ConfigGraphViz 📊

Visualize the structure of your configuration files! ConfigGraphViz parses common config formats (INI, YAML, TOML) and generates Graphviz DOT language output, allowing you to create diagrams of your configuration hierarchy.

This helps understand complex configurations by showing sections, nested keys, lists, and values.

Features

• Parses INI (.ini), YAML (.yaml, .yml), and TOML (.toml) files.
• Generates Graphviz DOT language output (.dot file content).
• Visualizes nested structures (dictionaries/sections, lists).
• Distinguishes sections, keys, lists, and values using different node shapes/colors (customizable via DOT attributes).
• Simple Python API.
• Requires Python 3.8+.
• Type Hinted.

Installation

Install from PyPI:

# Install required parsers too
pip install configgraphviz PyYAML "tomli; python_version < '3.11'"

Or install directly from GitHub:

# Make sure to install dependencies manually or via the repo's requirements
pip install git+https://github.com/manyan-chan/ConfigGraphViz.git 
pip install PyYAML "tomli; python_version < '3.11'" # Install deps separately

For local development:

git clone https://github.com/manyan-chan/ConfigGraphViz.git
cd ConfigGraphViz
pip install -e .[test] # Installs PyYAML, tomli, pytest etc.

Note: You need Graphviz installed separately to render the generated .dot files into images (e.g., PNG, SVG). See Graphviz Download Page.

Usage

1. Import:

   from configgraphviz import parse_config, build_dot_graph
   from pathlib import Path
   

2. Parse your configuration file:

   config_file = "path/to/your/config.yaml" # Or .ini, .toml
   
   try:
       parsed_data = parse_config(config_file)
       print("Successfully parsed config.")
   except (FileNotFoundError, ValueError, ImportError) as e:
       print(f"Error parsing config: {e}")
       exit()
   

3. Generate the DOT graph string:

   graph_name = Path(config_file).stem # Use filename as graph name
   dot_string = build_dot_graph(parsed_data, graph_name=graph_name)
   print("\nGenerated DOT output:")
   print(dot_string)
   

4. Save and Render (using Graphviz):

   output_dot_file = f"{graph_name}.dot"
   output_image_file = f"{graph_name}.png" # Or .svg, .pdf, etc.
   
   with open(output_dot_file, "w", encoding="utf-8") as f:
       f.write(dot_string)
   print(f"\nDOT graph saved to {output_dot_file}")
   
   # Optional: Use subprocess to call Graphviz's 'dot' command
   import subprocess
   import shutil
   
   if shutil.which("dot"): # Check if 'dot' command exists
       try:
           subprocess.run(["dot", "-Tpng", output_dot_file, "-o", output_image_file], check=True)
           print(f"Rendered graph image saved to {output_image_file}")
       except (subprocess.CalledProcessError, FileNotFoundError) as e:
           print(f"\nFailed to render graph using 'dot' command: {e}")
           print("Please ensure Graphviz is installed and in your system's PATH.")
   else:
       print("\nGraphviz 'dot' command not found.")
       print(f"You can render the DOT file manually: dot -Tpng {output_dot_file} -o {output_image_file}")
   

Example Output

Given a simple YAML file (example.yaml):

server:
  host: "192.168.1.100"
  port: 8080
  retry_options:
    attempts: 3
    delay: 5

database:
  type: postgresql
  connection: "..."
  pool_size: 10

users:
  - name: alice
    role: admin
  - name: bob
    role: user

Running the script will produce a .dot file. Rendering example.yaml.dot with dot -Tpng example.yaml.dot -o example.yaml.png produces the actual graph image using Graphviz.

The following Mermaid diagram illustrates the concept of the generated structure (Note: This is Mermaid syntax for demonstration in Markdown, not the DOT output itself. Actual layout by Graphviz may differ):

graph LR
    root("(root)") --> server("server\n(section)");
    root --> database("database\n(section)");
    root --> users("users\n(list)");

    server --> server_host("host: '192.168.1.100'");
    server --> server_port("port: 8080");
    server --> server_retry("retry_options\n(section)");
    server_retry --> server_retry_attempts("attempts: 3");
    server_retry --> server_retry_delay("delay: 5");

    database --> database_type("type: 'postgresql'");
    database --> database_connection("connection: '...'");
    database --> database_pool("pool_size: 10");

    users --> users_0("0\n(item)");
    users --> users_1("1\n(item)");
    users_0 --> users_0_name("name: 'alice'");
    users_0 --> users_0_role("role: 'admin'");
    users_1 --> users_1_name("name: 'bob'");
    users_1 --> users_1_role("role: 'user'");

Testing

Uses pytest. Install test dependencies (pip install -e .[test]) and run:

pytest

Contributing

Contributions welcome! Please submit issues or PRs on GitHub.

License

MIT License - see LICENSE file.