codebase-digest 0.1.40

Consolidates and analyzes codebases for insights.

Codebase Digest

Codebase Digest is a command-line tool written in Python that helps you analyze and understand your codebase. It provides a structured overview of your project's directory structure, file sizes, token counts, and even consolidates the content of all text-based files into a single output for easy analysis with Large Language Models (LLMs).

Table of Contents

• Features
• Installation
• Usage
• Configuration
• Ignore Functionality
• LLM Prompts for Enhanced Analysis
• Contributing
• License

Features

• 📊 Directory Tree Visualization: Generate a hierarchical view of your project structure
• 📈 Codebase Statistics: Calculate total files, directories, code size, and token counts
• 📄 File Content Consolidation: Combine all text-based files into a single output
• 🚫 Flexible Ignore System: Support for custom patterns, defaults, and .gitignore files
• 🎨 Multiple Output Formats: Choose between text, JSON, Markdown, XML, or HTML
• 🌈 Colored Console Output: Visually appealing and informative summaries
• 🧠 LLM Analysis Support: Comprehensive prompt library for in-depth codebase analysis

Installation

Via pip (Recommended)

pip install codebase-digest

From source

git clone https://github.com/kamilstanuch/codebase-digest.git
cd codebase-digest
pip install -r requirements.txt

Usage

Basic usage:

cdigest [path_to_directory] [options]

Examples:

1. Analyze a project with default settings:

   cdigest /path/to/my_project
   

2. Analyze with custom depth and output format:

   cdigest /path/to/my_project -d 3 -o markdown
   

3. Ignore specific files and folders:

   cdigest /path/to/my_project --ignore "*.log" "temp_folder" "config.ini"
   

4. Show file sizes and include git directory:

   cdigest /path/to/my_project --show-size --include-git
   

5. Analyze and copy output to clipboard:

   cdigest /path/to/my_project --copy-to-clipboard
   

Configuration

Option	Description
path_to_directory	Path to the directory you want to analyze
-d, --max-depth	Maximum depth for directory traversal
-o, --output-format	Output format (text, json, markdown, xml, or html). Default: text
-f, --file	Output file name
--show-size	Show file sizes in directory tree
--show-ignored	Show ignored files and directories in tree
--ignore	Patterns to ignore (e.g., '*.pyc' '.venv' 'node_modules')
--keep-defaults	Keep default ignore patterns when using --ignore
--no-content	Exclude file contents from the output
--include-git	Include .git directory in the analysis
--max-size	Maximum allowed text content size in KB (default: 10240 KB)
--copy-to-clipboard	Copy the output to clipboard

Ignore Functionality

Default Ignore Patterns

The following patterns are ignored by default:

DEFAULT_IGNORE_PATTERNS = [
'.pyc', '.pyo', '.pyd', 'pycache', # Python
'node_modules', 'bower_components', # JavaScript
'.git', '.svn', '.hg', '.gitignore', # Version control
'venv', '.venv', 'env', # Virtual environments
'.idea', '.vscode', # IDEs
'.log', '.bak', '.swp', '.tmp', # Temporary and log files
'.DS_Store', # macOS
'Thumbs.db', # Windows
'build', 'dist', # Build directories
'.egg-info', # Python egg info
'.so', '.dylib', '.dll' # Compiled libraries
]

Custom Ignore Patterns

You can specify additional patterns to ignore using the --ignore option. These patterns will be added to the default ignore patterns unless --no-default-ignores is used.

Patterns can use wildcards (* and ?) and can be:

• Filenames (e.g., 'file.txt')
• Directory names (e.g., 'node_modules')
• File extensions (e.g., '*.pyc')
• Paths (e.g., '/path/to/ignore')

Example:

cdigest /path/to/my_project --ignore ".txt" "temp" "/path/to/specific/file.py"

.cdigestignore File

You can create a .cdigestignore file in your project root to specify project-specific ignore patterns. Each line in this file will be treated as an ignore pattern.

Overriding Default Ignores

To use only your custom ignore patterns without the default ones, use the --no-default-ignores option:

cdigest /path/to/my_project --no-default-ignores --ignore "custom_pattern" "another_pattern"

LLM Prompts for Enhanced Analysis

Codebase Digest includes a comprehensive set of prompts in the prompt_library directory to help you analyze your codebase using Large Language Models. These prompts cover various aspects of code analysis and business alignment:

Use Cases

1. Codebase Mapping and Learning: Quickly understand the structure and functionality of a new or complex codebase.
2. Improving User Stories: Analyze existing code to refine or generate user stories.
3. Initial Security Analysis: Perform a preliminary security assessment.
4. Code Quality Enhancement: Identify areas for improvement in code quality, readability, and maintainability.
5. Documentation Generation: Automatically generate or improve codebase documentation.
6. Learning Tool: Use as a teaching aid to explain complex coding concepts or architectures.
7. Business Alignment: Analyze how the codebase supports business objectives.
8. Stakeholder Communication: Generate insights to facilitate discussions with non-technical stakeholders.

Prompt Categories

1. Code Quality & Understanding

   • Error and Inconsistency Analysis
   • Risk Assessment
   • Documentation Generation

2. Learning & Knowledge Extraction

   • User Story Reconstruction
   • Mini-Lesson Generation

3. Code Improvement & Transformation

   • Best Practice Analysis
   • Language Translation
   • Refactoring Suggestions

4. Testing & Security

   • Unit Test Generation
   • Security Vulnerability Analysis

5. Business & Stakeholder Analysis

   • Business Impact Analysis
   • Stakeholder Persona Generation
   • Business Model Canvas Analysis
   • Value Proposition Canvas Analysis
   • SWOT Analysis
   • Jobs to be Done (JTBD) Analysis
   • Lean Canvas Analysis
   • OKR Analysis
   • Customer Journey Map Analysis
   • Value Chain Analysis

For detailed instructions on using these prompts, refer to the individual files in the prompt_library directory.

Contributing

Contributions are welcome! Please feel free to submit a Pull Request.

License

This project is licensed under the MIT License - see the LICENSE file for details.