What is the Synapse Visual Studio Code extension?

  • Article

The Synapse Visual Studio Code extension supports a pro-developer experience for exploring Microsoft Fabric lakehouses, and authoring Fabric notebooks and Spark job definitions. Learn more about the extension, including how to get started with the necessary prerequisites.

Visual Studio (VS) Code is a one of the most popular lightweight source code editors; it runs on your desktop and is available for Windows, macOS, and Linux. By installing the Synapse VS Code extension, you can author, run, and debug your notebook and Spark job definition locally in VS Code. You can also post the code to the remote Spark compute in your Fabric workspace to run or debug. The extension also allows you to browse your lakehouse data, including tables and raw files, in VS Code.

Prerequisites

Prerequisites for the Synapse VS Code extension:

After you have installed the required software, you must update the operating system properties.

Windows

  1. Add JAVA_HOME to the environment variables and point it to the directory where java 1.8 is installed. For example, if you install JDK at this path C:\Program Files\Java\jdk-1.8, set the JAVA_HOME variable value to that path.

  2. Add both %JAVA_HOME%/bin and the condabin subfolder of the Conda installation to the system path directory. The condabin folder could be in the folder where you installed and in this format \miniconda3\condsbin. Open the environment variables and add %JAVA_HOME%/bin and condabin paths to the Path variable.

macOS

Run the conda.sh in the terminal:

  1. Open the terminal window, change the directory to the folder where conda is installed, then change to the subdirectory etc/profile.d. The subdirectory should contain a file named conda.sh.

  2. Execute source conda.sh.

  3. In the same terminal window, run sudo conda init.

  4. Type in Java --version. The version should be Java 1.8.

Install the extension and prepare your environment

  1. Search for Synapse VS Code in the VS Code extension marketplace and install the extension.

  2. After the extension installation is complete, restart VS Code. The icon for the extension is listed at the VS Code activity bar.

Local working directory

To edit a notebook, you must have a local copy of the notebook content. The local working directory of the extension serves as the local root folder for all downloaded notebooks, even notebooks from different workspaces. By invoking the command Synapse:Set Local Work Folder, you can specify a folder as the local working directory for the extension.

To validate the setup, open the extension settings and check the details there:

Screenshot of the Settings screen, showing the selected local working directory.

Sign in and out of your account

  1. From the VS Code command palette, enter the Synapse:Sign in command to sign in to the extension. A separate browser sign-in page appears.

  2. Enter your username and password.

  3. After you successfully sign in, your username will be displayed in the VS Code status bar to indicate that you're signed in.

    Screenshot of the VS Code status bar, showing where to find your sign-in status.

  4. To sign out of the extension, enter the command Synapse: Sign off.

Choose a workspace to work with

To select a Fabric workspace, you must have a workspace created. If you don't have one, you can create one in the Fabric portal. For more information, see Create a workspace.

Once you have a workspace, choose it by selecting the Select Workspace option. A list appears of all workspaces that you have access to; select the one you want from the list.

Screenshot of VS Code Explorer, showing where to find the Select Workspace option.

Current Limitations

  • The extension under the desktop mode doesn't support the Microsoft Spark Utilities yet
  • Shell command start with "!" is not supported.

Related content

In this overview, you get a basic understanding of how to install and set up the Synapse VS Code extension. The next articles explain how to develop your notebooks and Spark job definitions locally in VS Code.