Code formatter for JupyterLab.
- JupyterLab installed under Python3.6+ environment
- Mainly because Black requires 3.6+, and I like using Black the most;
- Typing support requires 3.5+, and I like typing;
jupyter labextension install @ryantam626/jupyterlab_code_formatter
pip install jupyterlab_code_formatter
jupyter serverextension enable --py jupyterlab_code_formatter
# Remember to install one of the supported formatters (it's in the next section)
conda install -c conda-forge black
jupyter labextension install @ryantam626/jupyterlab_code_formatter
conda install -c conda-forge jupyterlab_code_formatter
jupyter serverextension enable --py jupyterlab_code_formatter
Make sure to restart jupyter server after installation.
For jupyterlab_code_formatter
to work, you would also need some formatters, three are currently supported:
- Black;
- YAPF;
- Autopep8;
- Isort;
Install at least one of them through pip install black
for example (or its equivalent in other package management systems).
Assuming you do have one of the formatters installed in the Python environment that runs JupyterLab, some extra options would appear under JUPYTERLAB CODE FORMATTER
, use those to apply formatting to current codecell (no multiple code cell application atm).
There are some formatter config exposed through the Jupyter Lab Advanced Settings Editor, have a browse and change it if you wish.
Add an extra entry to your keyboard shortcuts settings with something like
{"jupyterlab_code_formatter:black":{
"command": "jupyterlab_code_formatter:black",
"keys": [
"Ctrl K",
"Ctrl M"
],
"selector": ".jp-Notebook.jp-mod-editMode"
}}
This basically says "Under edit mode (detected through the selector), using the chord Ctrl K + Ctrl M, invoke the jupyterlab_code_formatter:black
command". And there you have it 🎉
For Autopep8, YAPF or Isort, simply do autopep8/yapf/isort instead of black.
You might see some errors about writing grammar tables when using the black formatter on Windows, simply manually create the directory as suggested in here and you should be good to go.
For a development install (requires npm version 4 or later), do the following in the repository directory:
npm install
npm run build
jupyter labextension link .
pip install -e .
To rebuild the package and the JupyterLab app:
npm run build
jupyter lab build