Databricks Bundle Manager

A simple Python Dash web application for managing Databricks bundles.

Features

• Simple Configuration: Just set bundle path, target environment, and profile
• Smart Initialization: Creates bundles if path doesn't exist, detects existing bundles
• Bundle Commands: Initialize, validate, deploy, run, and destroy operations
• Custom Commands: Execute any databricks bundle command
• Real-time Output: View command execution results

Prerequisites

• Python 3.7+
• Databricks CLI installed and configured

Installation

1. Install dependencies:

   pip install -r requirements.txt

2. Ensure Databricks CLI is installed and configured:

   # Install Databricks CLI if not already installed
   pip install databricks-cli
   
   # Configure with your Databricks workspace
   databricks configure --token
   
   # Optional: Configure additional profiles
   databricks configure --token --profile myprofile

Usage

1. Start the application:

   python app.py

2. Open your browser: http://localhost:8050

3. Configure:

   • Bundle Path: Path to your bundle directory (existing or new)
   • Target Environment: Environment name (e.g., dev, staging, prod)
   • Databricks Profile: Profile name or DEFAULT

4. Use commands:

   • Initialize: Creates bundle if path doesn't exist, detects if exists
   • Validate: Validates bundle configuration
   • Deploy: Deploys bundle to target environment
   • Run: Executes a job (specify job name)
   • Destroy: Removes deployed resources
   • Custom Command: Execute any databricks bundle command

How It Works

Bundle Path Logic:

• Path doesn't exist: Creates new bundle with databricks bundle init
• Path exists, no bundle: Initializes bundle in existing directory
• Path exists with bundle: Uses existing bundle for all operations

Command Execution:

• Initialize: Runs from appropriate directory to create bundle
• Other commands: Changes to bundle directory and executes command
• All commands: Automatically adds --target and --profile parameters

Examples

New Bundle:

1. Set Bundle Path: /Users/me/my-new-project
2. Click Initialize → Creates new bundle at that path

Existing Bundle:

1. Set Bundle Path: /Users/me/existing-project
2. Click Validate → Runs validation from that directory

Commands Generated:

• databricks bundle init default-python --output-dir my-project --target dev --profile myprofile
• databricks bundle validate --target dev --profile myprofile
• databricks bundle deploy --target dev --profile myprofile

Troubleshooting

• Command not found: Install Databricks CLI
• Authentication errors: Configure CLI with databricks configure --token
• Bundle not found: Check bundle path points to correct directory
• Permission errors: Ensure proper workspace permissions