zed
zed copied to clipboard
zed-industries
Jupyter Notebook
Check for existing issues
- [X] Completed
Is your feature request related to a problem?
I'm not sure "Feature Request" is the right thing to do, but I think this is a bit more than just a syntax highlight + LSP.
Every now and then, I go back to VS Code to work with Jupyter notebooks, and it would be nice to stay on Zed (+ collaboration 😙👌). This task is probably a monumental one, as it also involves a different buffer view and possibly something that could be done by the community (?), but I still want to put it on the radar.
Describe the solution you'd like
It would be great to open Jupyter notebooks directly in Zed and be able to interact with it (adding new blocks, running code, etc.), similarly to what VS Code does.
If applicable, add mockups / screenshots to help present your vision of the feature
Is jupyter support with the ability to run the cells going to be implemented ?
I would suggest Hydrogen's approach to interface with Jupyter local or remote kernals: https://github.com/nteract/hydrogen
I would suggest Hydrogen's approach to interface with Jupyter local or remote kernals: https://github.com/nteract/hydrogen
Hydrogen is definitely a killer feature, it was really awesome to use Hydrogen in Atom! I would love to see it in Zed.
Jupyter notebook support would open up a huge community in Zed, I would certainly be part of the bandwagon if Zed introduced it
@mikayla-maki @JosephTLyons, reading the "room temperature gauge" of the data science community, the interest is pretty strong in Zed, generally speaking, re: (https://github.com/zed-industries/extensions/issues/157) , (https://github.com/zed-industries/extensions/issues/136), & (https://github.com/zed-industries/extensions/issues/145) ... doing a first take on jupyter integration with already supported Python in Zed might pave the way for the community for a future feature
The only reason I haven't moved to Zed is because Jupyter notebooks aren't supported. I'd love to see this :0
I'm looking for people to start working on a integration of nteract Jupyter but I'm a really Rust beginner.
VS Code created some incompatibilities when trying to implement their own version of the notebook frontend resulting in some plotting libraries & widgets not working - among other inconveniences.
I'm not sure if this is possible, but maybe the jupyter installed in the venv could be used to render the notebook?
Jupyter notebook support would bring large number of new users to try out Zed. There's probably no other feature with potential to bring so many users.
I would love to use zed for all of my work but because jupyter notebook is not present not moved to zed fully. it is difficult to go back and forth from vs code to zed.
I just installed Zed this morning and love the style and speed,
but after finding that Notebook is not supported...
I have to go back to VSCode.
Ipython notebooks are also an essential part of how many academics (in Physics, Chemistry, Biology, etc) interact with programming, data analysis and presentation. It would definitely be necessary to incorporate ipython notebooks to make the switch to Zed viable for any of them.
installed Zed. super impressed. found out ipynb not supported. disappointed. back to VS. that was quick. LOL
Hang in tight! We're working with a contractor to build some of this out!
Btw, I saw that Nteract is working on a new version of their Jupyter notebook, using Rust and TypeScript. https://github.com/nteract/nteract-next
Hi! 👋 I'm one of the nteract leads and now working on Zed support for notebooks. I'm prioritizing notebook support in Zed above the nteract-next work.
Not having integrated jupyter notebooks is the only thing that prevents me from using zed. Glad to hear about the progress!
This is blocking me from using Zed. Also, to leapfrog VS code, if you could add syntax highlighting for SQL written in Python str, you'd have something that might cause folks to switch over.
syntax highlighting for SQL written in Python str
That's a really good idea. If you haven't already, please open a new issue for this!
Is there a timeline on the implementation of this feature ?
No timeline at this time. The tracking issue for implementation and pre-requisites is https://github.com/zed-industries/zed/issues/9778.
👋 I recently started using Zed and am greatly enjoying it! My work involves using notebooks near-daily, so I was very glad find this issue and see that @rgbkrk is developing support for Zed. I was also curious to learn more about the codebase and thought that the project sounded very interesting, so I started working on a POC as well. @rgbkrk I don't expect that you'd necessarily have use for this, as it may be redundant to/unaligned with your work. However, if there's any part of it that you think could be made useful to you, please do let me know! I'd love to contribute. The kernel client functionality is currently provided by the jupyter_client Python package though pyo3 bindings -- I'd be happy to try hooking it up to runtimed.
https://github.com/zed-industries/zed/assets/11862683/9ee2be84-c7be-4fb6-a436-9bd496798877
Sweet @davidagold!
I'm finally back to working on this again. I like this notebook UI.
In https://github.com/zed-industries/zed/tree/runtimes I'm working with the protocol directly in Rust. I'd rather not introduce python as a development dependency, though it is super cool to see how you wrapped it with PyO3.
I think what might be best to do to ease collaboration and get things moving is for me to put the initial runtime work behind a flag in settings.
Here's the draft PR for runtimes: https://github.com/zed-industries/zed/pull/12062
I think what might be best to do to ease collaboration and get things moving is for me to put the initial runtime work behind a feature flag.
Sounds good! I will look into revising the notebook to use the runtimes functionality included in #12062. Thank you very much for taking a look, @rgbkrk 🙏
