Feature Request: Implement Output Insertion as Comments for Evaluated R Code

[ipstone]

Title: Feature Request: Implement Output Insertion as Comments for Evaluated R Code

Description:

I propose adding a feature to vscode-R that allows users to evaluate selected R code and automatically insert the output as comments in the editor. This feature, inspired by a similar functionality in Nvim-R, would enhance the data analysis workflow by allowing immediate inspection and documentation of code outputs within the editor.

Reference:

• https://github.com/jalvesaq/Nvim-R/blob/5f3876a53fdceae4ee6c022bf94d24ca3d6bc1a2/doc/Nvim-R.txt#L378
• https://github.com/jalvesaq/Nvim-R/blob/5f3876a53fdceae4ee6c022bf94d24ca3d6bc1a2/R/start_r.vim#L827

Key Points:

• Select and evaluate R code in the editor.
• Capture the output and insert it as a comment directly below the evaluated code.
• Useful for iterative exploration and documentation in data analysis.

Implementation Suggestion:

• Extend existing 'send code to terminal' functionality.
• Add a method to capture terminal output and format it as a comment in the editor.

Impact:
This feature would streamline the R coding process in VS Code, making it more efficient for data analysis and similar tasks.

[benz0li]

IMHO That is what R Notebooks were (are?) meant for.
ℹ️ Unfortunately, VSCode Notebook API support was never implemented because the R kernel depends on zmq.

[benz0li]

There is the Jupyter extension. Jupyter Notebooks paste output below the R code cells.

The Quarto extensions shows output side by side in a Jupyter interactive console.

[eitsupi]

ℹ️ Unfortunately, VSCode Notebook API support was never implemented because the R kernel depends on zmq.

This is off-topic, but there are other kernels for R besides IRkernel.
https://github.com/melff/RKernel
https://github.com/jupyter-xeus/xeus-r

[ipstone]

thanks @benz0li @eitsupi

The point of running the code and getting output into the Rscript, is really Rscript/text based/focused data analysis cycle, it's a lot of faster than jupyter, and oftenly are done in an terminal environment. I have been using juyter and Rmarkdown, but it's more for situations where we need to present the data; for explorative data analysis, a plain Rscript is usually leaner and getting the job done faster.

ideas/suggestions on sending codes to R and wrap the output back into the editor (or even getting the output to the clipboard is helpful)