.NET Interactive is an engine that can run multiple languages and share variables between them. Languages currently supported include:
- C#
- F#
- PowerShell
- JavaScript
- SQL
- KQL (Kusto Query Language)
- HTML*
- Mermaid*
*Variable sharing not available
As a powerful and versatile engine, .NET Interactive can be used to create and power a number of tools and experiences such as:
- Polyglot Notebooks
- REPLs
- Embeddable script engines
Since .NET Interactive is capable of running as a kernel for notebooks, it enables a polyglot (multi-language) notebook experience. When using the .NET Interactive kernel, you can use different languages from one cell to the next, share variables between languages, and dynamically connect new languages and remote kernels within a notebook. There's no need to install different Jupyter kernels, use wrapper libraries, or install different tools to get the best experience for the language of your choice. You can always use the best language for the job and seamlessly transition between different stages of your workflow, all within one notebook.
For the best experience when working with multi-language notebooks, we recommend installing the Polyglot Notebooks extension for Visual Studio Code. While the full .NET Interactive feature set is available in Jupyter, many features are only usable via code, whereas the Polyglot Notebooks extension provides additional features including a language/kernel picker for each cell, enhanced language services, a multi-kernel variable viewer, and more.
There are several ways to get started using .NET Interactive with Jupyter, including Jupyter Notebook, JupyterLab, and nteract.
.NET Interactive can be used as the execution engine for REPLs as well. The experimental .NET REPL is one example of a command line REPL built on .NET Interactive. In addition, .NET REPL can actually be used to set up automation for your Polyglot Notebooks.
We support running on devices like Raspberry Pi and pi-top [4]. You can find instructions here.
For more information, please refer to our FAQ.
The multi-language experience of .NET Interactive is truly a collaborative effort amongst other groups at Microsoft. We'd like to thank the following teams for contributing their time and expertise to helping light up functionality for other languages.
- PowerShell Team: PowerShell support
- Azure Data Team: SQL and KQL support
Telemetry is collected when .NET Interactive is started. Once .NET Interactive is running, we collect hashed versions of packages imported into the notebook and the languages used to run individual cells. We do not collect any additional code or clear text from cells. The telemetry is anonymous and reports only the values for a specific subset of the verbs in the .NET Interactive CLI. Those verbs are:
dotnet interactive jupyter
dotnet interactive jupyter install
dotnet interactive http
dotnet interactive stdio
The .NET Interactive telemetry feature is enabled by default. To opt out of the telemetry feature, set the DOTNET_INTERACTIVE_CLI_TELEMETRY_OPTOUT
environment variable to 1
or true
.
The .NET Interactive tool displays text similar to the following when you first run one of the .NET Interactive CLI commands (for example, dotnet interactive jupyter install
). Text may vary slightly depending on the version of the tool you're running. This "first run" experience is how Microsoft notifies you about data collection.
Telemetry
---------
The .NET Core tools collect usage data in order to help us improve your experience.The data is anonymous and doesn't include command-line arguments. The data is collected by Microsoft and shared with the community. You can opt-out of telemetry by setting the DOTNET_INTERACTIVE_CLI_TELEMETRY_OPTOUT environment variable to '1' or 'true' using your favorite shell.
To disable this message and the .NET Core welcome message, set the DOTNET_INTERACTIVE_CLI_TELEMETRY_OPTOUT
environment variable to true
. Note that this variable has no effect on telemetry opt out.
You can contribute to .NET Interactive with issues and pull requests. Simply filing issues for problems you encounter is a great way to contribute. Contributing code improvements is greatly appreciated. You can read more about our contribution guidelines here.