languageserver VS jupyterlab-lsp

I'm not familiar with data science, but I would start with setting up a language server for R: https://github.com/REditorSupport/languageserver . Kate already have a default client configuration for it, so it should work out of the box.

Symbol outline by LSP Client: It can connect to any LSP "server" (runs locally) which implements LSP protocol (https://microsoft.github.io/language-server-protocol/), and supports "stdin" communication. A number of LSP server configuration is provided by default (including R), but user can extend, and override this configuration. As long as the LSP servers are properly installed (kate will start them) and the Kate LSP client configuration is set up properly should work. LSP also provides other IDE-like features (like code completion, jump to symbol definition, diagnostic/linter messages) etc. I found an LSP server for R (https://github.com/REditorSupport/languageserver), you may want to give it a shot.

You'd need: - hrsh7th/nvim-cmp with set up LSP source (hrsh7th/cmp-nvim-lsp). - neovim/nvim-lspconfig. - R language server installed. This is a regular R package, can be installed from CRAN.

Installed languageserver in R.

Lsp server for R like py-right for python. Check this

I have downloaded the language server from here https://github.com/REditorSupport/languageserver, but don't know where it put it. so form now it's in the "R\R-4.1.3\bin".

Besides a common modern Neovim "IDE setup", some other useful R-specific tools: - languageserver R package for LSP integration. - styler R package for code formatting. Somewhat slow and lacks some features, but seems to be the best current solution.

First of all, make sure that you actually use r_language_server. For that, at least install languageserver R package (run R from command line and execute install.packages("languageserver")) and set up r_language_server with nvim-lspconfig. I think after that you should be able to see some diagnostic information.

The truth is that most editors will use R language server underneath ( https://github.com/REditorSupport/languageserver ) so it's more what layout you prefer than what functionalities an editor has to offer.

Other plugins/utilities can fill in some of the other gaps. In the screenshot below, I'm using NvimTree to browse files, and I'm also currently using Neovim's native LSP client (in Neovim 0.5.0) with r_language_server installed (configured with the Neovim team's plugin nvim-lspconfig).

I've recently seen JupyterLab LSP that bring the static analysis aspect of LSP to the notebook, and been wondering if there is anything similar that try to bridge a running kernel back into Neovim.

JupyterLab is optimised for handling R, Python and Julia. Code intelligence-wise it requires installing jupyterlab-lsp to get all the best features.

Pylance is really a great LSP in VS Code and it's pretty fast, and Jupyter Lab has an LSP that is also great depending on your use case. VS Code has remote host support for Docker, VMs, etc. depending on what your plans are.