January 5, Project Jupyter announced the release of “JupyterLab 3.0”, a next-generation Notebooks interface. This release brings great enhancements, including a debugger bundled by default. It is available from the project website.
Being the successor to Jupiter Notebook, JupiterLab is a web-based interactive IDE that offers features like notebook, terminal, text editor, and file browser. JupyterLab can be extended using npm packages and APIs. JupyterLab 3.0 is the latest major release after JupyterLab 2 series, released in February 2020.
This release includes a visual debugger by default. In order to use it, you need a supporting kernel like xeus-python. It also brings table of contents extensions. It allows you to see and navigate the structure of a document. You can now set the display language of the user interface. You will need to separately install the language pack as a Python package from the GitHub language pack repository.
The Simple Interface mode is improved. Also known as the Single-Document Mode, it is now more streamlined and document-oriented. The mode can be switched by toggling from the status bar. Its support for mobile has been improved with a more compact layout. Now it automatically switches to the Simple Interface mode when the window is resized down.
Extensibility is also improved as JupyterLab extensions can now be distributed as prebuilt extensions. You don’t have to rebuild JupyterLab nor install Node.js. Prebuilt extensions can be distributed as Python packages using package managers like pip, conda, and mamba. Extensions author can use the TypeScript extension cookiecutter to develop prebuilt extensions by default.
This release includes many other great improvements.