Replicheck

A Python tool for detecting code duplications and code quality issues in your projects. Replicheck analyzes your codebase to identify similar or identical code blocks, high complexity functions, large files/classes, and TODO/FIXME comments, helping you maintain code quality and reduce redundancy.

Features

• Detect high cyclomatic complexity functions (Python, JavaScript/JSX, Typescript/TSX, C#)
• Detect large files and large classes by token count (Python, JavaScript/JSX, Typescript/TSX, C#)
• Detect duplicate code blocks (Python, JavaScript/JSX, Typescript/TSX, C#)
• Detect Unused imports and vars (Python)
• Find TODO/FIXME comments across your codebase
• Support for Python, JavaScript, JSX, Typescript, TSX and C# files
• Configurable similarity threshold and minimum block size
• Ignores virtual environments and user-specified directories
• Detailed reporting in text, JSON, or Markdown format
• Summary section at the top of every report
• Hyperlinkable file paths in Markdown and supported terminals
• 90%+ test coverage and robust error handling

Installation

1. Clone this repository
2. Install dependencies:

pip install -r requirements.txt

Usage

Basic usage:

python main.py --path /path/to/code

Options

• --path: Directory to analyze (required)
• --min-similarity: Minimum similarity threshold (0.0 to 1.0, default: 0.8)
• --min-size: Minimum code block size in tokens (default: 50)
• --output-format: Output format (text, json, or markdown, default: text)
• --output-file: Path to save the report (optional)
• --ignore-dirs: Space-separated list of directories to ignore (default: .git .venv venv env ENV build dist)
• --complexity-threshold: Cyclomatic complexity threshold to flag functions (default: 10)
• --large-file-threshold: Token count threshold to flag large files (default: 500)
• --large-class-threshold: Token count threshold to flag large classes (default: 300)
• --top-n-large: Show only the top N largest files/classes (default: 10, 0=all)

Example:

python main.py --path ./src --min-similarity 0.85 --min-size 30 --output-format markdown --output-file report.md --ignore-dirs .git .venv node_modules --complexity-threshold 12 --large-file-threshold 800 --large-class-threshold 400 --top-n-large 5

Report Output

Summary Section

Every report starts with a summary, e.g.:

## Summary
- 14 high complexity functions (0 Critical 🔴)
- 9 large files (7 Critical 🔴)
- 1 large classes (0 High 🟠)
- 0 unused imports/variables ✅
- 12 TODO/FIXME comments
- 0 duplicate code blocks ✅

Markdown Output Example

# Code Duplication Report

## Summary
- 2 high complexity functions (1 Critical 🔴)
- 3 large files (1 Critical 🔴)
- 1 large classes (1 High 🟠)
- 2 unused imports/variables
- 4 TODO/FIXME comments
- 1 duplicate code blocks

## High Cyclomatic Complexity Functions
- [src/example.py:42](src/example.py#L42) process_data (complexity: 18) [Critical 🔴]
- [src/utils.py:10](src/utils.py#L10) calculate (complexity: 12) [Medium 🟡]

## Large Files
- [src/huge_module.py](src/huge_module.py) (tokens: 2100) [Critical 🔴]
- [src/medium_module.py](src/medium_module.py) (tokens: 800) [High 🟠]

## Large Classes
- [src/huge_module.py:5](src/huge_module.py#L5) BigProcessor (tokens: 700) [Critical 🔴]
- [src/medium_module.py:20](src/medium_module.py#L20) MediumProcessor (tokens: 350) [High 🟠]

## Unused Imports and Vars
- [main.py:204](main.py#L204) [F841] local variable 'js_files' is assigned to but never used
- [main.py:207](main.py#L207) [F841] local variable 'cs_files' is assigned to but never used


## TODO/FIXME Comments
- [src/example.py:99](src/example.py#L99) [TODO] Add error handling
- [src/utils.py:25](src/utils.py#L25) [FIXME] This is a hack, refactor later

## Code Duplications
- Clone #1: size=50 tokens, count=2 (cross-file)
    - [src/foo.py:10](src/foo.py#L10) - src/foo.py:10-20
    - [src/bar.py:30](src/bar.py#L30) - src/bar.py:30-40
    Tokens: def foo ( ) : x = 1 y = ...

• In Markdown reports, file paths are clickable and link to the relevant file/line (works in GitHub, VSCode, etc.).
• In supported terminals, file paths are clickable if your terminal supports OSC 8 hyperlinks.

Development

To run tests:

pytest --cov --cov-report=html

JavaScript/JSX Support

Replicheck now supports duplicate detection, large file/class detection, and cyclomatic complexity analysis for JavaScript and JSX files. Cyclomatic complexity for JS/JSX requires Node.js and the npm package typhonjs-escomplex (see Development section).

JavaScript/JSX Development

To enable JS/JSX metrics, install Node.js and run:

cd utils
npm install typhonjs-escomplex

This is required for cyclomatic complexity analysis of JS/JSX files.

C# Support

Replicheck now supports C# files for:

• Duplicate code block detection
• Large file and class detection
• Cyclomatic complexity analysis (based on tree-sitter and custom static heuristics)

No external dependencies are required for C# analysis. Just include .cs files in your scan directory and Replicheck will handle the rest.

Pre-commit Hooks (Recommended)

You can use pre-commit to automatically format and lint your code before each commit. This is free and easy to set up:

1. Install pre-commit:

   pip install pre-commit

2. Install the hooks:

   pre-commit install

3. Now, every time you commit, the hooks defined in .pre-commit-config.yaml will run automatically.

If you want to run the hooks manually on all files (RECOMMENDED Before Commit):

pre-commit run --all-files

Contributing

We welcome contributions! Please see CONTRIBUTING.md for guidelines before submitting issues or pull requests.

License

This project is licensed under the Apache License 2.0.