Merk
Tilgang til denne siden krever autorisasjon. Du kan prøve å logge på eller endre kataloger.
Tilgang til denne siden krever autorisasjon. Du kan prøve å endre kataloger.
To build data and AI apps with Databricks Apps, you can use any IDE that supports Python, such as PyCharm, IntelliJ IDEA, or Visual Studio Code. Azure Databricks recommends using the Databricks extension for Visual Studio Code, but you can also edit your code in the Databricks notebook and file editor.
The Databricks Apps environment automatically sets several environment variables, such as the URL of the Azure Databricks workspace running the app and values required for authentication. Many apps also need custom configuration, such as a specific command to run the app or parameters for accessing a SQL warehouse. Use the app.yaml file to define these settings.
Note
App files can't exceed 10 MB. If any file in the app directory exceeds this limit, deployment fails with an error.
To develop an app locally:
- Build and test your app in your preferred IDE.
- Run the app locally at the command line and preview it in your browser.
- When the app is complete and tested, move the code and required files to your Azure Databricks workspace.
Use the following topics to support your app development:
| Topic | Description |
|---|---|
| App environment | Learn about available binaries, environment variables, and resource defaults. |
| Define the app runtime | Specify how your app runs, including the command, arguments, and default settings. |
| Configure compute size | Select the CPU and memory for your app to match workload requirements. |
| Add resources | Include static files or other assets that your app depends on at runtime. |
| Define environment variables | Configure environment variables for authentication, workspace access, or app-specific settings. |
| Manage dependencies | Define Python dependencies using pip or uv, and Node.js dependencies using npm. |
| Connect using token authentication | Connect to an API app from a local development environment using token authentication. |
| Databricks AppKit | Accelerate app development with a Node.js and React SDK that provides type-safe database queries, pre-built UI components, and a plugin system. |