dataform-dependency-visualizer 0.2.2

Visualize Dataform table dependencies as interactive SVG diagrams

Dataform Dependency Visualizer

Generate beautiful, interactive SVG diagrams showing dependencies between Dataform tables.

Features

• 📊 Interactive SVG Diagrams - One diagram per table showing all dependencies
• 🎨 Color-Coded by Type - Tables, views, and operations visually distinct
• 🔍 Master Index - Browse all tables in a single HTML interface
• 📁 Schema Organization - Automatically organized by database schema
• ⚡ Pure Python - No Graphviz required, pure SVG generation
• 🎯 Smart Layout - Clean orthogonal routing for professional diagrams
• 📝 Text Wrapping - Long table names automatically wrapped
• 🧹 Cleanup Utilities - Built-in tools to fix common Dataform issues

Installation

pip install dataform-dependency-visualizer

Prerequisites

You need Dataform CLI installed:

# Install globally
npm install -g @dataform/cli

# Or in your project
npm install @dataform/core

Quick Start

1. Generate dependency report from your Dataform project:

cd your-dataform-project
dataform compile --json > dependencies_report.txt

2. Convert to text format for visualization:

poetry run python -m dataform_viz.dataform_check > dependencies_text_report.txt

3. Generate diagrams:

# For all schemas
dataform-deps --report dependencies_text_report.txt generate-all

# For specific schema
dataform-deps --report dependencies_text_report.txt generate dashboard

# Generate browsable index
dataform-deps --report dependencies_text_report.txt index

4. View results: Open output/dependencies_master_index.html in your browser

Command Reference

# Generate for specific schema
dataform-deps --report FILE generate SCHEMA_NAME

# Generate all schemas
dataform-deps --report FILE generate-all

# Generate master index
dataform-deps --report FILE index

# Open index in browser
dataform-deps --report FILE index --open

# Custom output directory
dataform-deps --report FILE --output my_output generate-all

# Cleanup Dataform issues
python -m dataform_viz.dataform_check --cleanup

Python API

from dataform_viz import DependencyVisualizer

# Initialize with report
viz = DependencyVisualizer('dependencies_text_report.txt')
viz.load_report()

# Generate for specific schema
count = viz.generate_schema_svgs('dashboard', output_dir='output')

# Generate for all schemas
total = viz.generate_all_svgs(output_dir='output')

# Generate master index
viz.generate_master_index('output')

Output Structure

output/
├── dependencies_master_index.html  # Interactive browser interface
├── analytics/
│   └── analytics_customer_summary.svg
├── dashboard_wwim/
│   ├── dashboard_wwim_table1.svg
│   └── ...
└── datamart_wwim/
    └── ...

Understanding the Diagrams

Each SVG shows:

• Center (Yellow): The table being viewed
• Left (Blue): Dependencies - tables this reads FROM
• Right (Green): Dependents - tables that read FROM this
• Schema Labels: Shows schema for each table
• Type Badges: table, view, incremental, operation

Cleanup Utilities

Fix common Dataform compilation issues:

python -m dataform_viz.dataform_check --cleanup

This removes:

• database: lines from config blocks
• database: from ref() calls
• Replaces constant references with actual values
• Cleans up dependencies.js files

Common Issues

"wwim_utils is not defined"

Run the cleanup utility:

python -m dataform_viz.dataform_check --cleanup

Empty SVG output

Use text format report, not JSON:

python -m dataform_viz.dataform_check > dependencies_text_report.txt

Requirements

• Python 3.10+
• Dataform CLI
• Node.js (to run Dataform)

Links

• GitHub: https://github.com/OshigeAkito/dataform-dependency-visualizer
• Documentation: GitHub README
• Issues: GitHub Issues

License

MIT License - See LICENSE file for details.

Changelog

v0.2.0 (2026-01-13)

• Added cleanup utility for database references
• Added constant replacement functionality
• Improved error handling
• Enhanced documentation

v0.1.0

• Initial release
• Basic SVG generation
• Master index generator
• Command-line interface
• Color coding - Tables (blue), views (green), operations (orange)

Master Index

The master index organizes all tables by schema with:

• Clickable table names that open their SVG
• Expandable/collapsible schemas
• Type badges (table/view/incremental)
• Search functionality

Requirements

• Python 3.8+
• Dataform project with compiled dependencies

How It Works

1. Parse dependencies_report.txt generated by Dataform
2. Extract table dependencies and metadata
3. Generate SVG diagrams with orthogonal routing
4. Create master index HTML for easy navigation

Project Structure

output/
├── dependencies_master_index.html    # Main entry point
└── dependencies/
    ├── schema1_table1.svg
    ├── schema1_table2.svg
    └── ...

License

MIT License - see LICENSE file for details.

Contributing

Contributions welcome! Please open an issue or PR on GitHub.

Links

• GitHub Repository
• Issue Tracker