How to open Julia files as a Jupyter Notebook locally in Visual Studio Code
In this page we will show how to open *.jl files (e.g., index.jl) stored in T4AJuliaTutorials repository as a Jupyter Notebook in VS Code. This instruction might be useful when one wants to update/modify/fix our files showing them in the form of Jupyter Notebook and get clear and meaningful diffs when doing version control.
The operations described below are demonstrated on Windows; however, these operations are the same across other operating systems such as macOS and Linux.
Since .ipynb files are in JSON format, they can be difficult for humans to read and compare difference between modified files and the original one. They happen to contain large image files, very long outputs and binary data which causes increasing the size of files. Generally speaking, pushing large files to Git repository is a bad practice. Git repository should contain source files, not results. If you commit large (exceeds 100MB) files to repository, GitHub may refuse your changes.
See GitHub Docs About size limits on GitHub to learn more. GitHub Docs says:
We recommend repositories remain small, ideally less than 1 GB, and less than 5 GB is strongly recommended. Smaller repositories are faster to clone and easier to work with and maintain.
Certainly, it is possible to establish a workflow or protocol to clear the output cells of a Jupyter Notebook before commiting/pushing their changes. In reality, however, researchers, scientists, and programming novices may not always be concerned with the protocol. Therefore, making protocols contains manual operation won't succeed in reality.
Instead of tracking .ipynb, we track source code as .jl files while this can be converted to .ipynb files via jupytext
jupytext allows to save notebooks in the plain text, getting meaningful diffs when doing version control.
In Visual Studio Code, also known as VS Code, you can install various extensions that enhance the workflow by allowing developers to edit files in a Jupyter Notebook format while continuing to store them as *.jl files.
Open your terminal(Windows PowerShell if you are using Windows) and the following command:
PS> code --install-extension ms-toolsai.jupyter --install-extension julialang.language-julia --install-extension congyiwu.vscode-jupytextYou will get:
Installing extensions...
Installing extension 'congyiwu.vscode-jupytext'...
Installing extension 'julialang.language-julia'...
Installing extension 'ms-toolsai.jupyter'...
Extension 'congyiwu.vscode-jupytext' v0.1.2 was successfully installed.
Extension 'ms-toolsai.jupyter' v2024.6.0 was successfully installed.
Extension 'julialang.language-julia' v1.103.2 was successfully installed.
Extension 'ms-toolsai.jupyter-keymap' v1.1.2 was successfully installed.
Extension 'ms-toolsai.jupyter-renderers' v1.0.18 was successfully installed.
Extension 'ms-toolsai.vscode-jupyter-cell-tags' v0.1.9 was successfully installed.
Extension 'ms-toolsai.vscode-jupyter-slideshow' v0.1.6 was successfully installed.
You can also install extension from VS Code GUI. Open VS Code by running the following commands:
PS> cd <path/to/T4AJuliaTutorials>
PS> code . Then type Ctrl + Shift + X (on macOS, Cmd + Shift + X) to open extensions view: In you want to install an extension for Julia, type Julia in the search box in extension view.
Then click Install button to install Julia's extension named julialang.language-julia.
To see what extensions are installed, type @installed in the search box. You would see the following output if you followed the previous instructions code --install-extension blah blah....
See Visual Studio Code documentation Browse for extensions to learn more.
If you are using macOS or Linux, skip this section.
Since Jupytext is written in Python, you need to have a Python runtime to execute Jupytext's functionalities. This assumption, having Python runtime, is satisfied on Linux and macOS by default. Unlike most Unix systems and services, Windows does not include a system supported installation of Python. To make Python available on Windows, open Windows PowerShell and type
PS > python3
It will launch Microsoft Store and display store making us to install Python 3.12. Click Install button to install Python.
PS > python --version
Python 3.12.4Open your terminal and open T4AJuliaTutorials directory:
PS> cd path/to/T4AJuliaTutorials
PS> code .If you want to open index.jl as a Jupyter Notebook, right click and select Open as a Jupyter Notebook.
You will get the following results:
There is another approaching. On the top right in VS Code. You should see label icon Open as a Jupyter Notebook. Then click this.
To run Jupyter Notebook with a Julia kernel in VS Code, open the command palette in VS Code and type Notebook: Select Notebook Kernel
Then, choose the Julia release channel.
This action will display the Run All icon. Clicking this icon will execute all cells in the notebook.
As a result, you will see the output as expected:
Refer the documentation: Working with Jupyter Notebooks in Visual Studio Code