Dashboard Export Tool

Build a command-line tool that exports Superset dashboards with all their dependencies to a portable ZIP archive format.

Problem Description

Your tool should accept a dashboard ID and export the complete dashboard including all its dependencies (charts, datasets, and database connections) to a ZIP file. The export should use YAML format for metadata and preserve dependency relationships.

Requirements

Input

• Accept a dashboard ID as a command-line argument
• Accept an output file path for the ZIP archive

Export Process

1. Retrieve the dashboard metadata by ID
2. Identify and collect all dependencies:
   • Charts used in the dashboard
   • Datasets used by those charts
   • Database connections used by those datasets
3. Export each resource to YAML format
4. Package all YAML files into a ZIP archive
5. Ensure proper dependency ordering in the export

Output Format

• Create a ZIP file containing:
  • dashboard.yaml - Dashboard metadata
  • charts/*.yaml - One file per chart
  • datasets/*.yaml - One file per dataset
  • databases/*.yaml - One file per database connection
  • metadata.yaml - Export metadata including version and timestamp

Error Handling

• Handle missing or invalid dashboard IDs gracefully
• Validate that all dependencies can be retrieved
• Provide clear error messages for any failures

Implementation

@generates

API

def export_dashboard(dashboard_id: int, output_path: str) -> None:
    """
    Export a dashboard with all its dependencies to a ZIP archive.

    Args:
        dashboard_id: The ID of the dashboard to export
        output_path: Path where the ZIP file should be created

    Raises:
        ValueError: If dashboard_id is invalid or dashboard doesn't exist
        IOError: If output_path cannot be written to
    """
    pass

def get_dashboard_dependencies(dashboard_id: int) -> dict:
    """
    Retrieve all dependencies for a given dashboard.

    Args:
        dashboard_id: The ID of the dashboard

    Returns:
        A dictionary containing:
        - 'charts': List of chart IDs
        - 'datasets': List of dataset IDs
        - 'databases': List of database IDs

    Raises:
        ValueError: If dashboard_id is invalid or dashboard doesn't exist
    """
    pass

Test Cases

Basic Export Test { @test }

• Given a valid dashboard ID (e.g., 1), when export_dashboard is called, then a ZIP file is created at the specified output path @test

Dependency Resolution Test { @test }

• Given a dashboard with 2 charts, 2 datasets, and 1 database, when get_dashboard_dependencies is called, then it returns all dependency IDs correctly @test

Invalid Dashboard Test { @test }

• Given an invalid dashboard ID (e.g., 99999), when export_dashboard is called, then it raises a ValueError with a clear error message @test

ZIP Structure Test { @test }

• Given a successful export, when the ZIP file is opened, then it contains the correct directory structure with dashboard.yaml, charts/, datasets/, databases/, and metadata.yaml @test

Dependencies { .dependencies }

apache-superset { .dependency }

Provides the data models, export commands, and APIs for dashboard management.