Code Review API

A Python FastAPI service that provides endpoints for creating and retrieving AI-powered code reviews, with asynchronous analysis using Anthropic's API.

Features

• Create and retrieve code reviews
• Asynchronous code analysis workflow
• MongoDB storage with schema validation
• Anthropic API integration for intelligent code analysis
• Structured logging with configurable outputs
• Health check endpoint with database connectivity status
• Standard sets management for code review rules
• Classification system for technology-specific standards
• Git repository analysis and cloning
• Comprehensive test coverage

Prerequisites

• Python 3.12.8+
• MongoDB 6.0+ (running in Docker)
• Anthropic API key
• Git (for repository analysis)

Installation

0. Ensure you have the correct Python version:

# Install Python 3.12.8 if not already installed
pyenv install 3.12.8

# Set local Python version
pyenv local 3.12.8

1. Clone the repository:

git clone <repository-url>
cd code-review-api

2. Create a virtual environment and activate it:

python -m venv .venv
source .venv/bin/activate  # On Windows: .venv\Scripts\activate

3. Install dependencies:

pip install -r requirements.txt

4. Create a .env file in the project root:

See .env.example for required variables

Running MongoDB

Start MongoDB using Docker Compose:

# Start MongoDB
docker-compose up -d mongodb

# Check MongoDB status
docker-compose ps

# View MongoDB logs
docker-compose logs -f mongodb

# Stop MongoDB
docker-compose down

The MongoDB data will persist in a named volume code_review_mongodb_data.

MongoDB Schema Validation

The database uses schema validation to ensure data integrity:

• Classifications: Stores technology/language classifications
• Standard Sets: Contains predefined sets of standards with repository links
• Standards: Individual standards with text content and classifications
• Code Reviews: Review requests and results with compliance reports

Schema validation is automatically applied during database initialisation.

Running the API

Start the FastAPI server:

uvicorn src.main:app --reload

The API will be available at http://localhost:8000

API Endpoints

Code Reviews

# Create a Code Review
POST /api/v1/code-reviews
Content-Type: application/json

{
    "repository_url": "https://github.com/user/repo",
    "standard_sets": ["python", "security"]
}

# List All Code Reviews
GET /api/v1/code-reviews

# Get a Specific Code Review
GET /api/v1/code-reviews/{_id}

Standard Sets

# Create a Standard Set
POST /api/v1/standard-sets
Content-Type: application/json

{
    "name": "python",
    "repository_url": "https://github.com/org/standards",
    "custom_prompt": "Optional custom prompt for analysis"
}

# List All Standard Sets
GET /api/v1/standard-sets

Classifications

# Create a Classification
POST /api/v1/classifications
Content-Type: application/json

{
    "name": "python"
}

# List All Classifications
GET /api/v1/classifications

Health Check

GET /health

Documentation

API documentation is available at:

• Swagger UI: http://localhost:8000/docs
• ReDoc: http://localhost:8000/redoc

Development

Project Structure

src/
├── __init__.py
├── config.py           # Configuration management
├── database.py         # MongoDB connection
├── logging_config.py   # Logging setup
├── models/
│   ├── __init__.py
│   └── code_review.py  # CodeReview model
├── services/
│   ├── __init__.py
│   └── anthropic.py    # Anthropic API integration
├── api/
│   ├── __init__.py
│   └── v1/
│       ├── __init__.py
│       └── code_reviews.py  # API endpoints
└── main.py            # FastAPI application

Running Tests

pytest

Error Handling

The API uses standard HTTP status codes:

• 200: Successful GET request
• 201: Successful POST request
• 400: Invalid input
• 404: Resource not found
• 500: Internal server error

Logging

The application uses structured logging with the following format:

%(asctime)s - %(name)s - %(levelname)s - %(message)s

Logs are written to stdout and can be redirected as needed.

License

[License Type] - See LICENSE file for details