fabric-jumpstart 0.1.9

Get started quickly with pre-built solutions, tutorials, demos, and accelerators—automated, high-quality, and open-source.

Fabric Jumpstart accelerates Microsoft Fabric adoption with ready-to-run accelerators, demos, and tutorials that install directly into your workspace in minutes via fabric-cicd.

Install the Library

Requirements: Python 3.10–3.13 and access to a Microsoft Fabric workspace.

pip install fabric-jumpstart

List and Install a Jumpstart

Run inside a Fabric notebook (or any Python environment with Fabric credentials):

import fabric_jumpstart as jumpstart

# Renders an interactive catalog
jumpstart.list()

# Copy the install command from the catalog, past in another cell and run!
jumpstart.install("stateful-streaming-lakehouse")

Notes

• workspace_id is optional when you run in a Fabric notebook; it auto-detects the current workspace. Specify to deploy to another target workspace.
• install() accepts extras like item_prefix and unattended=True if you prefer console logs over HTML output.
• Jumpstarts that include file upload configuration will automatically upload small data files to a Lakehouse's Files area after deployment — no extra arguments needed.

Handling Name Conflicts

If items with the same name already exist in your workspace, Fabric Jumpstart will detect conflicts and provide resolution options:

1. Overwrite existing items:

   jumpstart.install("spark-structured-streaming", overwrite=True)
   

2. Auto-generate a prefix to avoid conflicts:

   jumpstart.install("spark-structured-streaming", auto_prefix_on_conflict=True)
   

   This generates a prefix like js3_sss__ (jumpstart ID + abbreviated name) and applies it to all deployed items.

3. Provide a custom prefix:

   jumpstart.install("spark-structured-streaming", item_prefix="demo_")
   

The prefixing strategy:

• Renames item directories (e.g., MyNotebook.Notebook → js3_sss__MyNotebook.Notebook)
• Updates all references to renamed items within configuration files
• Uses word-boundary matching to avoid double-prefixing if you re-run the same install
• Reuses existing prefixes from previous attempts to prevent js3_sss__js3_sss__ patterns

Testing a Jumpstart Before Registration

Use _install_from_github() to test a jumpstart directly from a GitHub repo before adding it to the registry. This method builds a synthetic config from the arguments you provide and runs the same install pipeline as install().

import fabric_jumpstart as jumpstart

jumpstart._install_from_github(
    logical_id="my-jumpstart", # sets name of root folder that items are deployed to
    repo_url="https://github.com/my-org/my-repo",
    repo_ref="v1.0.0",                           # tag or commit SHA — not a branch
    workspace_path="my-jumpstart/",              # defaults to "{logical_id}/"
    entry_point="GettingStarted.Notebook",
    items_in_scope=["Lakehouse", "Notebook"],
    workspace_id="<guid>",                       # target workspace (auto-resolves to the current ws in Fabric)
)

Common optional parameters:

jumpstart._install_from_github(
    logical_id="my-jumpstart",
    repo_url="https://github.com/my-org/my-repo",
    repo_ref="abc1234",
    entry_point="GettingStarted.Notebook",
    items_in_scope=["Lakehouse", "Notebook", "SQLEndpoint"],
    workspace_path="my-jumpstart/",              # defaults to "{logical_id}/"
    name="My Jumpstart",                         # display name (defaults to logical_id)
    workspace_id="<guid>",                       # target workspace (auto-detected in Fabric)
    files_source_path="my-jumpstart/data/",      # upload binary/data files after deploy
    files_destination_lakehouse="MyLakehouse",   # target Lakehouse for file upload
    files_destination_path="raw/",               # destination path in Lakehouse Files
    item_prefix="test_",                         # prefix all deployed item names
    unattended=True,                             # console output instead of HTML
)

Once the jumpstart installs successfully, add a YAML file to fabric_jumpstart/jumpstarts/ and switch to the standard jumpstart.install() flow.

Contributing

See the root contributing guide for shared guidelines (commit conventions, issue workflow, PR process), then follow the Python library contributing guide for development setup, quality checks, and the new jumpstart workflow.