What asset types are supported by Git folders?

Only certain Databricks asset types are supported by Git folders. This article details the supported asset types and how to use them with Git folders. When an asset type is supported in Git folders, this means it can be serialized, version-controlled, and pushed to the backing Git repo.

Asset types supported by Git folders

The supported asset types are:

Asset Type	Details
File	Files are serialized data and can include anything from libraries to binaries to code to images. For more information, read What are workspace files?
Notebook	Notebooks are specifically the notebook file formats supported by Databricks. Notebooks are considered a separate Databricks asset type from files because they are not serialized. Git folders determine a notebook by the file extension (such as .ipynb) or by file extensions combined with a special marker in file content (for example, a # Databricks notebook source comment at the beginning of .py source files).
Folder	A folder is a Databricks-specific structure that represents serialized information about a logical grouping of files in Git. As expected, the user experiences this as a “folder” when viewing a Databricks Git folder or accessing it with the Databricks CLI.
Query (Public Preview)	Databricks SQL (DBSQL) queries can be committed as IPYNB notebooks (extension: .dbquery.ipynb). Git support for DBSQL queries requires enabling the new SQL Editor. Queries created with the new SQL Editor feature disabled can be placed in a Git folder but cannot be committed to the remote repository.

When working with your assets in Git, observe the following limitations in file naming:

• A folder cannot contain a notebook with the same name as another notebook, file, or folder in the same Git repository, even if the file extension differs. (For source-format notebooks, the extension is .py for Python, .scala for Scala, .sql for SQL, and .r for R. For IPYNB-format notebooks, the extension is .ipynb.) For example, you can’t use a source-format notebook named test1.py and an IPYNB notebook named test1 in the same Git folder because the source-format Python notebook file (test1.py) will be serialized as test1 and a conflict will occur.
• The character / is not supported in file names. For example, you can’t have a file named i/o.py in your Git folder.

If you attempt to perform Git operations on files with names that have these patterns, you will get an “Error fetching Git status” message. If you receive this error unexpectedly, review the filenames of the assets in your Git repository. If you find files with names that have these conflicting patterns, rename them and try the operation again.

Unsupported asset types

note

You can move existing unsupported assets into a Git folder but cannot commit any changes made to them to the remote repository.

Databricks asset types that are not supported in Git folders include the following:

• Alerts
• Dashboards (including legacy dashboards)
• Experiments
• Genie spaces

Notebook formats

For more information about notebook formats for Git folders, see Notebook formats.