there are two meta issues on nbdime listing actions needed to help with this:

• [required actions] [Meta issue] Integration in @jupyterlab/git jupyter/nbdime#680
• [other points] Migration to jupyterlab 4.0 jupyter/nbdime#659

Please assign yourself to an issue when you start working on it so that we can split efficiently the effort between those issues.

[RafaelClovr]

Hi, one question, if pyhton 3.7 is deprecated, should be removed from CI?

[benz0li]

@fcollonval Is there an approximate timeframe for the release that will support JupyterLab 4?

[welcome]

Congrats on your first merged pull request in this project! 🎉

Thank you for contributing, we are very proud of you! ❤️

[krassowski]

@benz0li it will likely happen once final nbdime version is released. In the meantime you can help by testing the latest prerelease (currently v0.50.0a0) and report any issues you see.

[benz0li]

@benz0li it will likely happen once final nbdime version is released.

@krassowski Thanks for the info. This is perfectly fine.

In the meantime you can help by testing the latest prerelease (currently v0.50.0a0) and report any issues you see.

I did very basic smoke testing and it worked OK.

---

Unrelated to this issue but worth mentioning: There is also jupyterlab/jupyterlab_pygments#31: Progress seems to have stalled.